Experiencing Wages: Social and Cultural Aspects of Wage Forms in Europe since 1500 9780857456847

Citation preview

EXPERIENCING WAGES

International Studies in Social History General Editor: Marcel van der Linden, International Institute of Social History, Amsterdam Volume 1 Trade Unions, Immigration and Immigrants in Europe 1960–1993 Edited by Rinus Penninx and Judith Roosblad Volume 2 Class and Other Identities Edited by Lex Heerma van Voss and Marcel van der Linden Volume 3 Rebellious Families Edited by Jan Kok Volume 4 Experiencing Wages Edited by Peter Scholliers and Leonard Schwarz Volume 5 The Imaginary Revolution Michael Seidman Volume 6 Revolution and Counterrevolution Kevin Murphy Volume 7 Miners and the State in the Ottoman Empire Donald Quataert

EXPERIENCING WAGES

Social and Cultural Aspects of Wage Forms in Europe since 1500

EDITED BY PETER SCHOLLIERS AND LEONARD SCHWARZ

Berghahn Books New York • Oxford

SCHO-SCW-prelims-pi-viii.qxd:SCHO-SCW-prelims-pi-viii.qxd

7/10/08

3:48 PM

Page iv

First published in hardback in 2003 by Berghahn Books www.BerghahnBooks.com First paperback edition printed in 2004 Reprinted in 2006 © 2003, 2004, 2006 Peter Scholliers and Leonard Schwarz All rights reserved. No part of this publication may be reproduced in any form or by any means without the written permission of Berghahn Books. Library of Congress Cataloging-in-Publication Data Experiencing wages : social and cultural aspects of wage forms in Europe since 1500 / edited by Peter Scholliers and Leonard Schwarz. p. cm. -- (International studies in social history : v. 4) Includes bibliographical references and index. ISBN 978-1-57181-546-0 (hbk, alk. paper) -- ISBN 978-1-57181-547-7 (pbk, alk. paper) 1. Wage payment systems--Europe--History. 2. Wages--Europe--History. 3. Compensation management--Europe--History. I. Scholliers, Peter. II. Schwarz, L. D. III. Series. HD4927.E85E96 2003 331.2'94--dc21

2003052476

British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library. Printed in the United States on acid-free paper. ISBN 978-1-57181-546-0 (hardback) 978-1-57181-547-7 (paperback) Cover illustrations: Left section: © AMSAB, Institute for Social History (Ghent, Belgium). Poster of the Christian Workers’ Movement, circa 1900 (no reference number). Right section: © Hulton-Getty Archive (London), Pay day, 3rd. June 1871. The Brickyards of England – Paying the children at the inn.

CONTENTS

List of Figures and Tables

vii

Introduction 1. The wage in Europe since the sixteenth century Peter Scholliers and Leonard Schwarz

3

Custom, Wages and the Market 2. Institutional and cultural change in wage formation: port labour in Antwerp (sixteenth – eighteenth centuries) Harald Deceulaer 3. When labour hires capital: evidence from Lancashire, 1870–1914 Michael Huberman 4. Giving notice: the legitimate way of quitting and firing (Ghent, 1877–1896) Patricia Van den Eeckhout

27

53 81

Changing Pay Systems and Wage Forms 5. Wage forms, wage systems and wage conflicts in German crafts during the eighteenth and earlier nineteenth centuries Reinhold Reith

113

6. Wage forms, pay systems and labour control in nineteenth-century agriculture. Evidence from the Dutch province of Groningen Henny Gooren and Hans Heger

139

7. Cash, wages and the economy of makeshifts in England, 1650–1800 Craig Muldrew and Steven King

155

vi

Contents

Age, Gender and Wages 8. Gendered wage systems and industrialisation in Finland in the late nineteenth and early twentieth centuries Sakari Heikkinen

183

9. Engendering the experience of wages: the evolution of the piecework system at the Spanish Tobacco Monopoly, 1800–1930s Lina Gálvez-Muñoz

201

10. Age, gender and the wage in Britain, 1830–1930 Paul Johnson

229

11. At what cost was pre-eminence purchased? Child labour and the first industrial revolution Jane Humphries

251

Notes on Contributors

269

Index

273

LIST OF FIGURES AND TABLES Figures 2.1.

New members of the guild of the transport workers per year, 1700–1794 (10 – yearly averages)

40

2.2.

Number of workers paying kaarsgeld to the transport-workers guild, 1720–1789

41

2.3.

Receipts of the offerings during the guild’s mass on St. Andrew’s day (in guilders)

43

4.1.

Conflicts brought before the Ghent Conseil de Prud’hommes 1877–1896

85

4.2.

Outcome of cases ending in reconciliation, Conseil de Prud’hommes, Ghent, 1877–1896

91

4.3.

Length of the period of notice at stake, Conseil de Prud’hommes, Ghent, 1877–1896

95

6.1.

Gowler’s model of a wage payment system

141

8.1.

Average wages of female and male manufacturing workers by age, Finland, 1902–1910

190

8.2.

The correlation between piece rate frequency and the proportion 194 of women in the labour force, by manufacturing industries and occupations, Finland, 1902–1905

10.1.

Male earnings in 1833, Lancashire cotton and Wiltshire wool

239

10.2.

Male earnings by age, 1830s, 1880s and 1930s, English industries

241

10.3

Female earnings by age, 1830s and 1930s, English industries

242

10.4.

Female earnings as a proportion of male earnings, 1830s and 1930s, English industries

243

11.1.

Cumulative percentage of working children by age, all children, England, early-nineteenth century

255

11.2.

Cumulative percentage of working children by age, according to the father’s occupation, England, early-nineteenth century

257

viii

List of Tables

Tables 3.1. 3.2. 3.3. 3.4. 3.5. 4.1. 4.2. 6.1. 6. 6. 6. 7.1. 8.1. 8.2. 8.3. 8.4. 8.5. 8.6. 8.7. 9.1. 10.1. 11.1. 11.2. 11.3.

The cotton-textile industry in Bolton and Oldham, 1850–1914 Demographic, labour force and religious make-up of Bolton and Oldham, 1851–1911 Poor Law statistics, 1870–1907 Poor Law statistics, 1907 Rings versus mules: Bolton and Oldham Cases related to termination of employment, Conseil de Prud’hommes, Ghent, 1877–1896 Outcome of cases related to the custom of giving notice, Conseil de Prud’hommes, Ghent, 1877–1896 Ratio of arable land and pasture in Groningen, nineteenth century Appendix 1. Farmers’ families, labourers’ families and servants in the different Groningen regions, 1862 Appendix 2. Average numbers of nonresident farm labourers and servants per farm in the Groningen regions, 1862 Appendix 3. Working population in Groningen agriculture in 1909 and its increase since 1810 and 1860 Forms of circulating currency in the eighteenth-century economy, England Female–male segregation of occupations and manufacturing industries, Finland, 1902–1910 The ratio of female urban outdoor workers’ wages to male workers’ wages, Finland, 1851–1910 (percentages) The ratio of female manufacturing workers’ wages to male workers’ wages, Finland, 1902–1910 (percentages) Wages of manufacturing workers by sex, age and branch, Finland, 1902–1910 (average male wage in the branch = 100) The age and work experience of manufacturing workers, Finland, 1902–1910 Wage systems in manufacturing industries, Finland, 1902–1910 (percentages) Male and female wage systems in the tobacco, textile and printing industries, Finland, 1902–1905 (percentages) The evolution of piecework system, Compañia Arrendataria de Tabacos, Seville, 1917–1933 Estimates of percentage deviation of regional wages from level in Lancashire, 1833 and 1867–1870 Age at starting work by father’s occupation, England, early nineteenth century Trends in age at starting work, England, early nineteenth century Participation rates for children ten to fourteen, world data, 1950–2000

56 57 69 70 72 84 90 143 149 149 150 159 186 188 188 189 190 192 193 219 234 256 258 265

INTRODUCTION

CHAPTER 1

THE WAGE IN EUROPE SINCE THE SIXTEENTH CENTURY

Peter Scholliers and Leonard Schwarz

An archaeology of this volume Historical consideration of wages and prices emerged in Europe in the second half of the nineteenth century and by 1900 it was growing firmly, together with the general contemporary interest in the wage. The research considered issues as varied as wage and price movements, wage forms and methods of payment, working time, labour conflicts, as well as wage theories.1 However, interest in the movement of (real) wages exceeded interest in the more qualitative aspects of the wage. This has been explained by the nature of the sources (e.g., tariff lists, account books, insurance documents) and by the growing influence of statistics in those days.2 Thus, the main concern of the pioneers of wage research, for example C. von Tyszka in Germany, A.L. Bowley in Britain and E. Levasseur in France, was to collect hourly, daily or weekly wages and to estimate the cost of living, hoping to obtain insights into the overall evolution of the living standards of the working classes, particularly since the 1850s. During the 1930s, influential scholars such as J. Kuczynski and F. Simiand continued to study wages, prices and real wages, situating their research within a broad theoretical and increasingly sophisticated quantitative framework.3 At the same time the International Committee on Price History coordinated the compilation of very long-run wage and price series, with contributions from eminent scholars as W. Beveridge, M. Elsass, N. Posthumus and A. Pribram.4 As a consequence, by 1940 historical interest in wage forms, wage systems and shop-floor relations (or rapports salariaux) had declined. This contrasted with the attention given to the subject by (labour) economists who were zealously discussing pay methods, wage theories, the effect of high wages on output and the like.5

4

Wages in Europe since the sixteenth century

When, after the Second World War, the study of wages, prices and real wages was taken up again, the approach of many historians remained resolutely quantitative, enticing numerous historians to incorporate more and more data, tables and graphs, to consider diverse occupations, to study various regions and to apply refined calculation formulae. In each decade, at least one or two papers provoked a lively debate among economic and social historians, thus leading to a non-stop production of papers, revisions, critical notes, rejoinders and comments. The debate on the standard of living in Britain during the industrial revolution is without doubt the best example.6 Around 1980 the qualitative approach of the wage made a modest reappearance, due to ever-persistent criticism of the methodology and meaning of real wage studies,7 to the accession of the so-called new institutional economics 8 and particularly to a ‘cultural turn’ within the social historiography of the early modern period.9 This is not to say that qualitative wage studies were completely lacking in the 1960s and 1970s. For example, Hobsbawm considered the view of workers with regard to the ‘rules of the game’ of wages, giving custom a central position in wage formation.10 Nonetheless, such an approach remained rather exceptional and quantitative analysis predominated. To illustrate this, we may refer to a paper by Schwarz, who in 1985 advocated the study of the ‘social history of the wage’ but, deploring the lack of information in this regard, he then had to drop the matter entirely.11 It was within this context that an international colloquium was organised in Brussels in March 1988. The primary aim was to compare historical series of real wages in various countries since the late eighteenth century, but the intention was also to contribute to the methodology and the theory of the real wage. Implicitly or explicitly, most participants assumed that real wages did inform – at least to a considerable extent – our understanding of the living standards of the greater part of a nation’s population, that they reflected the overall economic performance of a country, and/or that their movements influenced workers’ social and political behaviour. It was there that we, the editors of the present volume, met for the first time. It appeared that we represented two distinct paths of historiography which, surprisingly, had only rarely converged. Scholliers dealt with the nineteenth and twentieth centuries and wished to construct and compare trustworthy long-run series of real wages; Schwarz focussed on the eighteenth and early nineteenth centuries and considered the role of the wage, incomes in kind, and perquisites. During the colloquium, both paths became quite apparent via a different language and lively discussion. Although the volume reflects both approaches very well, it appeared that such mixed views were particularly valuable because of the more balanced and complete study of the wage that they provided.12 Nonetheless, the concluding ‘Suggestions for further research’ pleaded for the integration of the ‘social history of wages’ into the study of the movement of the real wage. During the 1990s, the huge attention to movements of (real) wages diminished. Some attention remained, particularly among a group of

Peter Scholliers & Leonard Schwarz

5

Anglo-American economic historians with a special interest in econometric history.13 This work has not prevented some social scientists and even economic historians from continuing to use the ‘old’ and often out-ofdate series, placing these firmly within labour market theories (about wage formation and/or differentials).14 Some economic historians adhere to a global view, seeking to compare the real wages of various countries and continents.15 Many social and economic historians have taken on a more refined and even relativist view of the movements of (real) wages, focussing on wage differentials with regard to region, gender or occupation.16 Most likely, this shift was caused by the ongoing critique of real-wage series and the ensuing proposal of alternative measurements for the standard of living (e.g. nutrition, height, literacy, mortality), that attracted some earlier students of the real wage.17 The ‘social’ approach to the wage developed swiftly. As mentioned above, this was not new, but it gradually became more widespread. It was directly linked to an overall shift of interest within labour history that started among social historians of the early modern period and that was, in its turn, influenced by ‘new directions’ in history writing.18 In the second half of the 1980s, several books and papers were published where the movement of (real) wages was discussed as part of a broader social and cultural view of the wage. Thus, attention was paid to the precise nature of the work that was undertaken and the wage for undertaking it, the hiring of workers, the place of the wage in total income, small-scale wage conflicts (between one employee and the employer) alongside the study of larger-scale conflicts, the role of guilds and unions, the influence of custom on wage formation, the meaning of the wage, the development of wage earning, etc.19 In particular, non-monetary forms of the wage (goods, entitlements) were given the attention they rightly deserved and questions were asked about the (low) supply of cash up to the middle of the nineteenth century, the evolution of payments in kind or the extent of credit.20 Fortunately, this new approach spread to the historiography of the industrial period too and historians were soon analysing wage forms, pay systems, wage theories and shop-floor wage relations in the nineteenth and twentieth centuries.21 Would it be appropriate to bring together wage and labour historians with widely varying views, interests and methodologies and to seek to cross borders between the quantitative and the qualitative, between time periods, between countries or between structuralist and post-structuralist approaches, in order to discuss the ‘social history of the wage’? We believed it would, when we proposed a so-called C-session on ‘Wage forms, entitlements and industrialisation in Europe, eighteenth – nineteenth centuries’ to the Twelfth International Economic History Congress (August 1998). This was accepted, but alas, the session never met. Fortunately, in a rare clairvoyant moment we had foreseen to organise a post-conference on the same topic and this did meet in April 1999 in Le Domaine des Treilles, a marvellous site in the Provence, that offers all possible conditions for fruitful talks. Papers were presented and discussed at

6

Wages in Europe since the sixteenth century

length.22 The meeting was a success and the publication of the papers seemed to impose itself. Some authors withdrew, while others were asked to join, but all readily revised their text (eventually), and all chapters have been thoroughly rewritten especially for this volume. This book covers a varied range of places and periods, but it does not provide a geographically or chronologically well-balanced survey. Such a survey may be possible when statistics of (real) wages are compared, but to present a European survey of wage forms, wage systems, perquisites, entitlements, wage differentials and the like since the sixteenth century is simply unrealisable. So, the reader will find a very unbalanced book in terms of time and place, but one that reflects research innovations in a traditionally important field. With this book, we wish indeed to contribute to a new approach of the study of the wage.

Old and new questions about the wage The advantage of the turn toward the ‘social history of the wage’ is that wages are increasingly situated within the social history of work, with the ensuing fundamental questions being about the precise significance of wage earning and the wage. Such was the underlying issue of the Treilles conference. This was hardly new. In 1981 Woodward questioned the use of the Phelps Brown and Hopkins series, then widely quoted and considered a reasonable indicator of wages, prices and living standards in the very long run.23 His critique touched the very core of the problem about the significance of the wage. The author argued that it was far from clear who actually received the wage (contractors, subcontractors, gang leaders), how the wage was divided between the workers of the gang, what the ‘social status’ of the workers was (independent businessmen, subcontractors, wage earners), the costs that these craftsmen had to bear (raw materials, tools), whether they had income from other sources (farming, for example) that would influence their attitude toward wage labour and whether non-monetary wages (drink, lodging) were important. Ultimately, Woodward concluded, the long-run series inevitably link (frequently) selfemployed businessmen in pre-industrial times to modern wage earners, which makes them misleading measures of the standard of living of a wage earner.24 We would add that some of these comments are also valid for the modern period, when the actual income was – and is – composed of a monetary wage, combined with a non-monetary wage (e.g., coal for miners; meal vouchers) and with social benefits of various kinds (e.g., poor relief, pension). Similarly, the social status of a worker was (and is) not always clear (cottage workers, craftsmen). Not surprisingly, some historians have considered it an unconditional necessity to focus on such questions before tackling the movement of wages, whether monetary or real.25 In short, the turn toward the social and cultural history of the wage is a complex venture, but one that is very rewarding and animated. We hope that this volume contributes to the process, and does so without

Peter Scholliers & Leonard Schwarz

7

ignoring that wages are a production cost, which means that, alongside the social and cultural history of the wage, there is of course an economic history. In fact, the basic question of this book is surprisingly simple and naïve: how were people paid for their work? The problem of ‘how much’ is of course linked to this question, which will tend to appear in this book under the form of wage differentials rather than under the form of wage movements. Right away, the problem of defining both ‘work’ and ‘pay’ appears. With regard to ‘work’, it has been argued that paid labour was and is marginal when seen against the whole of labour (household, voluntary, slavery, compelled, self-employed, etc.) and that people often combined and alternated wage labour with other forms of work and income.26 Wage labour refers to (legally) free labour done by a person for another person or an institution, and since the Middle Ages this type of work seems to have expanded gradually. The process was certainly not linear and occurred with considerable regional and temporal differences. Nevertheless, wage relations were at the core of big social transformations (commodification, urbanisation, industrialisation, migration, etc.), especially, but not exclusively, since the middle of the eighteenth century.27 Moreover, contemporaries saw wages and wage movements as important issues.28 The money wage incited workers to protest, strike, petition or join an organisation, while it drove employers to apply new technology, to enlist female labour, to demand or to reject State intervention. Wage labour structured the world of an ever-growing number of people. Surely, no one would disregard its role.29 The history of wage labour is not central here. The main issue of this volume is that of payment for specific work. Despite the many historical wage studies, there remains an endless list of questions and themes that, in our opinion, urgently need (renewed) attention. For the sake of clarity this book shall group these themes into three general categories which have manifested themselves in the more recent literature and within which we may, albeit hesitantly, situate each chapter of this volume. We shall thus apply the following labels: ‘Custom, wages and the market’, ‘Changing pay systems and wage forms’, and ‘Age, gender and wages’. In so doing, we must stress that a rigid distinction cannot of course be upheld: one chapter may consider payment methods related to gender in a market economy full of rules and conventions. Having said this, it is quite obvious that not all questions and themes related to wage history will be dealt with here. Quite the contrary. For instance, issues related to the calculation of the wage (measurement of output, negotiation of tariffs), the timing of the wage (how, where and especially when was the pay given to the worker), or the perception of the wage (how did the community see a wage worker; how did workers and employers perceive various wage systems and forms; what was their view on unemployment or migrant workers) are lacking. Clearly, there is still plenty of work to be done.

8

Wages in Europe since the sixteenth century

Custom, wages and the market One theme that has increasingly come forward in wage studies is the relationship between custom, the market and the wage. This is not new. 30 Phelps Brown and Hopkins addressed the core of this problem in rather plain terms in their 1955 study on English building wages from 1264 to 1954. They noted, as the most salient feature, the extent of the rise of the wage, but they also found an ‘extraordinary absence of falls’ in daily wages, very long periods of unvarying daily wages and a ‘remarkable stability’ in the wage differential between rates of craftsmen and labourers. In their attempt to explain these phenomena, they questioned the applicability of the laws of supply and demand to the wage: they thought it unlikely that ‘supply and demand remained exactly balanced at the ruling price; rather it must have been that their movements were not wide enough to overcome the inertia of convention’, particularly inertia vis-àvis downward pressures.31 After the Great War, wage gaps tended to diminish because of the effects of inflation and education.32 In later work on the nineteenth and twentieth centuries, Phelps Brown and Browne presented a framework of factors affecting the course of the wage, including the wage earner’s desire for higher wages, custom (that inhibited change), phases of the business cycle, market environment, conditions on the labour market, productivity, the capital-output ratio and workers’ vigour (individually or via unions). ‘Custom inhibited change, but was broken by the trade cycle’, the authors wrote, thus allocating a different position to custom compared to early modern times.33 Slowly but surely, their conclusions about the absence of wage falls, the long periods of unvarying wages and the stability of wage differentials in the early modern period came to be revised.34 Nonetheless, the notions of ‘custom’, ‘tradition’ and ‘convention’ gradually gained attention in the study of the wage, whether directly influenced by Phelps Brown’s work or not. These notions appeared to be very complex, more so than was captured by Hobsbawm’s innovative paper in 1964. M. Sonenscher noted for eighteenth-century France that ‘custom’ implied a sense of wage differentials and work hierarchy, entitlements and drinking money, credit and wages in kind, celebration meals and other ‘customs oral in character’.35 That this sense had actual consequences, is shown by J. Rule for England around 1760, where the ‘sense of a legitimate right’ caused a conflict over wages involving five hundred weavers who struck to ‘support our ancient custom’.36 One may ask – and historians are asking – when and how ‘custom’ emerged in language and practice, when and how it was fought for, and when and how it eventually disappeared. These questions link up with the notion of ‘moral economy’ that is debated among eighteenth-century social (particularly English) historians. On the whole, we feel that the concept is more likely to retard than enhance the study of changing wage forms during this period. Originally used by Edward Thompson to describe the behaviour of bread rioters in England during the second half of the eighteenth century, the concept has

Peter Scholliers & Leonard Schwarz

9

been attacked and defended; some have expanded it into ever wider areas, with its meaning becoming ever vaguer in the process; others, chief among them Thompson himself, have taken care to control its original conceptual sharpness. ‘Moral economy’ as used by Thompson was not solely a matter of tradition or of some degree of opposition to the market economy. There needed to be a legal background that could be used to justify direct action; furthermore, the action needed to be accepted by broad circles of the population, not merely those undertaking the action. Thompson came to accept – or was at least ‘more than half persuaded’ – that there was a strong case for extending the concept from an explanation of bread riots to a wider description of a communitarian culture, such as existed in the West of England clothing towns during the eighteenth and earlier nineteenth centuries – ‘a dense texture of trade rituals and customary usages, endorsed by community sanction’. He was, however, reluctant to go further, and was distinctly unhappy with W. Reddy’s use of the term to describe ‘a set of values and moral standards that were violated by technical and commercial change’.37 By Thompson’s original standards, the wage was therefore not part of the moral economy. Unlike bread prices, it did not affect everybody. Its legal and communitarian backing was far weaker than – say – grain prices; any attempt by the authorities to fix wage rates by law was almost invariably intended as a ceiling to wages, however rapidly the ceiling might become a floor in practice. In fact, by its flexibility and its capacity to attract individuals and to divide communities, the money wage was potentially as deeply inimical to tradition as was inflation, its cousin from the same world of markets and money.38 Yet, as many studies in this volume and elsewhere have shown, there were definite elements of tradition in the wage. This was not only the nineteenth-century formula of ‘a fair day’s work for a fair day’s pay’ – Hobsbawm’s rules of the game – but particular attributes of the wage that attained more widely in eighteenth-century Europe, some of them gradually disappearing at different speeds in different regions. In this context we will refer to four themes that need to be kept in mind when considering the ‘traditional’ or ‘pre-modern’ wage. In the first place, as King and Muldrew have shown in their chapter in this volume, the specie necessary for a flexible labour market was usually not available. Secondly, there was the notorious tendency of some wages to stick at a particular monetary level for a century or more. For most of the eighteenth and earlier nineteenth centuries the pay of women in agriculture was set at 6d. a day over most of England, irrespective of price movements, but also irrespective of a surplus or shortage of female labour.39 Building wages are a better-known but less striking example. However, usually wages, even building wages, did move – at any rate for men. The building workers examined by Reith in this volume had to cope not only with (slowly) changing money wages but also with changing entitlements, such as meal times. Thirdly, there is the question of whether there was ever anything particularly ‘moral’ in the level of wages received by women. Despite the

10

Wages in Europe since the sixteenth century

research by Pam Sharpe on the stickiness of women’s agricultural wages during the eighteenth century, it is very tempting to look for the full application of the market economy, red in tooth and claw, to working women from the Middle Ages onwards – and it should be noted that the eighteenth-century wage stickiness included the inflationary second half of the century. Fourthly, and related to the third point, wage-earning men notoriously fought as hard as they could to defend themselves from the full rigours of a flexible labour market. The greater their skills – in other words the greater their market power – the greater their success. Concluding a detailed examination of three industrial disputes in eighteenth-century England – shipwrights in a small Devon port in 1766, Exeter wool sorters in 1787 and the weavers and wool combers in a wide area of Devon in 1726 – Rule concludes that ‘industrial disputes in the eighteenth century were complex phenomena, neither entirely within or without the reach of the moral economy’, and the same conclusion would apply to some of the eighteenth-century Central European industrial disputes discussed by Reith in this volume.40 It is worth recalling Randall and Charlesworth’s comment on Rule’s essay: ‘the moral economy market model was…not an alternative to a capitalist market but a model of a capitalist market subject to careful regulation’.41 Up to the present day, the rhetoric of community and morality has been more readily paraded in a defensive than in an aggressive industrial action.42 It is this persistent model of a regulated market economy, however little observed in practice, that needs to be remembered. It is this that puts the money wage at the borders of ‘modernity’ and at the centre of questions regarding the extent to which a labour market existed and of the attitude of the labour force to the money wage. In fact one of the intriguing aspects of the wage is its mixture of cultural and economic baggage. There was a labour market, labouring families had growing aspirations (or at least, during the price rise from the 1750s they wished to maintain their living standards), and the wage reflected this. There were also cultural norms and traditions that somehow had to coexist with the wage and with the market (the issue of how custom is shaped and changed by workers and employers is addressed by Van den Eeckhout in this volume). In the rare event of the norms being enshrined by a bureaucracy – as in Britain’s naval dockyards – a complex system of traditional payments and perquisites coexisted with an even more complex system of overtime pay, difficult for any outsider to understand and even more difficult for the Admiralty to break. The role of the state cannot of course be ignored. States operated on many levels, including areas of obvious relevance to this discussion such as the regulation of market conditions, the supply of specie, labour relations and legislation and social welfare schemes. This introduction will consider three of them. The first is the provision of specie. This was an area in which most states had only a partial success until the later eighteenth or even the nineteenth century. As Muldrew’s work has shown, the shortage was important in a market economy as developed as late seventeenth-century England, while coin clipping and forgery were common in

Peter Scholliers & Leonard Schwarz

11

England until the 1780s.43 For the purpose of paying wages it was the low-value copper coinage or what in France was called ‘billon’ – copper mixed with silver – that was more important. Copper, however, was heavy and governments tended not to produce enough of it. Between 1782 and 1789 the value of copper coinage produced by the French mints was less than half of one percent of the combined gold and silver coinage that they produced, but this weighed almost one-third of the weight of the gold and silver coinage produced. It is easy to see why the French government was happy to give Boulton a contract to use his steam engines to produce large amounts of copper coinage.44 Shortages of all kinds of coin were endemic in many parts of Ancien Régime Europe – the Netherlands being an exception – and shortages of copper coins were more endemic.45 As King and Muldrew show in this book, a shortage of coin compelled the payment of exchange entitlements (i.e. wages) in kind or credit. A second role for the State was as a large industrial employer (for instance with naval shipyards in some British or French ports, the Arsenal in Venice or – later – railways and coalmines. In this volume L. GalvezMuñoz describes the case of the Seville factory of the Compañía Arrendataria de Tabacos, owned (or at least controlled) by the government, which needed to take special care as so much of the welfare of Seville depended upon its employment and production processes. In such cases a complex network of entitlements grew up, whose existence appears in the records only when a hard-pressed State seeks to reduce them. The British Admiralty had permanent problems with its dockyards; in Toulon the dockyard workers were unable to prevent the Ancien Régime from large scale subcontracting in 1786, but they took their opportunity to riot in 1790 and the authorities more or less retreated.46 The rulers of Venice always watched the expenses of their Arsenal carefully, but until the Venetian state itself ended they took good care to ensure that the supply of free wine to the workers at the Arsenal was maintained.47 The third role of the State that will be briefly mentioned in this context was the relief of poverty, or – in twentieth-century terms – the operation of social welfare. Some states, such as Ancien Régime England or revolutionary France, sought to play a central role in the relief of hardship. The absence of such relief in Spain made it necessary for the government to take especial care with the Compañía Arrendataria de Tabacos. The influence of poor relief and charity on wages was all-pervasive and where State agencies played a central role in the relief of hardship as in England or Holland – albeit in such a decentralised manner that to see this as part of the role of the State is problematic – the outcome is a mass of documentation of the effects on wages. The contribution of the Poor Law to wages in southern England between 1795 and 1834 is well known and generally put under the generic (if rather misleading term) of Speenhamland. However, King and Muldrew show here the importance of poor relief from the 1790s, while in his recent book King discusses the Poor Law’s contribution to wages over the entire century and a half between 1700 and 1850 – a contribution that extended far beyond the Speenhamland

12

Wages in Europe since the sixteenth century

counties, and also drew in the ‘deserving poor’ who might be employed explicitly in order to keep them off poor relief.48 Such employment applied especially to women – which goes some way to explain the stickiness of agricultural wages for women noted above. But it was also liable to pervade the entire local economy. It is a large theme, implicit in virtually all the chapters of this book, particularly in the discussion of the Spanish Tobacco Monopoly by Gàlvez-Muñoz, where the absence of alternative schemes of welfare was an important constraint on the ability of the State’s freedom of manoeuvre with the labour force. King and Muldrew specifically refer to the problem – ‘there have been few studies [in England] of alternative earning avenues or their relationship to the payment of poor relief and wages, but it is nonetheless important to investigate these variables as a precursor to dealing with the communal welfare system’. The question obviously requires further study. In this volume three chapters focus on the tension between the market for wage labour and the conditions of work expected by the workers. Harald Deceulaer links the fluctuating transport volumes of the Antwerp harbour in the long seventeenth century to changing wage formations, noting that rules, norms and values were transformed by all the actors and institutions involved. Michael Huberman detects distinct work relationships in two cotton communities in Lancashire in the nineteenth century, one of bonds between worker-firm (paternalism) in Bolton and worker-worker (co-operation) in Oldham; these traditions led to different practices and different economic performances. Using the court cases of the Conseil de Prud’hommes between 1875 and 1900, Patricia Van den Eeckhout found a shift from an accustomed ‘fortnight’ to an ‘eight days’ period for giving notice in Ghent. This suggests a trend toward flexibility within work relations, while it also emphasises the flexibility that was attained within a rather rigidly regulated legal framework. She also shows how ‘custom’ operated in the daily practice of labour relations. Huberman and especially Deceulaer show that the ‘market’ had an influence on wage formation and work practices when one industrial activity is studied (textiles; harbour work), while Van den Eeckhout does not (and cannot) refer to ‘one’ market or ‘the’ business cycle because of the great variety of occupations with which she is confronted in Ghent.49 Crucially, these authors do not accept ‘usage’ or ‘tradition’ as given, but as constantly negotiated, contested and, paradoxically, adaptable. They also show that it is impossible to draw clear borderlines between the eighteenth and the nineteenth century with regard to the influence of ‘custom’. As suggested by Hobsbawm, it was perhaps the spread of scientific management around 1900 (incentive pay schemes, time and motion studies, etc.), together with its inevitable and understudied concomitant, the generalisation of the money wage, that replaced the sense of custom in wage claims by a sense of progression.50 Huberman’s argument will doubtless prove controversial, but those who object will need to set their own arguments within a wide social and economic context and to this extent it is to be hoped that it will set the terms for future discussion of the issues that he raises.

Peter Scholliers & Leonard Schwarz

13

Changing pay systems and wage forms Hobsbawm’s suggestion leads us to the second set of problems: the pay system. In his 1964 paper, he deplored the absence of research on this issue.51 In their 1993 study of the development of pay systems in the agriculture of Groningen during the nineteenth century H. Gooren and H. Heger were still regretting the absence of such research.52 Quite recently, R. Reith has repeated this criticism.53 Why such silence? Would it not be important to study wage systems? Is it possible that for many centuries the distinction between a piece rate and a time wage was not particularly important, because ‘workmen produced an accepted output for an accepted remuneration from an accepted working week’?54 In this volume Huberman compares the different remuneration methods for two towns within the Lancashire cotton industry and draws out their significance. Distinct work and labour relationships in different Lancashire towns implied distinct systems of payment. In Bolton’s fine spinning mills, a tariff list appeared in 1813 and stipulated prices per pound of yarn; this list did not include a speed clause. Earnings accordingly varied with the age of the machine: the more recent the machine, the higher the output and the wage. In Oldham’s coarse spinning mills, the tariff list dated from 1872 and provided for the piece rate by dividing the normal output on a given mule by the standard weekly wage (our italics); later, a speed clause was introduced. The Oldham list aimed at equal earnings. These were truly opposite procedures for calculating both wages and earnings, and both methods implied different organisations of labour relations and of work, that in turn characterised their communities’ social and political life. It would of course be erroneous to suggest that pay systems have never been examined, but on the whole their developments have been studied in very general terms.55 R. Duplessis, for example, constructs the following picture for early modern Europe. Piece wages – i.e, the payment to an individual or a work gang for a specific task or quantity of output – were generally preferred by employers because these prompted the workers’ efforts, and because of the limited need for supervision. But workers also saw an advantage in piece rates, because they could, within limits, set their own pace and regulate their earnings. Time wages – i.e., the payment to an individual or a work gang for a task during a specific period of time – were paid to workers who worked together in a multistage process (the classic example being the building industry), to workers who performed standardised jobs (such as tailoring) and to workers in new industries (such as ironmaking).56 Generally, piece wages prevailed over time wages and with the development of industrialisation, and particularly sweated outworking, piece wage methods were increasingly applied.57 In many sectors, there were surpluses to the wage when the labour market was tight, the work was urgent or particularly difficult.58 Hobsbawm wrote that pay systems became more complex after 1850. He advanced three reasons for this: the growing interest of economists in methods of payment, the extension of payment by results and incentive

14

Wages in Europe since the sixteenth century

schemes, and finally the increase in money wages. Around 1900, he continued, there was the spread of scientific management, aiming at increasing labour efficiency and profits and accompanied by increasingly complicated pay methods (bonus systems named after their inventors – Rowan, Halsey or Bedaux).59 It is of course possible that the apparent growing complexity of wage systems from the nineteenth century is merely a reflection of the greater availability of source material and that pay systems were indeed equally complex prior to 1850. Nevertheless, we can document the growing complexity of pay methods in the nineteenth century by the example of the Ghent machine building industry. Up to 1850, daily wages prevailed, while only a few categories of workers had weekly wages. These simple time wages faded during the 1850s. Then, the travail à l’entreprise appeared, which was a combination of a (theoretical) hourly wage and a time period for the manufacturing of a particular machine part. Time and wage were negotiated on the basis of experience and tradition. Also, the possibility for adept workers to earn a moderate premium was included. With the introduction of machine tools in the 1880s productivity rose and hence the possibility of increasing the premium. However, a new pay system was introduced in the 1890s that foresaw a more sophisticated premium system as well as a penalty system when the output norm was not reached. This norm was no longer the result of an agreement between workers, overseers and managers, but of a study by the firm’s technical bureau. Furthermore, cuts of the (theoretical) hourly wage were applied – the so-called rate busting. Finally, around 1905 an extremely complex premium scheme was introduced (to the despair of unions), that adopted elements of scientific management and that implied the utter technicality of wage calculation.60 Other modern industries offer similar examples.61 Whether a worker was paid by time or piece and received a bonus, he or she could obtain money or another form of payment. The form of the wage has retained the historian’s attention to a greater extent than was the case with pay methods. For example, G. Schmoller stressed the importance of the non-monetary wage up to 1800, noting that its rigid nature in terms of quantity of goods automatically led to a stable real wage. He thus connected wage forms to custom and to habit.62 Since the pioneers’ labours historians have paid attention to wage forms but, as noted above, the theme came fully to the fore in the 1980s. For example, when considering the alleged shift from payment in kind to payment in cash in eighteenth-century France, M. Sonenscher stressed that there was no linear trend, that both systems coexisted and were dependent upon changes in prices, production schedules and labour-market conditions.63 If such a shift between monetary and non-monetary wage forms had indeed occurred, many questions need to be solved: when, where, how and why did this happen and what were the consequences for the level of money wages? There seems to have been a direct link between cash payment, payment by results, and growing industrialisation in the nineteenth century, leading to the gradual decline of payment in kind, perhaps

Peter Scholliers & Leonard Schwarz

15

indeed forcing its decline. Yet, in some so-called advanced economies, like the Northern Netherlands in the sixteenth century, the wage appeared only under a monetary form.64 At the other extreme, the extent to which wages in kind had disappeared in nineteenth-century England, the heartland of industrialisation, remains debatable.65 If payment in food, beer or lodging would appear to have diminished from the later eighteenth century, other forms emerged, spread or endured in the nineteenth and twentieth centuries such as coal, ‘chips’, subsidised meals and – later – meal vouchers, profit sharing, the use of a car, etc. We know hardly anything about the development of wages in kind during the nineteenth and twentieth centuries. The list of elements contained within non-monetary wages may also contain credit. If the latter may be defined as ‘postponed’ wage, it is clear that part of our present-day indirect wage (pension, sick care and paid holidays) may also be classified under the heading of credit. Of course, linking directly both forms of wages to each other is entirely a-historical. Causes and effects differed totally: social security schemes of the post-1945 era contrast widely with the credit allowed to the bon ouvrier by his seventeenth-century employer. For early modern Europe, Schwarz noted that ‘a large and immediate entitlement for a man with a job was credit. This might be from his employer… or it might be from a retailer or a publican’.66 ‘Entitlements’, or perquisites, lead us back to the world of rights, custom, habits, traditions and conventions. In this book, three chapters focus on the problem of pay systems and wage forms, and all three emphasise the many bonds with other features of society. Thus, Reinhold Reith considers the study of wage forms and pay methods as the point of departure for understanding the history of wage labour and wage conflicts in general. He deals with eighteenth- and nineteenth-century Germany, linking wages in kind to the labour market. With workers frequently living (and more frequently dining) in the master’s household, he is able to show that many labour conflicts took place around the issue of in-kind wages, disputes about the quality of meals being a typical example. He found that the piece wage spread gradually, although it did not simply replace the time wage but was added as a temporary compensation. While Reith suggests that piece rates were not so much applied from disciplinary aims, but were used particularly for managing periods of intensified demand, Henny Gooren and Hans Heger find that for Groningen agriculture in the nineteenth century the spread of the incentive wage was directly linked to increasing control over the work. This was needed, they argue, because of changes in labour organisation and work frequency that entailed the hiring of more casual workers. Because payment in kind was given exclusively to regular and resident workers, in-kind wages disappeared gradually. Both chapters illustrate the huge complexity and diversity of payment systems in the eighteenth and nineteenth centuries, as well as the impossibility of marking a distinct moment of transition. In their chapter on England between 1650 and 1800, Craig Muldrew and Steven King emphasise the lasting shortage of cash (up to the early nineteenth century), the interrelation between a

16

Wages in Europe since the sixteenth century

formal labour market and a complex world of makeshifts, a remuneration system full of customary entitlements, and the existence of credit networks. Lack of cash would lead to irregular payments, truck systems and all kinds of means to make ends meet, including charity. The poor law made this even more complex. They argue that the meaning and the level of the wage cannot be understood without incorporating all these elements. To us, the three chapters prove, once again, the central, irrefutable position of wage forms and payment methods in the general history of labour, involving the labour market, cash availability, work conflicts, charity, family decisions, credit and managerial tactics.

Age, gender and wages So far, issues of gender have hardly been dealt with. Yet wage forms, wage systems and wages have a clear gendered dimension. Gooren and Heger show that in Groningen in-kind wages for resident workers consisted of wool for the maids and the right to graze a sheep for the men. In nineteenth-century cotton mills, male overseers and maintenance workers received a weekly wage, while female reelers, spoolers or warpers were paid by the piece, with consequences for the working time, the status of men and women, as well as the work effort. The biggest difference between male and female wages, however, was that women were generally paid less than men. Such inequality appears to have been ‘traditional’ and can be encountered from the fourteenth century.67 Wages related to gender appeared in historical research quite recently via two routes.68 First, there was the big question ‘Are women underpaid?’ that was asked by (feminist) social researchers in the 1970s. Historians picked up the problem and started to investigate the money wages of male and female workers especially in the nineteenth century. They concluded, among other things, that there was no systematic discrimination against women, at least when occupational discrimination is disregarded.69 Recently, J. Burnette provided a significant addition to this research. She suggested that piece-rates and time-rates have often been misinterpreted, which has led to an underestimation of women’s wage rates. Correcting for differences in hours worked, payments in kind (particularly board) and misreporting of home work would lower the male-female wage ratio. The remaining wage differential may be explained by productivity differences. This is not to say that wage discrimination against women did not exist, nor to deny that custom played a role, but to limit the extent of both of these.70 Nevertheless, other researchers have reached conclusions that include a clear discrimination against women (doing equal jobs and having had equal schooling). For example, in the Dutch diamond industry during the 1890s, women replaced men at fine stone cutting, leading to an immediate decline of the wage rate.71 It may be that Burnette’s argument places too much faith in the working of the labour market; it will doubtless give rise to a lively literature.

Peter Scholliers & Leonard Schwarz

17

The second way of introducing gender into wage studies, originated out of a reaction against the assumption that adult male wages were representative of all wage earners and that male wages thus inform about the general standard of living, the production costs or the economic performance of a country. Not surprisingly, this criticism has been heard loudly in the course of the debate on the standard of living during the industrial revolution in England. In 1992 S. Horrell and J. Humphries, referring to the paradox between a relatively optimistic consensus based on real wage trends for male workers and pessimistic indicators based on data for consumption and height, insisted that family income should be considered in its entirety. Data from household budgets allowed these authors to revise trends in the standard of living, stressing that between 1787 and 1865 family incomes rose less than did male earnings while inequality between families grew more than might have been expected on the basis of male wages.72 Similar findings had appeared with regard to Belgium in the nineteenth and early twentieth centuries,73 thus underlining the general need for a cautious approach towards regarding the male wage as the socio-economic proxy. Writing about the English industrial revolution, J. Mokyr sought to reconcile the contradiction between stagnant average per capita consumption for many important items and rising male wage rates. He suggested that the self-employed, who formed ‘perhaps’ as much as forty percent of the English labour force in 1845, were lagging behind wage earners; in addition the census – for what it is worth – suggested that women formed about thirty percent of the labour force in 1851.74 The significant point remains that even in England by the mid-nineteenth century reasonably regular payment by means of a money wage cannot be said to have been relevant to at least one-third of the adult male labour force, while a significant number of the remaining two-thirds were working in agriculture, many of them in low-wage agriculture, where there may have been regular supplements to money incomes. Jan de Vries’ concept of an ‘industrious revolution’ aimed to tackle directly the disparity between real wage rates as reflected by the indices – usually declining from the mid-eighteenth century – and the growth of consumer demand during this period, a reconciliation that he suggested was brought about by increased effort and self-exploitation from households.75 Despite the many investigations into women’s work and wages over the last decade, among which questions about the paid labour of married women and the ideology of the male breadwinner have figured most prominently,76 many problems remain to be solved. In this volume, Sakari Heikkinen tackles the wage in Finnish agriculture and industry during the decades around 1900. Important wage gaps between male and female workers, as well as distinctive wage systems between both, point very clearly at a gendered wage. The wage gap may be explained by women’s shorter work experience, a presumed lower productivity and indeed discrimination, while the difference with regard to payment methods in industry (women were paid more by the piece than men) would be the ‘solution’ for a contradiction that was formed between custom and the

18

Wages in Europe since the sixteenth century

needs of the market. Lina Galvez-Muñoz deals with a similar problem, but focussing on one industry, the Spanish tobacco factories from the early nineteenth century to the 1930s. She asks whether employers’ choices with regard to a particular remuneration method did indeed have a gendered dimension. She shows that neither the (labour) market nor customary forces can explain the choice of pay systems, but that gender, ‘as a principal explanatory tool’, helps to interpret the forms of remuneration. Attention to family budgets includes the consideration of paid labour by children. Needless to say that child labour has been hugely debated since about the 1840s, leading to fierce indignation by both bourgeois and leftist observers and to legal restrictions in many countries. It is only recently that historians have sought to situate child labour within the context of the family economy and income. Indignation has often remained, but the starting point has differed. Thus, some historians have studied the share of the mother, the father, the children and other household members in total family income often linking it to the standard of living over the family life cycle.77 In this line of research child labour began to be viewed as an income source that influenced, for example, family structures, debt or consumption.78 With this interest in the family income over the family life cycle, the study of the wage in relation to age emerged. Already around 1900 some social researchers had studied this79 and recently household budgets have been used again to study the work and wages of older men in London around 1930.80 In this volume Paul Johnson tackles the question whether custom or market determined wage differentials; he does so not by looking at pay differences by skill or by industry, but by examining pay differences by age. He compares earnings surveys for three periods (1833, the late 1880s and 1929–1931), notes a ‘long-run stability in the relationship between age and earnings’, and concludes that a traditional relationship between age and earnings was so strongly imprinted on the social and cultural environment that a century of economic change and technical innovation failed to undermine it. The stability clearly requires explanation; and Johnson is clearly seeking to escape from the catch-all term ‘custom’, a term that historians are tempted to use when all else fails. The implications of this argument are considerable and the concern for free riding and collective action is apparent elsewhere in this volume. Finally, for the period of the industrial revolution in Britain, Jane Humphries stresses the particularities of child labour that was clearly unfree and hidden. She studies the age of children at the start of work, connecting this to estimated earnings, apprenticeship and family circumstances, such as the death of an older brother. Overall, she finds an ‘exceptionally intensive use’ of child labour, that she connects to the ‘industrious’ revolution, thus putting the child worker (back) at the centre of the industrialisation process. This chapter might at first appear to be more concerned with the nature of the industrial revolution than with wages as such; our argument, of course, is that such a distinction is tenuous.

Peter Scholliers & Leonard Schwarz

19

A final word Our classification of the papers into three groups inevitably impairs the richness of the themes and chapters. Indeed, all the chapters cover more than merely the issues of custom, gender, age or managerial tactics with regard to pay methods. In fact, most of them address all these issues. Moreover, most chapters touch upon topics that could hardly be mentioned in our introduction: individual conflicts over firing and the payment of wages due (Van den Eeckhout), collective wage conflicts (Reith), work organisation and control (Deceulaer, Gooren and Heger, Gàlvez-Muñoz), workshop regulations (Van den Eeckhout, GàlvezMuñoz), or the wage bill and industrialisation (Humphries). Furthermore, some chapters situate their theme firmly within a broad historical view, incorporating the long term or a comparative perspective (albeit within ‘national’ borders) and referring to technological change, credit, collective actions, family economy, child labour, the industrial revolution, disputes brought to court, guilds, or wage differentials of varying natures. In addition, theories are referred to: wage theories, of course, but also theories about the labour market, a model of wage payment systems, neoinstitutional economic theory, and theories about the development of paid labour. Yet, empiricism dominates the great majority of the chapters, which has tempted one author to oppose (neoclassical) economic theory to sound historical research. To us, all this confirms the centrality of the wage with regard to the history of labour. Also, it confirms the need to incorporate economic, social, cultural and ideological matters in an open way. We firmly believe that the history of the wage awaits a bright and shining future. Finally, some words of thanks. First, we wish to thank all the participants at the Treilles colloquium of April 1999: those who presented a paper, as well as those who commented, discussed, argued and disagreed. The papers most certainly gained a lot in the process. Second, we thank the anonymous referee of the International Institute for Social History for useful and relevant criticism, as well as the two referees of Berghahn Books for pointed comments and many suggestions. Also, we wish to thank Hugo Soly, director of the project ‘Labour 1500-2000’ (financed by the Flemish Fund for Scientific Research) for financial support. Last but not least, we particularly wish to thank the members of the Board of La Fondation des Treilles for the opportunity to meet in Les Treilles. Mmes. C. Bachy and S. Melkoyan were extremely helpful in organising the meeting, while the Domaine des Treilles proved to be the most exquisite place for fruitful discussions. It is not only a place to work hard and to concentrate, but also where one can enjoy working hard. Brussels and Birmingham, August 2002

20

Wages in Europe since the sixteenth century

Notes 1. An excellent example is G. Schmoller, ‘Le mouvement historique des salaires de 1300 à 1900 et ses causes’, Revue Internationale de Sociologie (1904), pp. 3–21; after having presented a broad survey of the wage and price development in Europe, the author (professor of Political Economy at Berlin University) paid attention to the form of the wage, the organisation of work, the role of institutions and the role of custom , all of which had changed tremendously. See the chapter by R. Reith in this volume, dealing with the German Historische Schule. 2. Reinhold Reith, Lohn und Leistung. Lohnformen im Gewerbe 1450 – 1900 (Stuttgart, 1999), pp.24–32. 3. Kuczynski and Simiand had broad historical interests, as indeed did Keynes. For another example, see Elizabeth Gilboy, ‘The cost of living and real wages in eighteenth-century England’, Review of Economic Statistics (1936), pp.134–43. 4. On the history and achievements of this Committee, see O. Dumoulin, ‘Aux origines de l’histoire des prix’, Annales Economies, Sociétés, Civilisation (1990), pp. 507–22. 5. See, e.g., the classic by Maurice Dobb, Wages (Cambridge, 1929 – 14 reprints up to 1960). 6. As landmark papers of the 1950s and 1960s, we may refer to the survey by Eric Hobsbawm and R. M. Hartwell, ‘The standard of living during the industrial revolution: a discussion’, Economic History Review (1963), pp. 120–46; for the 1970s Michael Flinn, ‘Trends in real wages, 1750–1850’, Economic History Review (1974), pp. 395–411, and for the 1980s Peter Lindert and Jeffrey Williamson, ‘English workers’ living standards during the industrial revolution: a new look’, Economic History Review (1983), pp. 1–25. A mention of the Dutch debate (‘Would high wages between 1800 and 1850 be responsible for slow industrial growth?’ to which, e.g., Joel Mokyr, Jan de Vries and Richard Griffiths contributed) may suffice to show that (real) wages were also discussed widely outside Britain (see, e.g., Jan Luiten Van Zanden, ‘Lonen en arbeidsmarkt in Amsterdam, 1800–1865’, Tijdschrift voor Sociale Geschiedenis (1983), pp. 3–27). 7. For a survey of the critique of real wages, see Joel Mokyr, ‘Editor’s introduction’, in Joel Mokyr (ed.), The British industrial revolution. An economic perspective (Boulder – Oxford, 1993), pp. 118–30. An inventory of fausses pistes with regard to early modern (real) wage research may be found in Denis Morsa, ‘Salaire et salariat dans les économies préindustrielles (XVIe-XVIIIe siècles). Quelques considérations critiques’, Revue Belge de Philologie et d’Histoire (1987), pp. 751–84. 8 The NIE devoted theoretical attention to organisations, firms, contracts, customs, rules, etc in relation to economic development within neoclassical economics; see P. G. Klein, ‘New institutional economics’, in B. Bouckaert and G. De Geest (eds), Encyclopaedia of law and economics (Aldershot, 2000), pp. 456–89. 9. By a ‘cultural turn’ we refer to new directions or new approaches that emerged in the 1980s and became ‘undeniable‘ in the 1990s. These do not form a coherent paradigm, but are related to a wide and perhaps incoherent variety of new approaches to economic and (especially) social history writing. For an impression of the directions and discussions, see, e.g., ‘Wirtschafts – und Sozialgeschichte – Neue Wege?’ Vierteljahrschrift für Sozial – und Wirtschaftsgeschichte (1995), vols. 3 and 4; Social History (1995), vols. 1 and 3; Storia della Storiografia (1995); Historiens et géographes (1995), no.350; Tijdschrift voor Sociale Geschiedenis (1997), vol. 2. 10. Eric Hobsbawm, ‘Custom, wages and work-load in nineteenth-century England’, in A. Briggs and J. Saville (eds), Essays in Labour History in Honour of G. D. H. Cole (London, 1960), and subsequently reprinted in Eric J. Hobsbawm, Labouring Men (London, 1972), pp. 334–70. Hobsbawm suggested that employers and employees had to learn the rules of modern industrial relations, which occurred partially (up to the 1850s) and then completely (towards the 1880s). 11. Leonard Schwarz, ‘The standard of living in the long run: London, 1700-1860’, Economic History Review (1985), pp. 24–41. 12. Peter Scholliers (ed.), Real wages in nineteenth and twentieth century Europe. Historical and comparative perspectives (New York – Oxford – Munich, 1989).

Peter Scholliers & Leonard Schwarz

21

13. At the time of writing, its most recent outcome has been the appearance of Feinstein’s real wage index for Britain between 1770 and 1882, where the author is well aware of the technical criticisms that have been made of previous indices and which is probably more robust than its predecessors: Charles H. Feinstein, ‘Pessimism perpetuated: real wages and the standard of living in Britain during and after the industrial revolution’, Journal of Economic History, 58 (1998), pp. 625–58. This is unapologetically quantitative. 14. E.g., Olivier Marchand and Claude Thélot, Le travail en France (1800-2000), s.l., 1997, pp. 155–172, use the work by Jean Lhomme [‘Le pouvoir d’achat de l’ouvrier français au cours d’un siècle 1840-1940’, Le Mouvement Social 1968), pp. 41–70], who relied heavily on J. Kuczynski’s contested estimates. 15. In the 1990s, the most innovative work was indeed no longer about Europe, but about a world-scale comparison; see, e.g., Jeffrey Williamson, ‘Real wages, inequality and globalization in Latin America before 1940’, Revista de Historia Econòmica (1999), pp. 101–42. For a pre-industrial European comparative perspective, see Jan Luiten Van Zanden, ‘Wages and the standard of living in Europe, 1500 – 1800’, European Review of Economic History (1999), pp. 175–97. 16. E.g., J.-M. Chanut, J. Heffer, J. Mairesse and G. Postel-Vinay, ‘Les disparités de salaires en France au XIXe siècle’, Histoire et Mesure, 10, (1995), pp. 381–409; Edward H. Hunt, Regional wage variations in Britain 1850–1914 (Oxford, 1973) was a particularly early example. 17. For example, Nick Crafts, an eminent contributor to the debate on real wages up to the early 1990s, concluded in 1997 that it was necessary ‘to move beyond real wages’; Nicholas Crafts, ‘Some dimensions of the ‘quality of life’ during the British industrial revolution’, Economic History Review (1997), p. 617. See, too, François Crouzet, ‘Les niveaux de vie en Europe à la fin du XVIIIe siècle: l’apport qualitatif ’, in B. Etemad, J. Batou and T. David (eds), Pour une histoire économique et sociale internationale. Mélanges offerts à Paul Bairoch (Genève, 1995), pp. 137–67. 18 With regard to early modern labour relations, e.g., this shift was emphasised in the collection of essays edited by Catharina Lis, Jan Lucassen & Hugo Soly, Before the unions. Wage earners and collective actions in Europe, 1300-1850, International Review of Social History, Supplement 2 (Amsterdam, 1994). 19. E.g., Michael Sonenscher, Work and wages. Natural law, politics and the eighteenthcentury French trades (Cambridge, 1989). 20. E.g., Leonard Schwarz, London in the age of industrialisation (Cambridge, 1992), pp. 161–8. 21. E.g., Henny Gooren and Hans Heger, Per mud of bij de week gewonnen. De ontwikkeling van beloningssystemen in de Groningse landbouw, 1800-1914 (Groningen,1993), pp. 19–79; Reith, Lohn und Leistung, pp. 48–54. 22. It is a great pleasure to thank the commentators who not only gave a manifest surplus value to the debates in Les Treilles, but who also inspired greatly the authors of the papers as well as the editors: Ad Knotter (Maastricht), Henny Gooren (Nijmegen), Pat Hudson (Cardiff), Gilles Postel-Vinay (Paris), Iori Prothero (Manchester), Carmen Sarasua (Barcelona). 23. The Phelps Brown and Hopkins series run between the thirteenth and the twentieth centuries, see Henry Phelps Brown and Sheila Hopkins, A perspective of wages and prices (London, 1981), which includes most of the original papers published in the 1950s. 24. Donald Woodward, ‘Wage rates and living standards in pre-industrial England’, Past and Present (1981), pp. 28–45. See too, Donald Woodward, Men at work: labourers and building craftsmen in the towns of northern England, 1450-1750 (Cambridge, 1995). 25. E.g., Martin Daunton, Progress and poverty. An economic and social history of Britain, 1700-1850 (Oxford, 1995), p. 421 (chapter 16 is significantly labelled ‘The standard of living and the social history of the wage’). 26. Chris Tilly and Charles Tilly, Work under capitalism (Boulder, 1998), chapter 2 (Worlds of work).

22

Wages in Europe since the sixteenth century

27. See, e.g., Robert Castel, Les métamorphoses de la question sociale. Une chronique du salariat (Paris, 1995), pp. 109–22. 28. Stressed by, among others, Schwarz, London in the age of industrialisation, pp. 165–6. 29. In view of the criticisms by P. van den Eeckhout ( ‘Wage formation: some more problems’, in Scholliers, Real wages, pp. 40–47) it might be safer to limit this comment to ‘money wage labour’, but this begs the question of this book. 30. According to Ricardo, habit and custom contributed to the workers’ consumption norm, which in turn influences wages: David Ricardo, On the principles of political economy and taxation (ed. R.M. Hartwell, Harmondsworth, 1971 – orig. 1817), pp. 118–9. 31. Later, ‘wage rigidity‘ and ‘wage stickiness‘ have been put within the concept of the ‘efficiency wage‘, which directly links (high) wages to (high) productivity: see, e.g., Pierre Cahuc and André Zylberberg, Economie du travail. La formation des salaires et les déterminants du chômage (Paris – Brussels, 1996), pp. 180–227. 32. Phelps Brown and Hopkins, A perspective of wages and prices, p. 8. 33. E.H. Phelps Brown and M.Browne, A century of pay. The course of pay and production in France, Germany, Sweden, the United Kingdom and the United States of America, 1860–1960 (London – New York, 1968), pp. 151–2. 34. In this volume, Reith summarises the critiques. For Holland between 1500 and 1620, Jan de Vries and Ad Van der Woude found rapidly rising wages and highly fluctuating wage differentials; yet, some wage rates remained unchanged for two hundred years between 1600 and 1800 (The first modern economy (Cambridge, 1997), pp. 609–19). The wages paid to seventeenth-century London building workers have been examined in detail by Boulton: the modal wage rate does indeed show elements of long-term stability (though not usually for more than two or three decades), but there was a large variation around the mode. To some extent, therefore, the stability of building wage rates is a statistical creation of historians seeking to find patterns in the mass of data: J. Boulton, ‘Wage labour in seventeenth-century London’, Economic History Review (1996), pp. 275–83. The tradition of seeking the mode has been particularly strong in England; compare with, for instance, the work on Germany and Belgium: M.J. Elsass, Umriss einer Geschichte der Preise und Loehne in Deutschland (Leiden, 1936–49); H.J. Gerhard, Löhne in vor-und frühindustriellen Deutschland (Göttingen, 1984); C. Verlinden, J. Craeybeckx and E. Scholliers (eds), Documents pour l’histoire des prix et des salaires en Flandre et le Brabant, esp. volume 2 (Brussels, 1965). It is fair to note that during the last thirty years even English scholars have been pointing to the dispersion around the mode: see the references in Leonard Schwarz, ‘The formation of the wage: some problems’, in Scholliers, Real wages, n.29. 35. Sonenscher, Work and wages, chapter 6 (the quotation on p.206). 36. John Rule, The labouring classes in early industrial England 1750-1850 (London – New York, 1991), p. 116. 37. E. P. Thompson, ‘The moral economy reviewed’ in E. P. Thompson (ed.), Customs in Common (London, 1991), pp. 337–41; W. Reddy, The rise of market culture (Cambridge, Mass., 1984), pp. 331–4. Thompson was more than half persuaded by Adrian Randall, ‘The industrial moral economy of the Gloucestershire weavers in the eighteenth century’, in J. Rule (ed..), British Trade Unionism 1750-1850. The Formative years (London, 1988), pp. 29–51. 38. P. Linebaugh, The London hanged: crime and civil society in the eighteenth century (London, 1991) has been much criticised by reviewers, but it is almost unique in recognising the divisive power of a money wage. Linebaugh’s stress on the 1790s as a pivotal decade for the advance of the money wage in London is also debatable, but the impact of the inflation of that decade on social relations remains to be fully studied. 39. Pam Sharpe, ‘The female labour market in English agriculture in the industrial revolution: expansion of contraction?’, Agricultural History Review, 47:3 (1999), pp.11–81. 40. John Rule, ‘Industrial disputes, wage bargaining and the moral economy’ in A.J. Randall and A. Charlesworth (eds), Moral economy and popular protest: crowds, conflict and authority (Basingstoke, 1999), pp.166–86.

Peter Scholliers & Leonard Schwarz

23

41. A. J. Randall and A. Charlesworth, ‘The moral economy: riot, markets and social conflict’ in A.J. Randall and A. Charlesworth (eds), Moral economy and popular protest, p. 17 (their italics). 42. The defensive communitarian rhetoric is particularly in evidence when de-industrialisation threatens vital regional employment in coal, steel, cars, etc. 43. Craig Muldrew, The economy of obligation: the culture of credit and social relations in early modern England (London, 1998); John Styles, ‘ “Our traitorous moneymakers”: the Yorkshire coiners and the law, 1760–83’, in J. Brewer and J. Styles (eds), An ungovernable people: the English and their law in the seventeenth and eighteenth centuries (New Brunswick, 1980). 44. Angela Redish, Bimetallism: an economic and social analysis (Cambridge, 2000), pp. 149–58, 167–89. See also L. Dermigny, ‘Une carte monétaire de la France’, Annales ESC, 19 (1955), p. 481: the figures given above omit the recoinage of 1786: if this is included the copper currency was worth only 0.2 % of the combined gold and silver but weighed 30 %. 45. de Vries and van der Woude, The First Modern Economy, pp. 87–90. 46. Malcolm Crook, Toulon in war and revolution (Manchester, 1991), pp. 47, 95. 47. Robert G. Davis, ‘Venetian shipbuilders and the fountain of wine’, Past and Present, 156 (1997), pp. 55–86. 48. Steven King, Poverty and welfare in England 1700-1850. A regional perspective (Manchester, 2000), pp. 59, 118. 49. This is a suitable point to warn about any attempts to draw a direct comparison between M. Huberman and P. Van den Eekhout: textile workers form only a minority of the Ghent database. 50. Hobsbawm, ‘Custom, wages and work-load’, p. 360. 51. ‘We know so little about systems of management and wage payment that it is dangerous to generalize about them, especially in view of the incredible complexity of the industrial scene’, Hobsbawm, ‘Custom, wages and work-load’, p. 357. 52. Gooren and Heger, Per mud of bij de week gewonnen, p. 2. 53. Reith, Lohn und Leistung, p. 29. 54. Rule, The labouring classes, p. 121; Hobsbawm, ‘Custom, wages and work-load’, p. 349. 55. Significantly, historians have to refer to ‘classics’ of labour economics, such as David Schloss, Methods of industrial remuneration (London, 1898); Maurice Dobb, Wages; Tom Lupton (ed.), Payment systems (Harmondsworth, 1972); Angela Bowey (ed.), Handbook of salary and wage systems (Aldershot, 1982). 56. Robert Duplesssis, Transitions to capitalism in early modern Europe (Cambridge, 1997), pp. 267–8. 57. Rule, The labouring classes, p.126. 58. This is illustrated with regard to the Antwerp port labour by Harald Deceulaer, ‘Arbeidsregulering en loonvorming in de Antwerpse haven, 1585 – 1796’, Tijdschrift voor sociale geschiedenis (1992), pp. 29–31. 59. Hobsbawm, ‘Custom, wages and work-load’, pp. 356–61. 60. Peter Scholliers, ‘L’identité des ouvriers-mécaniciens gantois au XIXe siècle’, Histoire, Economie et Société (1987), pp. 96–7. 61. E.g., Patrick Fridenson, Histoire des usines Renault. 1. Naissance de la grande entreprise 1898-1939 (Paris, 1972), pp. 38-9, 77. 62. Schmoller, ‘Le mouvement historique’, p. 13. 63. Sonenscher, Work and wages, p. 194. 64. Jan de Vries, ‘An employer’s guide to wages and working conditions in the Netherlands, 1450–1850’, in Carol Leonard & B. Mironov (eds), Hours of work and means of payment: the evolution of conventions in pre-industrial Europe (Milan, 1994), p. 56. 65. Schwarz, London in the age of industrialisation, p. 164. 66. Schwarz, ‘The formation of the wage’, p. 37. 67. Duplessis, Transitions to capitalism, p. 20. 68. Evidently, wage differences between men and women were well known right from the beginning of the wage research; yet, this did not form a great concern. For instance, in

24

69. 70. 71. 72.

73.

74. 75.

76. 77.

78.

79.

80.

Wages in Europe since the sixteenth century a survey on the German working classes, the general wage development in the textile industry was rightly seen as ‘inadequate’ because of the steadily growing proportion of women; hence, female wages were neglected [W. J. Ashley, The progress of the German working classes in the last quarter of the century (London, 1904), pp.96–100]. E.g., Donald Cox and John V. Nye, ‘Male-female wage discrimination in nineteenthcentury France’, Journal of Economic History (1989), p. 903. Joyce Burnette, ‘An investigation of the female-male wage gap during the industrial revolution in Britain’, Economic History Review (1997), 257–281. Corrie van Eijl, Het werkzame verschil. Vrouwen in de slag om de arbeid, 1898–1949, (Hilversum, 1994), p. 308. Sara Horell and Jane Humphries, ‘Old questions, new data, and alternative perspectives: families’ living standards in the industrial revolution’, Journal of Economic History (1992), pp. 849–80. Peter Scholliers, ‘Family income, needs, and mother’s wages. A critical survey of working-class budget inquiries in Belgium, 1853–1929’, in Toni Pierenkemper (ed.), Zur Ökonomik des Privaten Haushalts (Frankfurt – New York, 1991), p. 169. Joel Mokyr, ‘Is there still life in the pessimist case? Consumption during the industrial revolution, 1790-1850’, Journal of Economic History (1988), p. 88. Jan de Vries, ‘Between purchasing power and the world of goods: understanding the household economy in early modern Europe’, in J. Brewer and R. Porter (eds), Consumption and the world of goods (London, 1993), p. 85–125. For a view of the debate, see, e.g., Colin Creighton, ‘The rise of the male breadwinner family: a reprisal’, Comparative Studies in Society and History (1996), pp. 310–37. E.g., Michael Haines, ‘Industrial work and the family life cycle, 1889–1890’, in P. Uselding (ed.), Research in economic history (1979), pp. 289–356; Patricia van den Eeckhout, ‘Family income of Ghent working-class families, c. 1900’, Journal of Family History (1993), pp. 87–110. Elyce Rotella and George Alter, ‘Working class debt in the late nineteenth-century United States’, Journal of Family History (1993), pp. 111–34; Sara Horrell and Jane Humphries, ‘The exploitation of little children: child labor and the family economy in the industrial revolution’, Explorations in Economic History (1995), pp. 485–516. E.g., Louis Varlez, Les salaires dans l’industrie gantoise. 1. Industrie cotonnière (Brussels, 1901), pp. 80-1 (presenting the wage of female and male cotton workers whose ages ranged from 13 to over 70). Dudley Baines and Paul Johnson, ‘The labor force participation and economic well-being of older men in London, 1929–1931’, in Timothy Guinane, Paul Johnson and Clara E. Nuñez (eds), The microeconomic analysis of the household and the labour market, 18801939 (Seville, 1998), pp. 125–35.

CUSTOM, WAGES AND THE MARKET

CHAPTER 2

INSTITUTIONAL AND CULTURAL CHANGE IN WAGE FORMATION: PORT LABOUR IN ANTWERP (SIXTEENTH–EIGHTEENTH CENTURIES)1 Harald Deceulaer

In 1617, an unruly Antwerp transport worker claimed that his guild could not punish him after a dispute about wages, as he had drunk with the guild’s dean, Simon Lostaert. According to testimonies, he had placed a pint of beer before the dean during the guild’s feast. Unaware who had offered the drink, Lostaert said ‘God bless’, and gratefully drank it.2 Disputes were normally settled by drinking together. As the parties had accidentally drunk together, this was used as an argument that a settlement had been reached. This anecdote reveals the importance of shared values in a group culture, where regularly collective drinking confirmed the group’s identity in both work and leisure. It also points to the importance of social and cultural attitudes, habits and beliefs in the way an early modern guild structured and sanctioned the organisation of work and wages. It is well known that early-modern guildsmen regularly feasted, that they were involved in civic and religious rituals, and engaged in sociability. These activities were intertwined with symbolic exchanges of cultural markers and signs, through which they imagined the boundaries of their communities, fashioning and refashioning their sense of identity, honour and status.3 The dynamic struggle over rank, classification and taxonomy undoubtedly had important cultural meanings, but may also raise social and economic questions. It is not clear how these symbolic practices were related to guilds as economic institutions and how they evolved over time

28

Changes in wage formation

in form, intensity and meaning in the early modern period. Did these social and cultural practices matter (much) for processes of economic and institutional change? Was the organisation of work and the wage level in a port (and, thus, part of the transaction costs of merchants) influenced by guild feasts, processions and their meanings and values? Most studies about early-modern work and wages have not focussed on this problem. Research about early-modern wages has generally been limited to the construction of series of building-workers wages, due to the lack of sources and to the complexity of early-modern wage formation. As these nominal early-modern wages seem to have been extremely stable in the long run, real wages were perceived to be primarily determined by the cost of living, i.e. the price of basic food. The construction of time series of real wages has certainly been an important analytical tool in the study of living standards.4 These techniques led to an influential body of research, in which several hypotheses were built, e.g. about the nature of aggregate demand or about social relations.5 As wages were supposedly paid in a static, customary way and as waged work was considered to comprise only a small part of the early-modern population, the food riot was often considered to be the dominant form of early-modern protest, as an expression of an early-modern ‘moral economy’.6 The alleged longterm stability of early-modern nominal wages also fitted very well in an old static vision of the Old Regime and the world of the guilds. Still, most early-modern workers and artisans did not receive a stable time wage, but were paid for specific tasks in highly different forms. In this field, clear-cut quantitative indices are much more difficult to assemble, and institutions and organisations should be included in the analysis. The study of wages must be linked to economic evolutions, social relations, institutional change, entrepreneurial strategies, political structures and cultural norms and values. In this chapter I will not study wages as quantitative series, but as indications of concrete social relations and practices within a broad cultural framework. On the basis of a case study of transport workers’ wages, I will argue that wages were not static, and that the composition and the ways of calculating the wage mattered for the income, and even for the social identity of the workers. Institutional and organisational change were therefore of paramount importance and were also related to changes in the meaning and perception of labour. Therefore, attention will be directed to cognitive processes, the system of values, and the construction of a social reality.7 I will try to show how general seventeenth- and eighteenth-century cultural transformations influenced the economic institutions of the Antwerp transport workers. The multiple interactions of institutional, economic and cultural changes are thus the broader problem addressed in this chapter. The problem of the interactions between ‘economics’ and ‘culture’ are increasingly on the international research agenda. Obviously, bringing the concept of ‘culture’ into economic history is a fascinating, but far from self-evident or unproblematic challenge, for which historians have given many different answers. ‘Culture’ has perhaps been employed

Deceulaer

29

most fruitfully on the demand side, e.g., in the important work on consumption and taste since the late seventeenth century.8 However, at least since Max Weber, there has been a tradition to look at cultural influences on the supply side as well. Although Weber himself actually presented a more complex and group-specific analysis of the interaction between culture and economics than is often thought, many of his followers have been less subtle. ‘Protestant’, ‘European’ or even ‘AngloSaxon’ values have been assigned an important place in explaining the pre-eminence of the West in the world economy.9 These ideas have often been highly tautological and were based on a homogeneous, undifferentiated concept of culture, which apparently permeated society as a whole. Such reading of culture, of course, clashes with many insights from social history (and, more recently post-colonial theory), which presented accounts of an ‘elite’ and a ‘popular’ culture, or a ‘dominant’ and a ‘subordinate’ one. Needles to say, these dualistic views have been highly influential, for example in descriptions of the allegedly traditional, risk-averse mentalité of the early-modern ‘peasant’ or guildsman vis-àvis the modern profit-calculating entrepreneur or homo economicus. More recently, however, this dichotomous view of culture has also been rejected as too simplistic.10 We have to look more closely to interactions and appropriations between different groups and individuals, which led to intriguingly hybrid, socially mixed forms, usages and meanings of culture.11 In this respect, institutions offer interesting fields of negotiation (and battle), as they were not merely imposed ‘from above’, but were also shaped by the practices and interpretations of many different actors: ‘as social phenomena, they do not exist independently of ways of thinking about and (sometimes in contrast) performing the routines they prescribe’.12 Some recent theoretical work may offer fruitful tools to work out connections between culture and economics. In institutional economics, Douglas North makes a distinction between institutions (rules) and organisations. In his model, rules consist of formal rules, informal rules and enforcement mechanisms. Organisations are political bodies, economic bodies, social bodies and educational bodies. Organisations and institutions are embedded in a political and cultural setting: ‘If markets are incomplete, the information feedback is fragmentary at best, and transaction costs are significant, then the subjective models of actors modified both by very imperfect feedback and by ideology will shape the path’.13 In a recent debate Bo Gustafsson and Douglas North stressed the necessity to integrate the historical study of organisations and culture. The identity, performance, emergence and eventual disappearance of institutions were explicitly set on the research agenda.14 Relative changes of prices and culture are both considered to be important elements explaining institutional change, but their exact mix remains unclear: ‘We need to know much more about culturally derived norms of behaviour and how they interact with formal rules to get better answers’.15 These interactions may be better grasped with a few sociological insights by Bourdieu,

30

Changes in wage formation

Granovetter and others, related to the status of a group, distinction, ‘embeddedness’ and social closure. The presence or the absence of a group in an official taxonomy depends on its ability to carve out a recognised position. Cultural goods are seen to be important weapons in the struggle to obtain and monopolise positions.16 Therefore, a specific style of life is expected from aspiring members of a status group: ‘(…) les membres des groupes fondés sur la cooptation, comme sont la plupart des corps que protège un numerus clausus explicite ou tacite (médecins,..) ont toujours en commun plus et autre chose que les caractéristiques explicitement exigées’.17 Although I do not intend to reduce culture to ‘ideological struggle’, I will view the group culture of specific Antwerp transport workers as promoting the collective status and the internal cohesion on the one hand, and as a symbolic barrier to nonmembers on the other. In this chapter, I will apply and test some of these concepts on the organisation and payment of port labour in Antwerp from the sixteenth to the eighteenth century. Antwerp’s transport labour underwent important organisational and institutional changes in this period, which has to be linked to broader economic, social and cultural transformations. During the sixteenth-century economic boom, the guild-like (sub)organisations of transport workers were abolished, but they re-emerged when transport volumes declined after the capture of the town by Spanish troops in 1585. The institutional setting influenced both the transaction costs of merchants and social inequalities among the workforce. Furthermore, cultural strategies, representations and practices strengthened the institutional organisation of port labour. The gradual change of the organisations and institutions in the eighteenth century went hand in hand with a change in cultural practices. Although predominantly a piece of social history, this chapter also touches upon a few classical themes in the cultural history of the early-modern period, e.g. the influence of large cultural transformations such as the Counter-Reformation and the Enlightenment on broader segments of the population.18 By studying the shifting negotiation between work, wages, representations and cultural strategies on a micro-level, we may see that macro-transformations as the Counter-Reformation and the Enlightenment were partly constructed by social and cultural grassroots appropriations, practices and conflicts, which were related to the allocation of economic resources, political positions and power. Transport workers were a numerically important group in urban society, and it seems that the organisation of labour markets and wages often bear signs of resemblance in many port towns. I focus on Antwerp because this town was the most important port town in the Southern Netherlands, and because the successive periods of spectacular economic growth and relative decline enable study of the evolution of institutions in different contexts. Furthermore, the Antwerp City Archives posses a rich set of records which permit an extraordinary in-depth analysis of labour relations.

Deceulaer

31

Guilds, ‘nations’ and workers in the Antwerp port The labour market in the port was organised by three guilds: the transport workers, the grain porters and the peat porters. Two smaller groups existed independently: the wagon drivers and the cranemen (who worked with the city’s crane). Fines and rules were often decided collectively, with a majority vote.19 The rules concerning work and wages had to be issued by the local authorities, and merchants were usually consulted. The guilds of the transport workers held political offices on the Maandagse Raad and Brede Raad, (the ‘Monday Council’ and the ‘Large Council’), the local representative institutions (which, amongst other things, had to approve of the taxes). These political positions enabled the guilds to pressure the local government.20 The transport-workers guild also participated in civic and religious rituals, by marching in the processions and parades. More importantly, the large guild of the transport workers was composed of several subgroups (called naties, or ‘nations’), that monopolised the transport of particular goods, or that controlled all work at certain quays. Often, they helped the city government in the collection of the excise duties. The naties were groups with a fixed number of workers. When there was excessive work, they employed members of the guild who were not part of these naties through subcontracting arrangements, or even casual workers, who were not member of the guild at all. To become a member of a natie, one had to buy a position from a resigning member (often someone who had become too old or weak for the job). These sums were higher than one could normally afford from modest earnings alone, and many new members had borrowed money to purchase membership. On top of this, at entry a small sum had to be paid to the town, and another to the guild. Some naties also required a meal for all members, or a payment of a small sum to each member. Members of the guild who were not part of a natie never bought an official position, but only paid the smaller entrance-barrier to the town and to the guild. Likewise, their claim to work in the port was less strong. Therefore, the ‘wage’ of a member of the naties can partly be seen as a rent on an investment of capital. This rent was not the only advantage, however, because they were able to sell their membership when they were older.21

The rise, fall and re-emergence of the naties in the sixteenth century In the early sixteenth century, the vigorous economic expansion of the Low Countries and the redirection of international trade routes made Antwerp the centre of the European market economy.22 With the growth of port traffic, conflicts arose with merchants who wanted to reduce transaction costs. In 1544, Charles V abolished the naties, and only one large general guild was kept. Merchants were free to hire any worker, as long

32

Changes in wage formation

as he was a member of the guild. Access to this guild became very easy.23 Charles V had taken similar measures for the transport workers of Ghent, as a part of a constitutional reorganisation of the town.24 The work organisation in the Antwerp port between 1544 and 1585 anticipated what was going to happen in the port of Amsterdam, when transport volumes grew from the early seventeenth century onwards.25 In the late 1560s however, the economic tides turned. Domestic economic problems, international trade policies and religious troubles weakened the economy of the Southern Netherlands. The traffic on the river Scheldt declined strongly, especially after 1585, when the Dutch closed the river after the separation of the Northern and Southern Netherlands. Ships from the Southern Netherlands could not gain access to the sea, but had to be transshipped into Dutch boats in the small village of Lillo. Foreign ships heading for Antwerp had to transfer their cargo into Zealand boats at Middelburg or Vlissingen, and taxes had to be paid.26 This new political border clearly increased the transaction costs, and port traffic was strongly reduced after 1585.27 The guild system, and especially the system of guild subgroups, was able to re-establish itself. Again, jobs in these subgroups became restricted and were bought and sold. Merchants or shippers could not simply employ any worker they liked. The subgroups, or naties, had a strong monopoly, which was asserted by law and defended through litigation. They could employ other workers through subcontracting if there was too much work to do themselves. Work was done collectively, they were also paid collectively. Sometimes, groups of workers received money on an individual basis, but they were supposed to hand this over to the natie, and each received his share at the end of the week. The re-emergence of the naties was a process that went hand in hand with both external and internal reorganisations of social relations. Shippers and merchants on the one hand, and ordinary workers on the other had to be forced to accept the renewed power of the naties. Formal sanctions by the city government strengthened the position of the natie members vis-à-vis that of the shippers, merchants or entrepreneurs. Shippers who organised the unloading of the ships themselves were regulated by the city government in 1592, 1611, 1613, 1625 and 1635.28 The longterm conflict between the soap boilers and the butter carriers is instructive for the changes in balances of power in the Antwerp port. In the larger part of the sixteenth century, soap boilers were free to organise autonomously the transport of their wares, but in the seventeenth century they had to employ the butter carriers.29 The stronger position of the naties can be illustrated by the remark of two shippers in 1642. According to them, the workers ‘do not follow the ordinances of the magistrate, but solely their own heads and minds’.30 A merchant woman declared bitterly in 1662 that the butter porters ‘have become such great masters that they don’t simply try to extort more money than they are entitled to, they even try to prevent others from working’.31

Deceulaer

33

This last remark points to the internal reshaping of social relations. It goes without saying that the re-emergence of the naties after 1585 strengthened the segmentation of the workforce, which created tensions with ordinary workers. The first natie to reassert itself after 1585 was that of the butter porters. They strengthened their monopoly through a similar series of conflicts with ordinary members of the transport-workers guild. When in 1587, the worker Guiliam Verhulst was charged with carrying butter, he protested that ordinary workers (guild members) had been transporting butter in small quantities for the last thirty years.32 In 1596, ordinary guild members questioned the legitimacy of their exclusion in a petition to the city magistrate. They argued that they had to have the same rights, as they were all members of the same guild.33 These efforts failed, however. A strict control on the excise duties was of crucial importance for the revenues of the town, and the transport of these goods was therefore only entrusted to a small group of workers. In the course of the seventeenth century, a growing segmentation of the workforce could be witnessed. In 1608, several regulations were set up which confirmed the privileges of a few naties.34 The workers at the ‘Wharf ’ and the Bierhoofd (two Antwerp quays) limited their numbers in 1608 and 1615.35 The carriers of coal were officially recognised as a natie in 1635.36 Similar processes were probably going on in other parts of the Southern Netherlands. In the same period, a growing segmentation of the workforce in the ports of Ghent and Bruges can also be registered. The fact that the guilds of the transport workers in Menen and Geraardsbergen also date from the seventeenth century points at the same direction.37 It is clear that formal enforcement mechanisms were of crucial importance in the institutional reorganisation of the Antwerp port. Yet informal sanctions were also embedded in a strong group culture. When in 1631 the worker Adriaen van Lippenveldt had worked on his own, and refused to share the wage, he was not only sanctioned by the guild. After having insulted a dean and fought with a colleague, his fellow workers refused to drink with him. A few days later, five colleagues dropped a heavy barrel on him.38 Whether this kind of violence was often used on the Antwerp quays is not clear. Among guild members, fights were fined, with different tariffs according to the degree of violence.39 The remark of a grain porter in 1693 that ‘a wrestle or a blow among artisans is not of the same importance as among persons of a higher standard’,40 may suggest regular minor violence. However, natie members who were threatened by violent unorganised competitors on the Antwerp quays argued that ‘they did not wish to defend their rights with fighting or violence’, as ‘this should not be tolerated in a well policed commercial town’.41 Strong group norms were especially important in the smaller naties with a fixed number of members who did the work themselves, where each member received his share of the income at the end of the week. As they were sometimes paid individually, but shared their income, deceit was an ever present danger. In 1769, the workers of the Laag Bierhoofd excluded their fellow worker Johannes Regemortel, as he had repeatedly withheld

34

Changes in wage formation

money. His exclusion was presented as ‘an example to all others, since good faith has always been and must remain the foundation of the natie, in order to prevent prejudice to each one and to the natie as a whole’.42

Work and wages in the Antwerp port during the apogee of the Guild Regime, 1585–ca 175043 When a shipper or a merchant wanted goods to be transported after 1585, he had to address himself to the guild or the natie which held a privilege on the goods involved. The dean of the guild or the one in charge of the natie, then, appointed a team of workers. This was done by throwing dice or by complex rotation systems. The shipper or the merchant could not employ the workers of his choice. The nonmembers of the natie which could be employed in busy periods, had to be paid the same wage as the regular workers, but they were obliged to hand over half of their wage to the natie involved (which distributed this to the regular members). When wages were paid individually, natie members had to put the money in the natie box. Wages were also paid collectively to the natie by the merchants. At the end of the week, each worker received his share. In the house of the natie, notes about the work of each member were made with chalk on a beam, or with a tally stick. As in most early-modern ports, wage formation in Antwerp was highly complex. First of all, it is not clear whether the payment to the formal transport workers can be considered as a clear-cut wage. The official transport workers were employed by merchants, but they could employ other workers as well. Although they were not ‘masters’, they were members of a guild, and they probably regarded themselves as independent workers.44 Workers were not paid by the hour or by the day, but according to a specific task. Detailed tariffs existed for different products and different loads. For the workers of the Sint Jans Vliet for example, an ordinance of 1768 classified 441 types of wages. For some groups of transport workers, the place where the goods were to be delivered influenced wages too. The grain porters, for example, received a lower wage in the area within sixty steps of the main grain market. The town was divided into three ‘circles’ with a different rate. Extra strenuous efforts were often compensated with supplementary payments, e.g., for butter porters when they had to get the goods out of a cellar, or for the grain porters when they had to walk over other ships. Likewise, different tariffs existed for goods which had to be delivered to the first, second or third floor.45 The wage in the port of the Old Regime cannot simply be regarded as a straightforward payment for work. Rather than an exact reflection of the reality of everyday work, the tariffs were a yardstick, a set of rules and norms around which practices and conventions fluctuated. The favourable tariffs for some customers, e.g., merchants in grain and salt,46 were not the most complicating factor, nor were the payments in kind, which some workers received and which they often sold.47 More importantly, it must

Deceulaer

35

be remembered that labour in the Antwerp port of the Old Regime was not a commodity, but an officially recognised guild privilege. Holders of this privilege had invested in membership, and were compensated with a rent. Merchants who employed non-guild members to unload their goods, violated the privilege of the official workers, and therefore, had to compensate this by paying their ‘wages’ (for work they had not done).48 Official workers also received a part of the wage paid to casual labourers. Another element which renders a reconstruction of the wages nigh impossible, is the fact that the prescribed tariffs were not always applied in actual transactions. Workers often claimed and received more than their official wage, and this was not limited to an occasional pint of beer.49 In 1645 a few merchants stated that they had always paid fifty stuyvers for the transport of 14,400 pounds of coal, while according to the tariff, they only had to pay thirty stuyvers.50 In the late 1570s, the coal carriers claimed that the formal wage rise had already been paid earlier, ‘not according to an ordinance, sed tacito populi consenti, as the community found that we could not survive on such a small salary’.51 Of course, these payments were not simply paid by consent; they were the result of conflict, negotiation and struggle. In 1611 a merchant complained that the porters of lime and stone refused to work without a tip of three guilders.52 Often, customers refused to pay the (higher) conventional rates, and wished to return to the official wages prescribed in the ordinances.53 These conflicts resembled conventions and practices between masters and journeymen in other sectors, e.g. in the textile industry.54 The wage in the port of Antwerp was therefore not ‘traditional’, remaining fixed during the seventeenth and eighteenth centuries. Formal wage rises did not occur after the 1640s, but that did not preclude fluctuations in actual payments, due to changes in supply and demand in the labour market.55 The sum of the wages the transport workers received at the end of the week did not equal exactly the money which customers had paid for the transport service. Many naties had general costs, which were paid by collective receipts. Most of them supported colleagues who were ill, or who had had an accident.56 Others had to buy or rent equipment, to maintain their house or to feed horses (e.g. the wagon drivers).57All naties also compensated customers when their goods were damaged during transport.58 In some naties extra sources of income existed, that were related to the guild organisation. When the yearly balance of the incomes and expenditures of the guild of grain porters was positive, the result was distributed among the members.59 Among the butter porters, each new member had to give six guilders to every other member.60 All these elements exclude a precise quantitative reconstruction of the (real) wage level in the port. Contemporaries were well aware of this. When in 1759 the grain porters were condemned to pay a part of the wages of a few casual workers for the past year, they cried out ‘How can we possibly make an account of the bill of the past, as this is formed by hundreds or thousands of little parts? Furthermore, they continued, it is well known that ‘in one year there is much more work than in the other’.61 Thirty years earlier, the coal

36

Changes in wage formation

workers had already said, ‘Most of the work is done during three months of the year’.62 Port labour was indeed notoriously irregular. Shipping was often impossible in winter. Until the 1860s, the Antwerp port was dominated by sailing ships, which were of course sensitive to weather conditions.63 All this endorses the view that ‘precise measurement of real wages in the eighteenth century is almost impossible’.64 In 1611 and in 1759, contemporaries claimed that lime and stone carriers and grain porters earned about 300 guilders a year.65 According to this source, the knechten (i.e. servants or men who were employed via subcontracting) of the grain porters would earn 80 to 90 guilders a year. These sums equalled respectively 500 and 133 to 150 (summer) percent of the daily wages of an unskilled worker in the building trades.66 It is hard to assess whether these figures are realistic or not. In the incomplete Antwerp tax rolls of 1747 (with information for only six of the thirteen quarters of the town), a distinction is made between those who were too poor to pay tax, those who only paid a small per capita tax, and those who paid taxes according to their income (from 1 to 300 guilders). For the highest categories, it is estimated that the tax rate amounted to an average of three percent of the income. For poorer groups, however, this rate was probably lower.67 In this source, 110 transport workers were present. Generally, it is not possible to differentiate between official and casual workers. Sixty-one percent of all registered transport workers paid nothing or only the per capita tax, and can thus be considered poor. With the low and irregular incomes of most transport workers in mind, it comes as no surprise that they combined port labour with other activities, although this is difficult to demonstrate. In 1652, a transport worker stated that ‘everybody knows and it is daily observed that a transport worker is not able to live on his trade alone, so one is an ale-house keeper, the other a small mercer and another a baker’.68 In 1692, a few casual labourers who were employed by the grain porters were also said to work as a soldier, a bleacher, a carpenter and a journeyman-brewer.69 Not all multiple occupations were born out of necessity, however. Twelve members of naties were registered to have a double profession. Nine were also active as shopkeepers, one was an innkeeper, one had a job with the office of tolls collecting, and another was a night watchman.70 They were the richest workers in the port. It is highly probable that they engaged other workers via subcontracting.

Cultural practices and representations The cultural or ideological legitimacy of the social inequality between natie members, ordinary guild members and nonmembers may have sometimes been problematic. Carrying goods was not a job for which elaborate skills or an apprenticeship were needed. The acquisition of a natie membership was enough to be assured of a steady job, and to pocket half of the wages of the workers employed via subcontracting. In most part of

Deceulaer

37

the seventeenth and early eighteenth centuries, they even had to be paid when a merchant had organised the work himself. Therefore, creating distinction through symbols, behaviour and ritual may have been crucial for the natie members. My earlier work on the Antwerp and Ghent clothing trades suggested different types of investment in distinction and status among artisans working on different market segments. Artisans with highly personalised skills or positions were generally more eager to develop status strategies.71 I will briefly discuss a few crucial social and cultural elements, which could be appropriated for status purposes in the early-modern period: religion, buildings and rituals. The transport workers guild as a whole had their altar at which a mass was read each Sunday morning. In the course of the seventeenth century, their liturgical goods and ceremonial vestments expanded significantly.72 On the yearly feast day, they distributed prints with the image of their patron saint. These practices are hardly surprising in the context of the triumphant Counter Reformation, and one must be careful not to interpret them simply as status strategies. Other elements, however, do point in that direction. The organisations of the transport workers seem to have invested strongly in buildings, which were important carriers of meaning. The transport workers guild had one building, but the naties of the butter carriers, the peat porters and the coal porters had a house of their own. The grain porters even owned two buildings at the most important grain market of Antwerp.73 Among all the guilds in sixteenth-century Ghent, the transport workers guild owned one of the most expensive houses.74 Cultural practices were not simply developed to enhance collective status, but also to differentiate themselves from competitors. The re-emergence of the naties and the concomitant reorganisation of work, wages and social classifications were accompanied by changes in ritual and representations. Parallel with the improvement of their positions on the labour market after 1585, the butter porters had gained prominence over the ordinary workers in the processions, in the governance of the guild and in the representative positions in the City Council. In the oath, new ‘ordinary’ guild members had to swear not to unload goods that were the privilege of the butter porters (butter, oil, soap and almost all goods which were indirectly taxed). The swearing of the oath had a political and religious character. New members had to swear allegiance to the Duke of Brabant (the monarch) and to the city of Antwerp. In the 1650 the ordinary guild members tried to reduce this symbolic inequality. When the magistrate decided that the workers’ dean and the one of the butter porters would alternate yearly in fulfilling these functions, the latter circumvented this measure by appointing new deans each year (which kept the prominence in their hands). It was unacceptable to them that a new butter porter would have to swear an oath to a dean who was maybe only an ordinary worker. Therefore, they asked to swear their oaths to the mayor himself, which was granted.75 In 1650, an ordinance by the central government confirmed the prominence of the butter porters in processions and

38

Changes in wage formation

in the city council.76 The butter porters also constructed a symbolic distinction vis-à-vis death and the afterlife. In 1673, a contract with the guild chaplain stipulated that if one of them died, six priests had to sing the Miserere and the Dies Ire from the Church’s High Choir during the Requiem mass.77 The ordinary Requiem mass – customary for all guildmembers – was clearly not enough. Gestures, cries and conspicuous forms of consumption may also have expressed and reinforced existing differences between several groups of workers in the port. When they were offered work, the natie members of the ‘Maeygat’ yelled three times ‘Meat’ in the gate near the quay, to warn their colleagues.78 In the poorer neighbourhood near the river Scheldt, this must have been a powerful symbol of status and prosperity. A similar symbol, wine, was drunk by the cranemen during their yearly feast day, and by representatives of naties and merchants to solemnly drink to agreements.79 Individual natie members also drank wine: in 1624 two workers of the Bierhoofd quay went to drink ‘a pint of wine or two’.80 Sources about individual consumption habits are scarce, but in three probate inventories of early-modern transport workers wineglasses were registered.81 Clothing may have constituted another element in the representation strategies of the natie members. In 1611, a merchant stated that the lime and stoneworkers ‘dressed themselves in satin and other costly attire’.82 A few early-seventeenth-century probate inventories of natie members indeed register silver buttons, large amounts of shirts, etc.83 The strong development of the clothing trades in Antwerp at the time probably offered possibilities for distinction and appropriation.84

Recontracting and subcontracting: towards a new institutional setting, ca 1750–1800 Although earlier research has shown that transport volumes in the port of Antwerp did not grow in the second half of the eighteenth century,85 the organisation of work and wages did not remain static. In the course of the eighteenth century, two (possibly interrelated) phenomena can be sketched concerning work and wages. First, a rise in casual labour, and second, an attempt by merchants and manufacturers to lower the wages of transport workers. Both developments went hand in hand with changes in cultural practices and representations. In the eighteenth century, the number of new transport workers (who were members of the guild) did not grow (Figure 2.1). In the account books of the guild, we may find the receipts for a small yearly tax of four stuyvers, which each worker in the harbour had to pay to the guild (called kaarsgeld, literally ‘candle money’). The casual labourers in the port probably had to pay this tax too. Although we hardly possess any detailed information about the collection of this kaarsgeld, it seems a reasonable hypothesis that the fluctuations of this small yearly tax reflected the irregular segment of the labour market more than the number of new guild

Deceulaer

39

members. As can be seen in Figure 2.2, the receipts of this tax rose after 1771. I extrapolated the number of workers who probably paid this tax. The rise after 1771, and the concomitant decline of new entries to the guild of transport workers could suggest that a smaller group of guild members put out work more often to a larger group of casual workers. This informal system of subcontracting is extremely difficult to study, but limited evidence indicates a growth of this phenomenon in this period.86 In a list made up in 1794 the Sint Jans Vliet workers were all called baes (i.e. ‘boss’),87 a term supposed to have arisen only in the 1830s.88 This clearly suggests subcontracting. A clear-cut evolution of incomes in port labour is difficult to present. It is not very likely that incomes from port labour grew in the eighteenth century. Several small extras came increasingly under attack. The coal carriers were no longer given any ‘bag money’ (for the maintenance of their bags) in 1723. A year later, a small payment for the work in deep ships was abolished. In 1732, a small yearly sum for the ladders of the grain porters was cancelled, as was the annual payment of sixty guilders as a compensation for their help in the collection of the excise duties.89 Full payments to the privileged workers when the work was done by others were slowly eroded. Already in 1686 the magistrate had decided that the workers of the Laag Bierhoofd quay were not entitled to receive any compensation when they had been too busy to do the job themselves. In 1725 a similar arrangement was put into practice for the cranemen.90 These measures were probably a sign of a increasing criticism of the organisation of work in the port. In 1740 two masons asked for a new ordinance, ‘which would stop the naties from playing the master’.91 Fourteen years later, the Antwerp sugar refiners even stated that ‘when the naties would not exist in our city (…), one would find very easily a large quantity of honest people who were to be content with half of the wages, and of whom one could get with courtesy what the workers now refuse with brutality’.92 In 1754 it was stated that the grain porters were obliged to hire and pay non-guild members when they could not handle the job (another indication of the growth of subcontracting). Twenty years later, a new ordinance was promulgated which gave the merchants the right to have their goods transported by their domestics. As the formulations in the ordinance were highly subtle and cryptic, it comes as no surprise that earlier historians have quoted the measure as a mere confirmation of existing practices (and have therefore underestimated the element of change in the eighteenth-century port).93 Contemporaries, however, were well aware of what was at stake. The guild stated that this ordinance destroyed their organisation completely. As they were members of the City Council, this new measure could not be enacted without offending the City’s constitution. Furthermore, they almost openly threatened with disorder and riots: ‘many calamities are to be feared, since there are some five hundred members of the guild’.94 Their threats were probably in vain because, at the end of the Old Regime, merchants were said only to pay a small fine when they hired non-guild members.95

1700–09 1710–19 1720–29 1730–39 1740–49 1750–59 1760–69 1770–79 1780–89 1790–94

Figure 2.1. New members of the guild of the transport workers per year, 1700–1794 (10-year averages)

0

5

10

15

20

25

30

40 Changes in wage formation

1763

1760

1757

1754

1751

1748

1745

1741

1738

1735

1732

1729

1726

1723

1720

Figure 2.2. Number of workers paying kaarsgeld to the transport-workers guild, 1720–1789

0

50

100

150

200

250

300

350

400

450

Deceulaer 41

1787

1784

1781

1778

1775

1772

1769

1766

42

Changes in wage formation

Merchants also tried to lower the existing tariffs. Several customers refused to pay the regular wages in 1713, 1714, 1718, 1723, 1727, 1741, 1783 and again in 1784.96 In a more subtle way, they tried to pay the same wage for more work.97 As most tariffs were not paid according to the weight, but per package, barrel or bag, this left room for manoeuvring. A group of workers complained in 1786 that the bags of tobacco suddenly weighed 1500 pounds, instead of the traditional 900. Similar tactics were followed by the sugar refiners and wood merchants in 1756 and 1778.98 In several towns in the Southern Netherlands, merchants and entrepreneurs tried to increase their control over the transport sector in the last quarter of the eighteenth century.99 These efforts, combined with the decline in traffic in the Antwerp port and the general rise in prices, may have resulted in a lower income in the late eighteenth century. In 1777, the deans of the guild of the transport workers stated that ‘they are almost all very poor people, who have been obliged to spend the greater part of what they have earned during their glory years’.100 There are signs that the growth in subcontracting and casual work eroded the traditional group culture of the naties. Detailed data on the absentees during processions of the grain porters show that these were higher among non-working members who organised subcontracting.101 Ceremonial activities and sociability likewise declined. In the seventeenth century, about forty percent of the transport workers guild expenses were spent on collective meals and drinks, but these expenses diminished to some twenty percent in the eighteenth century. Religious spending by the guild also declined. In the late seventeenth and early eighteenth century, twenty to thirty percent of the guilds’ expenditures were spent on masses, religious artefacts and the like. Between 1776 and 1785, this amounted only to seven percent. Of course, these figures do not reflect the personal attitudes of the guild members. The account books of the guild, however, could inform us about these, as they register a part of the receipts of the offerings in the yearly mass on St Andrew’s day, the patron saint of the guild. The low values in the 1580s probably reflect the religious confusion after the defeat of the Calvinist Republic (1578–1585) by the Spanish troops. The high receipts between 1590 and 1629 must be related to the heyday of the Counter Reformation.102 The amount declined slowly in the second half of the seventeenth century, only to reach a climax in the 1690s, a decade of dramatically high food prices, starvation and high mortality. During the eighteenth century, the values remained low, although a slight rise can be registered between the 1750s and the 1770s. Whether these figures point to a general weakening of the religious fervour among the transport workers is not clear, but the religious activity within the framework of the guild unquestionably declined. Admittedly, these evolutions were probably related to broader cultural transformations which may have reduced the power of religion and ritual in the eighteenth century. Yet, these cultural developments may have eroded the already fragile social cohesion among the transport workers. The cultural practices

1760–69

1750–59

1740–49

1730–39

1720–29

1710–19

1700–09

Figure 2.3. Receipts of the offerings during the guild’s mass on St. Andrew’s day (in guilders) 103

0

5

10

15

20

25

30

Deceulaer 43

1790–94

1780–89

1770–79

44

Changes in wage formation

which had traditionally strengthened the distinction and the legitimacy of the privileged naties were slowly crumbling.

Institutions, efficiency and change The (in)efficiency of institutions is a strongly debated topic in institutional economics.104 The question arises whether the changing role of the Antwerp naties during the early-modern period can be linked to changes in efficiency. Although complaints among shippers and entrepreneurs were often heard, the growing segmentation and a strengthening of the guild organisation in the late sixteenth century need not to be seen as a sign of a fossilisation of the Southern Netherlands’ economy. The naties need not necessarily be depicted as inefficient, rigid price makers who dictated the market. First, there is evidence that the naties invested in an improvement of transport facilities. In 1585, the cranemen claimed not to have any wagons, but in 1634 they had several ‘sledges’ and one wagon. Ten years later, they had bought several new small wagons.105 In 1648 the wood porters and the workers of the ‘Bierhoofd’ changed their old sledges for new small carts, which increased their transport capacity.106 In 1773, the workers of the Platte Steger and the Berderenwharf together even financed the construction of a crane, which cost 1,175 guilders.107 Second, the organisation of port labour in Antwerp was not that different from other towns, even in ‘modern’ economies as in England or the Dutch Republic. The guild(-like) organisations of transport workers reduced economic uncertainty for merchants, by offering an available, reliable and experienced workforce. Their monopoly was upheld by formal and informal sanctions, but small breaches of the rules were often informally settled by the payment of drinks and a small sum, well below the official fines in the ordinances.108 Furthermore, the fact that they could compensate damages, if necessary, may have been important. Some of the goods were of high value, especially in the eighteenth century when more colonial goods were unloaded on the Antwerp quays. In situations of asymmetrical information and uncertainty, institutions and organisations such as guilds could reduce transaction costs and offer solutions in the forms of quality control. Theoretically, merchant control of workers engaged on the spot could fulfil the same function. This could reduce the high collective wage to a lower individual wage. However, control costs were probably higher in the long run,109 at least when transport was irregular, and goods costly. Merchants could only think of exerting control themselves, when their operations reached a certain scale. This may have been the case in sixteenth-century Antwerp, in seventeenth-century Amsterdam, or in late-eighteenth-century London, but not in most other European ports. In this sense, the formal and informal enforcement mechanisms of the transport workers and the ‘embeddedness’ of their behaviour in a strong group culture could generate trust or discourage malfeasance.110 The transport workers emphasised their personal qualifications

Deceulaer

45

and their local networking.111 To be accepted in a natie, a candidate had to have a good reputation. The workers of the Wharf stated in 1785, ‘the workers of the natie should not only be capable of doing the heavy work, they should also be of very good behaviour and have a good reputation, as the merchants have to entrust their goods, their houses and their warehouses to them’. The peat porters said in the early 1780s: ‘It is not good to employ ordinary people whom one meets on the street to carry the peat or the coal, as they have to be carried into the houses of citizens, therefore it is absolutely necessary that this is done by people who are known as honest and trustful’.112 Third, the growing segmentation of the workforce did not only create new organisations, it also equipped the Antwerp port with an institutional framework for negotiations and recontracting. The naties and the transport workers guild often participated in ‘conferences’ with merchants and the city government, to adapt certain rules and tariffs to new circumstances. In 1620 the grain merchants stated that they had always been consulted when changes in pay were planned. In 1645 the secretary of the city organised a series of negotiations with different naties to discuss the allocation of the unloading of fruit. In 1701 the new wages of the workers of the Vlasmarkt were only registered after the consultation of several merchants. A close examination of some tariffs indeed suggests a rather flexible organisation of work and wages, in which merchants were not the helpless victims of greedy guilds. Demands for a rise in pay were often turned down, e.g., in 1644, 1652, 1666 and 1676.113 Some tariffs were even lowered in the seventeenth and eighteenth centuries.114 The framework of negotiations between organisations of transport workers, merchants and the city government provided a context which made possible new bargains and compromises between the players.115 The influence of merchants and of the city government probably checked the wage demands of the natie members. The aggressive cries and campaigns to reform the naties may give the impression that these institutions were hopelessly inefficient. However, although merchants certainly tried to contain the wages of the natie members, they mostly did not favour the abolition of their organisations. In fact, in the early nineteenth century, the Antwerp naties were even re-established after the abolition of all guilds. In most European ports, similar guild-like organisations continued to play a role until the middle of the nineteenth century,116 which seem to suggest that they provided useful services.

Conclusion: economics, culture and institutions The early-modern wages of Antwerp port labourers were not static. As the heterogeneity of their wage was so important, both workers and merchants could influence small extras, drinking money and the calculation of wages. Therefore, the real wages of these workers cannot simply be computed on the basis of the prices of foodstuffs. If these observations could also be

46

Changes in wage formation

applied to other groups of early-modern workers or artisans a few hypotheses about early-modern society may need qualification, e.g. the customary character or the “stickiness” of early-modern wages, the rising real wages in the second half of the seventeenth century and the predominance of the food riot in social protest. Both in the sixteenth and in the eighteenth centuries, the institutional setting of the port was not destroyed, but adapted. Early-modern institutions and organisations were probably less rigid and more open to change than is usually thought. As is becoming clear, incremental institutional changes have been very important in the early-modern period. Rules were reinterpreted in new contexts, economic and social practices changed and norms and values were transformed. In many studies, guilds are now seen as flexible organisations, which were part and parcel of the European market economy.117 How do we explain the institutional change in the Antwerp port? Obviously, transport volumes are crucial elements. The abolition of the naties in 1544 and their re-emergence after 1585 must probably be explained by the radical changes in transport volumes. Similar evolutions were registered in other ports, which expanded vigorously. In seventeenth-century Amsterdam, the Old Regime grain porters formed an unlimited group, and new members were simply appointed by the magistrate after the payment of a relatively low sum. Merchants were free to hire any worker they liked. The importance of the Amsterdam grain and staple market, and the strong influence of merchant capital in the Dutch Republic, probably explain this situation.118 In late eighteenth-century London as well, the porters’ monopoly was questioned.119 After 1585, the organisation of work and wages in the Antwerp port again resembled the organisations and institutions of smaller ports. Privileged labourers provided the port with a permanent and reliable workforce who could compensate for damages. A group of privileged workers was able to become small subcontracting entrepreneurs. These entrepreneurs, their guild-like organisations, the merchants and the town government participated regularly in negotiations and recontracting. In the late sixteenth and early seventeenth centuries, cultural practices reinforced and perpetuated existing relations and positions. Ways of drinking, praying and marching strengthened internal and external hierarchies and relations. It is no surprise that struggles about privileges and positions were not only fought politically or legally, but also culturally. Transport volumes were not the only important variable leading to change. As the Antwerp port did not grow in the second half of the eighteenth century, other elements may explain the slow erosion of payments for privileged work, the attack on the drinking money and the lowering of some tariffs. Changes in the competitive environment may have played a role. The increase of road transport (and the decline of road transport costs),120 the growth of the port of Ostend, and the attempts of the central government to promote transit after 1760 certainly were stimuli to implement lower wages and transaction costs. Furthermore, the definitions of work and wages themselves were open to change. The social, political,

Deceulaer

47

legal and cultural battles about the definition and the composition of the wage clearly influenced transaction costs in the early-modern Antwerp port and the livelihood of many of its workers. Perceptions, ideology and legitimacy clearly mattered. The growing role of subcontracting and more general evolutions probably undermined the group culture and the collective cultural practices and representations of the transport workers. Collective ritual and religious ceremony declined, which weakened the traditional legitimacy of the privileged workers. Obviously, the changes in cultural practices of Antwerp transport workers were not simply rooted in group culture. They can of course be related to more general cultural transformations (the triumph of the Counter-Reformation in the early seventeenth century or a certain secularisation in the eighteenth century). Nevertheless, these transformations on a macro-level provided different opportunities for appropriation by specific groups on a micro-level. The question whether economic change caused cultural change or vice versa, is perhaps not the right one. This question is essentially based on a closed model with easily tested and quantifiable variables. A blurring of the boundaries between ‘economic’, ‘social’ and ‘cultural’ history is probably a more interesting alternative. These limitations were constructed as analytical devices in order to improve our understanding of past societies. When these boundaries block or limit our understanding of the lives and the world of early-modern people, we must question their usefulness. Concepts such as interdependence and ‘circular causation’ may be more attractive in a time where some of the more adventurous economists and historians defend interdisciplinary ventures and a cross-fertilisation of multiple traditions.121

Notes 1. All archival material is to be found in the City Archives of Antwerp. Abbreviations of the records: GA: ‘Gilden en Ambachten’, P: ‘Processen’, PS: ‘Processen supplement’, PK: ‘Privilegekamer’. 2. GA 4137, 10 and 17 February 1617. 3. Steven L. Kaplan, ‘Social Classification and Representation in the Corporate World of Eighteenth-Century France: Turgot’s ‘Carnival’ ‘, in Steven L. Kaplan and Cynthia Koep (eds), Work in France. Representations, Meanings, Organisations and Practice (Ithaca – London, 1986), pp. 176–228; J. Farr, Artisans in Europe 1300–1914 (Cambridge 2000), pp. 4–6, 22. 4. See among others Herman Van der Wee, ‘Prices and Wages as Development Variables: A Comparison between England and the Southern Netherlands, 1400–1700’, in Herman Van der Wee (ed.), The Low Countries in the Early Modern World (Aldershot, 1993), pp. 223 – 41. The literature on early-modern real wages is discussed critically by Jan de Vries, ‘Between Purchasing Power and the World of Goods: Understanding the Household Economy in Early Modern Europe’ in John Brewer and Roy Porter (eds), Consumption and the World of Goods (London – New York, 1992), pp. 89–98. 5. Some historians argued that the decline of food prices after 1650 led to a growth of real wages, which influenced demand for industrial goods, e.g. for fashionable clothing. Cf. Jan de Vries, The Economy of Europe in an Age of Crisis (Cambridge, 1976), p. 84;

48

6.

7. 8.

9.

10.

11.

12.

13. 14.

15. 16.

17. 18.

19.

20.

Changes in wage formation H. Van der Wee, ‘Industrial Dynamics and the Process of De-Urbanisation in the Low Countries From the Late Middle Ages to the Eighteenth Century. A Synthesis’, in Herman Van der Wee (ed.), The Rise and Decline of Urban Industries in Italy and the Low Countries (Leuven, 1988), p. 364. Edward P. Thompson, ‘The Moral Economy of the English Crowd in the Eighteenth Century’, Past and Present, 50 (1971), pp. 76–136; John Walter, ‘Grain Riots and Popular Attitudes to the Law: Maldon and the Crisis of 1629’, in John Brewer and John Styles (eds.), An Ungovernable People. The English and their Law in the Seventeenth and Eighteenth Centuries (London, 1980), pp. 47–84. A critical review of the concept is presented by John Bohstedt, ‘The Moral Economy and the Discipline of Historical Context’, Journal of Social History (1992), pp. 265–84. As called for by Knut Sogner, ‘Recent Trends in Business History’, Scandinavian Economic History Review, 45 (1997), pp. 62–4. See amongst others, Brewer and Porter, Consumption; M. Berg and H. Clifford (eds), Consumers and Luxury. Consumer Culture in Europe 1650-1850 (Manchester – New York, 1999). Recent examples are given by Peter Temin ‘Is it kosher to talk about culture?’, Journal of Economic History 57 (1997) pp. 267–84, and David S. Landes, The Wealth and Poverty of Nations: Why Some Are So Rich and Some So Poor (New York, 1998). Critiques of the latter have been abundant, see amongst others Joel Mokyr, ‘Eurocentricity Triumphant’, American Historical Review, 104 (1999), pp. 1241–6 and in the same volume Charles Tilly, ‘A Grand Tour of Exotic Landes’, pp. 1253–7. See for a critique concerning rural history Philip T. Hoffman, Growth in a traditional Society. The French Countryside 1450-1815 (Princeton, 1996), pp. 12–20, and about guildsmen Farr, Artisans. Roger Chartier, ‘Culture as Appropriation: Popular Cultural Uses in Early Modern France’, in Steven L. Kaplan (ed), Understanding Popular Culture. Europe from the Middle Ages to the Nineteenth Century (Berlin – New York – Amsterdam, 1984), pp. 229–53. The quote is from Lauren Benton, ‘From the World-Systems Perspective to Institutional World History: Culture and Economic Theory in Global Theory’, Journal of World History, 7 (1996), p. 288. Douglas.C. North, Institutions, Institutional Change and Economic Performance (Cambridge, 1990), pp. 4–5, 95. Bo Gustafsson, ‘Some Theoretical Problems of Institutional Economic History’, Scandinavian Economic History Review, 46 (1998), p. 28; Douglas C. North, ‘Comment on Bo Gustafsson’, Scandinavian Economic History Review, 46 (1998), p. 65. North, Institutions, p. 140. Pierre Bourdieu, La Distinction. Critique sociale du jugement (Paris, 1979), p. 557; Raymond Murphy, Social Closure. The Theory of Monopolisation and Exclusion (Oxford, 1988), pp. 163, 182. Bourdieu, La distinction, p. 114. See among others, Robert Darnton, The Literary Underground of the Old Regime (Cambridge, Mass. – London, 1982), p. viii and passim; Roy Porter, The Enlightenment (London, 1990), pp. 42–50; about the Counter-Reformation in Antwerp, Alfons K.L. Thijs, Van Geuzenstad tot katholiek bolwerk. De Contrareformatie in Antwerpen (Turnhout, 1990). GA 4150, 1603 (rules of the butter porters accepted after vote; PK 922 1668 ordinance of the workers of the Bierhoofd and the wharf prescribes that differences have to be settled with majority votes, GA 4160, 1766. Harald Deceulaer, ‘Guilds and Litigation: Conflict Settlement in Antwerp’, in Marc Boone and Maarten Prak (eds), Individual, Corporate and Judicial Status in European Cities (late Middle Ages and Early Modern Period, Proceedings of the Colloquium Ghent, October 12th-14th (Leuven-Apeldoorn, 1995), p. 187.

Deceulaer

49

21. The usage of the capital of a sold nation place is not altogether clear. Impressionistic evidence suggests that some ex-porters became shop or innkeepers, while others bought a small house; GA 4170, 11 may 1754, PS 395, 1730, P A 1184, 1732. 22. Herman Van der Wee, ‘The Low Countries in Transition: From the Middle Ages to Early Modern Times’, in Van der Wee (ed.), The Low Countries, pp. 17–21. 23. Gustaaf Asaert, ‘From Wharf to Commercial Metropolis’, in Fernand Suykens and Gustaaf Asaert (eds), Antwerp; A Port for All Seasons (Antwerp, 1986), pp. 142–5. 24. Peter Stabel, ‘Schippers, wagenvoerders en kruiers. De organisatie van de stedelijke vervoerssector in het laatmiddeleeuwse Vlaanderen’, Bijdragen tot de Geschiedenis, 82 (1999), p. 168. In many southern German towns, Charles V also weakened the constitutional position of guilds, Pierre Jeanin, ‘Holders of Power and Economic Activity in German Merchant Towns in the Sixteenth and Seventeenth Centuries’, in Herman Diederiks, Paul Hohenberg and M. Wagenaar (eds), The Visible Hand and the Fortunes of Cities. Economic Policy in Europe since the Late Middle Ages (Leicester–London–New York, 1992), p. 41. 25. Milja Van Tielhof, ‘Stedelijke regulering van diensten op de stapelmarkt: de Amsterdamse korengilden’, in Leo Noordegraaf (ed.), Ondernemerschap en bestuurders. Economie en politiek in de Noordelijke Nederlanden in de late Middeleeuwen en de vroegmoderne tijd (Amsterdam, 1999), pp. 491–524. 26. Alfons Thijs, ‘The River Scheldt closed for two centuries, 1585–1790’, in Antwerp. A Port For all Seasons, pp. 203–204. 27. Paul M. M. Klep, ‘Het Brabantse stedensysteem en de scheiding der Nederlanden’, Bijdragen tot de Geschiedenis, 73 (1990), pp. 117–9. 28. GA, 4137, 24 December 1592, GA 4160, 16 September 1611, 12 April 1613, GA, 4137, 5 February 1625, GA 4130 19 July 1626, 27 October 1626, 8 May 1627, 10 February 1694, GA 4130, 19 January 1635. 29. In 1544, Charles V had given the former the right to transport their products by their domestic servants. At the end of the sixteenth century, the city government simply overruled this imperial ordinance, and forced the soap boilers to employ the butter carriers. In 1600, the Archdukes Albrecht and Isabella restored the soap boilers in their rights to organise their own transport. Yet, conflicts continued, and in 1611 the city government again proclaimed that they had to employ the butter porters. In 1639, 1640 and 1647 the two groups struggled about the organisation of the labour market once more, and in 1659, the butter porters were finally confirmed in their privilege. This relation is certainly maintained until 1738, because in an inquiry of that year, the soap boilers stated that they employed the butter porters on a regular basis, cf. E. Dilis, De Antwerpse naties. Geschiedkundig overzicht van het ontstaan, de ontwikkeling en de inwendige inrichting dezer arbeidersvereenigingen vanaf de XIVde eeuw tot op onze dagen (Antwerp, 1906), pp. 34–5; GA 4138, 4 September 1599; GA 4505, 1600 GA 4129, 29 March 1611, cited in the trial of 1640; GA 4138, 1639, GA 4505, 1640 ; P A 502, 1647 ; Ordinance cited in P S1176, 1736; Frans Smekens, ‘Schets van aard en betekenis der Antwerpsche nijverheid onder het Oostenrijks bewind’, Verzamelde Geschriften Frans Smekens, s.d., s.pl., p. 10. 30. PS 4800, 1642. 31. P A1150, 1662. 32. P A139, 1587. Four years later, a similar incident proves that the privilege of the butter porters was not undisputed in this period , P A823, 1591. 33. GA 4137, 10 October 1596. 34. GA 4160, 11 February 1608 ; GA 4162 bis, 11 February 1608. 35. PK 694, 40, 23 January 1608 ; PK 717, 210, 27 July 1622. 36. GA 4163, 31 December 1635. 37. Stabel, ‘Schippers, wagenvoerders en kruiers’, pp. 169, 171. 38. PS 7720, 1630. 39. GA 4150, 1603, 3 March 1646. 40. P R71, 1693. 41. PS 5872, 1610 and PS 3954, 1610.

50

Changes in wage formation

42. P A477, 1769. 43. Harald Deceulaer, De Antwerpse naties, 1585–1796 unpublished Masters thesis, Free University of Brussels, 1991, and by the same author, ‘Arbeidsregulering en loonvorming in de Antwerpse haven, 1585–1796’, Tijdschrift voor Sociale Geschiedenis, 18 (1992), pp. 22–47. 44. In 1642, some shippers complained that the workers of the Bijltjesbrug always did the work according to their own will (PS 4800, 1642). In 1760, the grain porters stated proudly that ‘the merchant should be content to be served in the way the grain porters have always served them’ (PS 3504, 1760). 45. GA, 4130, 9 December 1709; GA 4174, 15 September 1721; P S494, 1713. 46. GA 4081, 7 September 1641, 30 December 1687. 47. In 1618, the grain porters stated that they were sometimes paid in grain (PK 710, 29 March 1618). Coal carriers were allowed to buy the breeze in the ships, GA 4132, 8 April 1675. Compare to Walter M. Stern, The Porters of London (London, 1960), p. 64. 48. GA 4163, 17 December 1638, GA 4174, 28 November 1688; GA 4505, 1714. 49. E.g., in , PS 3117, 1611; PS 3957, 1652. 50. GA, 4162 bis, 31 December 1625, 1645. 51. GA 4172, undated document, probably 1578. 52. PS 3117, 1611. Similar complaints in 1755: GA 4163 bis, 24 September 1755. 53. E.g., GA 4150, 14 May 1671; P A484, 1724. 54. Alfons Thijs, Van werkwinkel tot fabriek. De textielnijverheid te Antwerpen (eind 15de – begin 19de eeuw) (Brussels, 1987), p. 37; John Rule, The Experience of Labour in Eighteenth-Century Industry (London, 1981), pp. 152–3. 55. A similar point is made by Reinhold Reith, ‘Wage conflicts and wage formation in early modern German guilds’, in Clara E. Nuñez (ed.), Guilds, Economy and Society (Seville, 1998), pp. 87–8. 56. GA 4503, 1591; 4160, 1594; 4163 bis, 10 May 1614; 4135, 25 October 1616, 1 October 1618. 57. Peat porters rented wagons, GA 4162bis, 19 July 1638. 58. PS 3954, 1610; GA 4161, 1711. 59. In the second half of the seventeenth century, this could amount to 8, 9, 10, or even 27 guilders, Deceulaer, ‘De Antwerpse naties’, p. 84. 60. GA 4171, 26 June 1759. 61. P A875, 1759. 62. GA 4162, 1723–1724. This can probably be explained by the nature of the goods they mainly transported. 63. In 1840, ninety percent of the ships in the Antwerp port were sailing ships. In 1865 this was reduced to sixty-five percent, Greta Devos, ‘De Antwerpse naties tijdens de periode 1815–1940’, in Gustaaf Asaert (ed.), De Antwerpse naties. Zes eeuwen actief in stad en haven (Tielt, 1994), pp. 148-9, 165. 64. Michael Sonenscher, Work and Wages. Natural Law, Politics and the Eighteenth-Century French Trades (Cambridge, 1989), p. 203. See also Denis Morsa, ‘Salaires et salariat dans les économies préindustrielles (XVIe – XVIIIe siècles)’, Revue belge de Philologie et d’Histoire, 65 (1987), p. 764. 65. PS 3117, 1611, P A875, 1759. 66. I used early-modern wages for Ghent, as Antwerp wages are not available for the seventeenth and eighteenth centuries, Johan Dambruyne, ‘De Gentse bouwvakambachten in sociaal-economisch perspectief (1540–1795)’, in Catharina Lis and Hugo Soly (eds), Werken volgens de regels (Brussels, 1994), p. 65. 67. Karel Degryse, ‘Sociale ongelijkwaardigheid te Antwerpen in 1747’, Bijdragen tot de Geschiedenis, 58 (1974), pp. 126–8. 68. PS 3117, 8723, 1651. 69. P A1107, 1692. 70. Harald Deceulaer, ‘De Antwerpse naties’, pp. 55–6.

Deceulaer

51

71. Harald Deceulaer, ‘Guildsmen, Entrepreneurs and Market segments. The Case of the Garment Trades in Antwerp and Ghent, sixteenth-eighteenth centuries’, International Review of Social History, 43 (1998), pp. 1–25. 72. This is clear from a comparison of the inventories of the altar, GA 4133 1617 and 1688. See for an account of artisan devotion Frederik Verleysen, ‘ ‘Prentense Confrerieën?’ Devotie als communicatie in de Antwerpse corporatieve wereld na 1585’, Tijdschrift voor Sociale Geschiedenis, 27 (2001), pp. 153–74. 73. GA 4133, 1640, 14 March 1750; GA 4168, 23 February 1721, GA 4163, 1749, Dillis, De Antwerpse naties, p. 45. 74. Johan Dambruyne, ‘Rijkdom, materiële cultuur en sociaal aanzien. De bezitspatronen en investeringsstrategieën van de Gentse ambachten omstreeks 1540’, in Catharina Lis and Hugo Soly (eds), Werelden van verschil. Ambachtsgilden in de Lage Landen (Brussels, 1996), p. 159. 75. GA 4132, 16 May 1652; GA 4130, 26 January 1653. 76. GA 4150, 19 May 1650. 77. GA 4130, 6 December 1673. 78. GA 4161, 1711. In the probate inventory of butter carrier Tielman Brons, a barrel with about 360 pounds of meat was found in the cellar, N 2552, 3 February 1639. 79. GA 4505, 30 August 1707; GA 4503, 28 September 1718; GA 4172, 22 March 1618 (agreement between grain merchants and corn measurers). 80. PS 5630, 1624. 81. N 2256, 21 May 1622; N 3389, 28 April 1635; N 3390, 26 November 1636. 82. PS 3117, 1611. 83. N 3389, 28 April 1636; N 2550, 31 December 1638; N 2551, 3 February 1639. 84. Harald Deceulaer, ‘Entrepreneurs in the Guilds: Ready-Made Clothing in sixteenth and seventeenth century Antwerp’, Textile History, 31 (2000), pp. 133–49. 85. Thijs, ‘The river Scheldt’, pp. 191–2. 86. Deceulaer, ‘Arbeidsregulering en loonvorming’, pp. 39–40. 87. GA 4161, 1794. 88. Karel Van Isacker, De Antwerpse dokwerker 1830–1940 (Antwerpen,1966), p. 20. 89. GA 4162, 1723; P A484, 1724; P A1184, 1732. 90. GA 4160,8juli 1686 ; GA 4505, 31 January 1725. 91. GA 4163 bis, 1740. 92. GA 4160, 1754. 93. Floris Prims, Het herfsttij van het corporatisme te Antwerpen (Brussels, 1945), p. 34. 94. GA 4137, 1772. 95. GA 4137, 14 December 1771, 1777. 96. PK 801, 13 December 1713, 10 January 1714; PK 806, 228; 8 November 1718; PK 812, 27 September 1723; PK 814; 5 October 1723; PK 816; 21 March 1727; PK 831, 7 September 1741, 9 September 1741; PK 876, 8 August 1783; PK 878, 13 August 1784. 97. Deceulaer, ‘Arbeidsregulering en loonvorming’, p. 43. 98. PK 926, 300 v. 16 February 1756; PK 869, 123, 27 August 1778; PK 880, 36, 27 March 1786. 99. Marc Jacobs, ‘Oude structuren en verse vis. Representaties van corporaties te Nieuwpoort in de tweede helft van de achttiende eeuw’, in Lis and Soly, Werken volgens de regels, p. 287; Piet Lenders, ‘De eerste kamer van koophandel te Gent (1729–1795) en de opkomst van de ondernemende burgerij’, in Jean-Marie Cauchies (ed.) Het einde van het Ancien Régime in België (Kortrijk-Heule, 1991), pp. 218–20. 100. GA 4139, 1777. 101. In 1683–1685, ninety-five absentees were registered for four processions, eighty of them were non-working subcontractors. Comparison of GA 4092, 1683–1685 and GA 4094, 1683–1685. 102. Thijs, Van Geuzenstad tot katholiek bolwerk, p. 40. 103. GA 4140–4143.

52

Changes in wage formation

104. M. Granovetter, ‘Economic action and social structure: the problem of embeddedness’ American Journal of Sociology, 91 (1985), pp. 494–506; North, Institutions, Institutional Change, p. 6 and passim; Gustaffson, ‘Some Theoretical Problems’, pp. 18–21. 105. GA 4503, 1591; GA 4505, 16 March 1634. 106. PS 5720, 1648. 107. GA 4164, 1773. 108. GA 4160, 1609. 109. Bo Gustafsson, ‘The rise and economic behaviour of medieval craft guilds’, in Bo Gustafsson (ed.), Power and Economic Institutions. Reinterpretations in Economic History (Aldershot, 1991) p. 94. 110. Granovetter, ‘Economic action and social structure’, p. 490. 111. Gustafsson, ‘The rise and economic behaviour’, pp. 89–92. This idea is developed into different types of constructing quality by Harald Deceulaer and Marc Jacobs, ‘Qualities and Conventions. Guilds in Eighteenth-Century Brabant and Flanders: An Extended Economic Perspective’, in Nuñez (ed.) Guilds, Economy and Society, pp. 91–107. 112. PK 875, f° 30, 879, 23 July 1785. 113. GA 4162 ter, 1644, 1652, 1666, 19 September 1676. 114. The price for the transport of 100 pounds of butter was fixed at two stuyvers in 1632, at 1,5 stuyver in 1657 and at 1 stuyver in 1672, GA 4150, 20 November 1632; GA 4174, 13 November 1657, 3 December 1672. The wages of the coal workers were lowered in 1666 and 1723, with the argument that the transport of coal had become much more regular, GA 4162, 1723; GA 4163, 4 February 1723. 115. Compare to North, Institutions, Institutional Change, pp. 89–90. 116. See, e.g., about Marseille, Roger Cornu, ‘Les portefaix et la transformation du port de Marseille’, Annales du Midi. Revue de la France méridionale, 86 (1974), p. 184; W.H. Sewel, Structure and Mobility. The Men and Women of Marseille 1820-1875 (Cambridge, 1985), p. 38. 117. E.g., Daryl M. Hafter, ‘Les jurandes textiles au XVIIIe siècle : institutions dépassées ou nouvelles créatures du négoce moderne’, in Jean Bottin and Nicole Pellegrin (eds), Echanges et cultures textiles dans l’Europe pré-industrielle, Actes du Colloque de Rouen, 17–19 mai 1993, Revue du Nord, Hors Série, 12 (1996), pp. 157–70; Lis and Soly, Werelden van verschil; Geofrey Crossick (ed.), The Artisan and the European Town, 1600-1900 (Aldershot, 1997); Nuñez, Guilds, Economy and Society; Farr, Artisans. 118. Van Tielhof, ‘Stedelijke regulering’, pp. 503–15. See for a similar situation among the Amsterdam peat porters, Sandra Bos, ‘Uyt liefde tot malcander’. Onderlinge hulpverlening binnen de Noord-Nederlandse gilden in internationaal perspectief (1570–1820) (Amsterdam, 1998), pp. 111–2, 124, 128–9. 119. Stern, The Porters of London, pp. 119–68 120. Bruno Blondé and Raymond Van Uytven, ‘Langs land- en waterwegen in de Zuidelijke Nederlanden. Lopend onderzoek naar het pre-industriële transport’, Bijdragen tot de Geschiedenis, 82 (1999), pp. 141, 150. 121. Les Annales, ‘Tentons l’experience’, Annales ESC, 44 (1989), p. 1320; Robert Boyer, ‘Economie et Histoire. Vers des nouvelles alliances?’, Annales ESC, 44 (1989), pp. 1417, 1419; Eyup Ozveren, ‘An institutional Alternative to Neoclassical Economics?’, Review, 21 (1998), pp. 497–8, 524. The notion of circular causation is by Gunnar Myrdal.

CHAPTER 3

WHEN LABOUR HIRES CAPITAL: EVIDENCE FROM LANCASHIRE, 1870–1914 Michael Huberman

According to 1999 estimates, employees in the United States controlled approximately 8.3 percent of the total corporate equity in the country.1 Comparable figures for Europe are only slightly higher. Wage systems such as employee ownership, the argument goes, harmonise the often discordant interests of the firm’s stakeholders: managers, shareholders and workers.2 Where workers have the financial incentives of stakeholders, they will act to increase shareholder value, engaging in productivity enhancing cooperation as opposed to productivity-reducing conflict. Despite these apparent benefits, and although some measures of employee ownership are increasing in the United States as elsewhere, the number of workers participating as stakeholders remains modest in most developed countries. Dow and Putterman have summarised the reasons why labour seldom hires capital.3 Workers may be limited in the quantity of credit they can obtain, or they may be risk averse and prefer to diversify their portfolios. Compared to the owners of capital, workers have heterogeneous preferences and they may not wish to hold shares over long-term horizons. Alongside these financial reasons, economists and historians have given weight to the free-rider problem in large firms (the 1/N problem) as a constraint on worker ownership.4 To fix ideas, consider the case of a large enterprise whose workers are organised in teams. Management decides to introduce a profit-sharing scheme. An additional dollar of output, when divided among the many workers in the organisation contributes only a trivial amount to any individual worker’s compensation. Moreover, the output of any single work team contributes only a trivial amount to the compensation of members of that unit. Thus, there is no clear reason why one team should not simply free ride on all other teams in the organisation. Given the size of the organisation, it is simply too costly, relative to

54

When labour hires capital

the size of potential gains, for a single worker or team to monitor the performance of others. This line of argument is found in the classic statement of Alchian and Demsetz who asserted that, while workers cannot be relied on to monitor each other, the owners of capital have no incentive to shirk because they keep the difference between revenues and costs.5 It follows, as Alchian and Demsetz observed,that labour seldom hires capital. In this chapter, drawing on evidence of profit sharing in the Lancashire textile industry, I give an example of a successful experiment of labour hiring capital. I contrast the experience of organisations in two cottonspinning towns, Oldham and Bolton. In Oldham after 1870 workers participated in profit-sharing schemes and acted as stakeholders in the direction of their enterprises, limited liability companies popularly referred to as ‘limiteds’. As elsewhere, these types of incentive schemes promoted high levels of worker cooperation and capital investment, and fast growth in various measures of productivity. Organisations in Bolton did not take to profit sharing and the town grew relatively slowly – ‘like a tree rather than a mushroom’.6 The problem is why these two towns, only fifteen miles apart, developed different wage systems and why if worker ownership was so good – and the idea widely known – was it not adopted in Bolton. My answer is that profit sharing was successful in Oldham because it was compatible with the emotions workers and residents in the town felt and expressed with regard to one another, compared to their distant and bitter relations with their bosses.7 This predisposition affected how freerider and moral-hazard problems were resolved. Workers in the town had developed a long history of trustworthiness – they would not let each other down – and this empathy overrode the free-rider problem.8 But if Oldham embodied the altruism of The Theory of Moral Sentiments, then Bolton personified the unabashed individualism of The Wealth of Nations. Spinners in the town were indifferent to the well being of their peers and their bosses. Trustful behaviour in Bolton was contingent. Ever vigilant, workers in the town monitored each other, firms workers and workers firms, and all parties used the threat of sanctions to uphold contracts. I argue that there were no gains to introduce profit sharing in Bolton because its imposition would come up against the free-rider problem. By trustworthiness, I do not refer exclusively to the type of cooperation organisations can instil by paying an efficiency wage or by threatening retaliation to eliminate shirking.9 These are short-term arrangements that are susceptible to breakdown when there are even small changes in the information environment.10 Trust in Lancashire extended beyond the factory gate; indeed the transfer may well have been in the other direction. I refer therefore to a ‘culture of trust’, using culture to signify in the broadest sense the social norms or the ‘distinctive attitudes and actions that differentiate groups of people’.11 More formally, we can define a culture of trust as a systematic bias in favour of cooperative resolutions to multiple prisoners’ dilemmas.12 Embedded in history, trustworthiness creates and is shaped by institutions and practices at the firm and community levels.

Huberman

55

Lancashire wage systems are a case in point. Bolton firms that implemented profit-sharing schemes quickly scrapped them because they did not have the support of community organisations or institutions. But a Bolton worker or firm moving to Oldham would have adapted to the latter community’s practices. I associate the culture of trust in Oldham with its unique industrial relations history. In this regard, I follow the lead of Foster.13 While labour unrest was almost unknown in Bolton around the mid-nineteenth century, Oldham was celebrated for its militancy. During strikes Oldham workers learned to trust one another and not to free ride, a predisposition that was exploited in limited liability companies. But it is also evident that Oldham’s unique wage system had deep cultural roots. Foster himself observed that Oldham workers before 1850 were more likely to marry within their ‘class’. Oldham empathy was thus supported by community networks that had a long history. It is not my intention to identify the origins of Oldham culture. Rather, my objective is to shed light on the feedback mechanism between Oldham’s industrial relations and community institutions that gave rise to and advanced the town’s peculiar wage system. The argument in this chapter is based on elimination. First, I compare the expansion of Oldham and Bolton and evaluate the conventional explanations of the towns’ divergent development paths. Human and social capital factors, technological considerations and risk preferences cannot explain the gap between Oldham and Bolton. I propose that Oldham’s success was associated with its adoption of a new wage system, profit sharing. I then show that profit sharing was a form of labour ownership. Workers were well suited to profit sharing in Oldham because of their empathy for one another. This empathy existed inside the mill in the form of negligible investments in monitoring and outside the mill in generous poor relief expenditure. In the final section of the chapter I speculate on the reasons why Oldham’s experiment with labour ownership did not last.

Economic expansion in Bolton and Oldham: the received view Tables 3.1 and 3.2 summarise several key aspects of Oldham’s and Bolton’s growth before the First World War. Up until 1850 or so, Bolton’s population was larger and it had more spindles (the most common measure of capacity) and spinners than Oldham. Although Oldham had many more mills than Bolton before 1850, they were generally much smaller (Table 3.1). A different picture emerges by the turn of the century. Oldham’s population surpassed that of Bolton’s around 1870 (Table 3.2). The number of spindles and spinners in Oldham doubled between 1870 and 1890; there were twice as many new companies started in the town and many of these new enterprises were large. All told, by 1914 Oldham housed twenty-nine percent of spindles in the United Kingdom and represented the largest concentration of spinning mills in the world.14

56

When labour hires capital

Table 3.1. The cotton-textile industry in Bolton and Oldham, 1850–1914

Average count spun, 1900 Number of mills 1850 1870 1890 1910 Companies promoted 1858–96 Number of spinners 1850 1900 Workers per mill 1870 1910 Number of spindles: mules and rings (‘000) 1870 1890 1910 Spindles per mill (‘000) 1900 Average mule size (spindles) Spinners’ weekly earnings 1860 1886 1906 1906

Bolton

Oldham

60s

32s

70 111 108 150

178 279 321 335

76

156

1,500 4,800

1,000 5,600

– 194.45 3,200 5,000 9,500 58.00 1,100 34s 10d 38s 45s 45s

82.43 107.54 6,000 12,010 19,278 38.81 1,224 28s 33s 2d 42s 42s

Sources: Mills, spindles and company promotion: D. Gurr and J. Hunt, The Cotton Mills of Oldham (Oldham, 1998); J. Longworth, The Cotton Mills of Bolton, 1780–1985: A Historical Directory (Bolton, 1987); spinners: A. Fowler and T. Wyke (eds), The Barefoot Aristocrats: A History of the Amalgamated Association of Operative Cotton Spinners (Littleborough, 1987), p. 242; mule size: J. Jewkes and E. M. Gray, Wages and Labour in the Lancashire Cotton Spinning Industry (Manchester, 1935), p. 205; earnings: G. H. Wood, The History of Wages in the Cotton Trade during the Past Hundred Years (Manchester, 1910).

Oldham specialised in coarse yarn and in this sector, according to reports from various contemporaries, productivity (measured in pounds per spindle per week) increased by about fifty percent from 1860 to 1900.15 Bolton was known for its medium and fine yarns and the rate of growth of output per spindle in this sector is believed to be about one-half of productivity growth in coarse yarn.16 Comparisons of output per worker across sectors have proven to be more difficult to make, because of changes in labour inputs over the period, but indirect performance measures can be gleaned from wage evidence. Because payment was by the piece, earnings’ growth approximately mirrored productivity improvements, including that of labour.17 Oldham spinners saw their wages increase faster than elsewhere in Lancashire, by about 50 percent between 1860 and 1906. Despite these wage gains, unit labour costs in coarse spinning fell by about 15 percent

Huberman

57

between 1876 and 1906.18 For a comparable period, unit labour costs fell by only 2 percent per annum in Bolton. Table 3.2. Demographic, labour force and religious make-up of Bolton and Oldham, 1851–1911 Bolton Demographic characteristics Population (municipal borough) 1861 70,395 1911 127,851 Male-female ratio 1881 0.90 Percentage of population born outside Lancashire 1861 11.82 Percentage of population born in Ireland 1861 4.95 Percentage of population > 40 years 1881 21.30 Children under one year (per ‘000 women, 15–44 years) 1881 121.88 Married women, 20–24 years, as percentage of total number of married women, 1881 9.62 Age of cotton spinners 1861 35.6 1891 37.0 Labour force participation rates, 1881 Percentage of male population (greater than five years old) occupied Percentage of female population (greater than five years old) occupied Occupational distribution, 1881 Percentage of occupied workers in cotton mills All workers Male workers Female workers Religious affiliation, 1851 Percentage of population attending church Religious affiliation (in percent) of population attending church (previous line) Established church Dissent Catholic

Oldham

72,333 147,483 0.92 19.38 5.22 22.22 114.69 9.99 36.0 34.2

75.08

77.23

40.50

43.46

41.06 26.50 65.15

50.54 35.32 74.87

29.3

26.4

21.1 3.7 4.8

14.8 7.1 1.5

Sources: British Parliamentary Papers (various years), except for ages of spinners which are from M. Huberman, Escape From the Market: Negotiating Work in Lancashire (Cambridge, 1996) and a sample of approximately 200 spinners in each town from the 1891 manuscripts.

Perhaps the most widely accepted explanation of the rapid expansion of Oldham rests on the relation between technological dynamism and

58

When labour hires capital

market opportunities. Some authorities have claimed that, because of Britain’s control over foreign markets like India, the potential for growth in the coarse yarn market was virtually limitless, while Bolton’s domestic and foreign markets were saturated.19 This argument is suspect. Bolton manufacturers were well aware of market opportunities that had presented themselves in China and India, and with incomes rising throughout Europe after mid-century Bolton ought to have experienced fast growth as well.20 Oldham’s share of the market was not certain – it was won.21 The international coarse-yarn market was highly competitive and the town’s success in capturing a large share of it was based on its ability to drive unit costs down. In both towns, spinners led a work unit that included two or three junior assistants or piecers responsible for removing bobbins of yarn and tying broken threads. Before 1850, Oldham had been using the self-actor (a semi-automatic mule) while Bolton had hand mules; by the 1870s, the self-actor was widely exploited in Bolton as well, although fine spinning still required a greater degree of skill. Lancashire’s engineers made constant improvements to the self-actor’s operations.22 Moreover, exploiting the mule, spinners increased their effort and that of junior workers who were part of the spinning team.23 These options were clearly available in both towns. Such was Oldham’s ability to cut unit costs that, by the end of the century, it began to spin finer yarns and to compete headto-head with Bolton.24 By 1893, about sixty Oldham firms were spinning counts greater than no. 50.25 The upshot is that product markets were integrated and competitive, and, given the same technology and work organisation, we have no reason to suspect that Bolton, because of its specialisation, was in any way disadvantaged in protecting or expanding its market share along the lines of Oldham. Human and social capital models provide alternative explanations of economic growth. The link between older and better-trained workers and higher levels of productivity is well established. But the average age of spinners in 1861 was about the same in Bolton and Oldham, about thirtysix years, and levels of education were also comparable.26 With regard to measures of social capital, economists have found that growth tends to be correlated with less ethnically diverse populations, high marriage rates and large family sizes (as measured by the number of siblings).27 Based on these indicators (Table 3.2), the towns were not very different. In its population makeup Bolton was more Lancastrian; still the proportion of the largest immigrant component, the Irish, was the same in the two towns. Birth rates were higher in Bolton, although women in the town married later.28 The percentage of older residents was about the same. The social capital approach has made much of the correlation between evangelical or dissident religious groups and growth.29 This may have been the case in Lancashire. There were more Dissenters and fewer Catholics and Anglicans in Oldham. But the overall rates of participation, based on the census of religious worship for 1851 from which these numbers were gathered, were not very different between the communities.

Huberman

59

The rise of the limiteds My explanation of the rapid rise of Oldham is its introduction of profitsharing schemes, a wage system that can be traced to the new forms of governance permitted by the Joint Stock Acts of 1855 and 1862.30 The Acts allowed companies to establish, along with ordinary share capital, a loan capital account. On these accounts, which performed a function similar to that of debenture stock or preference capital in the modern public company, firms paid out a scheduled rate of return, or dividend. Using these provisions, a group of skilled workers, who sought to gain through cooperation a greater equality of income and wealth, founded in 1857 the first limited, the Oldham Building and Manufacturing Company31. Under the name of the Sun Mill Company Limited, this company evolved into the largest mill in the district, paying an annual average dividend of 12.3 percent. Subsequently, there were three waves of expansion: 1870-75, 1883–84 and 1889–90. In all, between 1858 and 1896, Oldham formed 156 limited companies (see Table 3.1); by 1900 these companies operated about fifty percent of all spindles in the town and comprised 76 percent of the members of the Oldham Master Cotton Spinners’ Association.32 In what way were these limiteds examples of labour hiring capital? According to Dow and Putterman, in worker controlled and managed enterprises employees have a bundle of rights that include the right to acquire revenues, to determine how the firm is operated, and to transfer ownership to others.33 The Oldham limiteds satisfied these criteria. I first examine the degree of participation. The degree of participation in these concerns has been the subject of some debate.34 Throughout the period, worker participation was voluntary.35 At Sun Mill their investments were negligible, but they did gain experience as managers, purchasers of raw cotton and sellers of yarn. It is believed that worker involvement peaked in 1875 when about 75 percent of Oldham’s working class, or about 20 percent of the town’s population, held shares in the limiteds.36 The numbers of workers acting as stakeholders appear to have declined in the bear market of the early 1890s that saw a long slump in share values. But the Royal Commission on Labour reported that as of 1892 workers owned about 50 percent of share capital and sixty-three percent of the loan money of twelve representative companies.37 Moreover, as workers drew down their own direct investments, friendly societies, cooperative retailers and trade unions picked up much of the slack. As late as 1900, the Oldham Cotton Spinners’ Union invested in twenty-five limiteds.38 Individual workers as well as the union also had investments in cooperative societies that in turn held shares in the limiteds.39 Thus, workers, directly and indirectly, participated actively as stakeholders in the limiteds at least until the decade before the War. In all, worker participation lasted a little more that one generation.40 Benjamin Jones, a correspondent of the Co-operative News, captured the investing habits of this period:41

60

When labour hires capital

Every class of trade is represented, from the little piecer, with his 10s a week, to the banker, doctor and lawyer, including the competitive grocer, draper, butcher, cotton spinner, merchant and the timid old women with her stocking…no trade is too mean or respectable…Their morning and evening prayers are ‘premium, dividend’; and their Sunday holidays are often spent in search, not of a creed, but of a site upon which a model mill may be erected, where water is plentiful, good clay at hand, and a coal and a luggage station near to it.

Oldham’s designation as ‘Diviborough’ was well earned. Workers’ right to revenue was not questioned. They held equity and loans at their own mill and at competing concerns, a feature I will explore later. Loans, which in fact operated like deposit accounts, were favoured by the uninitiated. Companies generally issued shares of £5 denomination, which amounted to roughly three weeks pay for the average cotton spinner (30-35s per week). Returns on both equity and loans varied considerably, from 5 to 20 percent in the 1870s, but always more than could be earned at a local savings bank that paid 2.75 percent. The 5 percent minimum was fixed by custom and did not vary with the condition of the money market. With regard to decision making, workers as stakeholders accepted the basic organisational model of the cotton textile factory, where the line of direction ran from owners/directors to division managers, to foremen who oversaw operations, and down to spinners who supervised a work team of two or three junior workers. The limiteds were not kibbutzim and, in this sense, the limiteds were not a textbook example of a labour-managed firm – but they were not intended to be.42 Nonetheless, as stakeholders and occasionally as members of boards of directors, workers participated in the decision making of the company. Farnie wrote that many of the limiteds were governed on the principle of ‘one man one vote,’ a form of company governance unique in Britain.43 All shareholders, whether they be spinners, merchants or widows, were quite severe with managers who did not perform well. Managers of limiteds were less inclined to work short time than privately run mills and the new concerns were notorious for speeding up their machines and cutting wages at the first instance. Trade union leaders bemoaned that the limiteds had broken with the Lancashire tradition of keeping piece rates fixed and working fewer hours during recessions and, as a result, the number of disputes and strikes at the limiteds were higher than elsewhere.44 I will return to Oldham’s strike record in the next section, but it is clear that workers had double loyalties – as union members they would fight the speed-ups, as stockholders they would encourage them. In fact, at the annual meetings of Oldham’s spinners’ trade union, workers complained about the bad working conditions of the limiteds, only to suggest, when the agenda turned toward the union’s financial picture, that these same companies would provide sound investments for its members.45 Moreover, the union saw fit to hold onto their shares because the threat of selling them gave them added leverage in strikes.46 Workers traded their shares. The Cotton Factory Times, the weekly regional newspaper of spinning operatives that first appeared in 1885,

Huberman

61

reported price movements in the yarn and cloth market. On 15 March 1895, it warned workers not to take on any more risk. If available, it published profit statements of the limiteds and dividend announcements, such as on 29 December 1905 when it celebrated the 19 percent average annual profit rate of Oldham mills. It identified those firms that had bought the lasted vintage of self-acting mules or had switched to ring spinning. Finally, the newspaper summarised shareholders’ meetings and reported the value of shares bought and sold at the local exchange, the King’s Arms Hotel. There is both direct and indirect evidence of the overall performance of the limiteds. Toms has analysed the balance sheets of a small sample of limiteds. In line with estimates by Kruse of the effects of profit sharing, Toms found that their rates of return were only slightly above the average.47 There may be a number of explanations for this. The firms whose records are extant may not be representative, but it is also possible that private companies that survived were ‘whipsawed’ by the limiteds to perform better. The available balance sheets do not permit a direct comparison of firm performance in Bolton and Oldham, but the wage structures of organisations in the towns give support to the view that worker ownership had its expected results. We know that firms with effective incentive schemes like profit sharing tend to have a greater degree of wage compression than firms without similar personnel policies.48 This is exactly what we see in Lancashire. The ratio of spinners’ wages to that of junior workers’ was smaller in Oldham than in Bolton.49 Contemporaries gave support to these findings. Writing in the 1880s, Thomas Ellison, an historian of the industry, saw a clear connection between the rise in efficiency and worker ownership:50 The daily discussions which take place amongst the shareholders as to why dividends are small or otherwise have led almost every intelligent operative to become more economical with materials, more industrious, and to see what effect his individual [my emphasis] efforts have upon the cost of the materials produced. In fact, the bulk of the working class operatives of Oldham have more knowledge of the buying of cotton, working it up, and selling the manufactured goods than most private employers had ten years ago.

Before mid-century the typical Bolton firm was large and privately and family owned.51 Changes in limited liability rules left the town’s landscape untouched, even though workers and employers were aware of the reasons for Oldham’s success. According to the Bolton’s spinners’ union, the ‘attraction of the ‘divi’ or bonus’ incited workers to ‘give the greatest quantity of work in the time allotted, so that they may earn the highest possible wages… The voluntary high-pressure exertion of the operatives is unrivalled elsewhere’.52 But despite its record, Bolton’s union was disinterested, if not strongly opposed to profit sharing and it exercised tight control over its membership. When an employer in Bolton, forced new hires to purchase shares, the union reacted swiftly to reprimand the firm. With the support of private employers, the firm backed down on its

62

When labour hires capital

demand.53 It should come as no surprise that attempts to introduce profit sharing were modest and that they met with little success. At one firm, 160 workers were put on a profit-sharing plan but the scheme was abandoned in less than one year owing to its poor performance.54 In all, there were seventy-six limiteds created between 1860 and 1896; still eighty percent of these were ‘private’ enterprises (the corresponding figure for Oldham is fifteen percent) whose shares were neither widely distributed nor exchanged. Family concerns continued to dominate Bolton until the First World War.

Industrial relations and cooperative behaviour My objective in this section is to explain why Oldham’s workers were receptive to profit sharing while Bolton’s workers rejected it. I proceed by elimination. To begin, employee ownership had little to do with work organisation. A work team composed of one spinner and two to three piecers was dominant in both towns, as it was throughout Lancashire.55 Nor did it have to do with the fact that the town spun coarse yarn. In other towns spinning coarse yarn, like Stockport, there were few limiteds, while Ashton, which spun medium and fine yarns, had the greatest proportion of limiteds next to Oldham.56 With regard to worker characteristics, the evidence I have produced so far is aggregate. It seems that, on the whole, the demographic profiles and religious and social backgrounds of the average worker in the two towns were similar, but it may well be that there are unobserved differences among individuals between and within towns that have not been captured or controlled for in Table 3.2 that explain the predisposition of certain workers to profit share. This type of argument is always difficult to dismiss even in the presence of micro data. Having said that, we can eliminate the possibility that profit sharing in the two towns was associated with different attitudes toward risk. In principle, risk averse workers are not attracted to profit-sharing arrangements. The evidence suggests that in this regard the average Bolton and Oldham worker were equally likely to profit share. Bolton workers, like those in Oldham, already assumed a high degree of risk since both were paid by the piece, and their earnings moved with firm output which fluctuated regularly in both fine and coarse sectors.57 Risk aversion is often correlated (negatively) with income levels because poorer workers cannot insure themselves completely against shocks. But, in 1870, Bolton spinners earned the highest wages in Lancashire and they should have been more, and not less, likely to accept profit-sharing contracts. Risk aversion is often linked (positively) to workers’ investments in firm-specific skills. Workers may want to diversify their savings and invest elsewhere. On this score, Bolton’s skilled workforce would have had less of an interest in profit sharing, but it is difficult to imagine an investment opportunity in Lancashire that was not correlated with the cotton trade. Moreover, organisations in which workers are involved in decision making tend to

Huberman

63

have comparatively small employment fluctuations,58 and in the absence of a fully-developed unemployment insurance scheme it would be expected that workers with significant investments in firm-specific skills, like those in Bolton, would stand to gain more in cooperating with management in order to lower the risks of layoff. All told, risk aversion cannot explain the pattern of profit sharing found in Lancashire. My explanation of why workers in the two towns differed with regard to profit sharing is based on psychologists’ treatment of emotions that are revealed in different person-environment relationships.59 Individuals often develop pro-social and empathic feelings over the distress of others, particularly those who share similar backgrounds or concerns, like family members, neighbours or fellow workers. Individuals in these situations will have a predisposition to help out a co-worker more than a boss because of the narrower ‘social distance’ between them.60 They have a bias toward cooperative solutions in that they will not seek a greater share of the pie because they expect that other players will do so as well. In its extreme form, individuals are motivated to do well not so much by the direct pressure of their peers, but by feelings internalised toward their comrades. Empathy of this nature is built on and strengthens horizontal attachments. In other situations, individuals may lack empathy toward their peers, or at best be indifferent to them. They will only cooperate if there is a penalty for not doing so. In the language of game theory, they will not in the majority of cases chose freely the cooperative solution. In these cases, since cooperation is not certain, individuals will need to be monitored. I follow Lazear in evaluating workers’ responses to identical economic incentives assuming different patterns of social behaviour.61 Consider an employer who pays a straight time-wage to his workers. A worker who does not perform well, who shirks, will hurt the owners of capital. To motivate workers, the employer decides to introduce a profit-sharing scheme. What prevents workers from continuing to under perform? In the absence of empathy a worker may in fact continue to under produce; however, she takes the risk of being caught by the employer, a gamble that might be worth taking. Hence, there is little gain to profit share. In the presence of empathy there are potential gains, because workers care more about their colleagues than their bosses. They work hard not to let down their fellow workers. The upshot is that empathy overrides the free-rider problem, but in organisations lacking this predisposition cooperation will need to be monitored. It is precisely these monitoring costs that make cooperative behaviour more efficient – but less common – than non-cooperative behaviour.62 For psychologists the origins of this type of behaviour are obscure.63 Some have sought their origins in shame and guilt. While shame denotes some type of external enforcement, where workers are guilt-driven they will act by some internalised moral imperative to help each other.64 My data do not permit me to observe these traits, but I would suspect that Bolton and Oldham, only fifteen miles apart, did not differ much in this respect. Moreover, since my data are aggregate, I can only suppose that it was nurture and not nature that separated the two towns. Other psychologists

64

When labour hires capital

observe that there are often formative or critical events that tilt individuals toward one form of emotional response or another. Recognising the hazards of inferring individual behaviour from large social changes, I follow this approach here. My explanation of the divergence in social behaviour is based on the observation that, owing to divergent historical trajectories, workers in the two towns experienced differently the key years of industrial change before mid-century. Bolton was one of the first centres of cotton spinning in Lancashire. As early as 1811, the town housed thirty-three mills, the average mill employed about 150 workers. By 1841 the average mill employed approximately 200 workers and was nearly twice as large as the average coarse-spinning establishment. Disputes and strikes into the 1820s over the introduction of longer spinning mules and piece rate cuts were not unknown, but after the spinners’ strike in 1836, the bitterness that marked the early period of industrial relations dissipated. The political arena saw limited mobilisation against the introduction of the New Poor Law of 1834, and by 1837 Bolton radicals were complaining about working-class apathy.65 From mid-century, if not earlier, paternalism became the hallmark of worker-firm relations.66 Employers’ concern for the welfare of their workers was manifested in their active participation in factory schools, libraries, housing, picnics and other day trips, and Christmas parties. Employers hoped to strengthen workers’ attachments to them, or, as one employer put it, ‘to cement a feeling of…mutual interest’. The low rate of turnover of firms and their large size gave mid-nineteenth-century Bolton the appearance of a hierarchical, if not manorial community in which workers and firms held reciprocal obligations. Craft unionism flourished in this environment. The Bolton spinners’ union maintained its independence from employers. Because it was millbased, Bolton unions could boast of having organised practically every spinner in the district, and provided for their members benefits not covered by their paternalist employers.67 For the 1890s when records are available, expenditure per member was as much as £3 per annum in Bolton; whereas in Oldham it was about £2. 3s. The presence of unions would seem to be inconsistent with the view that paternalism is an alternative to unionisation. In weaving centres, such as Blackburn, paternalism went hand in hand with weak unions, and in the 1920s large firms in the United States, like Sears and Kodak, used welfare initiatives to deter unionisation. But in Bolton the union assured that promotion was based on seniority; it also safeguarded spinners’ responsibility for the maintenance and repair of their machines and adjusting operating speeds.68 Arrangements between employers and the union were codified in the establishment of a piece-rate list that fixed the rules for bargaining. Employers and the union were equally proud that strikes and lockouts were practically unknown before 1914. The spread of factory industry in Oldham was less portentous. Before 1850 Oldham firms tended to rent or share space and power in a larger mill. According to Farnie, small firms, many of which were run by former

Huberman

65

spinners, ‘accentuated the degree of competition, and increased the mortality rate among factory masters,’ and, in contrast with Bolton, their presence prevented ‘the family firm from establishing an hereditary monopoly of local industry under a separate caste of employers’.69 As late as 1841, the average coarse mill employed only hundred workers. Breaking with the ownership practices of Lancashire, Oldham became known as ‘the pariah of the cotton trade’, and a ‘frontier town’.70 ‘Whether or not Oldham was rough’, Joyce wrote, ‘it was certainly vulgar’.71 The town’s manufacture of coarse yarns confirmed its dubious reputation. ‘The refuse from all the other Lancashire towns is brought to [Oldham], and worked up into the coarsest and trashiest fabrics’.72 Faced by competitive pressures, employers introduced new technologies, like the self-acting mule in the 1820s, and slashed piece rates. Up until 1825 or so, the industrial relations histories of the two towns were roughly similar. Beginning in the early 1830s, a different dynamic emerged in Oldham. A core group of Oldham workers, including cotton spinners, machine builders and miners, had organised to defend their social and economic position. Although Foster’s account of the development of ‘revolutionary class consciousness’ in the town remains controversial – more recently historians have tended to emphasise the gradual transition from radicalism to liberalism – there is little dispute that relative to other Lancashire communities, Oldham saw a greater degree of mobilisation.73 The small scale of its establishments may have facilitated worker combinations, and the fact that many of these units were owned by former spinners who had broken rank may have exacerbated further tensions between workers and bosses. On the factory floor, periodic disputes over piece-rate cuts reinforced worker cohesion. The strike record of Oldham workers shaped social behaviour in the town. At a strike vote, workers did not know how many of their colleagues would show up on the picket line the following day. Strikes tested individuals’ predisposition to free ride.74 The pivotal years in Oldham began with the spinners’ strike in 1837 and terminated with the engineers’ dispute of 1852. It was during these recurrent battles that Oldham workers, supported and promoted by community networks and institutions, formed strong bonds among each other and learned not to let each other down. The narrowing of social distance between workers had social and economic consequences. In addition to finding evidence of a high degree of intermarriage within the working class, Foster presented data on the greater probability of labouring families living next to craft families in Oldham compared to other industrial towns.75 Oldham’s residents were receptive to socialist-utopian ideals, as expressed in Owenism, and they were among the first in Lancashire to set up cooperative stores.76 Despite the high degree of militancy, Oldham’s union was much weaker than Bolton’s. It was organised on a district rather than a mill basis and it was less successful in defending craft privileges. The union permitted employers to introduce joiner-minding, an employment arrangement in which the traditional spinner-piecer work unit was replaced by two

66

When labour hires capital

minders, each with joint responsibility for spinning operations.77 Oldham unions did not provide many benefits, and whereas leisure activities were often sponsored by mills in Bolton, it was individuals in Oldham who organised themselves in going-off clubs.78 We should not read from this evidence that Oldham militancy was mere rhetoric. In Bolton, because of divided worker loyalties, the perception was that unions posed less of a threat and, as a result, they were fully integrated into the operations of paternalist mills. In Oldham, employers had good reason to be tough because of the town’s radical bent. Among the founders of the Oldham Building and Manufacturing Co. Ltd., the first of the Oldham limiteds I have previously referred to, was William Marcroft, a grinder at a local engineering shop.79 Marcroft, a nonconformist, was a participant in the key political struggles of the 1840s and a supporter of Oldham’s radical political parties. After the defeat of the engineers in the strike of 1852, Marcroft turned his energies to cooperative production and retailing, a system in which workers as clients held equity and received a share of the profits. Exploiting changes in governance permitted by the new company legislation, Marcroft and others tapped into the basic empathy Oldham workers had for each other. The radical workers of Oldham had remade themselves into the most ardent capitalists. At first pass, this transformation seems inexplicable but the emotional responses of Oldham workers in both guises were similar. Behaviour was based on the moral imperative not to let a co-worker down, an essential ingredient in striking, cooperative retailing and profit sharing. Workers had changed their clothing, but not their intentions.80 My research strategy in the next two sections of this chapter is to examine the outcome of this type of social behaviour. I focus on the resources devoted to monitoring and enforcement in factories as well as in other institutions in the towns, like the provision of poor relief, where the potential to free ride existed. Recall that I have found that technologies, work organisations, market opportunities, risk preferences, demographic variables and levels of human and social capital were comparable across the two towns. Thus, if I find large variations in resources devoted to enforcement in the towns, I cannot reject the hypothesis that these differences were the result of divergent patterns of social behaviour.

Enforcement inside the factory By 1870 or so, attitudes of workers to each other and to their bosses had taken shape. Paternalist or vertical arrangements characterised Bolton, whereas in Oldham, because of strong horizontal attachments, workers saw that their interests were distinct from those of their employers. In both towns, firms and community institutions protected and perpetuated norms of social behaviour, but the nature of enforcement varied. There were in fact three levels of enforcement: by firms, workers and factory inspectors. On all counts, Bolton was the more censorious community.

Huberman

67

Firms in Bolton hired about one-and-a-half more supervisors and foremen to monitor activities than in Oldham.81 When called upon, Bolton firms also used fines, exactly what we would expect to see where there was no bias toward cooperation. Organisations needed to resort to punishment, often randomly applied, if only to demonstrate that breaking norms was a serious offence.82 Employers in the town blacklisted militant spinners who broke industrial peace and they demanded discharge notes when hiring.83 Bolton employers also monitored their competition. Owners of large mills gave financial support to workers who were out on strike against wage cutting at small, often rural mills in the Bolton region.84 In Oldham, managers’ supervision was more lax and they used fines less often.85 This was true regardless of mill size. The new limiteds after 1870 were about twice as large as mills built before mid-century, but there was no alteration in the number of overlookers per spinner. The 1/N problem was not a constraint on profit sharing because of workers’ inherent trustworthiness. Moreover, Oldham managers were less inclined to monitor their neighbours and, unlike Bolton firms, they did not engage in sanctioning their competition. Bolton workers were as vigilant as their bosses and they too engaged in monitoring and blacklisting. The nature of supervision of the work team was highly developed in Bolton. The ratio of piecers to spinners was greater in the town,86 until the First World War, if not later, Bolton spinners continued to exercise their control of hiring, negotiating and paying their own junior members of the spinning team, and they nominated them for advancement. Workers’ associations in Bolton devoted considerable resources to monitoring. The Bolton trade union blacklisted workers who did not pay their dues which was considered ‘a very serious matter…a disgraceful thing.’ The union spelled out the penalty of free riding: Here are men taking every advantage obtained by the self denial and exertions of their fellow workmen and too mean to pay their quota…The day is not too far distant, we hope, when it will be possible for us either to refuse to work with them or leave them so much in the cold that they will see the error of their way.

The union’s words and actions left their mark. In 1883, there were 115 spinners on the Bolton blacklist, but only 41 ten years later.87 In Oldham, workers monitoring workers was less common. Spinners had given up, through negotiation in the 1870s, their right to set the pay of junior team members, and there is no evidence of blacklists in the town. Third-party enforcement was also greater in Bolton. A detailed study of factory inspection found that the number of infractions for breaking health and safety codes and hours legislation was greater in Bolton, even though Oldham factories were notorious for their poor working conditions.88 Oldham firms were less disposed to follow administrative rules such as completing age certificates and the town’s magistrates were more lenient in applying the rules and handing out fines. We are now in a position to consider the effects on output if Bolton firms had imposed profit-sharing wage system on their workers. Since both

68

When labour hires capital

their union and employers provided them with benefits, Bolton operatives were indifferent between the lot of their co-workers and their bosses. In this regard, workers had a strong sense of independence or autonomy that was embedded in the town’s piece rate list which they defended vigilantly.89 If this autonomy was at all threatened, say by moving to a new system of team-based incentives, workers may well have reduced effort, provoking firms to increase levels of enforcement further and threatening any degree of trust that had been established. Kreps speculated that the use of efficiency wages would have similar results where workers exhibit a great degree of autonomy.90 He concluded: ‘[E]xplicit extrinsic incentives that are imposed may fight rather than complement preexisting incentives.’ The point is that the introduction of profit sharing along with a stiffer penalty for shirking or other types of multiple incentive schemes might not have had the expected effects on effort.91 Bolton organisations needed to match incentive structures and the type of social behaviour found in the town, and – without invoking optimality – this meant relying on piece rates, the tried and true method of eliciting effort in Lancashire. How then did Bolton survive confronted by Oldham’s advantage? I suspect that Bolton firms concentrated increasingly on their niche markets in fine yarn in the UK and elsewhere, leaving upstart Oldham to those count ranges in which there was overlap.

Enforcement outside the factory The fewer resources devoted to monitoring in Oldham may have been the result of a better designed piece-rate system and not the outcome of workers’ innate trustworthiness. But social behaviour in the two towns was reflected outside as well inside the factory. It would have been difficult for Oldham workers to trust each other in the factory setting, if they did not find the same level of support in community-based institutions. Many of Bolton’s governing bodies were closed or oligarchic, the hierarchic nature of the town’s factories being reproduced in its civic institutions which were controlled by a small group of influential mill owners. Democratic expectations were relatively low, if non-existent in the town.92 In Oldham, civic life was more open. The number of mill owners on the municipal council actually declined in the last part of the century, their positions being taken up by a broad range of professions.93 Decision making, in the words of Winston Churchill, the town’s MP before the War, was in the hands of ‘happy-go-lucky amateurs’.94 The empathy between Oldham workers on the factory floor was mirrored in their active participation in the cooperative retailing movement and the large number of friendly societies in the town.95 An illustration of the type of empathy present in the two towns was their delivery of social assistance. The administration of the Poor Law, an institution noted for its potential free-rider problems, left some room to manoeuvre at the parish level. There were important differences in the

Huberman

69

philosophies of the Bolton and Oldham unions. Henry Ashworth, mill owner and leading figure in the Bolton union at mid-century, advocated the strict application of the work-house test, while Oldham magistrates, during the cotton famine, asked to dispense with the labour test and, although its request was denied, granted relief anyway.96 Table 3.3 gives a portrait of the application of the Poor Law in Bolton and Oldham from 1870 to 1907. The total number of paupers on relief was always greater in Bolton. This is consistent with the stronger growth of Oldham’s economy or the possibility that its workers had other avenues of relief to draw on, such as personal or family savings or provident societies, for example. But the trend in the number of paupers is downward in Bolton, while it was relatively stable, if not increasing, in Oldham. This period in the history of the Poor Law was marked by the movement against outrelief. The idea here was that by reducing the proportion on outrelief, the total number relieved and the total cost of assistance would fall. This may explain the declining numbers in Bolton, although relief expenditure in the town did increase. The dynamic in Oldham seems to have been different. With the exception of 1900, relief expenditure in Oldham was greater than Bolton’s and those on relief were generally the most deserving, women, widows and children. Altogether the proportion of non-able bodied receiving relief was always greater than in Bolton. My reading of these statistics is that in Oldham there was greater propensity to take care of the most deserving and the town’s administrators were willing to make constant improvements to its workhouse. The Bolton union seems to have been less inclined to make improvements to its workhouse and its expenditures remained generally lower.97 Table 3.3. Poor Law statistics, 1870–1907 Bolton Year

1870 1880 1890 1900 1907

Oldham

Paupers Indoor/ Non-abled (per ‘000) outdoor bodied paupers adults as % of total relieved

23.04 33.06 18.36 20.48 17.57

0.26 0.27 0.26 0.29 0.53

7.97 7.16 11.63 6.13 13.01

Relief Paupers Indoor/ Non-abled Relief expenditure (per ‘000) outdoor bodied expenditure per pauper paupers adults as % per pauper (£/s) of total (£/s) relieved

2/10 2/3 1/8 2/14 3/7

13.27 25.36 16.35 19.39 16.54

0.49 0.41 0.47 0.49 1.19

11.38 12.39 15.35 9.17 16.37

3/1 3/8 2/6 2/9 3/13

Sources: Poor Law reports for the 1st of July of each year. PP 1872 (387) LI; PP 1881 (60) LXXVIII; PP 1890 (303) LXIII; PP 1900 (136) LXXIII; PP 1908 (250) XCII.

The Poor Law authorities conducted a detailed survey of 1907 that allows us to evaluate whether the generous features of the Oldham union led to abuse of the system (Table 3.4). This was not the case. About 30 percent of those on relief in Oldham received payments for less than one

70

When labour hires capital

month; the corresponding figure in Bolton was 25 percent. Not only were durations shorter, but so was the frequency of use. The number of repeat users was much lower in Oldham. This is exactly what one would expect to see in a society where there was a strong sentiment not to let one’s coworkers down. Workers in Oldham would not abuse or free ride because this would mean less would be available for someone truly deserving. Table 3.4. Poor Law statistics, 1907

Total number of persons relieved Duration < 1 week (percentage of total relieved) 1 – 4 weeks 26 – 52 weeks Repeat users Once Twice Five or more

Bolton

Oldham

8,315

5,959

543 (6.53) 1,518 (18.26) 1,278 (15.37)

674 (11.31) 1,071 (17.97) 828 (13.89)

6,835 (82.20) 961 (11.56) 77 (0.93)

5,327 (89.39) 512 (8.60) 14 (0.23)

Source: PP 1908 (250) XCII.

Rings and mules and the decline of profit sharing Two events marked the history of Oldham in the period from 1900 to 1914. First, was the decline in workers’ participation in the limiteds; second, was the decision of many Oldham firms to stick with mule spinning despite the availability of ring spinning.98 My objective in this section is to show that these two events were related. If labour ownership was so good, why did it come to an end? Dow and Putterman give some reasons for the instability of cooperation where both workers and nonworkers own shares in the same organisation and where shares are traded freely. There are fundamental asymmetries between capital and labour inputs. Workers generally hold only one job at a time, while the owners of capital can spread their resources more evenly. Workers, like those in Oldham, can and do buy shares in other companies to redress this imbalance, but it is certainly much easier for the owners of capital to diversify their portfolios.99 Moreover, where workers bring both their labour and capital into the organisation, their preferences are more heterogeneous than those of capital. Consider the case of a firm with some worker ownership wanting to introduce a new technology. Older workers on the eve of their retirement will be concerned about the value of their

Huberman

71

shares and may be more responsive to restructuring; they will have different preferences than middle-aged workers who may lose their jobs in restructuring, and younger workers who may agree to change if they can obtain shares in the first place.100 As a result of these imbalances and asymmetries, capitalist investors often buy out workers, but workers rarely buy out capitalists. Worker ownership would probably last no longer that one generation. Dow and Putterman conclude that worker participation is unstable. The turn of events in Oldham is consistent with this story. Beginning in the 1880s, ring spinning was being slowly introduced in Lancashire, a technique that in textile industries elsewhere, like the US and Japan, had displaced the male mule spinner from his privileged and highly remunerative position in favour of cheaper labour, mainly women.101 Acting as stakeholders, younger and middle-aged male spinners would have probably fought against this restructuring to preserve their firm-specific investments. There were few employment options that offered comparable wages for redundant middle-aged spinners and their unemployment benefits were meagre.102 Older spinners were faced with the dilemma of choosing between protecting the value of their shares (and supporting restructuring) and defending the interests of their co-workers. Their pension benefits were minimal and at some point they had to sell their shares. The average age of Oldham spinners declined over the period (see Table 3.2) because of the town’s expansion, and many of the newer generation of spinners probably did not have the savings required or could not borrow to buy these shares, although unions and cooperatives held onto theirs.103 Recall that in the 1870s investors could buy individual units at about £5; by 1900 initial offerings were by blocks of shares.104 Taking advantage of workers’ heterogeneous preferences and their control of the capital market, managers and other speculators began buying out individual spinners in the boom years after 1900.105 By 1914, Oldham’s experiment with worker cooperation had come to an end. The irony is that the purchase of spinners’ individual shares by mill managers and the like did not usher in a wave of new ring-spinning technology. The union and cooperative movement retained their role as stakeholders and opposed the new technology to preserve the jobs of their median members. Recall that many limiteds were run on a ‘one man one vote’ basis, which meant that the increasing concentration of shares would not have deterred organised workers from having their opinion heard and acted upon. But mill managers were also opposed to changing technology. Mule spinning was still profitable. Moreover, many of the new limiteds were speculative ventures and they had to be built quickly. The blueprints for mule mills were easily available and indeed many of the town’s architects and builders who specialised in constructing mills with mules held significant investments in the limiteds built after 1900 and sat on boards of directors.106 The rate of installation of rings gives credence to this story. In the period, 1896–1905, the number of rings almost tripled, but in the next ten years they merely doubled.107

72

When labour hires capital

Evidence on the geographical distribution of spinning technologies in Lancashire is consistent with the view that there was a strong preference for limiteds to retain mules. Everything else constant, Oldham should have been as likely, if not more so, to have made significant investments in rings which were best suited to lower grades of yarn. But various indicators reported in Table 3.5 suggest that fine-spinning Bolton was as prone to invest in rings. Leunig argued recently that owing to the transport costs of shipping warp yarn, the adoption of rings in individual towns was correlated with available, or co-located, weaving capacity.108 Bolton had double the weaving potential of Oldham and this is reflected in the proportion of ring output in the two towns (line 6 of Table 3.5), despite the fact that the former was known for its fine yarn. But this explanation is at best partial. Ashton spun on average counts between the upper and lower ranges found in Oldham and Bolton, and its weaving capacity, according to Leunig, was about seven times that of Oldham’s.109 It should have been more receptive to the new technology, but the actual take up of rings was about the same as in Oldham, about eight percent of output. Transportation costs cannot be the sole explanation of the choice of technique. Like Oldham, Ashton had a militant history, and the town saw a great expansion in the number of limiteds.110 In both towns, stakeholders, whether they were unions, managers or architects, preferred mules. In contrast, there was a clear positive relation between privately run firms and ring spinning. Stockport spun coarse yarn and it had a greater share of rings, because it had a smaller number of limiteds; in Rochdale, a town dominated by paternalist employers and spinning coarse yarn, the percentage of ring spindles was also high, amounting to a third of the number in Lancashire in 1905.111 Finally, in Bolton, where families retained control of organisations and limiteds were practically unknown, the move to rings met with less resistance. Table 3.5. Rings versus mules: Bolton and Oldham

Percentage of ring factories 1901 Percentage of ring spinners 1906 Percentage of ring spindles 1905 1915 1927 Percentage of ring output 1906

Bolton

Oldham

6.14

5.64

24.92

20.04

– 0.25 3.7

5.8 10 11

7.9

8.5

Sources: Lines 1 through 5 – PP 1903 (289) LXIV; PP 1909 (4545) LXXX; Gurr and Hunt, Cotton Mills, p. 30; private correspondence with Douglas Farnie; line 6 – T. Leunig, ‘The Myth of the Corporate Economy’ (unpublished Ph.D. thesis, University of Oxford, 1996), p. 92.

Huberman

73

Conclusion In this chapter I have attempted to address two interrelated puzzles, one theoretical, the other historical: if labour ownership is so good, why do we not see more of it, and why was the growth of Oldham’s cotton spinning industry so much faster than Bolton’s. As for the first problem, I have argued that profit sharing, as a wage system, is not widely observed because of its potential for free-rider problems. These problems are done away with where there exists some degree of empathy between workers. Contrary to Alchian and Demsetz, workers can monitor each other because under certain conditions the costs of doing so may be very small. In other situations, workers may not be as concerned with the welfare of their peers and, as a result, the free-rider problem will persist under profitsharing schemes, forcing firms and workers to expend a greater proportion of resources to monitor each other. It is the presence of these costs that make cooperation so good, but rare because cooperative behaviour is dependent on some degree of trust. I have drawn on these propositions to answer my second question about growth rates in the two towns. In Oldham, workers cared a great deal about each other to the extent that these feelings overrode free-rider problems. The timing of Oldham’s growth went hand-in-hand with its adoption of profit-sharing arrangements. In Bolton, similar incentive structures were impractical because workers were indifferent to the welfare of their peers and their bosses. These motivations were revealed in greater monitoring on the factory floor in Bolton and outside the factory in the distribution of poor relief. Because attitudes in Bolton were not compatible with profit sharing, its growth was slower than Oldham’s. My reading of the historical evidence suggests that the options available to organisations to elicit cooperation are limited. The two towns I have studied shared common technologies, work organisations and market opportunities, and capital and factor markets were integrated. Risk preferences were also comparable across towns, as were levels of education and work experience. Moreover, workers in the towns shared a common sociocultural background. Yet, despite these likenesses, the two towns exhibited a divergence in their patterns of social behaviour. Oldham workers had a more militant past and, supported by community institutions, the empathy built up in the period of conflict that lasted until 1850 resurfaced in the limiteds after 1860. Oldham’s culture of trust could not have been reproduced easily. This result is consistent with recent experiments conducted by Schotter who found that in profit-sharing games participants achieved the cooperative solution if and only if they had made some prior commitments with each other.112 Lacking a heritage of cooperation, it would have been impractical for an individual Bolton firm to divert resources to promote trust and raise effort levels of its workers, because there was little institutional support both within and outside the factory for this type of human-resource initiative. For Bolton, policy was no substitute for history. This result is consistent with the view advanced by Bowles and Gintis that

74

When labour hires capital

communities support the norms and values of individual and group behaviour in organisations.113 Oldham’s experience with labour ownership lasted for one generation. The fact that so few workers actually participate in wage systems like profit-sharing schemes today, as in the past and for long periods of time, is evidence that cooperation is difficult to obtain. Although there are examples of labour ownership scattered across time and space, researchers have found no empirical regularity that would permit generalisations about the conditions in which cooperation will actually manifest itself.114 The peculiar history of the Oldham limiteds speaks directly to this point.

Notes 1. National Center for Employee Ownership (www.nceo.org). For European evidence on worker participation, see J. Bonin, D. Jones and L. Putterman. ‘Theoretical and Empirical Studies of Producer Cooperatives: Will Ever the Twain Meet?’, Journal of Economic Literature, 31: 3 (1993), pp. 1290–1320; for Japan, see D.C. Jones and T. Kato, ‘The Productivity Effects of Employee Stock-Ownership Plans and Bonuses: Evidence form Japanese Panel Data’, American Economic Review, 85: 2 (1995) pp. 391–414. 2. D. L. Kruse, Profit Sharing: Does it Make a Difference? (Kalamazoo, 1993). 3. G. Dow and L. Putterman, ‘Why Capital (Usually) Hires Labor: An Assessment of Proposed Explanations’, in M. B. Blair and M. J. Roe (eds), Employers and Corporate Governance (Washington, 1999), pp. 17–57. 4. E. P. Lazear, Personnel Economics (Cambridge, Mass, 1995); M. Van der Linden, ‘Households and Labour Movements’, Economic and Social History in the Netherlands: Family Strategies and Changing Labour Relations, 6:1 (1994), pp. 129–144. 5. The idea here is that worker-monitors have the incentive to run down capital equipment: A. Alchian and H. Demsetz, ‘Production, Information Costs and Economic Organization,’ American Economic Review, 59:4 (1972), pp. 473–494. 6. D. A. Farnie, ‘The Emergence of Victorian Oldham as the Centre of the Cotton Spinning Industry’, Bulletin of the Saddleworth Historical Society, 12 (1982), p. 50. 7. McCloskey advocates that economists and economic historians need to incorporate emotions in their models; D. McCloskey, ‘Bourgeois Virtue and the History of P and S’, Journal of Economic History, 58:2 (1998), pp. 297–318. 8. I define trust as the commitment of resources to an activity where the outcome depends upon the cooperative behaviour of others. Trustworthiness is defined as behaviour that increases returns to people who trust you. 9. W. B. MacLeod, ‘Equity, Efficiency, and Incentives in Cooperative Teams’, in D. C. Jones and J. Svejnar (eds), Advances in the Economic Analysis of Participatory and LaborManaged Firms (Greenwich, Conn., 1988), pp. 5–23. 10. G. J. Miller, The Political Economy of Hierarchy (New York, 1992); M. Weitzman, ‘Incentive Effects of Profit Sharing,’ in H. Siebert (ed.), Do Participation and Cooperation Increase Competitiveness? (Siebeck, 1995), pp. 51–78. 11. P. Temin, ‘Is it Kosher to Talk About Culture?’ Journal of Economic History, 57:2 (1997), p. 278. 12. F. L. Galassi, ‘Culture as an Explanation of Italy’s Economic Dualism,’ European Review of Economic History, 5: 1 (2001), pp. 29–60. 13. J. Foster, Class Struggle and the Industrial Revolution (London, 1974). 14. M. Rose, Firms, Networks and Business values: The British and American Cotton Industries since 1750 (Cambridge, 2000) p. 171. 15. These figures are from W. Lazonick, Competitive Advantage on the Shop Floor (Cambridge, Mass., 1990). Productivity figures are for 32s twist yarn, the average count

Huberman

16. 17.

18. 19.

20.

21. 22. 23. 24.

25.

26.

27.

28.

29.

75

spun in Oldham. See also S. N. Broadberry, The Productivity Race: British Manufacturing in International Perspective, 1850-1990 (Cambridge, 1997). M. Huberman, Escape From the Market: Negotiating Work in Lancashire (Cambridge, 1996). The Oldham piece rate or wage list had a built-in time component and any relation between earnings and effort is approximate. Moreover, the Oldham list divided the benefits of investments in technology between workers and firms. The Bolton list assured that all benefits went to workers. On the wage lists in the two towns, see Huberman, Escape. J. Jewkes and E. M. Gray, Wages and Labour in the Lancashire Cotton Spinning Industry (Manchester, 1935), p. 42. A. Marrison, ‘Indian Summer, 1870-1914,’ in M. B. Rose (ed.), The Lancashire Cotton Industry: A History Since 1700 (Preston, 1996); P. Joyce, Work, Society and Politics: The Culture of the Factory in Later Victorian England (Brighton, 1980). For Bolton manufacturers’ assessment of market opportunities see the annual reports for 1885 to 1890 of the Bolton Master Cotton Spinner’s Association (Z/FE/1), Bolton Local Studies Service. D. A. Farnie, The English Cotton Industry in the World Market (Oxford, 1979). G. R. Saxonhouse and G. Wright, ‘New Evidence on the Stubborn English Mule and the Cotton Industry’, Economic History Review, 37:3 (1984), pp. 507–519. W. Lazonick, Competitive Advantage on the Shop Floor (Cambridge, Mass., 1990). Oldham’s encroachment into fine spinning led to a dispute, beginning in the early 1890s, between manufacturers and spinners in the two towns. The Oldham spinners’ union estimated that 25 percent of its membership in 1900 spun Bolton counts, nos. 40–80. Oldham spinners, supported by Bolton manufacturers, demanded to be paid by the Bolton list of prices (which had higher rates than the Oldham list at selected counts). The ‘fine-counts question’ was resolved by the intervention of Lloyd George in 1907. See, Minute Books and Annual Reports of the Oldham and District Operative Cotton Spinners’ Provincial Association (TU1/1/8–23) and the Reports of the Master Cotton Spinners’ Association (OLD6/8/2–15), John Rylands Library. The increase in counts spun in Oldham does not mesh with the calculations of L. Sandberg (‘Measurements in the Quality of the British Cotton Textiles, 1815–1913’, Journal of Economic History, 99:1 (1968), pp. 25–43) who, using aggregate data on exports, found that after a rise in quality in the early to mid 1890s, from 1898 to 1914 the general trend was slightly downward. The wage structures of the two towns corroborate the view that workers were comparable in quality. In both towns spinners’ wages were about twice that of common labourers, and about 20 percent less than highly skilled workers like engineers. That said, the demographic profile of spinners favoured human-capital accumulation in Bolton. The average age of spinners in Oldham declined over the period, while that in Bolton increased. E. L. Glaeser, D. Laibson, J. A. Scheinkman and C. L. Soutter, ‘What is Social Capital’, NBER working paper, no. 7216, 1999. The idea here is that social interaction between family members and within homogeneous groups promotes mutual assistance which is a correlate of growth. On models of social interaction, see R. Gibbons and M. Waldman, ‘Careers in Organizations,’ in O. Ashenfelter and D. Card (eds), Handbook of Labor Economics, Vol. 3B (Amsterdam, 1999). Kinship relationships may have differed between the towns. Women in Oldham had children earlier and had a higher rate of labour force participation because of the greater percentage of grandparents co-residing in their children’s households (S. Szreter, Fertility, Class and Gender in Britain, 1860–1940 (Cambridge, 1996), p. 490). On women’s labour force participation in Bolton, see P. Thane, ‘Women and Work in Britain, c. 1870–World War I’, in P. Mathias and J. A. Davis (eds), The Nature of Industrialization: Enterprise and Labour from the Eighteenth Century to the Present (Oxford, 1996), p. 55. An example is C. Goldin and L. F. Katz, ‘Human Capital and Social Capital: The Rise of Secondary Schooling in America, 1910–1940’, Journal of Interdisciplinary History, 29 (1999), pp. 683–723.

76

When labour hires capital

30. My account of the limiteds owes much to the work of Farnie, English Cotton Industry, W.A. Thomas, The Provincial Stock Exchanges (London 1973), and G. von SchulzeGaevernitz, The Cotton Trade in England and on the Continent: A Study in the Field of the Cotton Industry (London, 1895). The only other part of the textile industry where the limited company had the same hold as in Oldham was the Irish linen trade: P. Cottrell, Industrial Finance, 1830–1914: The Finance and Organisation of English Manufacturing Industry (London, 1980), p. 105. 31. R. Smith, ‘An Oldham Limited Liability Company’, Business History, 4:1 (1961), pp. 34–53; R.E. Tyson, ‘William Marcroft (1822–94) and the Limited Liability Movement’, Transactions of the Lancashire and Cheshire Antiquarian Society, 80:1 (1980), pp. 60–80. 32. After Oldham, Ashton and Rochdale were the other important centres of limited companies (Farnie, English Cotton Industry). 33. Dow and Putterman, ‘Why Capital (Usually) Hires Labor’. 34. Farnie, English cotton industry; B. Jones, Co-operative Production (Oxford, 1894); B. Potter, The Co-operative Movement in Great Britain (London, 1891); S. Procter and J. S. Toms, ‘Industrial Relations and Technical Change: Profits, Wages and Costs in the Lancashire Cotton Industry, 1880–1914’, Journal of Industrial History, 3:1 (2000), pp. 55–72. 35. There is only scattered evidence that share ownership was a condition of employment. The Cotton Factory Times (26 February 1904) reported that ‘minders who have gone after situations have been promised employment if they would put some money in the company, but not otherwise.’ 36. Farnie, English Cotton Industry, p. 250; Jones, Co-operative production, p. 294; PP 1895 (7458) LXXX, p. 595. 37. Cited in Tyson, William Marcroft, p. 80; see also PP 1892 (6705) XXXV, pp. 724–26; PP 1894 (7458) LXXX, p. 596; Thomas, Provincial Stock Exchanges, p. 147. 38. See the balance sheets found in the annual report of the Oldham and District Operative Cotton Spinners’ Provincial Association (TU1/1/8–26). In the mid 1890s, when Toms reports that there was little interest in holding shares, the Oldham union still held investments in fifteen mills totalling £2500, which amounted to about 25 percent of its total expenditures. On friendly societies’ investments, see PP 1874 (995) XXIII, p. 294. 39. In 1900 there were seventy cooperatives and friendly societies in Oldham. They had on average about 150 members (PP 1901 (698) LXXIV). The records of two societies have survived: the Greenfield Co-operative Society Ltd. (Balance Sheets, CGRe/4/3) and the Royton Industrial Co-Operative Society (General Committee Minutes, CR/1/11). The former had investments in ten limiteds in 1908 for a total value of £600. This represented about 70 percent of its investments. The companies invested in varied over time. At its 1906 meeting, the Greenfield Co-operative heard from a member who had attended the shareholder meeting of the Mossley Cotton Co. Ltd. The company’s performance was deemed ‘unsatisfactory’ and along with the other shareholders the co-op demanded the resignation of the mill manager. These records are available at the Oldham Local Studies Library. Below, I give further evidence on the co-operative movement. 40. I discuss the end of worker cooperation in the penultimate section of this chapter. 41. Jones, Co-operative Production, p. 294 42. Farnie wrote: ‘The limiteds therefore remained true to their nature as associations of capitalists dedicated to the making of private profit and compelled to compete with private firms and with other limiteds.’ (English Cotton Industry, p. 261). 43. Farnie, English Cotton Industry, p. 260–261. For supporting evidence, see CFT, 7 Sept, 1906. 44. See the testimony of union representatives in PP 1892 (6705) XXXV, pp. 724–26. The union’s criticism is often taken as evidence that the limiteds were not true worker democracies. See Potter. My point is that a worker owned and run limited did not look very different in its organisation from a capital-run enterprise. 45. For the United States plywood industry, B. Craig and J. Pencavel. (‘Participation and Productivity: A Comparison of Worker Cooperatives and Conventional Firms in the Plywood Industry’, Brookings Papers on Economic Activity – Microeconomics, 1995, 121–174) found a positive relation between poor working conditions and labourmanaged firms.

Huberman

77

46. Oldham Standard, 12 July 1873; Oldham Chronicle, 31 July 1875. 47. S. Toms, ‘Financial Constraints on Economic Growth: Profits, Capital Accumulation and the Development of the Lancashire Cotton-Spinning Industry, 1885–1914’, Accounting Business and Financial History, 4:3, (1994), pp. 363–383; S. Toms, ‘Growth, Profits and Technological Choice: The Case of the Lancashire Cotton Textile Industry’, Journal of Industrial History, 1:1 (1998), pp. 32–56. Toms actually compares two mills: one a limited whose shares were widely held, a second whose share ownership was more concentrated. The latter’s rate of return was greater. The evidence is conflicting. One authority (cited in B. Law, Oldham Brave Oldham (Oldham, 1999), p. 93), comparing Oldham’s profit-sharing and private enterprises, observed that the limiteds had forced down total units costs to 2.5 pence per lb of yarn spun, but private concerns needed at least 1 pence per lb more to survive, a considerable gap in an industry that survived on narrow profit margins. 48. E. P. Lazear, ‘Personnel Economics: Past Lessons and Future Directions,’ Journal of Labor Economics, 17:2, (1999), pp. 199–236. 49. G. H. Wood, The History of Wages in the Cotton Trade during the Past Hundred Years (Manchester, 1910). 50. T. Ellison, The Cotton Trade of Great Britain (London, 1886), p.138. 51. D. A. Farnie and S. Yonekawa, ‘The Emergence of the Large Firm in the Cotton-Spinning Industries of the World, 1883–1938’, Textile History, 19:2 (1988), pp. 171–210; J. Garrard, Leadership and Power in Victorian Industrial Towns, 1830–80 (Manchester, 1983); P. Joyce, Visions of the People: Industrial England and the Question of Class, 1848–1914 (Cambridge, 1991); P. Taylor, Popular Politics in Early Industrial Britain: Bolton 1825–1850 (Bodmin, 1995). 52. Bolton and District Operative Cotton Spinners’ Provincial Association, 1884, BSC1/1/1, p. 81, John Rylands Library. 53. Ibid., 1895, BSC1/1/4, p. 63. 54. The scheme failed because ‘[m]ost of the workers in a cotton mill being paid by the piece, there [was] not the same opening for increased and improved work as where timewages are paid.’ PP 1894 (7458) LXXX, p. 628. This begs the question why firms that pay by the piece also see the need for introducing profit sharing. It is not uncommon for firms to combine incentive schemes (P. Milgrom and J. Roberts, ‘Complementarities and Fit: Strategy, Structure and Organizational Change in Manufacturing’, Journal of Accounting and Economics, 19, 1995, pp. 179–208). Under payment by the piece, workers have an incentive to overuse capital and produce poor quality output. Profit sharing would deter workers from these abuses because of incentives to protect capital investments and promote product quality. 55. I observe some differences between Oldham and Bolton with regard to spinner-piecer relations later in this chapter. 56. Farnie, English Cotton Industry, p. 260. 57. Bolton workers probably faced a higher degree of risk, because the town’s piece rate list did not have a built in time component (Huberman, Escape). Still, other towns that used the Bolton list, like Ashton, had a greater proportion of limited companies. 58. Dow and Putterman, ‘Why capital (usually) hires labor’. 59. P. Ekman. and R. J. Davidson (eds), The Nature of Emotion (New York, 1994); R. S. Lazarus, Emotion and Adaptation (New York, 1994). 60. G.A. Akerlof, ‘Social Distance and Social Decisions’, Econometrica, 65:5 (1997), pp. 1005–1027. 61. Lazear, ‘Personnel Economics’; see too J. M. Barron and K. P. Gjerde, ‘Peer Pressure in Agency Relationships’, Journal of Labor Economics, 15:2 (1997), pp. 234–254. 62. MacLeod, ‘Equity, Efficiency’. 63. Ekman and Davidson, Nature of Emotion. 64. The classic account is Max Weber’s comparative study of religion and economic growth in Europe. The anthropologist Ruth Benedict used shame and guilt to compare Japanese and Western ‘cultures’; R. Benedict, The Chrysanthemum and the Sword: Patterns of Japanese Culture (Boston, 1946). The approach lost favour in the 1960s and 1970s; G. Piers

78

When labour hires capital

and M. Singer, Shame and Guilt Cultures (New York, 1971). Recently, the concepts have resurfaced in classical studies: B. Williams, Shame and Necessity (Berkeley, 1993); legal studies: D. M. Kahan, ‘What Do Alternative Sanctions Mean?’ University of Chicago Law Review, 63:2 (1996), pp. 591–653; development economics, D. Lal, Unintended Consequences: The Impact of Factor Endowments, Culture, and Politics on Long-Run Economic Performance (Cambridge, Mass, 1998); and in political science: J. Elster, Alchemies of the Mind: Rationality and the Emotions (Cambridge, 1999). 65. Garrard, Leadership. 66. Taylor (Popular Politics, p.126) wrote: ‘The vulnerability of labour to the activities of innovating employers…played an important role in the greater subordination of labour to capital after the Chartist period’. See, also A. Howe, The Cotton Masters, 1830–1860 (Oxford, 1984), p. 142. Historians of paternalism have disputed its continuity. Joyce (Work, Society and Politics) asserted that paternalism after 1850 was a new initiative; Taylor (Popular Politics) and Brian Lewis (The Middlemost and the Milltowns: Bourgeois Culture and Politics in Early Industrial England (Palo Alto, 2001) have stressed its origins before mid-century. 67. The evidence on union membership and expenditures is from the Webb Trade Union Collection (Library of the London School of Economics and Political Science), Vol. XXXV, pp. 10, 22; Vol. XXXXVI, p. 35. 68. M. Savage, The Dynamics of Working Class Politics: The Labour Movement in Preston (Cambridge, 1987), p.237; P. Bolin-Hort, Work, Family and the State: Child Labour and the Organization of Production in the British Cotton Industry, 1780–1920 (Lund, 1989), pp. 158–161. 69. Farnie, English Cotton Industry, p.246. 70. Farnie, ‘Emergence’, p. 44. 71. Joyce, Visions, p. 30. 72. Farnie, ‘Emergence’, p. 45. 73. The standard critique of Foster is G. Stedman Jones, Languages of Class: Studies in English Working Class History, 1832–1982 (Cambridge, 1983). See, as well, R. A. Sykes, ‘Some Aspects of Working Class Consciousness in Oldham, 1830–1842’, Historical Journal, 23:1,1(980), pp. 167–80; D. Gadian, ‘Class Consciousness in Oldham and Other North-West Industrial Towns’, Historical Journal, 21:1 (1978), pp. 161–172, and S. L. Price, ‘Consensus Politics and Community in Oldham and Rochdale, 1790–1837’ (Unpublished Ph.D. thesis, University of Birmingham, 1997). For the debate on liberalism, see D. Gadian, ‘Radicalism and Liberalism in Oldham: A Study of Conflict, Continuity and Change in Popular Politics, 1830–52’, Social History, 21:3 (1996), pp. 265–280, and M. Winstanley, ‘Oldham Radicalism and the Origins of Popular Liberalism’, Historical Journal, 36:3 (1993), pp. 619–643. 74. G. Friedman, State Making and Labor Movements: France and the United States, 1876–1914 (Ithaca, 1998). 75. For a critique of Foster’s statistics, see R. Penn, Skilled Workers in the Class Structure (Cambridge, 1985). 76. J. K. Walton, Lancashire: A Social History, 1558–1939 (Manchester, 1987). 77. Around 1870 in Bolton and Oldham, a spinner earned more than double his junior worker. With joiner-minding in Oldham, the pay of big piecers and minders was about equal, each minder earning less than a spinner. It would have been expected that joinerminding would be more common in regions such as Bolton, where there were fewer alternative employment openings for older piecers (Bolin-Hort, Work, pp. 159–60; Lazonick, Competitive Advantage, p. 107). 78. Law, Oldham, p.93. 79. Tyson, ‘William Marcroft’. 80. Recall that the limiteds also took hold in Ashton. Like Oldham, Ashton saw a great wave of industrial disputes before 1850: M. W. Steinberg, Fighting Words: Working Class Formation, Collective Action, and Discourse in Early Nineteenth-Century England (Ithaca, 1999). 81. PP 1889 (5807) LXX, pp. 894–904.

Huberman

79

82. For examples, see the minutes of the Bolton and District Operative Cotton Spinners’ Provincial Association, 1884–1905, BSC1/1/1–22. 83. On the history of discharge notes or letters, see A. J. McIvor, Organized Capital: Employers’ Associations and Industrial Relations in Northern England, 1880–1939 (Cambridge, 1996), p. 39. On fines in Oldham, see PP 1892 (6705) XXXV, p. 734. 84. Bolton Chronicle 29 October 1836. 85. PP 1892 (6705) XXXV, p. 734. 86. PP 1889 (5807) LXX, pp. 894–904. 87. For examples, see Bolton and District Operative Cotton Spinners’ Provincial Association, 1884–1905, BSC1/1/1–9. The quote is from 1895, BSC1/1/4, p. 8. 88. R. Q. Gray, The Factory Question and Industrial England, 1830–1860 (Cambridge, 1996). 89. R. M. Ryan and E. L. Deci, ‘Self-Determination Theory and the Facilitation of Intrinsic Motivation, Social Development, and Well-Being,’ American Psychologist, 55:1 (2000), pp. 68–79. 90. D. M. Kreps, ‘Intrinsic Motivation and Extrinsic Incentives,’ American Economic Review, 87:2 (1997), p. 362. 91. MacLeod, ‘Equity’; Milgrom and Roberts, ‘Complementarities’. 92. Garrard, Leadership, pp. 164, 176; Howe, Cotton Masters, pp. 111, 146. 93. Law, Oldham, pp. 133–172. 94. Cited in Garrard, Leadership, p. 136. 95. While Bolton had forty-three cooperatives and friendly societies with about 5,000 members in 1874, Oldham had more than 200 such organisations with about 12,000 members (PP 1874 (995) XXII, pp. 278, 301). Over the period the number of cooperatives actually declined in Oldham as they became larger. See note 39. 96. On Bolton, see R. Boyson, The Ashworth Cotton Enterprise: The Rise and Fall of a Family Firm, 1818–1880 (Oxford, 1970), p. 187; for Oldham, see L. H. Lees, The Solidarities of Strangers: The English Poor Laws and the People, 1700–1948 (Cambridge, 1998), p. 236. 97. Lindert made a similar point in comparing British and continental poor-relief systems, P. Lindert, ‘Poor Relief Before the Welfare State: Britain versus the Continent, 1780–1880’, European Review of Economic History, 2:2 (1998), pp. 101–140. 98. For recent surveys of the debate between rings and mules, see Marrison, ‘Indian Summer’, and Rose, Firms. 99. The purchase of shares in limiteds other than their own gave Oldham workers only a small degree of portfolio diversification, because in a competitive industry, like cotton textiles, the fortunes of all firms tended to rise and fall together. My research suggests alternative reasons why workers bought shares in competing concerns: first, at the time of the initial offer of shares at their own firm, workers who faced some form of credit constraint might not have been able to purchase equity; second, Oldham workers held shares in competing concerns because of the strong bonds between workers across mills. 100. For a model of this type, see E. P. Lazear and R. Freeman, ‘Relational Investing: The Workers’ Perspective,’ NBER working paper, no. W5436, 1996. 101. Procter and Toms (‘Industrial Relations’) argue that rings did not have a wage advantage at a given count spun. 102. In fact, unemployment and retirement packages were more generous in Bolton. See Bolton and District Operative Cotton Spinners’ Provincial Association, 1900, BSC1/1/7, p. 6. Pensions in Bolton amounted to 5–10s. per week (that is, about 10–20 percent of a normal week’s earnings). 103. The spinners’ union in Oldham lent £2000 to a limited whose majority owner was John Bunting, the town’s leading speculator – and richest man – in 1896. Oldham and District Operative Cotton Spinners’ Provincial Association (TU1/1/26). 104. CFT, 12 February 1904. 105. Toms, ‘Financial Constraints’; Toms, ‘Growth, Profits’.

80

When labour hires capital

106. On architects and builders and their preference for mules, see R. N. Holden, ‘The Architect in the Lancashire Cotton Industry, 1850–1914: The Example of Stott & Sons,’ Textile History, 23:2 (1992), pp. 243–257. 107. Farnie’s numbers cited in D. Gurr and J. Hunt, The Cotton Mills of Oldham (Oldham, 1998), p. 30. 108. T. Leunig, ‘The Myth of the Corporate Economy’. Unpublished Ph.D. thesis, University of Oxford, 1996. 109. Leunig (Myth, pp. 92–93) gave these figures for the share of total output (weaving capacity) spun by sub-40 rings: Ashton: 7.9 percent (53 percent); Bolton: 7.9 (100); Oldham: 8.5 (7); Rochdale: 42.9 (100); Stockport: 22.7 (66). 110. Steinberg, Fighting Words. 111. Rochdale was known for its consumer cooperatives, but had few worker-run or controlled factories. In comparison with Oldham, the town faced fewer obstacles to make the switch to rings. There were more female workers in Rochdale’s textile factories and the trade union movement was weaker. 112. A. Schotter, ‘Worker Trust, System Vulnerability, and the Performance of Worker Groups’, in A. Ben-Ner, L. Putterman (eds), Economics, Values and Organization (New York, 1998), pp. 364–406. 113. S. Bowles and H. Gintis, ‘How Communities Govern: The Structural Basis of Prosocial Norms’, in A. Ben-Ner, L. Putterman (eds), Economics, Values, pp. 206–229. 114. Dow and Putterman, ‘Why Capital (Usually) Hires Labor’.

CHAPTER 4

GIVING NOTICE: THE LEGITIMATE WAY OF QUITTING AND FIRING

(GHENT, 1877–1896) Patricia Van den Eeckhout1

Belgian historians studying labour and wage relations in the nineteenth and twentieth centuries are not consumed by an overwhelming curiosity for what happened on the shop floor. The decision making process regarding social legislation and the political and institutional history of collective labour relations appear to figure much higher on the research agenda.2 Labour relations are thus predominantly perceived from the point of view of the political historian, more interested in shifting balances of political power than in the daily practices that shaped relations between employer and worker. Only a few studies descend to the level of industrial sector or shop floor and manage to treat labour relations as power relations with concrete and varying forms, instead of an abstract category with an unspecified content.3 The mentioned studies tend to focus on the most visible and explicit manifestations of labour relations, such as technological and organisational changes, strikes and wage tariffs. However, the fabric of wage relations is not merely shaped by more or less clear-cut formal arrangements. The importance of self-evident, informal and implicit understandings between workers and employers can hardly be underestimated, although they are more difficult to capture. Custom figures prominently among the more implicit components of nineteenth-century daily labour relations.4 This Chapter investigates the custom of giving notice, a practice related to termination of employment that has been hardly explored in historical research. It discusses how Ghent workers and employers activated and defined the custom and circumscribed the conditions allowing a legitimate refusal to apply it. Despite the steady permanence suggested by the word

82

Giving notice

‘tradition’, custom was a field of change and contest, so E. P. Thompson states.5 In that respect the custom of giving notice raises some questions. It appears that Ghent workers only discovered their ‘traditional’ right to a proper notice in the last quarter of the nineteenth century. Far more difficult than the references to the custom is the study of its history. Are we dealing here with a custom of recent invention or with workers’ appropriation of a tradition that, in the past, could only be claimed by their employers?

The Conseil de Prud’hommes The forum on which workers and employers presented their views regarding a proper termination of employment were the sessions of the Ghent Conseil de Prud’hommes, an institution whose jurisdiction played an important role in sanctioning the interlocutor’s interpretations. Kathlijn Pittomvils was the first Belgian scholar who explored the Conseil de Prud’hommes’ archives in her study on labour conflicts in Ghent in the first half of the nineteenth century.6 The Conseil de Prud’hommes was a kind of labour court introduced in 1810 under French rule. Ghent, the centre of the Belgian cotton industry, was the first Belgian town to have one installed. Other cities and industrial centres followed. One of the tasks assigned to the Conseil de Prud’hommes consisted in reconciling in minor conflicts between employers and workers individually. In the period considered by Pittomvils, workers were not represented as judges in the Conseil de Prud’hommes, only employers were, but in contrast to what one might expect, it did not merely serve the interests of the latter. Pittomvils’ findings present an important correction to the familiar dichotomy of a powerless worker and an omnipotent employer. The law of 7 February 1859 allowed ‘respectable’ workers to elect their representatives in the Conseils de Prud’hommes.7 The electorate was further broadened by the law of 31 July 1889 that dropped the precondition to be able to read and write and abolished the discretionary powers of local and provincial authorities in the elaboration of the list of voters.8 In the period under consideration in this paper, workers and employers were thus represented in equal numbers. Since the 1870’s the elections for the Conseil de Prud’hommes became one of the battlefields on which the emerging Ghent social democrat and Christian labour movement met. When either an employer or a worker had complaints regarding tout fait d’ouvrage, de travail et de salaire,9 they could bring their case before the Conseil de Prud’hommes. Within the latter a bureau de conciliation (verzoeningsbureel) was formed that held a session at least once a week. The bureau de conciliation was composed by two members, one member chosen among the elected representatives of workers, the other one chosen among the elected representatives of employers. Each case had to pass before the bureau de conciliation first. However, quite a few cases were resolved without any formality by the clerk of the court. The latter invited the two parties to attend the reconciliation session, in which all the possibilities of

Van den Eeckhout

83

compromise had to be exhausted. If it failed, the matter was taken before the bureau général (algemeen bureel), composed by an equal number of prud’hommes patrons and prud’hommes ouvriers. If the latter session was unsuccessful in its mediation, a sentence was pronounced, if necessary by default. In the meetings of the bureau de conciliation and the bureau général the parties involved explained their case, developed their argument and called witnesses if necessary. The minutes of the sessions of the bureau de conciliation are usually very succinct. They mention the plaintiff’s claim and the result of reconciliation. The minutes of the sessions of the bureau général, dealing with the cases in which no agreement was reached, are more elaborate. The arguments developed by the two parties are summarised, and so are the considerations on which the sentence of the bureau général is based. Although they are more explicit, the minutes of these sessions have a highly formalised character and they contain a lot of standard phrases. Moreover, we are not confronted with the words of the parties involved, but with the clerk of the court’s summary of their statements. The circumstances some of the arguments refer to are not always clear, given the use of euphemisms or very general descriptions of very particular situations. In some cases the reconciliation files contain more explicit information. The clerk of the court, who attended the sessions of the bureau de conciliation, made a note for the chairman of the Conseil de Prud’hommes who presided the session of the bureau général. In these notes the clerk summarised the discussions in the conciliation meeting in a less formal and sometimes even familiar tone. Occasionally, he made it very clear whose arguments he considered most legitimate. Additional information is sometimes provided by letters the parties addressed to the chairman of the Conseil de Prud’hommes. Despite their shortcomings the minutes of the sessions of the Conseil de Prud’hommes provide a glimpse of the lived experience of worker/employer relations. The arguments presented by the plaintiffs and defendants and the considerations on which the judgements of the bureau général were based, give an idea of what was considered legitimate in the fabric of mutual rights and obligations.

Conflicts regarding termination of employment 1877–189610 In the first half of the nineteenth century conflicts regarding termination of employment were an important source of disagreement.11 Between 1877 and 1896 the relative importance of this type of conflict decreased (see Figure 4.1), due to a growing number of complaints related to wage payment. Table 4.1 presents details regarding the specific character of the cases investigated. The majority of conflicts related to termination of employment had to do with the employers’ refusal to grant dismissed workers the customary period of notice. Column II lists the number of workers claiming due notice. Only in a small number of cases employers took the initiative for a straightforward request for eight or fourteen days’ notice

84

Giving notice

(column III). Sometimes they used indirect ways of reminding workers of their obligations. When workers introduced a case, asking for their livret, their wages or their tools, this could result in the employers’ request to fulfil their eight or fourteen days’ notice first (column VI). In a limited number of cases workers and employers did not make an explicit request for eight or fourteen days’ notice, but asked for the permission (column IV) or the obligation (column V) to finish a task. Column VII mentions the number of workers asking for their livret, a proper congé or their tools, but without any reference (in request or outcome) to the obligation of giving notice. Not every conflict concerning these matters was related to the latter. Table 4.1. Cases related to termination of employment, Ghent 1877–1896

1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896

I

II

III

IV

V

VI

VII

218 135 126 125 215 202 201 204 105 136 211 226 186 221 232 289 327 287 421 211

75 48 37 29 53 40 29 14 17 17 31 49 32 42 44 56 140 78 99 45

7 4 3 13 2 13 5 2 4 1 1 14 2 5 15 -

2 1 2 6 2 1 1 6 1 4 4 1 -

8 3 3 3 2 4 2 2 3 3 3 2 1 6 2 1 2 7 1

54 4 8 10 7 16 7 4 1 3 2 1 3 6 1 -

43 39 36 43 49 35 21 85 8 24 15 5 10 8 12 8 8 12 14 14

I: total number of cases brought before the Conseil de Prud’hommes; II: workers requesting 8 or 14 days’ notice (sometimes combined with the request to pay wage arrears); III: employers requesting 8 or 14 days’ notice; IV: workers requesting to be able to finish what they started; V: employers urging workers to finish what they started; VI: workers requesting their livret, wages and/or tools, resulting in the employer’s request for 8 or 14 days’ notice; VII: workers requesting livret, congé and/ or tools

In the years 1877–1896 mostly workers made complaints because the customary period of notice was refused (column II). This contrasts with the first half of the nineteenth century when workers seemed eager to leave as quickly as possible. In those years, the majority of cases introduced by factory workers before the Ghent Conseil de Prud’hommes dealt with the fact that employers refused to let them go.12 In the artisanal trades as well, the employers’ efforts to hinder the departure of their workers was an important source of conflict.13 In the statistical surveys of the Ghent

1889

1888

1887

1884

1883

1882

1881

1880

1879

1878

1877

termination of employment

1890

total number of cases

1891

1886

1885

Figure 4.1. Conflicts brought before the Ghent Conseil de Prud’hommes 1877–1896 (based on table 4.1)

0

50

100

150

200

250

300

350

400

450

Van den Eeckhout 85

1896

1895

1894

1893

1892

86

Giving notice

Conseil de Prud’hommes’ activities, available for the years 1844, 1845, 1846, 1849, 1850, 1851, 1859, 1861 and 1862, the category ‘giving notice’ was completely absent.14 Categories dealing with termination of employment referred to the employers’ refusal to allow workers to leave, either by denying them their livret, a proper congé or wages due. In other words, the employers’ practices hindering the departure of their employees were at the centre of these conflicts. Since the early 1870s (we have no records for the 1860s) the category ‘giving notice’ emerged. It appears that the right to a proper notice, a claim mostly made by workers, now figured prominently in the conflicts regarding termination of employment. How can this shifting focus be explained? Requesting period of notice was certainly not a new custom. K. Pittomvils mentions it occasionally for the first half of the nineteenth century, but always in connection with employers’ strategies to retain employees.15 Apparently workers ‘discovered’ that custom in the last quarter of the nineteenth century. In the actual state of research it is difficult to explain why they did not use it before. Are we dealing with changing power relations in the Ghent labour market, weakening workers’ positions? The study of the Ghent labour market might teach us more, if it were available. Perhaps the ‘democratisation’ of the Conseil de Prud’hommes since 1859 played a role. The inclusion of workers among its judges might have facilitated the introduction of the idea that obligations had to be reciprocal, encouraging more workers to complain when they were not granted what every employer expected: a proper notice.

The livret and the proper congé In the first half of the nineteenth century a lot of conflicts regarding termination of employment were related to the livret d’ouvrier. The statute of 9 frimaire XII (1 December 1803) decreed that every worker had to possess a livret.16 He or she had to hand it over to the employer, who kept it until the engagement ended. If the worker wanted to leave, he or she did not only require the livret but also a congé portant acquit de leurs engagemens, s’ils les ont remplis. Without the livret and the congé, testifying that he or she had fulfilled all the engagements towards the former employer, workers were supposed to be unable to find another job. The statute of 22 Germinal 11 (12 April 1803) prohibited employers of hiring workers without the livret and the proper congé.17 Although workers and employers often tried to circumvent this legislation, the employers’ refusal to return a livret or to deliver a proper congé was an important source of conflict in the first half of the nineteenth century.18 R. Steinfeld remarks, however, that the French regime of contract rules established a legal form of wage labour that was less coercive than English wage labour. Under the 1823 Master and Servant Act, English employers could have their workmen sent to the house of correction and held at hard labour for up to three months for breaches of their labour

Van den Eeckhout

87

agreements. The latter included quitting before one had served out one’s term or one’s notice.19 The act of 10 July 1883 made the livret optional.20 It depended from the arrangements in specific sectors and workshops whether it was requested. These legal changes explain the declining number of cases concerning the livret and congé brought before the Ghent Conseil de Prud’hommes after 1883 (column VII). The exceptionally high number of cases in 1884 (eighty-five) is due to the fact that seventy weavers at once accused their employer de Leener of withholding their livret. The plaintiffs withdrew their complaint.21

The contract between employer and worker The arrangements regarding termination of employment were part of the contract between worker and employer. Until 1900 Belgian legislation regarding contracts between these two parties was almost non-existent. The Napoleonic Code civil distinguished a contrat de louage de travail, and article 1780 specified that hiring someone’s services was only possible for specific tasks or a limited period of time.22 Contracts between workers and employers were regarded as ordinary civil contracts.23 The law of 10 March 1900 introduced the concept of a specific employment contract (contrat de travail). In its motivation the government referred to the fact that since the abolition of the guilds, labour relations had been governed by the very general principals of civil law regarding contracts and by local customs. The latter were claimed to be arbitrary and uncertain.24 By the players on the field the term ‘contract’ was not interpreted univocally, so it appears from the inquiry conducted in 1886 by a parliamentary commission.25 The answers to question 24 ‘how are contracts between employer and worker established in your industry?’ reveal that some of the employers associated a contract with a written document. Others identified a contract as any written or oral agreement between worker and employer. Given the absence of clear-cut definitions of what constituted an employment contract, the Conseils de Prud’hommes played a crucial role in defining an agreement as a relationship of employer and employee. For the Ghent Conseil de Prud’hommes both oral and written agreements were considered legitimate contracts. It also appears that according to the Ghent Conseil de Prud’hommes a contract was not necessarily limited to the oral or written agreement between an employer and each worker individually. Workshop regulations that had been visibly exposed in the workplace were also regarded as contracts.26 Those who worked there were supposed to have agreed with them at least implicitly. Although a worker’s freedom in his negotiations with a future employer was very relative, workshop regulations went a step further. They were clearly one-sided conventions, imposed by the employer, which were either to take or to leave.27

88

Giving notice

The Ghent Conseil de Prud’hommes had no problem accepting workshop regulations as legitimate contracts. Apparently they regarded them as ordinary civil contracts established by two allegedly free and equal individuals. In the arguments I find no reference to a decision of the Ghent bench of aldermen of 10 February 1810, stating that workshop regulations were associated with laws and thus were to be obeyed.28 Another element strengthening the legitimacy of workshop regulations was the fact that they were regarded as a valuable alternative for tradition. In two published reports regarding the activities in 1877 and 1881 the chairmen of the Ghent Conseil de Prud’hommes deplored that tradition was under a lot of pressure. Both the custom of giving notice (see further) and carrying a livret (presented here as part of tradition) were disappearing rapidly, thus undermining the durability of good labour relations. Fortunately, the chairmen argued, workshop regulations were becoming more numerous, which was the more beneficial since they replaced the uncertainty of oral agreements with unambiguous written rights and obligations. The Ghent Conseil de Prud’hommes introduced a register in which workshop regulations were recorded in order to give their content a public character.29 This attitude contrasts with the point of view of the French Conseils de Prud’hommes, albeit in the first half of the nineteenth century. The latter declared the war to these unilateral conventions, which had the tendency to surface when employers were unsatisfied with the ‘leniency’ of the Conseils de Prud’hommes’ jurisprudence.30 The law of 15 June 1896 making workshop regulations compulsory in Belgium for all companies employing at least ten (five since 1900) workers is in complete agreement with the Ghent Conseil de Prud’hommes’ practices in the years 1873–1896.31 Presented as an improvement of the workers’ legal security, the law codified the employers’ right to establish their own ‘local legislation’, although a procedure of consulting the personnel was introduced. The employer, however, had the last word. Most of the employers answering question 24 in the 1886 inquiry reported that in their establishment or in their industry, oral agreements were the rule. The length of these agreements was usually not specified and they were implicitly renewed every eight or fifteen days. From the cases presented before the Conseil de Prud’hommes, it appears that in Ghent also oral agreements predominated, but that some employers had introduced written workshop regulations. Even with a broad interpretation of the term, it was not always clear whether there was effectively a contract between worker and employer. The Ghent Conseil de Prud’hommes was confronted regularly, in different sorts of cases, with parties denying that there had ever been an agreement between them. Until 1883 the beginning or the existence of such an agreement could in principle always be proven by the fact that the employer was in possession of the worker’s livret, at least until it became optional in 1883. In sentences regarding the factory worker Pierre T’Sestig and the weaver Victor Van Beneden the Ghent Conseil de Prud’hommes stated that the act of handing over and accepting the livret

Van den Eeckhout

89

established the contract between employer and worker.32 Of course, not all employers and workers respected the livret legislation. One Ghent employer, the sugar factory La Ruche, even refused to accept the workers’ livret in an attempt to prevent them from addressing the Conseil de Prud’hommes. La Ruche was convinced erroneously of the fact that the jurisdiction of the council was limited to those whose contract had been sealed by the handing over of the livret.33 In the absence of a livret, the inscription in a personnel register or on wage lists could prove the mutual engagement. In all other cases the parties involved had to be confident that the oral agreement would not be contradicted. Ghent employers in general, associated with the Chamber of Commerce, were very much in favour of the livret, since it represented a constant moral and material reminder of the contract between worker and employer. As such it was the guardian of mutual rights and obligations and an antidote to irregularities and insubordination. Workers especially, so they argued, had to be reminded of their duties.34

The custom of giving notice An integral part of the agreement between worker and employer was the arrangement regarding termination of employment. The inquiry of 1886 reveals that in most regions and industrial sectors workers and employers had the obligation to announce in advance that the engagement ended. Both parties had to give notice either eight or fifteen days in advance.35 In Ghent the period of notice was eight or fourteen days (veertien dagen), although the sentences dressed in French speak of le préavis de quinzaine.36 Local customs formed the basis of this obligation, in Ghent and elsewhere. Neither plaintiffs nor defendants before the Ghent Conseil de Prud’hommes contested the overall legitimacy of these local customs. But quite a few argued that their particular case was a legitimate exception. In the advisory Conseil Supérieur du Travail preparing the law of 10 March 1900 regarding the employment contract, criminal law specialist Adolphe Prins suggested that the custom of giving notice tended to disappear in Belgium.37 Some twenty years earlier the chairmen of the Ghent Conseil de Prud’hommes had already deplored this development. The subsequent discussion in the Conseil Supérieur du Travail neither confirmed, nor contradicted this statement and focussed on the question whether giving notice should be made compulsory. While he asserted that a lot of workers were against it, the socialist Hector Denis defended the principle in the name of stability and morality. Adolphe Prins and Victor Brants, two followers of the conservative sociologist Frédéric Le Play, were in favour of the principle but argued that the objections of workers and employers made them change their mind.38 The general feeling was that one could not go counter the alleged habits and practices in the workshops. The reluctance to impose stringent regulations regarding dismissals is translated by article 19 of the law of 10 March 1900. It made period of notice of at

90

Giving notice

least seven days compulsory but the latter was subordinate to particular agreements, local customs or workshop regulations that stated otherwise.39 When workers or employers claimed they had not been granted proper notice or were not allowed to finish a task, they brought their case before the bureau de conciliation. Workers and employers of all kinds of trades addressed the Conseil de Prud’hommes. Most numerous were those of the textile and metal sector followed by printing, building, paper and tobacco. In the majority of cases dealt with by the Ghent Conseil de Prud’hommes in the years 1877–1896, parties were reconciled (Table 4.2). A sentence by the bureau général was the exception. In a certain number of cases there was no conciliation but not a sentence either.40 In most cases ending in reconciliation, the defendant either paid a financial compensation or agreed to grant eight or fourteen days’ notice (Figure 4.2). Since the 1880’s the proportion of cases ending in the payment of a financial compensation showed an increasing trend, which could point to a growing monetarisation of labour relations. Table 4.2. Outcome of cases related to the custom of giving notice, Ghent, 1877–1896

1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896

Total

Sentence

Conciliation

Unclear

146 59 52 55 66 79 45 23 22 22 34 55 36 51 54 76 153 87 121 46

27 22 9 6 10 16 2 4 3 0 1 12 3 12 8 9 40 9 2 6

103 31 36 41 32 45 37 15 12 15 15 29 27 29 29 48 65 58 59 34

16 6 7 8 24 18 6 4 7 7 18 14 6 10 17 19 48 20 60 6

If there was no compromise and a sentence was pronounced, the plaintiff always asked for financial compensation, since the relationship between the opponents was probably too tense to spend another eight or fourteen days together. In the sentence regarding the case of a fitter employed by La Gantoise, the Conseil de Prud’hommes stated that giving notice could only be understood in an atmosphere of good relations between the two parties. Since the fitter in question had cursed and

1886

1885

1884

1883

1882

1881

1880

1879

1878

1877

Figure 4.2. Outcome of cases ending in reconciliation, Ghent, 1877–1896

0

20

40

60

80

100

120

diverse financial compensation granted period of notice granted

Van den Eeckhout 91

1896

1895

1894

1893

1892

1891

1890

1889

1888

1887

92

Giving notice

threatened an overseer, he had himself undermined the conditions for a proper period of notice.41 For the parties involved and for the members of the Conseil de Prud’hommes the whole procedure regarding giving notice was so selfevident that they very seldom specified what was exactly meant by eight or fourteen days’ notice or the compensatory financial equivalent. The case of Adolph Reynaert against machine builder Phenix gives some clarification. On a meeting of the bureau de conciliation Phenix had agreed to pay Reynaert a financial compensation as an equivalent of eight days’ notice. We learn that Reynaert expected to receive wages for six working days at 3.30 franks a day. Only the working days of the period of notice were to be included in the financial compensation, so the bureau général stated. While Reynaert did not expect money for Sunday, he contested the fact that he was not paid for Monday. The bureau général argued that Monday, 19 July 1875 was a day off for all workers in Ghent given the annual Ghent fair and therefore he was only entitled to the equivalent of wages for five working days.42 When wood sawyer Desideer Claeys had barrel maker Lodewijk Van Damme summoned for sending him away without eight days’ notice, he specified that the requested financial compensation corresponded to a week’s wages.43 We thus learn that eight days’ notice usually refer to six working days and that the financial compensation for not granting eight days’ notice, was the equivalent of the wages that could have been earned in the working days of the corresponding period. It is not clear which elements determined whether an industrial sector or a company were in the habit of granting either eight or fourteen days’ notice. A particular sector or occupation cannot be associated with a specific period of notice. A sentence regarding a postman suggests a relationship between the frequency of wage payment and the period of notice. Since postman Pieter Vanderhaeghen was paid per month, he deserved a month’s notice, the Conseil de Prud’hommes argued.44 In a sentence regarding the weaver Jan Soudain employed in the factory of Baertsoen & Buysse, however, weekly payments appeared to be combined with fourteen days’ notice.45 The mechanic Alfred Murier was paid per month, but he requested the quinzaine.46 The length of the period of notice was probably determined by the duration of the engagement between worker and employer. The latter was implicitly renewed until notice was given. The frequency of wage payment could but did not have to coincide with the length of the period of notice. Anyhow, the proportion of cases in which eight days’ notice was requested instead of fourteen, increased in the course of the period we studied (Figure 4.3). The categories mentioned in the yearly statistical survey of the Ghent Conseil de Prud’hommes’ activities confirm this development implicitly.47 For the years 1875 to 1881 conflicts regarding eight and fourteen days’ notice were mentioned as two separate categories (quinzaine; huitaine). For the years 1882 to 1890 these conflicts were mentioned as one category (quinzaine ou huitaine). From 1891 on the notion ‘fourteen days’ notice’ disappeared from the statistical survey and only the category ‘eight days’ notice’ (huitaine) remained.48

Van den Eeckhout

93

We already referred to the complaint of the chairmen of the Ghent Conseil de Prud’hommes in 1878 and 1882 that tradition and more specifically the custom of giving notice was under pressure. The chairmen deplored ties were loosening between worker and employer. The shortening of the period of notice could be another symptom of more flexible labour relations. When notice was granted the clauses of the initial agreement had to govern the employment relationship until the very end. When Jean Van Moerkerke fired Pierre T’Sestig, the latter was not only entitled to fourteen days’ notice, but the working hours and wages in the latter period had to be those agreed upon in the initial engagement.49 Henry Andresen, a worker in Kaminski’s factory of gilt frames, was used to work for a contract wage. Andresen wanted to quit and was prepared to do his fourteen days’ notice. Kaminski, however, wanted to change the terms of the initial agreement and urged Andresen to conclude his term by working for a daily wage. The Conseil de Prud’hommes decided that the employer was no longer entitled to fourteen days’ notice and had to return the livret to Andresen, so that the latter was free to go.50 The machine maker Carels was sentenced to pay the engineer Julius Fonteyne compensation because he did not allow him to work as an engineer during his eight days’ notice. Instead he forced him to perform a humiliating task, which was even impossible to do in the winter season.51 Jan Van Hoorde and Theodore Hendricx, weavers at Monckarnie’s, were successful in claiming financial compensation for the fact that they were not allowed to do eight days’ notice at the wage they were accustomed to. Weaver Jacques Verspiete received compensation as well: Monckarnie had changed his assignment, which prevented him to earn his regular wage. In the argument presented by the Conseil de Prud’hommes, it appears that not the wage decreases as such were condemned, but rather the fact that Monckarnie had failed to announce it eight days in advance. In the Ghent weaving sector changes in working conditions and wage rates, so the conciliation board argued, always had to be negotiated eight days beforehand. It seems that changes in that respect were regarded as the beginning of a new contract between the two parties.52

Legitimate objections to giving notice Employers trying to fire their employees without proper notice or financial compensation had to come up with a credible story. Complaining about a worker’s performance was not enough. The coppersmith Isidore Vandeweghe’s objection that he was not satisfied with Victor Devylder’s work appeared to be unconvincing and the latter received the full financial compensation he requested.53 Almost twenty years later Isidore Vandeweghe was himself refused proper notice by his employer Willem Seebach, because he had delivered sloppy work. Work had been short in the days before he was fired, remarked the plaintiff Vandeweghe, and the Conseil de Prud’hommes granted him a financial compensation because his

94

Giving notice

boss did not offer to prove the mistakes he allegedly made.54 Cigar worker Pieter D’haese had the same experience and cigar manufacturer Pieter Boon had to pay him the compensation he asked for.55 Contradictory declarations also led to the dismissal of the argument. Brewer Van Mele fired brewer’s helper Mathieu Pauly without notice. He declared that Pauly had caused him serious damage. To strengthen his case he added that Pauly had initially accepted his wages and a certificate without requesting period of notice. The Conseil de Prud’hommes pointed to the contradiction in granting a worker a certificate and an allegedly bad performance, and Pauly received the financial compensation he asked for.56 Even when workers’ mistakes were proven, the Conseil de Prud’hommes reaction was not necessarily harsh, provided they were isolated occurrences and not committed in a spirit of insubordination. Camille Canderlier, workman in the factory of P. Smets, admitted that he had been sleeping on the job (less than half an hour though). The Conseil de Prud’hommes gave him a fine but judged that he was entitled to financial compensation for denying him eight days’ notice.57 Decoster, a wood merchant, had accepted Eduard Timmerman as an apprentice. He fired him without notice because he had cut the wood in wrong measures. One single mistake by an apprentice was not enough reason to deny him a proper notice, the Conseil de Prud’hommes argued, and Timmerman collected the full financial compensation he asked for.58 However, when railwayman Aloïs Kesteloot contested being fired without a month’s notice, customary in the Société du Chemins de fer de Gand-Terneuzen, he lost his case. The company could prove that it had sent him a letter, threatening to fire him if, after a series of grave errors, he committed one more mistake.59 In general, insubordination and a lack of respect were seen as legitimate reasons to deviate from the local custom of giving notice. There was no mercy for Joseph Pollet, a nightwatchman at Rey’s textile factory, who had been sleeping on the job. Two workers of Rey’s, reporting a conversation with Pollet, testified that the latter had been boasting that it was not the first time he took a nap and that he did not bother whether the master reprimanded him: the summer was near and work was abundant.60 Judocus Decleene was fired on the spot, because, so witnesses testified, he had been swearing at his master. The Conseil de Prud’hommes acknowledged the gravity of job loss, but argued that Decleene was also rightfully denied fourteen days’ notice. The council could not indulge Decleene to give a bad example for another fortnight and earning a wage in the course of it.61 A financial compensation was also out of the question. The weaver Francis De Belder, who did not only curse, but was also accused of mocking his master regularly, received the same treatment.62 The weaver Jan Stevens was not successful in his argument that a conviction by the Ghent police court for swearing at the overseer in Charles de Hemptinne’s factory, was not a sufficient reason for denying him a proper notice.63 Louis Mestdagh was accused of insulting and threatening one of the administrators of the spinning mill Filature du Nord in front of the other workers, after being reprimanded. He had no right to a proper notice, the

1885

1884

1883

1882

1881

1880

1879

1878

1877

Figure 4.3. Length of the period of notice at stake, Ghent, 1877–1896

0

20

40

60

80

100

120

140

160

180

8 days’ notice

14 days’ notice

1 month or more

not specified

Van den Eeckhout 95

1896

1895

1894

1893

1892

1891

1890

1889

1888

1887

1886

96

Giving notice

Conseil de Prud’hommes concluded.64 The brewer Alfons Kinsoen acted correctly, according to the Conseil de Prud’hommes, when he fired Adolf Debanck without notice since he could prove that the latter had been taken to court for ‘destruction of pets’.65 If proper notice was granted, it could be interrupted if the worker behaved badly. The baker Hendrik Goudenhoofd was sent away by the cooperative Volksvrijheid before his eight days’ notice were complete. He admitted to ‘committing deeds that disturbed the order in the shop and destroyed the understanding between boss and workman’.66 For all these allegations employers needed proofs, witnesses or confessions. Without them, they lost their case. Engineer Ferdinand Bracke was fired without notice by Victor Delin because he had been too drunk, so Delin stated, to perform his duties. Delin made no effort to prove his accusation and was sentenced to pay Bracke the compensation he requested.67 Not only swearing, sleeping and drunkenness but also the alleged refusal to obey orders was a legitimate reason to dismiss every right to a proper notice. The printer Adolf Hoste was justified to send typographer Julius Van Varenberg away without notice, since the latter refused to follow Hoste’s instructions.68 The paper factory Papeteries Gantoises had every right to fire engineer Jozef Desmet without notice. Witnesses, formerly employed by the factory, testified that Desmet had stopped the machines without permission and this in order to be able to leave the shop early.69 For some forms of disobedience there appeared to be more comprehension. Petrus Vanhaudenhove was sent away without the three months notice, required according to his written contract with the lemonade producer and trader Van Coppenolle. A message of the latter to the Conseil de Prud’hommes tells the story. The worker had asked his employer if he could stay at home on a Monday afternoon, since the July fair was in Ghent and his family had come over to visit him. The employer refused precisely because those days demand for lemonade was very high. Vanhaudenhove did not show up (in fact he surfaced on Wednesday after the fair had ended). As a result, so Van Coppenolle’s wife relates, her husband had to risk his life, leaving his sickbed to take care of business. Van Coppenolle’s tear-jerking letter did not make a big impression. In his summary of the case the clerk of the court described the facts as ‘a futility’. The Conseil de Prud’hommes seemed to agree with him and Vanhaudenhove received a substantial compensation for denying him three months notice.70 Usually, however, there was no tolerance for those who ignored their masters’ orders. The master himself, on the other hand, had to exercise his authority. Angelus Degrave, a worker employed by tool maker Nollet, protested because he was fired without notice for refusing to obey orders. Since three overseers had given contradictory instructions, so the plaintiff argued, he had chosen to obey his most direct supervisor. The Conseil de Prud’hommes concluded that Degrave did not have the intention of

Van den Eeckhout

97

neglecting his workman’s duties and that his master had failed to restore order in the workshop. Degrave received the compensation he requested.71 A specific category of disobedience was ‘leaving the shop without permission’. Plaintiffs accused of the latter, were unable to convince the Conseil de Prud’hommes that they were entitled to compensations. The sugar factory La Ruche blamed their worker Pius Desmet for leaving on his own initiative and as witnesses confirmed this allegation, he left empty handed.72 The argument was also used successfully against truss maker Karel Pichler73, weaver Jules Veniers74, tailor Victor Dauwe75, engineer Jacobus Bartholomé76 and brewer’s helper Bruno De Ronne who all learnt that their complaint had no ground.77 ‘Leaving the shop without permission’ is a rather elastic accusation that could hide a diversity of situations. In the case of Augustien Lejour, engineer in the sawmill of Job, Maes & Cie, the clerk of the court’s summary of the conciliation meeting relates that Lejour left his job on Monday morning together with other workers. When they returned to work on Tuesday they were sent away because they had broken their contract.78 Now and then we get a glimpse of a conflict after which the worker ‘took the initiative’ to leave the shop. Charles Lemayeur and four colleagues left the factory of Cosmel Jeune et Compagnie without permission and after ‘revolting against their master’.79 They were denied all compensations. That was also the case for the shoemakers Karel Lessenier and Hendrik Landuyt who left after having heard that their wages would be reduced.80 The weaver Edouard Brabandt and six of his colleagues argued they thought they could leave the shop, because they were unsatisfied with the result of the wage negotiations with their employer, but the Conseil de Prud’hommes did not agree.81 The factory worker Stefanie Mondal and thirteen of her colleagues were sent away without notice by La Filature du Nord for trying to leave the spinning mill without permission and before the machines had stopped. This could have caused great damage, so their employer complained. The clerk of the court’s summary of the case sheds some light on the conflict. A week or so earlier there had been an accident with a machine and during one or two hours the mill could not operate. The employer tried to make up for time lost. The first week workers did not complain about the requested overtime, but the second week they argued that the hours lost had been amply recuperated. The employer, backed by witnesses, claimed the workers had stopped all the machines at once and without warning, which could have caused accidents. The conduct of the workers, so the Conseil de Prud’hommes concluded, justified their immediate dismissal.82 Although the word ‘strike’ is not used in any of these cases, there was probably a thin line between committing acts which were associated with a strike and ‘leaving the shop without permission’. One employer tried to take advantage of this confusion. Factory owner Henri Beke requested a substantial sum from four female reelers whom he accused of striking and provoking the standstill of his factory. In fact Beke had announced he was

98

Giving notice

going to close the mill and had given his workers fourteen days’ notice. During this period, however, he was unable to guarantee them either regular work or a regular wage. The Conseil de Prud’homme concluded that in these conditions it was understandable that workers already looked for work elsewhere and that their absence had to be interpreted as an individual and not a collective initiative.83 Other circumstances providing a legitimate objection to eight or fourteen days’ notice, appear to be the fact that the worker was accepted on probation or on a temporary basis and thus not definitively engaged. Not in one of the cases on trial, however, the employer was able to convince the bureau général that these conditions were fulfilled. Since contractor Edmond Bevernage was unable to prove that the overseer Jan Antheunis had only been accepted on trial, he was sentenced to pay him compensation for firing him without notice.84 When the mechanic Alfred Murier summoned dredging operator Lutzer & De Swarte for not granting him proper notice, the latter argued that Murier had been on trial and that he could thus be sent away immediately. The council rejected this argument because Murier had been working there for almost two months. The first fortnight gave the employer ample opportunity to test a worker’s capabilities, the council argued. Since Murier worked there much longer, he was entitled to assume that he had been accepted and that the custom of giving notice was applicable to him.85 In the same vein, the thread twister Victor Jacquart was unsuccessful in sending the engineer Pieter Veeckman away without notice. In this case the employer reasoned that Veeckman was only accepted on a temporary basis in order to replace a sick workman. The Conseil de Prud’hommes concluded that since Veeckman had been working at Jacquart’s for five months and since the recovered workman was already employed elsewhere, Veeckman was correct to assume that he had been hired properly and that giving notice was applicable to him.86 Henri Delaconcorde was fired without notice by grain trader Brasseur De Crom with the argument that the former had been hired with the explicit precondition that he would have to leave without notice, if the sick workman he replaced returned to work. The plaintiff denied he had accepted this agreement. The Conseil de Prud’hommes contested the legitimacy of the alleged contract and the defendant had to pay.87 The wood merchants Matthys & De Vriendt tried to convince the Conseil de Prud’hommes that they could fire Lieven Meulenaere and three of his colleagues without notice because they were paid per task. It appeared, however, that the four workmen were paid per week and as a result local custom had to be applied.88 ‘Nation boss’ (natiebaas) Emiel Geirnaert had to pay weigher Hendrik Capiau a financial compensation for sending him away without eight days’ notice. The Conseil de Prud’hommes rejected Geirnaert’s argument that Capiau used to be steadily employed by him, but that this was no longer the case, that he only hired his services now and then, and that in these circumstances he was not supposed to give him eight days’ notice.89

Van den Eeckhout

99

Workshop regulations It seems clear that employers, trying to fire their employees without proper notice, had to prove that there were serious arguments to deviate from local customs. However, there was a way round local customs. The employer could dress workshop regulations explicitly stating that the local custom of giving notice reciprocally was not applicable. By the Ghent Conseil de Prud’hommes these regulations were regarded as contracts that replaced local customs entirely. Instead of convincing the Conseil de Prud’hommes that his employee did not deserve the application of local customs, the employer had to prove that the regulations existed and had been visibly exposed in the workshop. The smith Pieter Schepens was sentenced to pay his workers Jan Vercauteren and Karel Recollet a financial compensation, because he was unable to produce the workshop regulations he referred to and, in the absence of the latter, extraordinary reasons for firing the two without notice.90 The weaver Jan Van Heuverswyn argued successfully that he had been fired without proper notice, because the alleged regulations abolishing the local custom were not posted in the room where he had his loom. In fact none of the workers appeared to be aware of them. Van Heuverswyn accused his employer, De Staercke, of punishing him because he had refused to send his child to a catholic instead of a communal school. He convinced the Conseil de Prud’hommes who reprimanded De Staercke for firing a worker for reasons not related to his job or his attitude towards his master or fellow workers.91 Even if the existence of workshop regulations abolishing local customs could not be doubted, workers could successfully argue that they were not applicable to their case. Local customs were not of any use in such matters. Only a more powerful weapon could do the trick, namely an individual written contract with clauses that could substantiate the worker’s claim that he was a ‘special case’. Gustaaf Vande Casteele, a producer of vegetal fibres, thought he could fire hair spinner Francies De Backer without notice because his workshop regulations abolished the local custom. De Backer was able to produce a written contract, signed by the two parties, that stated that Vande Casteele hired his services for two years: he had still five months to go and he requested a substantial financial compensation for being sent away earlier. The Conseil de Prud’hommes followed De Backer’s argument, even if the literal clauses of the contract were not in his favour. The latter only mentioned the worker’s obligations, namely a substantial fine if he left before the given date. The Conseil de Prud’hommes argued that even if the inadequate contract did not grant De Backer any rights, he was justified to assume that obligations were reciprocal.92 Workers who were unable to produce contracts or other documents sustaining their claim that they were a ‘special case’, lost their plea. The weavers Paul Van Bambrugge and Hypoliet Vanderhaeghen were unable to convince the bureau général that workshop regulations of

100

Giving notice

Storme-Brasseur, abolishing the local custom of giving notice, were not applicable to their case. On the contrary, the Conseil de Prud’hommes explained, ignoring them would mean breaking a regular contract.93 Some workshop regulations specified that the custom of giving notice had been abolished for both workman and employer. This was the case in the candle factory of Caussemille & Cie and as a result Ferdinand Dhaeyer was refused a financial compensation.94 Spinner Lievin de Rycke had the same experience in the factory of Parmentier-Van Hoegaerden. ‘The regulations make a contract between the parties, against the old customs’, the bureau général stated in this case.95 Other employers dressed workshop regulations abolishing the custom one-sidedly. Four workmen of a telephone company, contesting the fact of being fired without notice after drunkenness, were confronted with these regulations which allowed a one-sided immediate dismissal.96 At the sugar factory La Ruche the custom was also abolished in favour of the employer only. The Conseil de Prud’hommes referred to these workshop regulations, when they declined the request of the engineer Edouard Windy for proper notice. When three other workers asked for their due wages La Ruche refused to pay them, referring to these same workshop regulations: they had left the factory abruptly after having declined to pay a fine. The Conseil de Prud’hommes decided that two of them would receive their wages, but only after deduction of a substantial fine for quitting without notice. The third workman had shown repentance for his conduct, so that La Ruche was prepared to pay the wage without fining him.97 Overseer Hippoliet Pauwels, however, was successful in fighting his dismissal without notice by that same company. Given the seasonal character of the activities in the sugar factory (a season was called campagne), Pauwels had been engaged for the duration of a campagne but with the promise of hiring him further for a job ‘compatible with his capacities and status’. The Conseil de Prud’hommes concluded that the definition of his duties distinguished him from La Ruche’s ordinary workmen and that, as a result, the one-sided abolishment of the custom of giving notice was not applicable to him.98 There is one example of workshop regulations that did not abolish giving notice entirely, but only in ‘specific circumstances’. Charles Renard and six other paper workers complained because they were fired without notice by the paper factory Les Papeteries Gantoises. An article of workshop regulations specified that giving notice was not required when shortage of work or particular circumstances occurred.99 The plaintiffs admitted that work was short, and they lost the case. Les Papeteries Gantoises should have stuck to that line of defence. When four workers of the same factory formulated the same complaint immediately after the former trial, the company lost the case. Again, reference was made to the article in workshop regulations, but instead of stressing shortage of work, Les Papeteries Gantoises complained that the four workers in question (four girls apparently related to two of the plaintiffs of the first case) had caused trouble. Not shortage of work but ‘particular circumstances’,

Van den Eeckhout

101

namely the disorder they caused, was the argument used against them. The notion ‘particular circumstances’ was much too vague, so the Conseil de Prud’Hommes concluded, while the gravity of the facts did not justify a dismissal without notice.100 Workshop regulations did not necessarily abolish local customs. The machine maker Carels referred to workshop regulations, requiring a reciprocal obligation of proper notice, when he reminded Camiel Bonnardeaux and three other workers of their duties.101 The idea that workshop regulations instead of local customs ruled labour relations could in some cases also appeal to workers. Twelve weavers of Van Heuverswyn’s factory, possibly involved in a collective action, asked for their livret, which their employer refused to return since they had not given fourteen days’ notice. The workers argued that they were entitled to leave the factory at any moment because there were no written workshop regulations explicitly requesting the fourteen days. The Conseil de Prud’hommes did not go along with their reasoning, because in the absence of workshop regulations or specific contracts local customs ruled.102 Even if there were no workshop regulations abolishing the local custom of giving notice, the employer could successfully argue that it was not applicable. What he needed in the latter case were witnesses testifying that this custom did not exist in his company and that all his workers were aware of it. The bleachers Filiep Vanden Holle and Jan De Raeve, asking for financial compensation lost their case when they were confronted with the declarations of three other bleachers, who asserted that in La Gantoise there was no obligation of giving notice and that everyone working there knew these conditions.103 The cement workers Jan Semey and Karel Bourgeois lost their case against the cement factory Picha because ‘two illiterate workmen’ testified that the custom had since long been abolished.104 The interrogation of the interested parties could also lead to the conclusion that a company did not know the custom of giving notice. The absence of the obligation of giving notice was thus presented as a custom specific for the company in question. Only seldom did the Conseil hear of sectors where it was claimed the custom was unknown. Mason Felix Wallijn requested a compensation for denying him eight days’ notice. The Conseil de Prud’hommes agreed with his employer that this trade did not know this obligation, neither for worker, nor for employer.105 Even when the principle of applying the local custom of giving notice was not questioned, a conflict could arise, for instance because one of the parties contested that notice had been given properly. Pierre Dardenne, an engineer in the factory of Leon Julien, accused the latter of firing him without notice. Witnesses were produced (former workers of Julien), testifying that notice had been given on the courtyard of the factory to all the workers collectively and on pay day individually. No one could have been more aware of the collective dismissal than Dardenne, so the council argued, because he had to prepare the machines for the period of inactivity. Dardenne left empty-handed.106 That was also the case for three

102

Giving notice

workers of the sugar factory Vercruysse-Bracq who claimed they had been sent away without proper notice. They had to admit, however, that their employer had announced at least eight days in advance that the factory would close for a while and that he would subsequently hire workers at his own discretion. The promise to reopen the factory raised doubts among the workers whether those who would not be engaged again, had received a proper notice. The Conseil de Prud’hommes did not follow their reasoning.107

The employer as plaintiff In most cases workers took the initiative to claim eight or fourteen days’ notice (see Table 4.1). However, now and then an employer also complained because workers left without warning. When the spinner Karel De Raeve left the factory of Van Heuverswijn without notice, the latter successfully claimed compensation for ‘the damage caused to his fellow workers’.108 Some employers, in conflict with their workers because they had left without notice, appeared to be reacting to a strike. In fact, going on strike was considered as a case of leaving without notice. The firm HooremanCambier refused to return the livret to the thirty-six weavers and spinners who requested them, because they had gone on strike before giving notice. The fact that, besides their livret, they also asked for the permission to do their period of notice ‘according to the old tariff, since long in existence in the workshop’, suggests that a conflict over wages might have provoked the strike.109 When seventeen workers of Kaminski’s factory of guilt frames summoned their employer for not granting a wage raise, the latter replied by requesting a large sum of money because the plaintiffs had stopped working without giving notice.110 In a conciliated case cigar manufacturer De Somer summoned eight cigar makers who had written him a note announcing they would cease work out of solidarity with other Ghent cigar workers, who were apparently on strike. De Somer reacted with a complaint because they had not given fourteen days’ notice. The defendants agreed to work the already prepared tobacco.111 The minutes of the session regarding printer Boterdaele, requesting financial compensation from typographer Jozef Mathieu and five other workers because they did not give notice, remain rather implicit regarding the cause of the conflict: the typographers in question were not satisfied with ‘changes introduced by their employer’. The clerk of the court note to the president of the Conseil de Prud’hommes, however, mentions that there had in fact been a strike regarding tariffs. Boterdaele’s summoning for not giving notice was thus a reaction to the latter. A note, in the margin of the clerk’s summary, gives the impression that the members of the conciliation board were divided on the question whether the local custom of giving notice had to prevail over the right to strike.112 Another strike regarding tariffs occurred in 1895.113 It was a failure and one of the employers, printer Adolf Hoste,

Van den Eeckhout

103

summoned ten of his typographers because they had not given fourteen days’ notice. He withdrew his complaint.114 In some cases information is too scarce to discern whether we are dealing with a strike.115 Probably quite a few cases involving a strike were presented without any reference to the obligation of giving notice. In 1895 for instance Ghent witnessed a lot of strikes and the total number of cases brought before the Conseil de Prud’hommes increased accordingly, but the number of cases explicitly connected with termination of employment rose only moderately (Table 4.1).116 During the discussion regarding the employment contract in the House of Representatives the problem of strikes caused the somewhat ambiguous standpoint of the socialists regarding dismissals. While in the discussions of the Conseil Supérieur du Travail the socialist member Hector Denis pleaded for a compulsory period of notice, he and the other members of the socialist fraction argued against compulsion if this ran counter to local customs.117 They referred to the thousands of miners who were on strike while the debate took place (April 1899) and could not be summoned by their employers because local customs did not include the obligation of giving notice. The socialists claimed that making the latter compulsory and requesting every worker to give notice individually would make strikes virtually impossible. However, the answers to the inquiry of 1886 raise doubts concerning the socialist allegation that giving notice was not customary in the mining regions.118 Employers were plaintiffs when they reacted to strikes, but also when home workers did not return their finished work. In these cases eight or fourteen days’ notice were not at stake but the custom of the so-called congéboom, meaning that the weaver had to make a last piece before the livret was returned. Weaver Florent Vermeire had to compensate Celestin Castiau from the town of Ronse because he had failed to do that.119 In a conciliated case, manufacturer Pieter Aerens summoned three weavers in Lokeren because they had not delivered ‘their chain with a last piece’.120 The following year Aerens made the same complaint.121 Silk producers in Deinze also had to remind some of their silk weavers of their obligations.122 Compared to the first half of the nineteenth century the number of conflicts involving home workers was marginal, which agrees with the declining importance of that occupation.123

Conclusion When in nineteenth-century Belgium an oral contract between worker and employer was established, very often only one clause was discussed, namely the wage. Frequently the other clauses of the engagement remained implicit. They were usually based on local customs, which does not mean that they were necessarily as arbitrary or uncertain as the legislators of the turn of the century claimed. There appeared to be a remarkable consistency within the Ghent Conseil de Prud’hommes’ considerations regarding

104

Giving notice

what Ghent local customs prescribed when employment ended. Quitting or firing had to be announced either eight or fourteen days in advance, both by worker and employer. Most of the sentences regarding dismissals explicitly referred to these local customs. Sanctions for the non-application of the local custom were not explicitly identified as local customs, but it was accepted as a rule that workers and employers had to pay for an unjustified disregard of the latter. There also appeared to be quite a consistent attitude regarding the circumstances that justified a denial of the customary period of notice: an intentional, repetitive or unrepentant neglect of one’s duties as well as disrespect and insubordination were legitimate reasons for immediate dismissal. Most of the discussions dealt with the problem of which acts could be qualified as such. Being on trial or being only temporarily engaged were also accepted arguments against the application of the custom. With its repeated references to the local custom of giving notice and the outline, time and again, of the circumstances that justified an immediate dismissal, the Ghent Conseil de Prud’hommes played an important role in defining the legitimate ways of terminating employment. In fact, it formulated what can be considered, in Alain Cottereau’s words, des quasi-législations locales du travail, a kind of local labour legislation.124 The repetitive reminding of the confines of legitimacy can be perceived as an offensive strategy. The Ghent Conseil de Prud’hommes that considered itself the guardian of local customs regarding labour relations, complained about an increasing undermining of the latter. Repetition and consistency were weapons in the struggle to maintain tradition and the relations of trust and respect between worker and employer that were associated with ‘the old ways’. The Conseil de Prud’hommes was rather successful in constructing local customs as a category which was all but ‘fluid and impermanent’.125 However, its enthusiasm for workshop regulations, whether they consecrated the custom of giving notice or not, suggests that the Conseil de Prud’hommes was less defending the intrinsic content of local customs than the certainties associated with tradition or unambiguous written workshop regulations. Uncertainty appeared to be the worst enemy of harmonious class relations. Even if the Ghent Conseil de Prud’hommes appeared to follow a rather coherent set of rules in its considerations, it had to pretend every case was entirely judged on its own merits. A socialist member of the Ghent Conseil de Prud’hommes, Pol Verbauwen, met with a lot of objections when he suggested in 1877 that, once and for all, the Ghent Conseil de Prud’hommes might decide whether, in the absence of workshop regulations, the custom of giving fourteen days’ notice was compulsory. While in the sentences a period of eight days’ notice was also referred to as customary, Verbauwen limited himself conveniently to the longer period.126 However, Verbauwen failed in his attempt to codify a local custom which, at least in the period studied in this article, appeared to be favourable to workers. The clerk of the court replied that the obligation of giving notice was not a law, but a custom, and the Conseil de

Van den Eeckhout

105

Prud’hommes was completely free in applying it or not. It was not the Conseil de Prud’hommes’ task to make legislation. The fact that the socialist Pol Verbauwen acknowledged that local customs were not applicable if there were workshop regulations is symptomatic of the Ghent Conseil de Prud’hommes’ attitude towards the latter. There appeared to be an implicit hierarchy in the different sources on which the Conseil de Prud’hommes based its jurisprudence. At the top was a specific contract between worker and employer (written or oral, but in the last case supported by witnesses), workshop regulations came second and, despite the warm feelings towards them, local customs came last. The latter were only applicable if not contradicted by the former. Hence the diligence of some employers in establishing their own ‘local’ labour legislation.

Notes 1. I would like to thank the editors of this volume and my colleagues Kathlijn Pittomvils, Harald Deceulaer and Sven Steffens for their valuable comments. 2. Patricia Van den Eeckhout and Peter Scholliers, ‘Social history in Belgium: old habits and new perspectives’, Tijdschrift voor Sociale Geschiedenis, 23 (1997), pp. 163, 176–177. 3. Ginette Kurgan-Van Hentenryk and Jean Puissant, ‘Industrial relations in the Belgian coal industry since the end of the nineteenth century’, in G. Feldman and Karl Tenfelde (eds), Workers, owners and politics in coal mining (Oxford, 1990), pp. 203–270; Sven Steffens, ‘Schneiderei, Konfektion, Heimarbeit. Aspekte der Zerfalls und der Umstrukturierung eines städtischen Handwerks in Belgien (19. bis frühen 20. Jahrhundert)’, Tijdschrift voor Sociale Geschiedenis, 20 (1994), pp. 428–460; Patricia Van den Eeckhout, ‘Onderaanneming en huisarbeid in Westeuropese hoofdsteden. Twee eeuwen flexibiliteit in de kledingindustrie (19e–20e eeuw)’, Tijdschrift voor Sociale Geschiedenis, 20 (1994), pp. 391–427; Peter Scholliers, Wages, manufacturers and workers in the nineteenth-century factory. The Voortman cotton mill in Ghent (Oxford, 1996); Bart De Wilde, Witte boorden, blauwe kielen. Patroons en arbeiders in de Belgische textielnijverheid in de 19e en 20e eeuw (Ghent 1997); Griet Van Meulder and Guy Coppieters, ‘Een sociale geschiedenis van de Limburgse mijnen 1917–1985’, in Tine De Rijck and Griet Van Meulder, De ereburgers. Een sociale geschiedenis van de Limburgse mijnwerkers (Berchem, 2000) pp. 425– 754. 4. Patrick Joyce, The historical meanings of work (Cambridge, 1987), pp. 20–21, 26–29. 5. E.P. Thompson, ‘Introduction: custom and culture’, in E.P. Thompson, Customs in common. Studies in traditional popular culture (New York, 1993), p. 6. 6. Kathlijn Pittomvils, ‘Alledaagse arbeidsconflicten in de Gentse textielindustrie. De praktijk van de werkrechtersraad in de eerste helft van de negentiende eeuw’, Tijdschrift voor Sociale Geschiedenis, 21 (1995), pp. 181–211. 7. People allowed to vote in theory: male workers with the Belgian nationality of at least twenty-five years old, who could read and write, and exercised their trade for at least four years. However, local and provincial authorities had the last word in the elaboration of the list of voters. Only three categories of workers were voters by right: those who had either a savings deposit of at least 100 francs, a diploma of courage and devotion or a diploma of morality and capacity installed by the royal decree of 7 November 1847, see Pasinomie, Règne de Léopold Ier, 3e série, t. 29. Brussels, 1859, 7 février 1859, n° 53.

106

Giving notice

8. Pasinomie, Règne de Léopold II, 4e série, t. 24. Brussels, 1889, 31 juillet 1889, n° 275. 9. Pasinomie, 7 février 1859, n° 53, art. 41. 10. The register of cases (Register van behandelde zaken) starts in 1877. We consulted the corresponding minutes of the sessions of the bureau de conciliation and bureau général (RAB, Verslagboek van de zittingen), the files regarding reconciliations (RAB, Verzoeningen), sentences (RAB, Vonnissen) and correspondence regarding them from 1877 on. We also used the minutes of sentences for the years 1873–1876. These archives of the Ghent Conseil de Prud’hommes are kept in Rijksarchief Beveren-Waas (hereafter RAB). 11. Kathlijn Pittomvils, ‘Arbeidsverhoudingen te Gent in de eerste helft van de 19de eeuw. Een studie van de conflicten voor de werkrechtersraad (1810–1858)’ , Unpublished masters thesis, Vrije Universiteit Brussel, 1994), pp. 104, 170 [hereafter, ‘Arbeidsverhoudingen’]. 12. Pittomvils, ‘Arbeidsverhoudingen’, pp. 117–130. 13. Pittomvils, ‘Arbeidsverhoudingen’, pp. 176–180, 182–184. 14. Rijksarchief Beveren-Waas. Provinciebestuur Oost-Vlaanderen, 4de Afdeling A, n° 496 and 1241; Stadsarchief Gent. Afdeling IV: de Werkrechtersraad, n° 240. We do not use these surveys for other purposes because they give no details on the cause of the complaints whose outcome was unclear and they include the cases in which the parties only came for advice. The information on the latter is very succinct or missing. 15. Pittomvils, ‘Arbeidsverhoudingen’, pp. 120–121, 124, 142, 178, 183–184. 16. Pasinomie, Consulat, 1e série, t. 12. Brussels, 1836, 1 décembre 1803, pp. 287–288. 17. Pasinomie, Consulat, 1e série, t. 12. Brussels, 1836, 12 avril 1803. 18. Pittomvils, ‘Arbeidsverhoudingen’, p. 120. 19. Robert J. Steinfeld, Coercion, contract, and free labor in the nineteenth century (Cambridge, 2001), pp. 39–191, 243. 20. Pasinomie, Règne de Léopold II, 4e série, t. 18. Brussels, 1883, 10 juillet 1883, n° 156. 21. RAB,Verslagboek van de zittingen, 11 August 1884. 22. Jean-Pierre Nandrin, ‘L’histoire du contrat de travail et la problématique de l’entreprise avant 1914’, Revue interdisciplinaire d’études juridiques, 15 (1985), p. 51. 23. Nandrin, ‘L’histoire du contrat de travail’, p. 51. 24. Pasinomie, Règne de Léopold II, 4e série, t. 35. Brussels, 1900, 10 mars 1900, n° 80, p. 82. 25. Commission du Travail, Réponses au questionnaire concernant le travail industriel. Vol. 1. (Brussels, 1887), pp. 304–315. 26. RAB, Vonnissen, 5 March 1877. 27. Alberto Melucci, ‘Action patronale, pouvoir, organisation. Règlements d’usine et contrôle de la main-d’oeuvre au XIXe siècle’, Le Mouvement social (1976), pp. 139–159. 28. Pittomvils, ‘Arbeidsverhoudingen’, pp. 114. 29. Rijksarchief Beveren-Waas. Provinciebestuur Oost-Vlaanderen, 4de Afdeling A, n° 1241. Verslag over de werkingen van den Werkrechtersraad van Gent gedurende het jaar 1877. (Ghent, 1878) and Verslag over de werkingen van de Werkrechtersraad van Gent gedurende het dienstjaar 1881. (Ghent, 1882.) 30. Alain Cottereau, ‘Justice et injustice ordinaire sur les lieux de travail d’après les audiences prud’homales’, Le Mouvement social (1987), p. 57. 31. Pasinomie, Règne de Léopold II, 4e série, t. 31. Brussels, 1896, 15 juin 1896, n° 214, pp.185–203. 32. RAB,Vonnissen, 10 February 1873, 10 April 1876. 33. RAB, Verslagboek van de zittingen, 10 January 1881. 34. University Library. University of Ghent, Vliegende bladen, II, L 83, Chambre de Commerce et des Fabriques de Gand, Livrets d’ouvriers de fabrique (Ghent, 1879). 35. Commission du Travail, Réponses au questionnaire, pp. 304–315.

Van den Eeckhout

107

36. RAB, Vonnissen, 10 February 1873. RAB, Verslagboek van de zittingen, 18 November 1878. Testimonies in Flemish in the same case speak of de veertien dagen, the fourteen days. RAB, Verslagboek van de zittingen, 10 January 1881, 19 September 1882. 37. Conseil Supérieur du Travail, Troisième session 1895–1896. Contrat de travail (Brussels, 1896), p. 45. 38. Conseil Supérieur du Travail, Contrat de travail, pp. 168–171. 39. Pasinomie, 10 mars 1900, n° 80, p. 82 40. This could be the result of the fact that the plaintiff withdrew his/her complaint, that the plaintiff, the defendant or both remained absent from the conciliation meeting, that no conciliation was reached but that the plaintiff did not request the defendant’s convocation for the bureau général, that the plaintiff him- or herself was accused of committing a mistake. 41. RAB, Verslagboek van de zittingen, 7 April 1891. 42. RAB, Vonnissen, 9 August 1875. 43. RAB, Verslagboek van de zittingen, 16 December 1890. 44. RAB, Verslagboek van de zittingen, 17 November 1879. 45. RAB, Vonnissen, 11 May 1874. 46. RAB, Vonnissen, 1882, Murier c/ Lutzer & de Swarte. 47. We do not use these surveys for other purposes because they give no details on the cause of the complaints whose outcome was unclear and they include the cases in which the parties only came for advice. The information on the latter is very succinct or missing. 48. Rijksarchief Beveren-Waas. Provinciebestuur Oost-Vlaanderen, 4de Afdeling A, n° 496 and 1241. 49. RAB, Vonnissen, 10 February 1873. 50. RAB, Vonnissen, 2 October 1876. 51. RAB, Verslagboek van de zittingen, 15 December 1879. 52. RAB, Verslagboek van de zittingen, 7 February 1877. 53. RAB, Vonnissen, 6 October 1873. 54. RAB, Verslagboek van de zittingen, 22 April 1890. 55. RAB, Verslagboek van de zittingen, 19 April 1892. 56. RAB, Verslagboek van de zittingen, 12 August 1890. 57. RAB, Vonnissen, 11 October 1875. 58. RAB, Vonnissen, 3 April 1876. 59. RAB, Verslagboek van de zittingen, 12 November 1889. 60. RAB, Vonnissen, 10 April 1876. 61. RAB, Vonnissen, 10 November 1873. 62. RAB, Vonnissen, 28 December 1874. 63. RAB, Vonnissen, 19 August 1875. 64. RAB, Verslagboek van de zittingen, 30 January 1894. 65. RAB, Verslagboek van de zittingen, 3 April 1894. 66. RAB, Verslagboek van de zittingen, 21 April 1891. 67. RAB, Verslagboek van de zittingen, 4 August 1885. 68. RAB, Verslagboek van de zittingen, 17 April 1888. 69. RAB, Verslagboek van de zittingen, 2 January1889. 70. RAB, Verslagboek van de zittingen, 5 April 1892; RAB, Vonnissen 1892 Van Haudenhove c/ Van Coppenolle. 71. RAB, Vonnissen, 20 November 1876. 72. RAB, Verslagboek van de zittingen, 9 July 1889. 73. RAB, Verslagboek van de zittingen, 6 August 1889. 74. RAB, Verslagboek van de zittingen, 30 May 1893. 75. RAB, Verslagboek van de zittingen, 1 August 1893. 76. RAB, Verslagboek van de zittingen, 19 September 1893. 77. RAB, Verslagboek van de zittingen, 31 July 1894. 78. RAB, Verzoeningen, 1891, Lejour c/ Job, Maes & Cie.

108

Giving notice

79. 80. 81. 82.

RAB, Vonnissen, 3 April 1876. RAB, Verslagboek van de zittingen, 19 September 1893. RAB, Verslagboek van de zittingen, 31 October 1893. RAB, Verslagboek van de zittingen, 6 September 1893; RAB, Vonnissen, 1893, Mondal & consort c/ La Filature du Nord. RAB, Verslagboek van de zittingen, 24 December 1877. RAB, Verslagboek van de zittingen, 6 March 1888. RAB, Verslagboek van de zittingen, 19 September 1882. RAB, Verslagboek van de zittingen, 27 February 1884. RAB, Vonnissen, 3 December 1875. RAB, Verslagboek van de zittingen, 28 October 1890. RAB, Verslagboek van de zittingen, 19 September 1893. RAB, Verslagboek van de zittingen, 15 January 1877. RAB, Verslagboek van de zittingen, 28 July 1879. RAB, Verslagboek van de zittingen, 7 February 1881. RAB, Verslagboek van de zittingen, 5 March 1877. RAB, Verslagboek van de zittingen, 27 February 1894. RAB, Verslagboek van de zittingen, 8 January 1877. RAB, Verslagboek van de zittingen, 4 April 1881. RAB, Verslagboek van de zittingen, 24 January 1881. RAB, Verslagboek van de zittingen, 27 December 1887. RAB, Verslagboek van de zittingen, 22 August 1893. RAB, Verslagboek van de zittingen, 22 August 1893. RAB, Verslagboek van de zittingen, 21 March 1881, 19 September 1881. RAB, Verslagboek van de zittingen, 24 February 1882. RAB, Verslagboek van de zittingen, 23 December 1890. RAB, Verslagboek van de zittingen, 26 March 1877. RAB, Verslagboek van de zittingen, 17 September 1895. RAB, Verslagboek van de zittingen, 18 November 1878. RAB, Verslagboek van de zittingen, 20 October 1896. RAB, Verslagboek van de zittingen, 19 May 1879. RAB, Verslagboek van de zittingen, 20 April 1877. RAB, Verslagboek van de zittingen, 26 March 1878. RAB, Verzoeningen, 1882, Affaire De Somer c/Castelein, Muller & cons. RAB, Verslagboek van de zittingen, 15 March 1892; RAB, Vonnissen, 1892, Boterdael c/ Mortier & Consort. A pamphlet published on 20 February 1892 by the Gentsche Boekdrukkersbond, an association of workers in the Ghent printing trade, denounced Boterdaele for trying to hire girls in order to avoid the new tariff. University Library. University of Ghent, Vliegende bladen, I, Imprimeurs 11. Avanti, Een terugblik, (Ghent, s.a.,) pp.565–566. RAB, Verslagboek van de zittingen, 13 November 1895. The case of printer Van Doosselaere who summoned seven printer’s helpers because they had failed to give fourteen days’ notice, the case of machine maker Carels who summoned six workers for not doing eight days’ notice. RAB, Verslagboek van de zittingen, 11 February 1880, 31 March 1892. Guy Vanschoenbeek, Novecento in Gent. De wortels van de sociaal-democratie in Vlaanderen (Antwerp, 1995) pp. 34–35; Arthur Verhaegen, De werkstakingen in 1895. De Gentsche Socialisten opgedragen (Ghent, 1896). Chambre des Représentants, Annales parlementaires, 1899, pp. 1029–1035, 1062–1065. Commission du Travail, Réponses au questionnaire, pp. 307–311. The jurisprudence referred to in ‘Contrat de travail’, Pandectes belges (Brussels, 1921), vol. 113, p. 881 also indicates that dans les charbonnages du bassin de Charleroi (1890) giving notice was customary. RAB, Verslagboek van de zittingen, 17 December 1877.

83. 84. 85. 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. 106. 107. 108. 109. 110. 111. 112.

113. 114. 115.

116.

117. 118.

119.

Van den Eeckhout

109

120. RAB, Verslagboek van de zittingen, 28 November 1879. 121. RAB, Verslagboek van de zittingen, 24 April 1880, 21 May 1880. 122. RAB, Verslagboek van de zittingen, 13 May 1881, 2 June 1881, 16 February 1883, 18 January 1884, 3 July 1884, 22 January 1886. 123. Pittomvils, ‘Arbeidsverhoudingen’, p. 305. 124. Cottereau, ‘Justice et injustice ordinaire’, pp. 51–53. 125. M. Sonenscher, Work and wages. Natural law, politics and the eighteenth–century French trades (Cambridge, 1989), p. 369. 126. RAB, Verslagboek van de zittingen, 3 September 1877.

CHANGING PAY SYSTEMS AND WAGE FORMS

CHAPTER 5

WAGE

FORMS, WAGE SYSTEMS AND WAGE

CONFLICTS IN

GERMAN CRAFTS DURING

THE EIGHTEENTH AND EARLIER NINETEENTH CENTURIES

Reinhold Reith

Wages and wage theories in the context of the ‘historical school of German economics’ In the late nineteenth century the ‘historical school of German economics’ dealt with the subject of wages, since the wage was a hotly disputed political issue.1 Scholars of political economy focussed upon the development of the standard of living and the development of nominal and real wages on the basis of prices and wage rates. The ‘historical school’ decisively shaped research in economic history, which concentrated particularly on quantitative aspects of wages. This was also undoubtedly a consequence of the available source material. Wage data and especially wage series had been recorded and preserved, particularly for the building trades and for urban day labourers. These data consisted of time rate wages, for the most part paid daily. This specific source material, as well as the wage series published later, gave rise to the impression that craft production had been remunerated exclusively with time rate wages. Nevertheless, the nature of the then available source material offers only one reason why the qualitative dimension of wages – the nature of the wage itself – has remained, for the most part, terra incognita. German political economists did not begin to treat the issue of wage forms intensively until 1900, a time when there was increasingly broad application of the concept of payment by results in industry. The systematic theory of wage forms which developed from 1895 to 1914 was essentially based upon the differentiation between time rates and

114

Wage forms, systems and conflicts

piece rates. The piece rate was defined as a direct efficiency wage whereby only the actual labour performed – that is, the output – was compensated. The time rate was defined as a specified wage paid for a specified period of time (day, week, year) regardless of the work actually performed.2 Ludwig Bernhard and Otto von Zwiedineck-Südenhorst were among the leading political economists who set about elaborating a systematic theory of wage forms during the discussions which followed the publication of the work of Frederick Winslow Taylor. Bernhard’s interest was practiceoriented, and he did not concern himself with questions of historical origins.3 Zwiedineck-Südenhorst, on the other hand, concentrated on sociopolitical aspects; he stood squarely in the tradition of the ‘historical school’ and also treated the historical dimension of wage policies, although his empirical investigations remained within a narrow range. He oriented his analysis on Karl Bücher’s ‘theory of economic stages’ including its sequence of ‘types of workshop systems’ upon which he based his own studies. Schematically, he assigned a wage form to each form of business. Indeed, he considered the piece rate to be an older form of compensation; historically, however, ‘its re-emergence was brought about by the form of labour organization used in large-scale production.’ Continuity was supposed to have been broken by a period of time rates. Therefore, he schematically assigned time rates to small-scale production.4 These hypotheses were adopted as established principles by standard texts in political economy. We find this view expressed by Eugen von Philippovich: ‘In light of the limited sales revenue and traditional techniques, there was no need for an economic incentive to raise efficiency’.5 Gustav Schmoller readily concurred with this opinion and viewed payment by results as a large-scale industrial phenomenon. Time rates were said to be the ‘time-honoured way’.6 Conrad’s Outline summarised this approach: previously, the use of time rates had been widespread, if indeed not general. A time rate, however, is a purely mechanical form of compensation: it does not take the personal output of individual workers into account, it deadens the worker’s enthusiasm for his task and leads to a slower, more apathetic way of working.7 Werner Sombart regarded developments in the wage that led to payment by results as one of the most important processes in the course of the nineteenth-century emergence of capitalism. He even spoke of the ‘irrationality of wage formation’ in the early modern period. He therefore considered it a waste of time to research further into this issue.8 The ‘historical school’ thus considered payment by results to be an element totally alien to the nature of the pre-industrial economy. Scholars did not even consider the possibility that it might have been widespread, so researching this issue seemed unnecessary. We can identify one reason for this assumption made by the ‘historical school’ in the fact that the significance of piece rates had again declined around the turn of the century. Political economists were discussing the ‘viability’ of small-scale crafts and trades.9 In contrast, payment by results had visibly increased its importance in large-scale production. Since political economists usually took contemporary problems as the point of departure for their historical research, they

Reith

115

were apparently incapable of imagining dynamic elements of small scale production, such as payment by results, in an historical dimension. However, the decisive point in explaining why political economists regarded payment by results as an achievement of the modern era was their assumption of a specific mentality on the part of economic actors. In the discussions that took place in the 1870s on the issue of shorter working hours, Lujo Brentano put forward the view that high wages resulted in the highest performance; however, this did not apply to the seventeenth and eighteenth centuries. During that period, workers acknowledged the power of tradition and led modest lifestyles with few material needs. Since then, the working class had changed and had attained a higher level of consciousness, together with higher aspirations.10 Brentano, therefore, did not even discuss payment by results in a historical perspective and was unable to establish a connection between payment by results and small-scale production either in the past or in the present. Small scale production was for him, as a consistent modernist, a form of business organization incapable of withstanding competition and condemned to extinction. He argued polemically against small firms: they were said to be economically inefficient operations employing less efficient personnel.11 Schmoller formulated his conception of the mentality of economic actors in his Outline. He defined natural, ‘primitive’ peoples as lethargic, carefree and short-sighted; their needs minimal, their effective demand virtually impossible to stimulate. He considered wage labourers at the start of the nineteenth centurry still to have been submissive; he went on to characterise them as unmanageable, indolent, freeloading, hostile to innovation, and obsequious. In his view, the higher standard of living of the working class was only the result of government-led reforms of the nineteenth century.12 The drive to pursue gainful employment developed rather slowly so, claimed Schmoller, payment by results could only be a modern phenomenon. The economic mentality presented here was perhaps most clearly put forth by Sombart in his most important work, Modern Capitalism. ‘Economic actors’ were said to have been simple people with powerful emotional drives, incapable of complete rationality, sluggish, undisciplined and lacking sharp business sense. They had, he believed, conducted business in a traditional manner, meaning ‘the way it had been passed down, the way it had been learned, the way to which one was accustomed’.13 In Sombart’s view, the reason for labour shortages in the pre-capitalist economy can be found in the ‘natural laziness, torpor and indolence of the masses…The attitude of the worker was like that of any man living in a natural state – laziness or at least a tendency to take it easy. The predominant opinion was to quit working if one had enough. The common man is, by nature, like a beggar, heartrendingly lazy, and he knows nothing of that industrious life that every proletarian leads today’.14 For Sombart, inspiring the work force with a capitalist spirit is the achievement of the entrepreneur. In early capitalism, it was said to have been expedient to pay low wages, since only in this way was it possible to motivate labourers even to bother to go to work. According to Sombart, it had been impossible to raise the intensity of labour by means of

116

Wage forms, systems and conflicts

wages. It was not until capitalism had reached a mature state that piece rates and wage premiums could be introduced.15 Well-known proponents of the ‘historical school’ therefore concurred that payment by results could not have had a historical dimension, since workers could not have been motivated by efficiency incentives until the arrival of mature capitalism. These elaborations on the mentality of economic actors were fraught with consequences. As late as the 1970s the concept of ‘proto-industrialisation’ was discussed with these basic assumptions, at least in Germany. The crux of this approach was the presumption of the mutually opposed strivings of economic actors – the assurance of one’s existence versus the maximisation of profits, or Nahrungsprinzip versus Gewinnprinzip. Concurring with Sombart, this concept proceeded from the assumption that both urban guild crafts as well as rural cottage industries were regulated by the ‘labour-consumer balance’ or Nahrungsprinzip. A producer purportedly cuts back output when his profit margin (or his wage) is high, since he regards his existence as assured even at a lower level of production. Higher wages are consequently translated into more leisure time and for this reason, output can only be increased, if it can be increased at all, by employing more workers. The pre-industrial supply curve of labour was definitely sloping backwards.16 The ‘historical school’ was thus extraordinarily influential with respect to the hypothesis of economic mentalities and motivations of economic behaviour. However, it offered no starting point for an historical analysis of the relation between wages and performance, or the analysis of wage conflicts in the past. Since the ‘modern worker’ – existing on a higher economic level than his ancestors and with considerable motivation to earn and therefore to work – was a product of mature capitalism, any historical perspective fell by the wayside. The wage theories of the historical school regarded tradition – the traditional wage – as the fundamental, positively determinative factor in wage formation, only regulated by fixed wage schedules established by government authorities. The wage of a freely negotiated labour contract was therefore only the product of centuries of development. Werner Sombart spoke pointedly of ‘the irrationality of wage formation’. He postulated that ‘from the sixteenth to the eighteenth centuries wages were set in a completely disorderly fashion’, their actual amount little affected by market forces since tradition and coincidence played decisive roles. Mechanisms to achieve equilibrium on a local basis were lacking.17 Certainly, as Otto von Zwiedineck-Südenhorst, conceded, demand and supply could be influential factors in wage formation, but only in industrial centres with high fluctuations (turnover) would such influences be particularly important.18 Gustav Schmoller, for his part, did not preclude the skill and the size of the labour force as factors of value, but he stressed the primary influence of social groups and institutions, along with the formal rules.19 Werner Sombart agreed with this concept of a labour market, in which supply and demand found no room, because ‘custom and pure chance’ dominated. For Sombart a regional or interregional balance (or level) was therefore out of the question.20

Reith

117

In this tradition Otto Brunner would maintain, more than fifty years after Sombart, that the economic history of pre-industrial times, in the sense of a history of markets, had only limited importance. Therefore not the market, but the ‘constitutive factors of the old order’ were of particular interest, chief among which was undoubtedly power (‘Herrschaft’).21 As late as 1994 Jan de Vries stated that: many historians deny that economic forces dominated the determination of what labourers were paid for their time, skill, and effort and are sceptical that economic objectives suffice to account for the behaviour of labourers. To some, culture and commonalty were far more influential than the market; to others, the market power of wage earners was simply too small, rendering the market one sided and tending to keep wages always at or near subsistence. In either case, economics is not the discipline of choice with which to approach the issue of labour conditions in pre-industrial society.22

The evidence for this point of view has to a great extent come from the building trades, where the research of Phelps Brown and Hopkins has been prominent. This has appeared to confirm the view that money wages remained ‘sticky’ for a long time, sometimes even throughout a century.23 The labour force seemed to move in a cage.24 But there is no doubt that labour conditions (daily wages, money wage) and the labour market in the building trades were not comparable with the labour markets of other crafts. These had their own labour markets, at least to a significant extent. A flexible labour market had been taking shape from the late Middle Ages. In the area of the crafts, an institutionalised form of labour migration developed through the movements of travelling journeymen, which had already begun in the fourteenth century, and was often compulsory from the sixteenth century on. The mobility of journeymen was a specific pattern of the Central European labour market. From a spatial perspective, the labour market had a supraregional dimension, although a closer analysis of individual professions reveals a considerable degree of differentiation.

The pre-industrial labour market and wage conflicts: historiographical traditions Not only wage theories but also conceptions of the pre-industrial labour market had considerable consequences for subsequent histories of labour conflicts. The hypothesis of the customary wage and of a regulated and local labour market which was not really considered to be a market at all, as well as the apparent ‘irrationality of the determination process’ rules out a priori any consideration of wages and wage conflicts and naturally concludes that strikes for higher wages are a particularly modern form of conflict! It would be helpful to provide a brief outline of the various German historiographic traditions in respect of wage conflicts.25 Those writing before 1914 had a common point of view, a view that was not confined to Germany and can be found in the Webbs’ History of Trade Unionism.

118

Wage forms, systems and conflicts

There were two common strands. On the one hand they accepted that disputes had existed in the past; on the other hand, with few exceptions, they tended to dispute their significance. Already in 1877 Georg Schanz had referred to the first wage dispute of the journeymen weavers and woolbeaters in Speyer in 1343: ‘with the stronger organization of journeymen, the fourteenth century opened up an era of wage struggles’.26 In 1909 Schoenlank and Schanz particularly emphasized the socio-political significance of the journeymen’s associations, regarding them as a model for unions, or the starting point of a line of continuity leading to them. ‘The journeymen’s movement had three primary areas of concern: wages, hours and job referral.’ They stressed that the issue of wages had already led to conflicts between masters and journeymen.27 Later, in 1929, Rudolf Wissell would concur with this position, documenting the ‘journeymen’s struggles’ spreading out from the journeyman woollen weavers’ wage conflicts in Constance in 1385. In his view, economic problems were the cause of the dispute.28 Eduard Bernstein was the foremost turn-of-the-century advocate of the opinion that journeymen had had ‘no progressive socio-political ideas’. It was not higher wages that they were after and their views of society had been conservative.29 Bernstein, in common with Karl Kautsky in Germany and the Webbs in England, did not see a line of continuity running from the journeymen’s movement to the labour movement, a point which constituted virtually the whole of his approach. In his view, there were no deep-seated social antagonisms or economic class differences underlying these occasional conflicts. Strikes came about spontaneously.30 Bernstein’s supporters such as Kampffmeyer objected to the traditions of the labour movement: ‘the ideas of the socialist wage-earning proletariat cannot spring up out of the conservative constitution of guilds bound together by formal articles of association’.31 Gustav Schmoller had in fact pointed out the similarities even in his early work. In his Outline of Political Economy, he reduced the ‘journeymen’s movement inflated into a proletarian class struggle’ back to ‘conflict between old and young within the same class’. He also conceded the existence of strikes since the late Middle Ages, but as a consequence of prohibitions and custom they were infrequent and did not play a major role: ‘it was only when the state permitted labour organizations in the second half of the nineteenth century that the labour force became a factor in the labour market’. Wilfried Reininghaus has already referred to the fact that Schmoller’s assessment has continued to shape opinions expressed in more recent discussions of this issue and has generally characterized recent scholarship. The more recent historiography shows its debt to its predecessors. Kaelble and Volkmann in 1972, fortifying themselves by reference to a 1952 article by Hobsbawm,32 summed up protest actions in pre-industrial and early industrial crisis situations as follows: ‘without organization and a tactical point of view, spontaneous protest was all that remained to those involved’.33 Hans-Jürgen Gerhard, referring specifically to the building trades, stated that pre-industrial and early industrial labour markets were characterized by ‘the almost complete lack of organization of those offering their labour for hire’. It was, therefore, by no

Reith

119

means surprising that ‘the number as well as the magnitude of wage rate changes remained strictly limited’.34 Christiane Eisenberg referred to this assessment as well: with regard to the power exerted by journeymen in the labour market she maintained that ‘nevertheless, in the eighteenth century, strikes were hardly appropriate means for journeymen’s associations to help their members to earn higher wages’.35 Even though Michael Stürmer acknowledges that the strikes of journeymen in the crafts and trades did indeed achieve success, he stresses that ‘the rules of the game governing wage strikes were unknown in the eighteenth century; they would be a laborious development of the Industrial Age’.36 Thus, Wolfgang Kaschuba speaks of a ‘traditional means of dealing with conflict’ during the age of the ‘old crafts and trades’. Reactions on the part of ‘journeymen workers’ to business cycle developments in the wage-price sphere were a recent phenomenon which he dates to the period shortly before and during the Revolution of 1848.37 More recent scholarship, the work of Schwarz and Grießinger in particular, has called attention to the fact that wage disputes and strikes for higher wages were widespread during the second half of the eighteenth century and especially during the 1790s.38 They interpret the actions of journeymen as a learning process, whereby they concur with Eric Hobsbawm: ‘The worker’s effort, or standard of output per unit of time, was determined by custom rather than market calculation, at any rate until he began to learn the rules of the game’.39 Arno Herzig follows this line of argument: in the 1790s, the Hamburg journeyman tailors had learned to tie their wage scale to the rate of inflation ‘and they had learned to implement the strike for the purpose of wage increases’.40 In his pioneering study, Grießinger sees ‘forerunners of modern strikes’ in journeymen’s labour actions during the period from 1743 to 1755 and considers the strikes – mostly successful actions aimed at the creeping decline in real wages – as an ‘indication of the learning process’.41 The erosion of traditional patterns of action went hand in hand with the acquisition of new patterns of behavior attuned to the realities of the market. When discussing this more recent scholarship, it should be made clear that all the approaches cited here seeking to explain the origin and spread of the wage strike refer to the workforce’s organization and processes of consciousness. Whereas Herzig sees successful learning processes and thus a certain degree of continuity, Grießinger speaks of a learning process which came to a dead end, although he has modified this position in a later paper on the strikes in the German building trades.42 Most efforts to explain the conflicts and to situate them within the business cycle have relied on the Labrousse cycle: crop failures led to price increases or to higher grain prices and to higher costs to households for basic foodstuffs for which demand was inelastic. The consequence was a decline in demand for commercial products. The effects upon journeymen throughout the crafts and trades were unemployment, a decline in real wages or a decrease in nominal wages. Grießinger also regards the price of grain or rye as the most reliable indicator of the social welfare of journeymen.43 Attempts to trace the development of journeymen’s wages

120

Wage forms, systems and conflicts

typically refer to the well-documented wages in the building trades. But these are not typical for crafts and trades as a whole since wages in this sector consisted in most cases of only a cash payment. Conclusions drawn in this way are meaningful only for some crafts and trades: they include only one form of payment (the day rate) and only one form of income (the money wage). In my opinion, the critical deficiency of this approach lies in the fact that it fails to take into account other forms of income and wages.

Systems of payment, wage forms and wage disputes in German crafts in the eighteenth century The point of departure for an analysis of labour and wage conflicts is the hypothesis that labour conditions and wage scales themselves, as well as the changes they went through, must be made an object of investigation in order to understand such conflicts in the eighteenth and nineteenth centuries. This might seem to be self-evident at first glance, but a perusal of the scholarly literature in the field shows that research has hardly focused on the question of payment systems and wage forms. An analysis of wage conflicts therefore presupposes an analysis of wage conditions. For the eighteenth and nineteenth centuries, we already have available supraregional surveys and comparative wage statistics that enable us to discern the distribution of payment-systems (time, hours of work, contract, premiums, reductions, measurement) and wage-forms (cash, natural and non-monetary components, benefits, mixed forms).44

Systems of Payment The information cited above as well as other sources provides us with a cross-section of the systems of payment during the second half of the eighteenth century.45 For instance, in most of the textile trades, encompassing a massive work force, piece rates continued to be the dominant wage form, but in the textile finishing trades time rates were more widespread. In the clothing and leather trades piece rates steadily increased their range of diffusion, catching on in major crafts such as the tailors and shoemakers. In woodworking trades, time rates may well have predominated although expanded sales gave rise to piece rate production here as well. In paper manufacturing, as in numerous other sectors, we encounter a mixed form: the premium time rate. In the book-printing trade, on the other hand, both time and piece rate wages were common, depending on the size of the workshop – these were characterized as Berechnen (calculation) and Gewißgeld (fixed base pay). In the metalworking trades the piece rate assumed increasing significance, though it was more widespread in the manufacture of small iron products like nails and rivets than in the production of luxury goods. Food manufacturers commonly calculated wages based on longer terms, so that the half-yearly wage was most frequently

Reith

121

found here. In the service trades as well, payment was almost exclusively according to time rate. Finally, in the building trades, the daily wage was typical. Where technical considerations prevented piece rate wages from being implemented the time rate remained the universal form of compensation. In the chemical technology sector, for example, only a time rate was possible: the intensity of work and duration of a process could hardly be controlled by the workforce, so that, at best, influence could be exerted upon production by the use of premiums. A precondition for piece rate wages is the ability to measure a given quantity of work, which in turn assumes a certain degree of standardization of a unit of output. Repair and single-piece work thus lent themselves to time rates. From a business perspective, increased demand, which could assume a variety of forms, constituted an important precondition for piece rate wages. Key factors that brought about piece rate production certainly included the demand by merchants for exports, sales at trade fairs and the supply of the armed forces; nevertheless, purchases by itinerant retailers as well as demand from the outlying rural areas and even local urban neighborhoods could bring about the start-up of piece rate production. Where demand was limited, as in crafts producing primarily for local markets, the discourse on the form of the wage was also a conflict over the allocation of the volume of work.46 The argument that piece rate wages made for ‘lazy journeymen’ who delivered ‘sloppy work’ was employed in this discourse as a strategic topos by opponents of this form of compensation, just as it was used in other cases to resist wage demands or to force wage reductions. Since piece rate wages appeared, as a rule, in conjunction with increased demand, and a clear connection can be established with business cycle upswings, we can thus maintain that it did not have a primarily disciplinary function, but was rather aimed at managing periods of intensified demand. On the other hand, this also made possible a certain contraction of craft production once demand had slackened again. Here, I would again draw attention to the various forms of piece rate: besides being implemented on a permanent basis, it could be used as a subsidiary form of compensation – and thus for a limited time – or in the same shop but only for certain workers, processes or products. Piece rate wages could ensure flexibility, elasticity of supply and, at least, for a limited time, could lead to increased productivity both in urban as well as rural crafts and trades. The spread of piece rate wages during the time frame under consideration here cannot be construed as a linear development. Significant factors accounting for this were its function and the process of adjustment of the form of wage to the development of the business cycle and to changing levels of demand. We are therefore only able to identify various phases or trends when piece rate wages assumed importance in small-scale production – for example, the period up to the middle of the sixteenth century, the second half of the eighteenth century and especially during the Seven Years’ War, as well as, finally, the period up the last quarter of the nineteenth century when it again lost significance in small scale production.

122

Wage forms, systems and conflicts

Nevertheless, we should not focus exclusively on piece rates; the time rate wage also proved to be an efficiency wage in many instances, although certainly in an indirect form. Where piece rates could not be implemented due to technical constraints, time rate wages were established in conjunction with premiums. In many cases, we find a demand for a budgeted output quota per day or week, combined with a schedule of deductions or even bonuses. If we take these mixed forms into account, we get the impression that virtually all wages were efficiency wages – even though in an indirect form.

Wage Forms Let us now look at wage forms and incomes. In the early modern period, a journeyman’s wage consisted of both cash and a wage paid in kind (room and board): that is to say, it had both monetary and non-monetary components. With the exception of the building trades and, to a certain extent, also the textile trades which permitted married journeymen and where cash wages were paid almost exclusively, journeymen were, as a rule, integrated into the master’s household and lived under the same roof – so non monetary wage components must also be taken into consideration. Ulf Dirlmeier points out that in the late Middle Ages, full room and board constituted at least 40 per cent of a journeyman’s total monthly wages and that the proportion was most often 50 per cent. 47 Analysing the crafts and trades journeymen in Saxony during the middle of the sixteenth century Helmut Bräuer assumes a relationship of 55 per cent for cash wages and 45 per cent for room and board.48 In summarizing the situation of unmarried crafts and trades journeymen in Bremen during the eighteenth century, Gerhard Schwarz indicates that ‘the money wage paid to unmarried journeymen in addition to room and board was constantly low’.49 In a comparative perspective the wage in kind was a specific consequence of a labour market which was based to a considerable extent on the migration of mobile journeymen in an area dominated by small and middling towns. Since the late Middle Ages a flexible labour market had taken shape in Central Europe and an institutionalised form of labour migration developed through the movement of travelling journeymen. Here lies a significant difference between conditions in Central Europe and those in England, the Netherlands, Italy and to some extent also France. This typically Central European form of labour migration distinctly characterised specific social forms whose characteristics were to a considerable extent formed by migration: workers living in the households of craftsmen. Compared with Central Europe, in England ‘the most important single source of industrial conflict was the money wage. In times of rising prices, most markedly in wartime, all workers with bargaining power made use of it, and for such periods it is meaningful to identify strike ‘waves’ and labour ‘movements’.50 But in England, as also in the Netherlands,51 journeymen and skilled workers were mostly married, so living in the master’s household was more the exception

Reith

123

than the rule. In France it seems that wages in kind were not, in the eighteenth century, as widespread as in Central Europe.52 This leads to the conclusion that wages included both a monetary and non monetary component. Here, I am in complete agreement with Patricia van den Eckhout that it does not make sense to reduce the concept of wage to the money wage alone.53 In the light of the significance of room and board, it is no surprise that food constituted a key point of conflict, and we must therefore consider these conflicts as wage disputes. They ranged from isolated disputes that involved only a single workshop all the way to work stoppages that encompassed an entire journeymen’s association or the whole guild. Thus, in analyzing action taken in times of conflict, it is important to know the role of payments in kind, as well as the consequences of price increases and price fluctuations. Rising prices for foodstuffs adversely affected both those journeymen who received most of their wages in cash – which is to say married journeymen – and masters.54 During periods of inflation the payments in kind received by journeymen who lived and ate in their master’s household represented a type of social welfare insurance. In a strict sense, their real wage rose; at the very least, we may speak of a balancing out of inflation. The extent (quantity and quality) of the board they received, however, could vary considerably. Since their income represented a combined value, conflicts affected single components of their total wage, or even several components.

Wage Disputes Let us first consider conflicts involving meals: Whereas a certain amount of reintegration of the journeymen into the masters’ households was becoming characteristic as early as the late sixteenth century,55 strong tendencies toward reversing this and doing away with the performance of certain duties can be observed from the late seventeenth century. During both the sixteenth and eighteenth centuries there is a noticeable trend towards replacing certain components of the board provided (rounds of drinks, breakfast) with a cash payment; that is, to compensate for them with a fixed sum of money. In the eighteenth century several work stoppages and conflicts also give the impression of a tendency to do away with the drink provided to journeymen.56 Along with the drink, breakfast was another case in which payment in kind was being replaced by cash allowances, although here references exist only to crafts and trades in middling and large cities. Whereas, for example, the journeyman goldsmiths in Hanover received in 1765 a breakfast ‘just as good as the one for the master himself ’, the master wig makers, coopers, furriers and locksmiths gave their journeymen only a sum of money with which they had to provide their own meals.57 In 1746 the journeyman joiners in Brunswick stopped work because the masters had reduced their breakfast: ‘they continued to serve only bread, whereas

124

Wage forms, systems and conflicts

during better times they had provided, of their own free will, butter, cheese, bread and fat’. Since the journeymen had nailed the cheese to the workshop wall just to spite them, the masters had to put an end to breakfast. In Dresden following the Seven Years War, the master joiners attempted to reduce the breakfast allowance from six Pfennigs to three. The conflict escalated and 150 journeymen went on strike. Following an investigation, the masters were instructed by the city administration to pay only three Pfennigs, whereupon the journeymen petitioned the elector. They pointed out ‘that in Leipzig, Berlin and other large cities, the journeymen had always received six Pfennigs for breakfast, and in other cities... the journeymen got as much bread as they could eat, along with either cheese or butter for breakfast.’ Since the journeymen also made reference to the rising price of grain – the loaves had become somewhat smaller – it was finally suggested to the masters that they ought to give ‘their journeymen a six-Pfennig loaf of bread for breakfast a little while longer’. Finally in 1805, the journeymen locksmiths of Augsburg demanded in connection with St Monday (when they were nursing hangovers from the previous night’s revelry) that ‘if the masters are prepared to permit the journeymen to get up an hour later and to improve the usual breakfast, then they (the journeymen) would also be ready to work on that day in the future’. The amount of food provided also varied to a great extent: that is, the journeymen did not necessarily receive ‘full board’ every day of the week. On the evening of St Monday, journeymen had no right to receive board. In the late eighteenth century conflicts also arose over the issue of meals on Sunday evening. Conflicts frequently flared up over the quality of the food. During Michaelmas Fair in 1763 in Leipzig, a conflict broke out between the journeymen tailors and their masters. The city administration finally ordered that a weekly wage should be paid in addition to the board provided – as for the latter, roast beef was to be served twice a week with soup, vegetables and meat as well as bread and butter on the other days. A journeyman reported in his autobiography that ‘it didn’t take long before sausage and even herring, which was called ‘tailor’s carp’ in Leipzig, began appearing on their plates’.58 A walk-out by journeyman joiners in Bremen in 1802 was caused by a similar conflict: three journeymen refused to work for a master because they considered a lunch consisting of large beans and herring to be ‘lousy grub’. If we attempt to establish a connection between the payment system and these disputes about the food served, we notice that the conflicts were, for the most part, limited to the amount of wages paid per amount of time worked (and weekly wages in particular). In a few cases, there was even a close relationship between board and hours: if the master reduced breakfast, the journeymen worked less or vice versa. Journeymen who worked on the basis of time rates (weekly wages) as a rule received board and lodging (as well as the performance of other services having the nature of a salary) together with their cash wage. For the masters, particularly those in crafts and trades in urban centers, the costs of this board came into play. The disputes sketched here were triggered by price fluctuations, to which the

Reith

125

response of the masters was a reduction of board, drinks or breakfast. Components of the board hitherto provided were replaced by compensation; that is, they were paid as a cash allowance. The journeymen thus became much more sensitive to price changes and the monetary portion of their wage assumed greater significance. This was especially the case where journeymen were paid according to a piece rate, or in sectors such as the building trades where the cash wage was virtually all that the journeymen received. But journeymen who were paid according to a piece rate and who received board in their master’s home had to pay the master an agreed sum (board fee, or kitchen fee) for meals received, a sum established according to the price changes of basic foodstuffs. Wages were based upon the amount of work performed and the journeymen paid for their board. We find evidence of board fee as early as the fifteenth century, above all in the textile sector. We can guage the precarious balance between the cash wage and the board fee from the conflicts that flared up on this issue among the lacemakers from the late sixteenth century. In Munich, the clashes over the amount of the board fee and the meals provided in return continued over a protracted period of time. In 1771, the masters and journeymen agreed on a board fee of sixty Kreuzer. After foodstuff prices had declined, however, the board fee would be set back to its original level of forty-five Kreuzer. Finally, in the summer of 1777, the journeymen demanded that this be done. When the masters refused to comply, the journeymen went out on strike en masse. The board fee was finally rolled back to its old level of forty-five Kreuzer. Nevertheless, the masters insisted upon receiving sixty Kreuzer because ‘even in these times of tight money’ they had to provide their journeymen with proper board in addition to their lodging. In 1806, several masters finally refused to serve dinner, despite the fact that the journeymen were demanding only soup as their evening meal. The masters were reminded of the 1771 agreement, but they now asserted that it was not the practice in Vienna and Prague to serve ‘full board’. Another initiative of the masters came in 1817, ‘due to the hard and pressing times’, with the demand that, in the future, only ‘breakfast and lunch as previously’ would be provided in return for the board fee.59 The close connection between piecework wages and board fee can best be illustrated using the example of the shoemakers. Among the latter, piecework wages had been paid since the fifteenth century and major jobs, particularly army contracts, were mostly paid in piecework wages.60 Beginning in the late seventeenth century, there were more and more references to the ‘producer’s wage’ (Macherlohn). In Leipzig in 1669 some of the masters paid piecework wages per pair of shoes; others demanded that payment be made on the basis of a day’s work.61 During the second half of the eighteenth century the practice of paying shoemakers piece-work wages became increasingly common in major urban centers. In Brunswick, Hanover, Leipzig and Dresden this had already been the case prior to the Seven Years’ War. In Frankfurt in 1770 the masters reported that in their craft labour was paid almost universally on a piecework basis, sometimes

126

Wage forms, systems and conflicts

with board included, sometimes not.62 In Munich during the last third of the eighteenth century, ‘full board’ was abolished with the introduction of piece-work wages; in Prague, Vienna and Brünn (Brno) in the 1780s lunch was the only meal with which journeymen were still provided , in return for board fee. In Königsberg, Prussia (1805) and in Berlin (1827), the journeymen worked ‘almost without exception on a piecework basis’, and it was no longer customary for them to be provided with board. Whereas disputes about board and working hours were typical for journeymen who worked for a weekly wage and resided in masters’ households, in the crafts and trades where piece-work wages (or daily wages) were paid, conflicts centered upon the board fee or the cash wage – ultimately, upon the amount of money paid.63 Turning our attention now to the conflicts having to do with cash wages, we notice various different causes for disputes, such as punctual payment of wages, the way in which wages were agreed upon, and the choice of the form of compensation. Highly skilled journeymen were better off receiving a piecework wage, assuming that demand was sufficiently high. Masters doing a brisk business preferred to pay piecework wages, and masters with small-scale operations preferred time wages. Demand was always the determinative element. Bonuses were an additional cause for conflict. Compared to the amount of salary, bonuses were certainly of secondary importance, but they nevertheless constituted a stable element of a journeyman’s income, and particularly in those cases where, in spite of the piecework wages, the labour contract provided for a certain quota (in a particular period of time) and the journeyman had a claim to a bonus once that goal had been reached – as was the case with the shoemakers. The Munich journeymen shoemakers went out on strike in 1739 and again in 1765 to demand the payment of the so-called ‘quarterly shoe’ (Vierteljahresschuhe) even if the master laid them off in the twelfth week. In Hamburg in 1795, the ‘food allowance’ as well as the ‘half-year’s shoe’ were among the demands of the journeymen shoemakers (including those who did not reside in the masters’ homes), while in 1796 in Leipzig, 200 journeymen demanded the ‘yearly shoe’. Conflicts also flared up about the ‘food allowance’ and the ‘half-year’s shoe’ in nineteenth century Hamburg, since in 1824 the smaller masters regarded these demands as ‘oppressive’. When the authorities tried to do away with both benefits in 1825, one hundred journeymen walked off the job, whereupon an agreement was reached on both issues, although these conflicts continued to flare up into the 1830s.64 Wage strikes could assume an offensive as well as a defensive character. For example, the Regensburg journeymen silk stocking makers reacted to a reduction of their piecework wage rates in 1740 with a work stoppage, Nuremberg journeymen wire makers responded to a reduction in wages by walking off the job in May 1801. Most documented strikes, however, aimed to achieve wage hikes. In most cases, a strike was only the last resort following a long series of confrontations, since it entailed high costs for both sides. Grießinger has pointed out the clever planning done by the

Reith

127

journeymen, who understood how to exploit times of high demand, for example, around trade fairs and labour shortages.65 Such wage conflicts, played out within shifting framework conditions, were mostly long, drawnout affairs. Wage conflicts, especially in northern Germany occurred above all during the Seven Years’ War in the context of a substantial rise in the price of basic foodstuffs as well as increasing debasement of the currency. The inflation led to wage conflicts among the shoemakers in Berlin in 1760; in December of that year, the journeymen’s piecework wage rates were increased ‘until the prices of said commodities dropped’. In Bremen, the debasement of the coinage and the ‘galloping inflation of food prices’ were the causes of a strike by the journeyman rope makers that forced the masters to raise wages in August 1761. With the demand for a wage increase ad duplum, the journeymen shoemakers of Hildesheim walked out in June 1762 and ultimately abandoned the town completely. In October 1763, during the Michaelmas Fair and the seasonal peak of demand, the Leipzig journeymen tailors left their masters ‘stranded with the necessary work waiting to be done’ and demanded to be paid ‘eight good Groschen’ instead of a Taler in bad money. A wave of offensive wage conflicts also occurred in the 1790s.66 In 1791 the Hamburg journeyman (daily paid) tailors demanded a wage increase and in 1792 they went on strike to force payment of those wage rates. Additional strikes occurred in 1794, 1795 and 1796 with the demand for wage hikes. With increased demand due to the war against France, the Cologne journeymen saddlers demanded a rise in 1792. In Frankfurt on the Main in February 1793 some of the journeymen shoemakers engaged in a work stoppage ‘because due to the current inflation they wanted an increase in the wage they had been receiving until then’. Higher piecework wages were demanded since the war had caused foodstuff prices to climb dramatically. After an investigation by the city council wages were finally raised. However in December 1798, when food prices had fallen, the masters rolled wages back again. In 1795, the Nuremberg journeymen nailmakers walked out and demanded ‘that, particularly in these terribly expensive times, and in light of the fact that they had to provide their own food, they be paid somewhat better wages.’ The Hamburg journeymen shoemakers asked for higher wages in May 1795, and since the masters only partially agreed to their demands, 300 to 400 journeymen walked off the job for a day. In a similar manner, the Nuremberg journeyman shoemakers went on strike in July 1795, also demanding a wage increase. In 1798, a substantial rise in the price of basic foodstuffs led to numerous conflicts. The Augsburg journeyman woollen weavers demanded a rise of a Kreuzer in March 1798, since ‘everything from head to toe has become dearer’ and in November 1798, the journeymen tailors in Dresden demanded a hike of their daily wage ‘not out of greed or just because they felt like it’, citing cited the tremendous inflation.67 In those place where it was customary to pay the tailors (as well as the joiners) a weekly salary, conflicts simultaneously flared up about working hours,

128

Wage forms, systems and conflicts

particularly concerning work by candlelight in the evening. 68 In Leipzig, the dispute had been brewing since 1797 and escalated in 1802, when the ‘week workers’ (paid on a weekly basis) walked out citing the shorter working hours of the day-journeymen (particularly in winter). Suffice it to say that strikes for higher wages were to be found, above all, among the tailors, shoemakers, cloth shearers, rope makers, lace makers and weavers. As a result of the constellation of their incomes consisting of piecework wages or even daily wages, the journeymen in these crafts and trades were bound more tightly to market conditions.

Wage disputes in the building trades This was true to an even greater extent in the building trades. As a rule, these journeymen did not live with their masters and they received their daily pay in the form of cash. With a longer work day in summer, they earned more than in winter. Their year could also be divided into three or more phases, depending upon the hours of daylight and the number of hours they were able to work. During the nineteenth century there was a tendency to subdivide the year even further (in Hamburg in 1846, the carpenters differentiated between eleven distinct phases, each with its own wage rate; in 1854, the masons and carpenters agreed upon seven), thereby manifesting the final stages of a trend toward hourly wages that finally prevailed in the second half of the century.69 In Nuremberg, for example, the rather small proportion formed by payments in kind had already been converted to a drink allowance by the end of the seventeenth century; in Augsburg, according to the wage schedule of 1718, a specified sum was given for beer and bread, or this sum was simply paid out. Thus, the essential inflation insurance function previously performed by payments in kind was omitted.70 A journeyman drew his monthly wages and, to the extent that he or his wife had no private garden plot to cultivate, was directly exposed to the risk of price fluctuations. Especially for married journeymen, monetary wages were crucial to their very existence. Up to now, historians have found very few references to strikes for higher wages in the construction sector during the second half of the seventeenth century. Nevertheless, such strikes constituted only the ultimate and most serious form of conflict. There were, to be sure, several other possible ways to induce the general contractors or the guild masters to grant wage increases without thereby provoking the authorities responsible for setting pay scales in the building trades and who, employers themselves, were interested in wage stability. During the second half of the seventeenth century, real wages in the building trades were relatively high and the demand for labour remained strong as a result of intense building activity necessitated by reconstruction following the Thirty Years War. The long-term loss of purchasing power of wages during the eighteenth century was the cause of numerous strikes; the concrete issues

Reith

129

involved were wages, the Meistergroschen (a fee paid by the journeymen to the masters) as well as the journeymen’s traditional right to lumber as a form of supplemental benefit. Above all, the strong inflation during the 1790s formed the background factor for numerous strikes in the building trades. In Frankfurt on the Main, a series of conflicts took place beginning in 1790 when the journeymen carpenters complained: ‘Hardly any of the other tradesmen experience current conditions to the same extent we do. Some of these artisans don’t have to pay for their own room and board as we do because their masters provide it. And some of these can change their wages at any time as they see fit and thereafter look on without batting an eyelid’.71 Wage demands made at the start of the building season offered the best promise of success in the construction trades. In the North the issue of wages most often led to conflicts, whereas in the South the concern was stronger control of the labour market.72 Different groups competed on this market: the married journeymen comprised a fixed (immobile) labour force supplemented by local unmarried journeymen, commuters from surrounding towns, seasonal workers and itinerant journeymen. As a whole, the building workforce was less mobile than those in other professions.73 For example, in 1765 Brunswick City Council passed a resolution that wages in the building trades should not yet be returned to their customary prewar levels; most of the foreign journeymen had left town and only the married ones remained. In the other crafts and trades, it was to be feared that the reintroduction of the old wage rates would cause the journeymen, most of whom were foreigners, to depart.74 To recapitulate our findings regarding forms of income and payment: for journeymen who worked for a weekly wage or who received their pay every six months and who received room and board in their master’s household, disputes were usually concerned with food and working hours, whereas for day labourers, piece workers and those whose wages were paid in cash, conflicts typically centered around board fees and the pay scale. The impact of price fluctuations of basic foodstuffs was completely different for these two groups. For those earning a weekly wage (with a high proportion of payment in kind), real wages rose during periods of rising prices (as long as the master did not reduce the amount of food served); real wages of piece workers sank during periods of inflation and rose when the cost of living went down. In the light of the connection between business cycle developments and strike frequency the inference which suggests itself is that strikes for higher wages on the part of piece workers and those whose wages were paid in cash characterized phases of rising prices for basic foodstuffs. However, strong demand together with fixed delivery deadlines opened up possibilities of bargaining. On the one hand, the fact of widespread strikes for higher wages in the late eighteenth century must be seen against the background of sharply rising prices; on the other hand, journeymen also exploited boom phases (such as army supply contracts) to achieve higher wages.

130

Wage forms, systems and conflicts

Wages and wage conflicts during the nineteenth century If we now take a look at nineteenth-century wage conflicts and strikes for higher wages, the information published to date makes it clear that the number of strikes declined sharply in comparison to the eighteenth century. Whereas a total of eighty-two work stoppages are documented for Augsburg from 1700 to 1806, there are records of only two (unsuccessful) actions during the period from 1806 to 1848, each of which was confined to a single shop.75 In Nuremberg, the city with the highest incidence of strikes during the 1790s, strikes became a rarity. ‘The journeyman hatters and tailors struck over wages in 1821 and 1865, respectively, and the journeymen brass moulders struck in 1835 to prevent the city from starting a hospital fund which might supplant their association’s own fund’.76 In Munich, wage demands led to a strike of the journeyman tailors and shoemakers in May 1848, but these failed.77 A higher incidence is shown in Hamburg with strikes by carpenters in the shipbuilding industry (1806, 1821, 1839), masons (1821, 1836, 1840), shoemakers (1825, 1835),78 goldsmiths (1835) and potters (1835).79 Most striking is the fact that the conflicts or strikes involved essentially the same occupational groups as in the eighteenth century. According to Ulrich Engelhardt, bookmakers, tailors, shoemakers and construction workers led the great wave of strikes in 1848; in the second strike boom of 1857, it was the construction workers and tailors.80 Although there had also been phases during the eighteenth century with a low incidence of strikes, and the 1790s can by no means constitute a standard of measurement, the low number of strikes during the first half of the nineteenth century nevertheless calls for an explanation. The supra-regionally coordinated termination of the journeymen’s funds at the end of the old Reich had certainly weakened the effectiveness of the journeymen’s guilds.81 The right of journeymen to form organized associations was greatly limited and they could no longer control substantial sectors of the labour market as they had done in the past. The local authorities could take repressive steps against journeymen’s guilds and their culture, as Neufeld has shown using the example of Nuremberg, although cities like Hamburg exercised greater moderation. Nevertheless, state control and repression alone are insufficient to explain the low incidence of strikes. It is also possible that a long-term shift in the labour market from a mobile, supra-regional to a regional labour market featuring a majority of native-born journeymen, together with the conflicts this would have brought with it, could have weakened the position of the journeymen’s guilds in the labour market. After all, an important factor upon which the power of the journeymen was based was their mobility and the social forms connected with this mobility. But the extent to which the abolition of mandatory tramping had an impact on migration and the labour market in the nineteenth century is still a grey area in scholarly research. Nor has there been much concrete quantitative data collected up to now about a tramping journeyman’s chances of finding employment

Reith

131

and since the role of the tramping artisan in the context of industrial and “modern” modes of migration has received virtually no attention the current state of research does not permit a comparative assessment of the labour market factor.82 However, besides official repression and the transformation of the labour market, it is important to focus on an additional factor – wages. How did wage forms and payment systems change?

Systems of payment and wage forms The Survey of Berlin Business Conditions in 1827 indicates that piecework wages were paid in thirty-six of fifty-eight occupational groups studied, and that they were the exclusive mode of compensation in fifteen of them. Room and board had declined sharply in significance, though with certain differences specific to particular occupations.83 Even taking into account Berlin’s somewhat special status, piecework wages were widespread in urban crafts and trades elsewhere , as illustrated by the example of Leipzig, where daily and piecework wage production in the major crafts and trades providing for basic human needs approached ‘salaried labour’.84 In Hamburg too, piecework wages had become more widespread. In 1848, the Hamburg archivist Lappenberg wrote that ‘most artisans work for piecework wages, the amount of which is freely negotiated between the masters and the journeymen. The latter’s earnings thus depend upon his physical power, his skill and endurance, as well as upon the number of customer orders to be filled, and finally upon the competition from equally skilled workers’.85 The 1847 Overview of the Weekly Wages of Workers Employed in Industrial and Commercial Businesses in the Royal District Capital of Augsburg shows that in numerous crafts and trades, cash wages only were paid.86 The report to the government of Upper Bavaria, filed by their expert in charge of commercial affairs in 1849, is also highly revealing: ‘in more and more crafts and trades,’ it is now the case ‘that masters and journeymen no longer wish the journeymen to be provided with room and board in the master’s home’. Journeymen, especially in crafts and trades in which piecework wages were paid rather than remuneration according to time worked, preferred not to work in the masters’ workshops, but rather at home.87 An 1851 survey conducted by the municipal authorities of Nuremberg, however, indicates that the majority of the journeymen still lived and ate in their masters’ homes, but board and other payments in kind ranged over a wide spectrum.88 At least in the case of crafts and trades in (large) cities, a number of trends can be observed. The piecework wage became more widespread, and payments in kind declined in significance. The latter were paid increasingly in monetary form, whereby the wage increasingly became a cash wage. This long-term trend did not proceed in linear fashion, and was repeatedly interrupted by conflicts. These disputes, having to do with the wage form and the payment system, were played out between the ‘large’ and ‘small’ masters, and ultimately focussed on the distribution of the

132

Wage forms, systems and conflicts

volume of work. The polarization of the masters can be said to have corresponded to a polarization of the journeymen. With regard to working conditions and wage rates, lines of conflict and interest groups also formed within the journeymen’s organizations, which were by no means homogeneous with respect to the wage forms and payment systems. Even when, as in the case of the Hamburg shoemakers, fixed wage scales were suspended (1832) and the establishment of wages was left to freely negotiated agreements between masters and journeymen, collective action was hardly to be expected.89 The example of the tailors can be used to provide a brief sketch of this process of differentiation. In the context of army contracts, there had been disputes in Frankfurt on the Main since the Seven Years’ War (1758) whether daily or piecework wages were permissible forms of compensation.90 But even after the war several masters voted to continue to allow the payment of daily wages. Around 1800, in any case, the daily wage had established itself along with the weekly wage as a payment system and had also gained importance in other large cities such as Breslau and Dresden. In the 1790s in Hamburg, the journeymen tailors on a daily wage had made their presence felt through several different wage actions. In 1797 in Bremen, journeyman tailors demanded to switch from a weekly to a daily wage in order to enable them to provide for their own board.91 But within the journeymen’s organizations as well, the payment system – and thus also the wage form – proved to be a line of division. In Regensburg, following conflicts between journeymen paid on a daily and weekly basis, day labour was once again abolished.92 In the 1820s in many different cities, however, an increasing number of day-wage and piecework-wage journeymen were working alongside those paid on a weekly basis. When a regulation forbidding piecework was issued in Leipzig in August 1820, workers walked off the job. The journeymen demanded a hike in the daily wage, and the masters were forced to acknowledge that ‘only a few of their journeyman tailors showed up for work’.93 The argument about the piecework wage flared up again in 1828, as several masters voted to suspend the ban with the argument that six pieceworkers could do the same amount of work as nine labourers being paid a daily wage. Other guilds in Leipzig as well as those in smaller towns permitted shoemakers, cabinetmakers, bag makers, glaziers and other artisans to work for a piecework wage.94 In 1840, 130 masters petitioned to have the ban suspended, arguing that piecework labour in Leipzig as well as ‘in all large and even middle-size cities in Germany is becoming increasingly prevalent’. The guild representative also put himself behind this effort since ‘production is on a piecework basis in the most important workshops here’.95 The spread of labour on a day and piecework basis did not remain limited to Leipzig. A survey by the municipal authorities in Augsburg came to a similar conclusion: in Augsburg itself in 1836, in the wake of protracted disputes, there were hardly any journeymen left working for a weekly wage.96

Reith

133

Three different payment systems, and thus also three different wage forms, coexisted or competed with each other. The pieceworkers and day labourers did not live in the master’s house as did those journeymen receiving a weekly wage, nor did they receive board; they could be laid off on shorter notice and had shorter working hours, but they could earn a higher wage, particularly the pieceworkers. The divergent working and living conditions corresponded to different communities of interest, which hardly gave occasion for concerted action. The research hypothesis that suggests itself is that it was only the long-term trend toward similar wage forms and payment systems that (re)established shared interests as a basis for joint action. After the middle of the century, arrangements whereby journeymen received room and board from their masters became less and less significant. The legal constraints to live in the master’s house were mostly repealed in the 1860s.97 Among journeymen, boarding in the master’s household became the object of increasing criticism, and even the largerscale masters preferred compensation based on cash payments. In Leipzig in 1810, a regulation was passed mandating that journeymen had to reside with their masters and again in 1835 the municipal authorities decided to continue that policy so that the journeymen be kept ‘under proper surveillance and subordination by the police and the political authorities’.98 But in 1848, the journeymen shoemakers could point to the ‘deplorable state of affairs’ whereby two journeymen had to share one bed, and they succeeded in having a policy implemented whereby each man would be provided with either his own bed or an adequate housing allowance in cash.99 During the business cycle boom of the 1870s, there were work stoppages in several crafts and trades where room and board was still customary with demands to do away with payments in kind and the room and board system.100 Nevertheless there remained some few crafts and trades, particularly in the food industry, where agitation against the room and board system did not begin until the 1890s, and this form of compensation was still widespread in towns and rural areas after the turn of the century.101 During the 1880s, the piecework wage became even more widespread among small businesses. The 1885 Survey of the State of Small Businesses in the Grand Duchy of Baden elaborated on the movement of workers to the cities ‘where their stay is much more pleasant and, because it has become rare for masters to provide room and board, also offers a greater degree of freedom’. Room and board was said to continue to be the rule in small towns, but the exception in larger cities.102 The Survey of the State of Small Business in the Administrative District of Mannheim, conducted in conjunction with the above-mentioned 1885 enquiry, confirmed that piecework wage production had spread out from the urban centres to encompass additional occupational groups. ‘Where work is paid by the piece – which is the rule among shoemakers and tailors, and not at all unusual among coopers, tinsmiths, basket weavers, lathe operators, saddlers, cabinetmakers and glaziers – the earnings of the journeymen are

134

Wage forms, systems and conflicts

generally somewhat higher, but it is precisely here that large fluctuations resulting from individual industriousness and ability become apparent.’ In numerous instances, ‘one and the same worker, depending upon whether a new job had come in or repair work had to be delivered, would work on a piecework or a time wage basis’. The survey concluded that, on balance, room and board in the master’s household was the exception rather than the rule.103 The Enquiry into the State of the Crafts and Trades with Particular Consideration of their Competitiveness with Large-Scale Industry that was conducted between 1893 and 1897 by the Association for Social Policy (Verein für Socialpolitik)104 distinguished between Neuproduktion – the manufacturing of new products – and the repair of products. It showed that among small-scale proprietors – those unable to keep up in the area of new production, or at the least among those where Neuproduktion had declined considerably, and who had accordingly shifted to repair and installation work and the production of special one-of-a-kind items – timebased wages had gained in importance, whereas the piecework wage was increasingly applied as a form of compensation in ‘new production’ in the light of enhanced possibilities where the division of labour called for larger operations. As a form of compensation in Neuproduktion it had lost considerable ground in small businesses, where time-based wages played an increasingly important role. After the turn of the century, small firms increasingly paid their compensation on the basis of an hourly wage. The piecework wage which, from an historical perspective, had been used in all fields of production, now developed into a form of compensation for larger firms. This also outlines the historical situation that caused scholars in the field of economics to assume that the piecework wage and compensation based on output were modern phenomena, a creation of ‘large-scale industry’. The situation at around the turn of the century must have created the impression, even among economists, that the piecework wage and compensation based on output could not possibly have been a form of compensation in use by small business. And this was precisely the point of departure for the historical studies produced by the ‘historical school of German economics’. Translated by Melvin A. Greenwald

Notes 1. Reinhold Reith, “Lohn und Leistung aus der Perspektive der Historischen Schule der Nationalökonomie. Zum Problem der Wirtschaftsmentalitäten”, in Friedrich Lenger (ed.), Handwerk, Hausindustrie und die historische Schule der Nationalökonomie. Wissenschafts- und gewerbegeschichtliche Perspektiven (Bielefeld 1998), pp. 78 – 104, 81 ff. 2. Rudi Schmiede and Edwin Schudlich, Die Entwicklung der Leistungsentlohnung in Deutschland. Eine historisch-theoretische Untersuchung zum Verhältnis von Lohn und Leistung unter kapitalistischen Produktionsbedingungen (3rd edition, Frankfurt/M., New York 1978).

Reith

135

3. Ludwig Bernhard, Die Akkordarbeit in Deutschland (Leipzig 1903). 4. Otto v. Zwiedineck-Südenhorst, Lohnpolitik und Lohntheorie mit besonderer Berücksichtigung des Minimallohnes (Leipzig 1900); v. Zwiedineck-Südenhorst, Beiträge zur Lehre von den Lohnformen (Tübingen 1904). 5. Eugen von Philippovich, Grundriß der Politischen Oekonomie, Vol 2, II. Teil (5th edition, Tübingen 1915), p. 334. 6. Gustav Schmoller, Die soziale Frage. Klassenbildung, Arbeiterfrage, Klassenkampf (Munich and Leipzig 1918), pp. 233 ff. 7. Johannes Conrad, Grundriss der Politischen Oekonomie, Vol I. Nationalökonomie - Allgemeine Volkswirtschaftslehre, 11th edition, ed. A. Hesse (Jena 1923), pp. 383 ff. 8. Werner Sombart, Der moderne Kapitalismus, Vol. III: Das Wirtschaftsleben im Zeitalter des Hochkapitalismus, 1st. edition (Munich and Leipzig 1927), pp. 670 ff. 9. Friedrich Lenger, “Die Gewerbegeschichtsschreibung der Historischen Schule. Einige zentrale Konzepte und ihr sozialpolitischer Kontext”, in: F. Lenger, Handwerk, Hausindustrie, pp. 9 – 18. See also Lenger, Sozialgeschichte der deutschen Handwerker seit 1800 (Frankfurt/M. 1988), pp. 110 ff. 10. Lujo Brentano, Über das Verhältniß von Arbeitslohn und Arbeitszeit zur Arbeitsleistung (2nd edition, Leipzig 1893), pp. 28, 32. 11. Ibid., p. 54. 12. Gustav Schmoller, Grundriß der Allgemeinen Volkswirtschaftslehre (Leipzig 1904), Vol. 2, pp. 266 ff. 13. Sombart, Der moderne Kapitalismus, Vol. 1, pp. 35 ff. 14. Sombart, Der moderne Kapitalismus, Vol. 1, p. 807. 15. Sombart, Der moderne Kapitalismus, Vol. 3, pp. 426 ff. 16. Peter Kriedte, Hans Medick and Jürgen Schlumbohm, Industrialisierung vor der Industrialisierung. Gewerbliche Warenproduktion auf dem Land in der Formationsperiode des Kapitalismus (Göttingen 1977), pp. 115, 138 ff. 17. Werner Sombart, Der moderne Kapitalismus, Vol. II: Das europäische Wirtschaftsleben im Zeitalter des Frühkapitalismus (2nd. ed., Munich and Leipzig 1916), pp. 829 ff. 18. Zwiedineck-Südenhorst, Lohnpolitik und Lohntheorie, p. 8. 19. Schmoller, Grundriß, Vol. 2, p. 305. 20. Sombart, Der moderne Kapitalismus, Vol. II, pp. 829 ff. 21. Otto Brunner, “Das “Ganze Haus” und die alteuropäische “Ökonomik””, in: Idem., Neue Wege der Verfassungs- und Sozialgeschichte (2nd edition, Göttingen 1968), pp. 103–127, 123 ff. 22. Jan de Vries, “How did the Pre-Industrial Labour Markets Function?”, in G. Grantham and M. MacKinnon (eds), Labour Market Evolution. The Economic History of Market Integration, Wage Flexibility and the Employment Relation (London, New York 1994), p. 39. 23. E. H. Phelps Brown and S. V. Hopkins, A Perspective of Wages and Prices (London 1981). 24. De Vries, ‘How Did the Pre-Industrial Labour Markets Function?’, p. 41. 25. For a European perspective, see: Catharina Lis and Hugo Soly, ‘An Irresistible Phalanx’: Journeymen Associations in Western Europe, 1300-1800, in: Catharina Lis, Jan Lucassen and Hugo Soly (eds.), Before the Unions. Wage Earners and Collective Action in Europe, 1300–1850, (International Review of Social History, Supplement 2), Cambridge 1994, pp. 11–52. 26. Ibid., p. 667. Georg Schanz, Zur Geschichte der deutschen Gesellenverbände (Leipzig 1877). 27. Bruno Schoenlank and Georg Schanz, “Gesellenverbände”, in: Handwörterbuch der Staatswissenschaften (3rd. edition, Jena 1909), Vol. 4, pp. p. 667. 28. Rudolf Wissell, Des alten Handwerks Recht und Gewohnheit, 2 vols, Berlin 1929, Vol I, pp. 458-500. (2nd ed., ed. Ernst Schraepler, 6 vols., Berlin 1971/88). Reinhold Reith, “The Social History of Crafts in Germany. A new Edition of the Work of Rudolf Wissell”, International Review of Social History, vol. 36 (1991), pp. 92–102. 29. Eduard Bernstein, Die Arbeiter-Bewegung (Frankfurt/M. 1910), pp. 17 ff.

136

Wage forms, systems and conflicts

30. Eduard Bernstein, Die Schneiderbewegung in Deutschland (Berlin 1913), Vol. 1, p. 13. 31. Paul Kampffmeyer, Vom Zunftgesellen zum freien Arbeiter (Berlin 1924), pp. 18 ff. 32. Eric J. Hobsbawm, “Economic Fluctuations and some Social Movements since 1800”, Economic History Review, 2, 1952, pp. 3–25. 33. Hartmut Kaelble and Heinrich Volkmann, “Konjunktur und Streik während des Übergangs zum organisierten Kapitalismus in Deutschland”, Zeitschrift für Wirtschafts- und Sozialwissenschaften, 92, 1972, pp. 528 f. 34. Hans-Jürgen Gerhard, Löhne im vor- und frühindustriellen Deutschland. Materialien zur Entwicklung von Lohnsätzen von der Mitte des 18. bis zur Mitte des 19. Jahrhunderts (Göttingen 1984), p. 13. 35. Christiane Eisenberg, Deutsche und englische Gewerkschaften. Entstehung und Entwicklung bis 1878 im Vergleich (Göttingen 1986), pp. 104 ff. 36. Michael Stürmer, Herbst des alten Handwerks. Zur Sozialgeschichte des 18. Jahrhunderts (München 1979), pp. 270 ff. 37. Wolfgang Kaschuba, “Vom Gesellenkampf zum Protest. Zur Erfahrungs- und Konfliktdisposition von Gesellen-Arbeitern in den Vormärz- und Revolutionsjahren”, in Ulrich Engelhardt (ed.), Handwerker in der Industrialisierung (Stuttgart 1984), p. 389. 38. Klaus Schwarz, Die Lage der Handwerksgesellen in Bremen während des 18. Jahrhunderts (Bremen 1975); Andreas Grießinger, Das symbolische Kapital der Ehre. Streikbewegungen und kollektives Bewußtsein deutscher Handwerksgesellen im 18. Jahrhundert (Frankfurt/M. 1981). 39. E. J. Hobsbawm, “Custom, Wages and Work-Load in Nineteenth-Century Industry”, in E.J.Hobsbawm (ed.), Labouring Men (London 1979), p. 348. See the critical comment of Michael Sonenscher, Work and Wages. Natural Law, Politics and the EighteenthCentury French Trades (Cambridge 1989), pp. 176 ff. 40. Arno Herzig, “Organisationsformen und Bewußtseinsprozesse Hamburger Handwerker und Arbeiter in der Zeit von 1790–1848”, in Arno Herzig, Dieter Langewiesche and Arnold Sywottek (eds.), Arbeiter in Hamburg (Hamburg 1983), p. 99. 41. Grießinger, Das symbolische Kapital der Ehre, pp. 316, 327 ff. 42. Andreas Grießinger, “Streikbewegungen im deutschen Baugewerbe an der Wende vom 18. zum 19. Jahrhundert”, in: II. Internationales Handwerksgeschichtliches Symposium, Veszprém 1982 (Veszprém 1983), Vol. 1, pp. 315–336. 43. Grießinger, Das symbolische Kapital der Ehre, pp. 290, 353. 44. Reinhold Reith, Lohn und Leistung. Lohnformen im Gewerbe 1450–1900 (Stuttgart 1999), pp. 66 ff. 45. Ibid., passim. 46. Ibid., pp. 300–306. 47. Ulf Dirlmeier, Untersuchungen zu Einkommensverhältnissen und Lebenshaltungskosten in oberdeutschen Städten des Spätmittelalters (Heidelberg 1978), pp. 233 ff. 48. Helmut Bräuer, Gesellen im sächsischen Zunfthandwerk des 15. und 16. Jahrhunderts (Weimar 1989), p. 55. 49. Schwarz, Die Lage der Bremer Handwerksgesellen in Bremen, p. 90. 50. C. R. Dobson, Masters and Journeymen. A Prehistory of Industrial Relations 1717–1800 (London 1980), pp. 28 ff. 51. Josef Ehmer, Heiratsverhalten, Sozialstruktur, ökonomischer Wandel. England und Mitteleuropa in der Formationsperiode des Kapitalismus (Göttingen 1991), pp. 163–184; Rudolf Dekker, “Labour Conflicts and Working-Class Culture in Early Modern Holland”, International Review of Social History, 35 (1990), pp. 377–420. 52. Sonenscher, Work and Wages, pp. 189 ff. 53. Leonard D. Schwarz, “The Formation of the Wage: Some Problems”, in: Peter Scholliers (ed.), Real Wages in Nineteenth and TwentiethCentury Europe (New York, Oxford, Munich 1989), pp. 21–39. See the critical approach of Patricia van den Eeckhout, “Wage Formation: Some more Problems”, in: Ibid., pp. 40–47. 54. Here, I will defer the question of the extent to which such households were self-sufficient with respect to foodstuffs.

Reith

137

55. Knut Schulz, Handwerksgesellen und Lohnarbeiter. Untersuchungen zur oberrheinischen und oberdeutschen Stadtgeschichte des 14. bis 17. Jahrhunderts (Sigmaringen 1985). Kurt Wesoly, Lehrlinge und Handwerksgesellen am Mittelrhein. Ihre soziale Lage und ihre Organisation vom 14. bis ins 17. Jahrhundert (Frankfurt/M. 1985). 56. All the conflicts cited here are documented in Reinhold Reith, Andreas Grießinger and Petra Eggers, Streikbewegungen deutscher Handwerksgesellen im 18. Jahrhundert. Materialien zur Sozial- und Wirtschaftsgeschichte des städtischen Handwerks, 1700–1806 (Göttingen 1992). 57. Reith, Lohn und Leistung, pp. 358 ff. 58. J.C. Händler, Biographie eines noch lebenden Schneiders, von ihm selbst geschrieben (Nuremberg 1798), pp. 32 ff. 59. Reith, Lohn und Leistung, p. 354. 60. Ibid., pp. 184–190. 61. Stadtarchiv Leipzig, II. Sektion S 116. 62. Reith, Lohn und Leistung, pp. 188 f. 63. Reith, Lohn und Leistung, pp. 327–340. 64. Eggers, Das Konfliktverhalten Hamburger Handwerker, pp. 210, 219. 65. Grießinger, Das symbolische Kapital der Ehre, pp. 389 ff. Andreas Grießinger, “Handwerkerstreiks in Deutschland während des 18. Jahrhunderts. Begriff – Organisationsformen – Ursachenkonstellationen”, in Engelhardt, Handwerker in der Industrialisierung, pp. 419 ff. Reith, Grießinger and Eggers, Streikbewegungen, pp. 16 ff. 66. These were offensive in the sense that the journeymen were not reacting to a reduction of their wages by the masters, although they were of course seeking to ensure that their wages kept pace with price rises. 67. Stadtarchiv Dresden, Innungsakten, S 132 Schneider. 68. Reith, Lohn und Leistung, pp. 328 ff. 69. William Gerber, Bauzünfte im alten Hamburg (Hamburg 1933), pp. 31, 211. 70. On payment in kind as an inflation insurance, see: Klaus Strolz, Das Bauhandwerk im Alten Zürich unter besonderer Berücksichtigung seiner Löhne, 1336–1798 (Aarau 1970), pp. 112 ff. 71. Grießinger, Streikbewegungen im deutschen Baugewerbe, p. 321. 72. Ibid., pp. 320 ff. 73. Ibid., pp. 326 ff. Reinhold Reith, “Arbeitsmigration und Gruppenkultur deutscher Handwerksgesellen im 18. und frühen 19. Jahrhundert”, Scripta Mercaturae. Zeitschrift für Wirtschafts- und Sozialgeschichte, 23, 1989, pp. 4–9. 74. Reith, Lohn und Leistung, p. 67. 75. Stadtarchiv Augsburg, Bestand 10, B 245, 29.1.1835. 76. ‘In each case the authorities clearly demonstrated their ability to break strikes, an ability which must account in part for the very low strike rate of the pre-trade-union era’: Michael J. Neufeld, “German Artisans and Political Repression: The Fall of Journeymen s Associations in Nuremberg”, Journal of Social History, 19, (1985/86) p. 493. 77. Michael Birnbaum, Das Münchner Handwerk im 19. Jahrhundert (1799–1868). Beiträge zu Politik, Struktur und Organisation des städtischen Handwerks im beginnenden Industriezeitalter (unpublished Ph.D. thesis University of Munich 1984), pp. 240 ff. 78. Eggers, Das Konfliktverhalten Hamburger Handwerker, passim. On Leipzig and Chemitz, see Horst Steffens, “Soziale Reaktionen auf technischen Wandel. Zum Streikverhalten deutscher Arbeiter im 19. Jahrhundert”, Ferrum, 65 (1993), pp. 20–30. 79. Arno Herzig, “Kontinuität und Wandel der politischen und sozialen Vorstellungen Hamburger Handwerker 1790-1870”, in: Engelhardt, Handwerker in der Industrialisierung, p. 307. 80. Ulrich Engelhardt, “Zur Entwicklung der Streikbewegungen in der ersten Industrialisierungsphase und zur Funktion von Streiks bei der Konstituierung der Gewerkschaftsbewegung in Deutschland”, Internationale wissenschaftliche Korrespondenz zur Geschichte der deutschen Arbeiterbewegung, 15 (1979), pp. 549 ff. 81. Grießinger, Das symbolische Kapital der Ehre, pp. 255–285.

138

Wage forms, systems and conflicts

82. Josef Ehmer, “Migration of Journeymen as Nineteenth-Century Mass Migration”, in R. Leboutte (ed.), Migrations and Migrants in Historical Perspective. Permanencies and Innovations (Bruxelles 2000), pp. 97–109. 83. The municipal officers put 98 questions to all crafts. Reith, Lohn und Leistung, pp. 428–432. 84. Ibid., pp. 171 ff.,194 ff. 85. Staatsarchiv Hamburg, CL XI Generalia, No. 2, Vol. 22. 86. Stadtarchiv Augsburg, Bestand 10, Nr. 1611 (1847). 87. Birnbaum, Das Münchner Handwerk, p. 166 f. 88. Rainer Gömmel, Wachstum und Konjunktur der Nürnberger Wirtschaft, 1815–1914 (Stuttgart 1978), pp. 101 ff. 89. Eggers, Das Konfliktverhalten Hamburger Handwerker, p. 219. 90. Reith, Lohn und Leistung, pp. 301 ff. 91. Schwarz, Die Lage der Handwerksgesellen in Bremen, pp. 305 ff. 92. Reith, Lohn und Leistung, pp. 177 ff. 93. Stadtarchiv Leipzig, II. Sektion S 2546. 94. Stadtarchiv Leipzig, II. Sektion S 2546, 14 July 1828. 95. Reith, Lohn und Leistung, p. 173. 96. Reith, Lohn und Leistung, p. 179. 97. Josef Ehmer, “Wohnen ohne eigene Wohnung. Formen des Zusammenlebens in städtischen Unterschichten des 18. und 19. Jahrhunderts”, in: J. Ehmer, Soziale Traditionen in Zeiten des Wandels. Arbeiter und Handwerker im 19. Jahrhundert (Frankfurt/M. 1994), pp. 71–82. 98. Reith, Lohn und Leistung, pp. 194 ff. 99. Ibid., p. 195; Eggers, Das Konfliktverhalten Hamburger Handwerker, p. 211. 100. Lothar Machtan, Streiks und Aussperrungen im Deutschen Kaiserreich. Eine sozialgeschichtliche Dokumentation für die Jahre 1871 bis 1875 (Berlin 1984), pp. 203, 255, 315. 101. Ehmer, “Wohnen ohne eigene Wohnung”, pp. 87ff. Friederike Föcking, Meister und ihre Gesellen. Arbeitskonflikte im Bäckereigewerbe Hamburgs 1890–1914 (Frankfurt/M. 1993), pp. 45, 50, 81 ff, 179. 102. Erhebungen über die Lage des Kleingewerbes 1885, veranstaltet durch das Großherzogliche Ministerium des Innern, (Karlsruhe 1888), Vol. 3, p. 12. 103. Erhebungen über die Lage des Kleingewerbes im Amtsbezirk Mannheim 1885, veranstaltet durch das Großherzogliche Ministerium des Innern (Karlsruhe 1887), pp. 6 ff. 104. Untersuchungen über die Lage des Handwerks in Deutschland mit besonderer Rücksicht auf seine Konkurrenzfähigkeit gegenüber der Großindustrie, 9 Vols, Leipzig 1895/96.

CHAPTER 6

WAGE FORMS, PAY SYSTEMS AND LABOUR CONTROL IN NINETEENTH-CENTURY AGRICULTURE. EVIDENCE FROM THE DUTCH PROVINCE OF GRONINGEN Henny Gooren and Hans Heger

For a long time historical research into wages has concentrated on variations in wage levels. The movement of nominal or real wages, the stickiness of wages in pre-industrial Europe and wage differentials have been studied extensively. The more qualitative aspects of the history of wages have, however, often been overlooked. The development of wage payment systems and of wage forms in particular, belongs to this neglected field of investigation. Little is indeed known about what kind of wage forms and systems were applied, and why these have changed – if they changed at all. In general one might anticipate that changes were closely connected to developments in labour and labour relations. Yet the way in which these developments interrelate historically is still largely unknown. This chapter tries to shed some light on that darkness. Our study focusses on the development of wage systems and wage forms, and on its relation with a specific and important aspect of labour and labour relations, labour control. Somehow, the amount of work or its quality had to be evaluated, so that the employer could match wages with labour (and the workers knew what to expect for their labour). Obviously, wage forms and wage systems played a role in this system of control and the main aim of this chapter is to investigate this role. To explore this we chose a specific field of study, namely agriculture in the Dutch province of Groningen between 1800 and 1914. To begin, we need to clarify how we interpret “wage forms” and “wage systems”. Simply put, “wage forms” refer to how labour was remunerated, and “wage systems” to how this remuneration was calculated. Labour

140

Wage forms, pay systems and labour control

economics and industrial psychology generally blend wage forms and wage systems into one category as ‘wage systems’ or ‘wage payment systems’.1 However, in this historical study it is more appropriate to make a clear distinction between the system by which the wage is calculated, and the form in which it is paid. This distinction, formulated long ago by N.G. Addens, stresses the fact that in the nineteenth century money was certainly not the only form of wage.2 Of course, present-day wage labourers may also receive some part of their remuneration in non-monetary benefits, but when analysing nineteenth-century agricultural wages, it is clear that a large part of the wage would have been paid in a non-monetary form. For this reason alone, the form of the wage deserves to be distinguished clearly from the calculation of the wage, the wage system. So, we need to pay attention to the monetary and non-monetary form of the wage, to question the relevance of shifts between both these forms, and in particular focus on the development of wage systems and labour control.

Wage systems and labour control: a theoretical note Since F. W. Taylor introduced scientific management methods to the industrial sector at the end of the nineteenth century, many complex systems of wages have been suggested and applied.3 For our purpose it is not necessary to look at the details of the systems in use at the beginning of the twentieth century; it should be emphasised that in general these systems were important issues in the bargaining between management and unions.4 It is clear that they can play a central role in labour relations; what needs to be explained is the manner in which they were so crucial for both employer and employees and their precise role in both industry and agriculture. Sociologist Dan Gowler dealt with such questions. According to him, ‘a wage payment system may be defined as a number of rules and procedures which relate effort to reward’.5 He presented a model of a wage payment system that clarifies and specifies this relation between effort and reward (Figure 6.1). The rules and procedures in this model describe the effort or labour (A) and the reward (B). The relationship between labour and reward (C) is determined by the rules and procedures (E) in such a way that a certain amount of labour results in a certain reward. Some rules and procedures (F) are intentionally set up to link the reward to the effort (D). Wage systems that include this reciprocity are generally known as incentive wage systems. Systems that lack this feedback (F and D) are usually described as time wage systems or fixed wages. The model shows two different roles for a supervisor. In a system with a feedback, the incentive wage system, the supervisor is primarily responsible for an accurate interpretation of the rules and procedures (G) in relation to the reward (H). His control of labour is above all limited to quality. It is the feedback (F and D) that functions as a so-called ‘silent supervisor’: it controls the quantity of labour. Such wage systems thus have built-in labour control. Systems without this feedback, such as time wages or fixed wages, result in

Gooren and Heger

141

a different type of supervision. Here, the interpretation of the rules and procedures (J) in relation to labour (I) is the most important. The supervisor has to ensure that the quantity and quality of the effort relate well to the fixed wage. Gowler makes clear that incentive wage systems are instruments of labour control that may be used in addition to personal surveillance. Time wages are not such ‘silent supervisors’. C

A Effort

E

B

Rules and procedures

Reward

F D I

J

Role supervisor Time wage system

G

H

Role supervisor Incentive wage system

Figure 6.1. Gowler’s model of a wage payment system

Because we are interested in the development of wage systems, this rather static model is not sufficient to explain dynamics in the long run. Nevertheless, it provides an important insight into the connection between wage systems and labour control. To what extent, then, is it possible to explain changes in wage systems from the point of view of labour control? In other words, to what extent did the evolution of wage systems connect with the extent to which farmers in nineteenth-century Groningen were able to control workers and labour processes? The issue of wage systems is clearly related to the organisation of farm work, and one can therefore hypothesise that the less farmers were in a position to oversee farm work, the more they would have been inclined to pay wages by performance. A precondition for the appliance of any incentive wage system, however, is that the work must be measurable. This applies to each separate activity as well as to the complete set of tasks of individual workers. On the whole, if the work process involves a single task, or a small number of tasks, it is easier to measure than when it consists of a variety of different and successive activities. So, it is obvious that a labourer with a single task was more likely to be considered for an incentive wage system than a jack-of-alltrades.6 The assumption, therefore, is that shifts in wage systems over time were linked to those alterations in labour organisation that affected the opportunities for labour control. These considerations therefore lead us to include in our analysis relevant developments in the quality of farm labour,

142

Wage forms, pay systems and labour control

the number of labourers per farm and the content of the task sets of the various types of farm workers. Thus, our approach of the developing wage systems will be from the perspective of the organisation of work.7

Wage forms As mentioned above, we need to take a closer look at the problem of the various forms in which wages were paid. Although in industrial psychology even job satisfaction may be considered as part of a wage, our historical exploration in this chapter is limited to the material remuneration of wage labour. In the nineteenth century, money became the most common form of the wage, but we know hardly anything about the significance, the spread, and the development of non-monetary wages. This has led Jason Ditton to describe payments in kind, perks and entitlements as ‘invisible wages’.8 The lack of visibility of non-monetary wages is indeed a major problem. One reason is that, in general, they are hard to trace in historical sources. If nonmonetary wages are not to be omitted in the study of the real wage – this is of course a point of great debate – the ‘qualitative history’ of the wage certainly needs to incorporate non-monetary wage forms.9 Recently, some studies have stressed the importance of investigating the total wage of individuals or households, rather than relying on the monetary wage only. For instance, Hatton and Williamson showed that a fifty percent nominal wage gap between farm and city in Michigan in the 1890s dropped to a real earnings gap of nine to thirteen percent, where it is the farm perquisites and cost of living differentials that account for almost all of the adjustment.10 P. King demonstrated the importance of a customary right with regard to the income of rural labouring poor in Gloucestershire: between 1750 and 1850 the gleaning of wheat might account for up to fourteen percent of the yearly income of a land labourer’s household; in households headed by widows it often contributed more.11 These examples indicate clearly that we must not underestimate those ‘invisible wages’, especially in agriculture. Still, a major problem remains. When we do not limit the wage to its monetary form and instead search for total remuneration, the question whether we have found this ‘total’ is difficult to answer. Because sources are rather silent, or at least modest, about rights, entitlements and payments in kind that were self-evident to contemporaries, we cannot be sure to have included all payments, particularly if one type of payment in kind would be added or disappear in the course of the nineteenth century. There is likely always to be a dark spot, and it is difficult to know its magnitude. This leads us to other questions with regard to the wage form. Were there major changes in wage forms since 1800, and how can they be explained? What is the relation between wage systems and different wage forms? Were non-monetary wage forms actually calculated as a money wage or were these partly viewed as mere perks that come with the job or as customary entitlements? If so, when did custom start or end, and why? Without doubt, development in nineteenth-century Groningen agriculture may shed light on ‘invisible wages’.

Gooren and Heger

143

Groningen agriculture and its wage systems and forms 12 Situated on the north coast of the Netherlands the province of Groningen provides a good case study for research into wage payment systems and forms of wages. At the beginning of the nineteenth century its agriculture was highly commercialised.13 Compared to other Dutch provinces there was a high input of hired labour. In addition, the province displayed strong regional differences. Furthermore and conveniently, many rich sources are available for our research. Account books of Groningen farmers, official surveys and contemporary literature give ample and precise information about the ways in which farm labourers were paid. The soil of Groningen was clay for the main part, although inland some regions were sandy or peaty. In general farms in Groningen, and especially in the clay area, were on a large scale (see the appendix). A particular system of tenure, called Beklemrecht, made fragmentation of the land almost impossible in many parts of the province. Around 1800 animal husbandry still played a major role in the local agriculture, but in the course of the succeeding century most farmers specialised in arable farming, while a minority specialised in animal husbandry. Still, even in regions where arable farming prevailed, some animal husbandry remained. Table 6.1 demonstrates this shift from pasture to arable land. In this changing agriculture, which we shall consider in more detail below, wage labour played an important role, and the larger farms in particular needed to call upon the labour market regularly. This market was populated by a labour force that had only very limited access to any land of its own. For most of its members land availability was restricted to renting parcels of land in order to grow fruit and vegetables and to keeping a pig or a goat.14 Table 6.1. Ratio of arable land and pasture in Groningen, nineteenth century15 Arable land in 1000 ha 1810 1825 1850 1880 1910

84 90 104 119 131

Pasture in 1000 ha 83 80 70 62 60

Total in 1000 ha 167 170 175 180 190

Arable land percent

Pasture percent

50 53 59 66 69

50 47 41 34 31

The waged labour force in Groningen agriculture can be divided into three main categories according to the nature of employment: the regular farmhands and maids who were living in (the servants), the nonresident regular labourers, and the casual or day labourers.16 The main differences in wage systems and wage forms ran according to these three categories.17 Research into wages often pays attention to differences according to gender, especially when focussing on wage gaps. Our study shows that gender made hardly any difference to wage systems, but gender did matter for wage forms. First, we shall deal with the living-in male and female servants. Such workers needed to be single. So in general the living-in servants were

144

Wage forms, pay systems and labour control

children of land labourers between about twelve and twenty-six years of age.18 A closer look at the wages of these servants reveals that several different components made up their wage and determined its form. Normally, servants received a money wage as well as payments in kind.19 Every servant in our study received board and lodging.20 Such payments were of course the consequence of the fact that servants were living-in, i.e. of the nature of their employment. Sometimes servants also received shoes, clothing and clothing materials.21 These payments in kind were mainly given during the first half of the nineteenth century. Other frequent allowances were linked to gender: wool for the maids, and the right to graze a sheep for the men. These payments seem to end almost completely around 1880. Furthermore, servants received some payments not so much for themselves as for their parents, making their status as children decisive in the work relation. These were, for instance, cheese, barley, straw and manure: all for the parental household and not given on a regular basis. In general one can conclude that the most important non-monetary wage forms for these servants were board and lodging. This conclusion seems obvious, but it is not. In a limited wage concept (all too simply described as ‘the wage is money’) it is not obvious at all. This concept led Daelemans, for example, to go as far as denying any non-monetary wage forms, payments in kind or entitlements for servants in Ernage, Belgium (and on the European continent in general).22 We cannot accept such a limited wage concept and prefer the total wage. Let us now turn to the monetary form of the wage. A rather peculiar wage for servants was the ‘earnest money’ (handpenning or Godspenning). Usually servants accepted employment for a period of one year, and tradition had it that work started in May or November.23 Farmer and servant agreed to this engagement a few months in advance. It was on this occasion that servants received the earnest money; for both farmer and servant this was therefore a kind of guarantee that the other would live up to the agreement. The farmer had ‘rewarded’ his employee, so it was unlikely that he would fail to keep his promise. If the servant did not keep the promise, the earnest money had of course to be returned.24 During the whole period under investigation this wage form was in use in Groningen, but not all servants were entitled to the one or two guilders earnest money. In fact, this money was only given to servants who started a new engagement. In 1834 Trijntje Filips entered the service of farmer Toxopeus at Lalleweer as a maid (voor groote meid), and she received 1.5 guilders as earnest money. She stayed until 1840, but never received the earnest money a second time.25 One could perhaps argue that earnest money is not a proper wage because it was not a remuneration for work done (it was not a prepayment), but a reward for a promise to accept a job. Nevertheless, the earnest money was still an entitlement or a customary right directly and solely linked with the particular labour relation of newly arrived servants. As such, it was a monetary wage form. The biggest monetary wage form was, of course, the ‘proper’ wage paid for the work. It is no surprise that the system by which this wage was

Gooren and Heger

145

calculated related directly to the agreed period of employment, i.e. a yearly wage. Throughout the province and during the whole period under study this was the most common wage system for living-in servants. There were only a few exceptions to this rule in the northern part of Groningen. There, both at the beginning and near the end of the century, some servants received half-yearly wages and others weekly wages. Incentive wage payments for resident workers were rare. From 1860 onwards there are indications that they were occasionally paid by piece for the reaping and threshing of grain, i.e. very specific (and probably additional) tasks. To sum up, it was common practice to pay living-in servants money wages on a yearly basis, to provide board and lodging and some other payments in kind, and to pay new servants a small sum of earnest money. The category of nonresident regular labourers presents a more complex picture.26 This complexity applies to their wage forms as well as to their wage systems. From the types of payment in kind they received, one can infer that most of them were adult males who were heads of households. They lived, for instance, in a house of their employer at a low rent or even no rent at all. This was particularly the case in the north and the west of the province. Of course this must be considered as quite a large payment in kind. Another wage form, which regular labourers might receive, was free board. This may have been quite important. For example, in the region of Westerwolde it was calculated in 1906 that a regular labourer had a yearly (monetary) wage of 223.50 guilders; his free board was calculated to cost the farmer another 130 guilders.27 Quite often they could use a small piece of land for growing carrots or potatoes. In addition, they were permitted to cut the grass on the side of ditches (the right of wallen); the grass was used to feed the labourer’s goat.28 Furthermore, regular labourers had the customary right to take some of the meat of dead cattle; if the cattle had died from sickness this might of course be risky. Another customary right for regular labourers was to glean after the harvest. Just as in Gloucestershire, gleaning was considered in Groningen as an important part of the income of a regular land labourer. In the 1880s it was estimated that gleaning made up between five and nine percent of the total wage.29 In the same period, however, farmers began to criticise the last two of these rights: taking meat from dead cattle and gleaning were now seen as counterproductive incentives that would encourage labourers to diminish their work effort.30 In 1906, for example, the right to glean was no longer granted in the region of Hunsingo.31 Another right consisted of the temporary use of the farmer’s horse and tools to work the land, for most regular labourers that had the use of a piece of land. And finally, farmers occasionally gave milk and old clothes to their regular labourers. As for the pay systems of the monetary wage of these labourers in the first half of the century, our sources are limited to the northern and eastern part of the province (the regions Hunsingo, Fivelingo and Oldambt). However, this restriction does not mean that the ways in which their money wages were calculated, were also limited. Compared to the servants, the regular labourers received more differentiated time wages

146

Wage forms, pay systems and labour control

prior to 1850: daily, weekly as well as yearly wages were paid. After 1850 yearly wages became quite exceptional for regular labourers, and daily and weekly wages remained the standard time rates. However, from 1880 onward regular labourers in the north of the province were sometimes paid by the hour for special tasks like the threshing of beans and peas with a steam engine. In the Oldambt (east of Groningen), on the other hand, wages by the hour were rather common: these were paid to regular labourers for overtime work in busy periods. Moreover, in a large part of Groningen this category of workers regularly worked at piece rates for certain clear-cut activities. In fact, our study shows a gradual regional expansion of incentive wage payments, mainly during the second half of the nineteenth century. Before 1850 piecework was limited to the northern part of the province, especially Hunsingo, and then it spread to the east (Fivelingo, Oldambt, Veenkoloniën). At the same time the number of tasks that were paid for by performance grew continuously. Most were related to harvest work in the fields, such as mowing, reaping and binding, as well as digging. In the southern, sandy part of the province incentive wages for regular labourers were not in use. The wages of day labourers show yet another pattern.32 This category of workers was rarely paid in kind. The only exceptions were free drinks and meals on the job in the field. Day labourers did not receive annual wages, and payments by the week were also rare. Throughout the nineteenth century day labourers were predominantly paid by the day. Only towards the end of the century did hourly payments came into use, initially for particular activities on the larger farms in the northern clay areas, such as weeding and threshing with steam engines. After 1900 the system of hourly wages spread to other Groningen areas and also applied to more and more types of work. Nevertheless, daily wages prevailed. At the beginning of the nineteenth century casual labour appears to have been paid occasionally by the piece. This was exclusively the case in the clay areas, and again only for specific tasks, such as the reaping and threshing of grain and coleseed, the cutting of grass and the digging of ditches. These were all activities for which in a later period regular labourers received incentive wages. In the course of the nineteenth century more and more casual labour in the clay areas was organised with incentive payments. These were activities like weeding, harvesting of flax, and digging up of potatoes. However it was only after 1890 that incentive payment systems spread all over the province.

Changing labour in the Groningen agriculture33 We have described how wage systems and forms developed for the three categories of workers. Now we need to ask in what way these developments were linked to alterations in the organisation of labour that in turn affected the opportunities for labour control. In other words, we need to explain the diverse shifts that occurred in the nineteenth century, particularly in

Gooren and Heger

147

the 1880s. Agriculture in Groningen underwent major changes, as already shown with regard to the use of land (Table 6.1). But more was going on. Around 1800 mixed farms were rather common. Farms in the clay areas were on average bigger than those in the sand and peat areas. In the course of the century, specialisation of this already highly commercialised agriculture expanded. In the central grassland area around the city of Groningen, in the southern part of the Westerkwartier (Western District) and in Westerwolde (the south-eastern part of the province) farmers concentrated on animal production for the market. Elsewhere, and particularly in the clay areas, arable farming grew in importance. Accordingly, the variation in crops diminished. In the clay areas cereal crops came to occupy a major position while the very labour intensive flax growing extended enormously. In the Veenkoloniën (the peat district of east and south-east Groningen) farmers specialised entirely in potato growing. In addition, cultivation intensified and mechanisation took off, especially after 1850. Machines for sowing, reaping and threshing were introduced and their usage became common on the larger farms in the clay areas. Such essential structural changes had effects on farm work. On mixed farms there was a wide range of labour activities that were staggered reasonably well over the year. As Table 6.1 shows, arable farming grew in importance during the nineteenth century. Because arable farming in general needs a larger labour input than animal husbandry, this specialisation led, between 1800 and 1850, to a large increase in the quantity of farm labour. In addition, new and more labour-intensive cultivation methods were introduced. At the same time farm work became less evenly spaced over the year, and farm activity became more seasonal. The latter development was reinforced during the second half of the century, mostly as a consequence of mechanisation and the diminishing numbers of crops. For example, threshing was typically done in winter and machine threshing significantly diminished this labour in terms of effort and time. Harvesting itself was concentrated into a shorter period of time, because of the diminishing variety of crops. This process was most pronounced in the clay areas (Hunsingo, Fivelingo, Oldambt and the northern part of the Westerkwartier). In the sandy regions (the southern part of the Westerkwartier, Woldstreek and Westerwolde) farm work was considerably less seasonal. However, in these areas we noticed a substantial increase in labour input at the end of the century. This was connected with reclamation, resulting in a vast extension of the arable area. The outcome of these various processes was that the demand for labour underwent major changes. Throughout the nineteenth century farms in the clay areas attracted more wage labour than elsewhere. For the year 1862 the Landbouwstatistiek (‘Agricultural Statistics’) offers very accurate statistics to prove this.34 Although the demand for labour was in general rising enormously up to the 1860s, this rise was by no means of equal magnitude for all categories of labourers. Because of the increasing seasonality of farm work, the numerical ratio between the various categories shifted in favour of non-resident labourers at the expense of live-in farmhands. From 1860

148

Wage forms, pay systems and labour control

until 1910 there was even a decrease in the absolute numbers of servants in Groningen agriculture. In addition, there was a growing demand for servants in the city. The number of nonresident labourers in agriculture kept growing at least until 1909.35 It seems highly probable that within the category of nonresident labourers casual labour grew in importance. The structural shifts in farming did not only change the demand for labour; they also affected the evolution of task sets of the individual labourers. From the point of view of labour control it is especially important to notice changes in the number of tasks that labourers generally had to perform. Because servants (and to a lesser extent, this included the nonresident regular labourers as well) had to carry out a great variety of tasks, while casual labourers were usual hired to perform only one or just a few specific labour tasks, one can infer from what we have stated before that between 1800 and 1914 types of tasks grew more limited in number.

Wage systems, wage forms and labour control To conclude, let us now turn to the effects of changing labour on both labour control and the use of wage systems and forms. The increase of farm work and its seasonal concentration into relatively short periods of time made personal surveillance by farmers difficult, indeed almost impossible. With regard to wage systems, farmers showed a growing preference for incentive pay systems. Their disposition was furthered by the diminishing variety of tasks that workers were called upon to perform. Farmers probably had an adequate idea of the quality of work done by regular labourers as they actually saw them all the year round. This was not the case with day labourers. Thus, it seems obvious that casual labour rather than regular work was paid by “silent supervision” incentive wage systems, as was suggested by Gowler with regard to modern industries. Unsurprisingly, the relative increase in casual compared to regular labour promoted the use of such wage systems. Our initial assumption that the various alternatives for controlling farm labour and the use of certain wage systems were interrelated, seems therefore to have been borne out by the historical evidence. At the same time this changing labour affected the use of wage forms. Non-monetary wage forms such as perks and customary rights were almost exclusively paid out to resident and nonresident regular labourers. With the relative increase in casual compared to regular labour, payments in kind diminished. This process appeared from the 1850s onward, but became particularly strong after 1890. So it seems that the exchange of entitlements between Groningen farmers and their employees was more and more limited to monetary wages.

Gooren and Heger

149

Appendix 1. Farmers’ families, labourers’ families and servants in the different Groningen regions, 1862; in absolute numbers and as a percentage of the total of the province36 Farmer’s families

Land labourer’s families

Servants

Region

Absolute As % of Absolute As % of Absolute As % of Numbers provincial Numbers provincial Numbers provincial total total total

Hunsingo C Fivelingo C Oldambt C Veenkolonien P Nwesterkwrt C Swesterkwrt S Woldstreek S Westerwolde S City of Groningen. Province of Groningen.

1,118 802 1,085 1,169 492 671 773 694 19

16 12 16 17 7 10 11 10 0.3

3,077 1,369 2,885 901 1,001 1,346 930 807 34

25 11 23 7 8 11 8 7 0.3

3716 1,856 2,787 1,210 1,419 901 1,002 843 28

27 13 20 9 10 7 7 6 0.2

6,823

100

12,350

100

13,762

100

C = clay; P = peat; S = sand Appendix 2. Average numbers of nonresident farm labourers and servants per farm in the Groningen regions, 186237 Region

Nonresident farm labourers per farm

Hunsingo C Fivelingo C Oldambt C Veenklniën P Nwesterkwrt C Swesterkwrt S Woldstreek S Westerwolde S City of Groningen. Total Province

4.18 2.59 4.04 1.17 3.09 3.05 1.83 1.77 2.72 2.75

Servants per farm

Total wage labourers per farm

3.32 2.31 2.57 1.04 2.88 1.34 1.30 1.21 1.47 2.02

7.50 4.90 6.61 2.21 5.97 4.39 3.13 2.98 4.19 4.77

C = clay; P = peat; S = sand Appendix 3. Working population in Groningen agriculture in 1909 and its increase since 1810 and 1860 38 Absolute numbers Total agricultural working population Total number of farms Average labour per farm

50.500 8.594 5.88

Increase in % since 1860

1810

11 26 -12

75 80 -3

150

Wage forms, pay systems and labour control

Absolute Increase In % of total Number numbers in % since working per 1860 1810 population farm Heads of farmer households 8.594 Farmers’ wives and daughters 6.703 Farmers’ sons 2.380 Total working population farmers 17.677 Heads of land labourers households and home-living sons 19.536 Land labourers’ wives and home-living daughters 5.197 Children under 16 years 1.500 Female servants 2.900 Male servants 3.690 Total working population of Land labourers 32.823

26 26 26 26

80 24 80 54

17 13 5 35

1 0.78 0.28 2.06

40

239

39

2.27

8 -32 -38 -38

73 -5 -6 -6

10 3 6 7

0.60 0.18 0.34 0.43

4

89

65

3.82

Notes 1. For instance, J. Dofny, Arbeid, techniek en beloningswijzen (Luxemburg, 1962); idem, Ontwikkeling der beloningswijzen (Luxemburg, 1962); D. Gowler and K. Legge, ‘The wage payment system: a primary infrastructure’, in D. Robinson (ed.), Local labour markets and wage structures (London, 1970), pp. 168-214; Prestatie en beloning. De prestatiebeloning in het verleden en heden. De te verwachten ontwikkeling. Raadgevend Bureau Ir. B. W. Berenschot NV (RBB) (Hengelo, 1964); H. Thierry, Beloningsmethodieken. Een onderzoek in elf bedrijven (Noordwijk aan Zee, 1966); H. Thierry, Loont de prestatiebeloning?: Een empirisch psychologische studie naar effekten van prestatiebeloning (Assen, 1968); F. Pot, Zeggenschap over beloningssystemen, 1850–1987 (Leiden-Amsterdam, 1988). 2. N. G. Addens, ‘Arbeid en loon in den landbouw’, Het Groninger Landbouwblad, 9 (1928). 3. For an early survey of wage systems in (British) industry, D. F. Schloss, Methods of industrial remuneration (London, 1892). 4. For wage systems bargaining and bargaining techniques in the Netherlands, F. Pot, Zeggenschap. 5. D. Gowler, ‘Socio-cultural influences on the operation of a wage payment system: an explanatory case study’, in D. Robinson (ed.), Local labour markets and wage structures (London, 1970), pp. 100–26. 6. ‘In the past a rule of thumb was often followed that all jobs in which output could not, or not easily, be measured should be put on time rates and checked by suitable supervision.’, in R. Weil, ‘Methods of wage payment’, Wage determination. Papers presented at an international conference (Paris, 1973), p. 166; see also T. Lupton and D. Gowler, ‘Selecting a wage payment scheme’, in T. Lupton (ed.), Payment systems (Bungay, 1972), pp. 239–76. 7. The three factors shown above represent the organisation of work. As such they are of crucial importance for the possibilities of labour control. Given the model we assume a strong relation between these changing possibilities and the development of wage systems. Of course there are many factors which, in their turn, have an influence on the organisation of work. For example: developments in the labour market, labour law and politics, agricultural competition from other regions, urbanisation and cultural

Gooren and Heger

8. 9.

10. 11.

12.

13.

14.

15. 16.

151

change. In this study on changes in payment systems we prefer to focus on the direct relation between the organisation of work, labour control and payment systems. Jason Ditton, ‘Perks, pilferage and the fiddle: the historical structure of invisible wages’, Theory and Society, 4 (1977), pp. 39–71. Leonard D. Schwarz, ‘The formation of the wage: some problems’ in Peter Scholliers (ed.), Real wages in nineteenth and twentieth century Europe. Historical and comparative perspectives (New York-Oxford-Munich, 1988/9), pp. 21–39; Patricia van den Eeckhout, ‘Wage formation: some more problems’, ibid., pp. 40-7; Frank Daelemans, ‘Wages of servants of the Cense–du-Sart in Ernage (Brabant), 1783–1849’, ibid., pp. 48-50. Tim J. Hatton and Jeffrey G. Williamson, ‘Wage gaps between farm and city: Michigan in the 1890s’, Explorations in Economic History, 28 (1991) p. 405. Peter King, ‘ Customary rights and woman’s earnings: the importance of gleaning to the rural labouring poor, 1750–1850’, Economic History Review, 44, 3 (1991) pp. 461–76. This part of the paper is mainly based on Henny Gooren and Hans Heger, Per mud of bij de week gewonnen. De ontwikkeling van beloningssystemen in de Groningse landbouw, 1800-1914 (Groningen, 1993). For the nineteenth-century economic and agricultural history of Groningen, Peter Priester, ‘De economische ontwikkeling van de landbouw in Groningen, 1800-1910. Een kwalitatieve en kwantitatieve analyse’, A.A.G. Bijdragen 31 (1991); and Richard Paping, ‘Voor een handvol stuivers. Werken, verdienen en besteden: de levensstandaard van boeren, arbeiders en middenstanders op de Groninger klei, 1770–1860’, Historia Agriculturae, 27 (1995). See also Henny Gooren, Hans Heger and Paul Klep, ‘Labour relations, culture and fertility in Dutch agricultural communities, 1888’, Economic and Social History in the Netherlands, 6 (1994), pp. 221–37. Jan-Luiten van Zanden, De economische ontwikkeling van de Nederlandse landbouw in de negentiende eeuw, 1800-1914 (Utrecht, 1985) table 5.1 and table 5.3. Information on the wage systems and forms of the three categories of workers was provided by several account books of Groningen farmers, official surveys and contemporary literature. See for the account books: Nederlands Agronomisch-Historisch Instituut (NAHI), Beene Hindriks (Perdok). Zijn boek, Losdorp 19 februari 1801; NAHI, Boekhouding van Jan Clasen 1765–1778; NAHI, Boekhouding van Pieter A. Toxopeus te Lalleweer 1827–1893/94; NAHI, Boekhouding van Jan Rijpkes Beukema te Zuurdijk en Bedum; NAHI, Bouwrekening van een boerenschuur en andere boekhoudkundige gegevens betreffende het Groningse kleibedrijf Doornbosch te Middelstum; NAHI, Lonen van werkzaamheden aan de dijk van den Negen Boerenpolder (en van inwonende knechten en meiden 1828–1917); Rijksarchief in Groningen (RAG), Archief Bouma te Beerta, inv. nr. 2 and 3; RAG, Archief Retzemaheerd te Hornhuizen, inv. nr. 1–5; RAG, Archief van de boerderij Terborg en haar bewoners de familie Steenhuis (1600–1926) te Wirdum, inv. nr. 87; RAG, Archief van de boerderij Zijlbrugge en haar bewoners de familie Wiersma, Ritzema en Ritzema van Ikema te Westernieland (1622–1919), inv. nr. 60; RAG, Archief van de boerderijen op Menneweerster Wierde en haar bewoners, inv. nr. 115; RAG, Archief van de families Barlagen, inv. nrs. 8, 12, 13 and 16; RAG, Archieven van de boerderij Jensemaheert en andere boerderijen onder Stedum en van de boerderij Cremersheert en andere boerderijen onder Loppersum en van de bewoners van deze boerderijen (1693–1900), inv. nr. 18. See for (official) surveys: N.G. Addens, ‘ De ‘vraagpunten’ der Groninger Maatschappij van Landbouw 1852–1941’, Agronomisch Historische Bijdragen 3 (1950); Landarbeiders: hun arbeidsduur en arbeidsverhoudingen. Rapport eener enquête gehouden door de Sociaal-democratische Studieclub (Amsterdam, 1909); ‘De landbouwkundige gegevens uit het Journaal der reize van den Agent van Nationale Economie der Bataafsche Republiek, J. Goldberg, en het Verbaal gehouden door den Commissaris van Landbouw, J. Kops, in het jaar 1800’, Historia Agriculturae, 5

152

17. 18.

19.

20. 21.

22. 23.

24. 25. 26. 27.

28.

29.

30. 31. 32.

Wage forms, pay systems and labour control (1959) 119–288; J.M.G. van der Poel, ‘De landbouwenquête van 1800’, Historia Agriculturae, 1 (1953) 48–194, II (1954) 45–233, III (1956) 105–170; ‘Landbouwstatistiek van de provincie Groningen over het jaar 1862’, Bijdragen tot de kennis van den tegenwoordigen staat der provincie Groningen, 5 (Groningen, 1870); ‘Onderzoek naar den zedelijken en materiëlen toestand der arbeidersbevolking ten plattelande, en van middelen om dien zoo veel mogelijk te verbeteren’, Verslag van het verhandelde op het vijfde Nederlandsche Landhuishoudkundig Congres, gehouden te Leiden, van den 10den tot den 14den Junij 1850; ‘Overzicht van de arbeidsloonen bij landbouw, veehouderij en tuinbouw in 1898’, Bijlage van het Verslag over den landbouw in Nederland over 1896 en 1897 (`s-Gravenhage, 1899); Staatscommissie voor den landbouw, ingesteld bij K.B. van 20 juni 1906, no. 72, Verslagen betreffende den oeconomischen toestand der landarbeiders in Nederland (`s Gravenhage, 1908–1912); Uitkomsten van het onderzoek naar den toestand van den landbouw in Nederland, ingesteld door de Landbouwcommissie, benoemd bij K.B. van 18 sept. 1886, no.28, I–IV (`s-Gravenhage, 1890). See appendices 1, 2 and 3 for the importance of each category of workers. See also I. J. Botke, ‘Het Schrijfboek van Marten Aedsges (1742–1806). Landbouwer te Zuurdijk. “Zoo nuttig in zijne kring als landbouwer” ’, Historia Agriculturae, 17 (1988), p. iv. Few exceptions include some young hands at the farm of P.R. Buurma in Midwolda; in 1834 one received only three pair of shoes and board and lodging; in 1845 and 1865 they worked for board, lodging and some clothes; see J. A. Kuperus, ‘Resultaten van een Groninger landbouwbedrijf (1832–1876)’, Historia Agriculturae, 3, pp. 175–261, ibid., p. 204; Gooren and Heger, Per mud, p. 22. See for the following: Gooren and Heger, Per mud, pp. 20-33. For instance, shoes were paid by farmer Beene Hindriks to his living-in servant Hendrik Nannes in 1802 and Freerk Eeurkes in 1812; see NAHI, Beene Hindriks. Zijn boek. Also the servants of farmer Toxopeus received shoes in 1827, 1830 and 1834; see NAHI, Boekhouding van Pieter A. Toxopeus. Daelemans, ‘Wages of servants’. For instance, C. J. Geertsema, ‘Beschrijving van den landbouw in de districten Oldambt, Westerwolde en Fivelgo’, Tijdschrift ter bevordering der Nijverheid, 31 (1868), pp. 287–288; and P. R. Bos, ‘ Arbeiders: hunne loonen, dagverdeeling, maaltijden, in de gemeenten Usquert, Warffum, Uithuizen en Uithuizermeeden’, Bijdragen tot de kennis van de provincie Groningen en omgelegen streken, I, eerste stuk, (1899), pp. 78-79. J. A. Kuperus, ‘Resultaten’, p. 205. NAHI, Boekhouding van Pieter A. Toxopeus te Lalleweer, 1827–1893/4. Gooren and Heger, Per mud, pp. 35–55. ‘Staatscommissie voor den landbouw, ingesteld bij K.B. van 20 juni 1906, no.72’, Verslagen betreffende den oeconomischen toestand, 1: Groningen (`s Gravenhage, 1908), p. 157. For instance, Geertsema, ‘Beschrijving van den landbouw’, p. 290; H. A. Wijnne, ‘Huishoudelijke toestand der arbeidende klassen in de provincie Groningen’, Bijdragen tot de kennis van den tegenwoordigen staat der provincie Groningen, 1 (1860), pp. 59–89, 75. This total wage also included the non-monetary forms. See ‘Antwoorden op vraagpunt 2’, Handelingen van het Genootschap van Nijverheid in de provincie Groningen (1882–1883), p. 91 and Uitkomsten van het onderzoek naar den toestand van den landbouw in Nederland, ingesteld door de Landbouwcommissie, benoemd bij K.B. van 18 sept. 1886, no. 28 (`s-Gravenhage, 1890), I, pp.19, 23. ‘Antwoorden op vraagpunt 3 van de punten van behandeling’, Handelingen van het Genootschap van Nijverheid in de provincie Groningen (1887–1888), p. 98. Staatscommissie voor den landbouw, Verslagen, 1, pp. 6–7. Gooren and Heger, Per mud, pp. 55–71.

Gooren and Heger

153

33. Ibid., pp. 107–86. 34. See appendix 1 and 2. Appendix 1 shows the numbers of farmers’ families, land labourers’ families and servants in the different Groningen regions in 1862, in absolute numbers and as a percentage of the total of the province of Groningen. Appendix 2 shows the average number of wage labourers per farm in the different Groningen regions in 1862. 35. Appendix 3 gives an overview of the Groningen agricultural working population in 1909 and its increase since 1810 and 1860. 36. Gooren and Heger, Per mud, pp. 165–170, and ‘Landbouwstatistiek van de provincie Groningen over het jaar 1862’. 37. Ibid. 38. Gooren and Heger, Per mud, pp. 170–5.

CHAPTER 7

CASH, WAGES AND THE ECONOMY OF MAKESHIFTS IN ENGLAND, 1650–1800 Craig Muldrew and Steven King

In his work on the artisans of eighteenth-century Paris, Michael Sonenscher has argued that ‘the relationship between work and wages was mediated by a variety of non-monetary customs and rights’. He was able to investigate the social context of wage payments and their value through exceptionally detailed court records.1 In 1989 Leonard Schwarz suggested that such factors needed to be investigated much more thoroughly for other parts of Europe as well if we are to properly understand what went into the ‘formation of the wage’, its level and its symbolism.2 Since then Peter Linebaugh has investigated the role that customary entitlements and negotiation played in relation to the wages earned by workers in the Royal Navy dockyards at Deptford, and Donald Woodward has investigated the payment systems of building craftsmen in northern England in the earlymodern period.3 From this work it is clear that wages were not simply a straightforward cash payment for work done. The level earned and what such earnings meant in terms of wealth and status were bound up with estimations of the value of factors such as entitlements to food or drink and other customary entitlements in kind, hours worked, skill, fines and compensatory payments for urban living conditions. The level and meaning of wages was also tied up with the place of the individual wage within the household economy, and with the place of wages earned through labour at household level in the total value of all sources of income available to the family through what has been termed an economy of makeshifts. Continental literature leads us to expect that in some places and at some times, ‘the wage’ might make only a limited contribution to family welfare so that by concentrating on wages either as an indicator of effort devoted to earning, or of wealth and poverty, we

156

Cash, wages and makeshifts

actually misunderstand the practical role and meaning of the wage as a component of a complex matrix of earnings and access to credit.4 English historiography has not moved so far, but we now know that contemporary pamphleteers and politicians recognised the components of the economy of makeshifts, and pauper letters confirm this reliance on a wide range of welfare avenues.5 In addition to these influences, the form and meaning of the wage was affected by the structure of credit networks and the illiquidity of monetary exchange within the economy. The issue of liquidity has had a tremendous effect on the way exchange was structured in England throughout its history. During the period under discussion here the number of gold, silver or copper coins in circulation was limited. Most buying and selling was done on credit and mutual debts would be cancelled against one another at various intervals. Farm servants often found that their wages were partly paid through exchange in kind or other customary entitlements such as beer, food or clothing, reducing the number and amount of cash payments that had to be made by farmers. Even wages for rural day labourers, which were supposed to be paid weekly in cash, posed a problem. Farmers often paid wages irregularly, so that the availability of credit played a large part in the way household earnings were formed on a weekly, monthly and yearly basis for this group. This is reflected in the nature of litigation, over small debts in towns. The amount of litigation over wages is dwarfed by suits for small sales credit, reflecting the fact that common practice in England allowed poorer households to buy ‘necessities’ on credit in anticipation of future wages or other household earnings. However, the practice emerged rather by default than engineering. For shopkeepers there was the constant risk that some of the debts would never be repaid (it was easy to spend more than was earned when total earnings were unpredictable) yet the poor were too numerous to be excluded en masse as consumers.6 Of course, negotiation was still central to the process of wage formation. The total value of the wage was measured in market price values, even if cash itself did not change hands and, while customary rights survived for very long periods, these were most prominent in areas like manors or mines where the law of equity allowed them to be used as a bargaining tool.7 Wage earners could draw on credit and entitlements as well as market earnings, but always had to negotiate for them from a disadvantageous position. They were not free to use ‘earning power’ in the modern sense, based on wages, but rather had to gain limited power through possible manipulation of credit or entitlements. It is on these three significant influences shaping the form and social meaning of the wage – liquidity, the economy of makeshifts and credit systems – that this chapter focuses. Before we move in these directions, however, it is necessary to reconstruct what contemporaries understood by ‘the wage’ and those that earned it, and to chart how this understanding changed over time. The word ‘wage’ itself seems to have been derived from ‘gage’ (a pledge to do something) from which the form to

Muldrew and King

157

wage law or battle or to make a wager derived. But the meaning evolved from making a pledge to constituting a reward to someone for a service after it was performed. Thus the wages of labour was its reward just as death was the wages of sin in the King James Bible. This understanding seems initially to have been used to refer to the payment of soldiers, as was salary, which came from the Roman payment to soldiers for salt.8 It is easy to see why it might have been commonly used in a military context, as medieval armies grew in size and it became common to hire soldiers for wages in large numbers from outside an estate to supplement knight service.9 Tracing the development of the term ‘wages’ to refer to a cash payment for a day’s work is much more difficult. As Ann Kussmaul has pointed out the use of the word ‘labourer’ to denote wage workers did not become a general term until the nineteenth century.10 It was also rare to find the word labourer used to refer to a social category before the end of the seventeenth century. Locke, of course, famously analysed the value of labour in the Second Treatise of Government but in his example of wage labour he used the term servant.11 Also Gregory King placed ‘labouring people’ together with ‘out servants’ in his table of income and expense.12 John Law, however referred to ‘the poor and other labourers’ as those who lived by earning ‘the Wages of Labour’, and by the time Adam Smith came to write the Wealth of Nations it was common to refer to wage earners and labourers as a social group.13 For most of the period before the late eighteenth century ‘servant’ was generally used as a catch all term referring to both day labourers and servants in husbandry hired for the year or a part thereof. Wages could refer to the contracted yearly payment, or to a day rate. The famous ‘Statue of Labourers’ was actually entitled De Servientibus, and in it labourers were referred to as operariorum, and the Latin terms used to refer to wages were most commonly stipendia, liberationes and salaria.14 The Subsidy of 1525 made reference to a difference between wealth and wages. In Norfolk churchwarden’s accounts from the early sixteenth century the term ‘wages’ was used to refer to day labour, but the words ‘stipend’ and ‘hire’ were also used. Most often the work done was simply stated with the pay given for the specific task mentioned. The same was true of churchwarden’s in Suffolk.15 Shakespeare used the terms ‘labourer’ (three times), ‘workman’ (nine times), and ‘wages’ (ten times) very rarely compared to ‘servant’, ‘service’ and ‘pay’ or ‘payment’ (used hundreds of times). In their farm accounts both Henry Best (1616–41) and Robert Loader (1610–20) also tended to use the term ‘wages’ only to refer to yearly servants’ wages or for harvest wages, while daily labour is referred to by the job done.16 In the 1680s the King’s Lynn Chamberlain’s accounts refer to odd jobs under the designation of ‘common charges’ while small wages are listed for continual tasks such as cleaning and looking after gates and locking doors.17 ‘Labourer’, however, became a common designation used in probate documents, although again this was more common in some places than others.18

158

Cash, wages and makeshifts

Liquidity and the form of wages The continued use of the term ‘service’, and the continued importance of the employment of servants in husbandry, had a basis in continuing shortage of cash in the economy in the early modern period. In turn, liquidity is a key variable for understanding the meaning of the wage and we need briefly to examine the degree to which the early modern economy was monetised. A lack of monetisation in the early Middle Ages is something that is considered to be characteristic of feudalism and its emphasis on obligations of service. Richard Britnell has estimated that between 1000 and 1180 the total circulating currency in England did not rise above £120,000 or only 1s. for each member of the population. The standard coin was the penny, which was sometimes cut in half or quartered but was otherwise of very little use in small daily transactions. This placed obvious limits on the monetisation of exchange and led to the payment of many tenurial obligations and dues in labour or kind rather than cash.19 With population growth in the thirteenth and fourteenth centuries the monetary supply also expanded, stimulating an increase in the commercialisation of English society. But despite the increased circulation of money most wage labourers on demesnes still received the majority of their remuneration in kind rather than cash.20 The amount of money fell again in the depressed years of the fifteenth century, and despite the influx of gold and silver from the New World into Europe, J.R. Wordie has argued that little stayed in England, and that by 1600 there was only £1.5 million in circulation, or only about £1 16s. for every household in the country.21 But, between 1540 and 1600 food prices also more than trebled, while industrial prices doubled, and the amount of goods being consumed on the market also, roughly, doubled.22 As a result, by the end of the sixteenth century the notional demand for money had probably increased by something like 500 to 600 percent, while the supply of coins hardly expanded at all. Even though the amount of gold and silver in England rose after 1600, the problems of shortage remained. In 1653, for instance, Ralph Josselin claimed that he was without even a penny for ‘divers dayes’.23 As a result of the bi-metallic nature of the coinage by the Restoration, the circulating currency was almost entirely composed of silver coins, and as clipping reduced the value of many coins, good silver money either left the country or was kept out of circulation by merchants.24 Samuel Pepys and William Petty both estimated that there was probably only £6–7 million worth of money in circulation in the late 1660s because so much was hoarded.25 This had to supply the demands of increased foreign trade, massively increased taxation, increasing poor rates, as well as wage payments. Liquidity was no better in the eighteenth and early nineteenth century. As the mint price of silver drove smaller silver coins out of the country in the early eighteenth century, the gold guinea was the standard unit of metallic currency.26 This can be seen in Table 7.1 which is an attempt to estimate roughly the availability of currency in the eighteenth-century

Muldrew and King

159

economy.27 While the amount of paper credit expanded considerably after the financial revolution, as did the total value of gold coins in circulation, this was of little use to pay weekly wages. Between 1700 and 1750 only one million pounds worth of silver coins were minted, and this was probably the maximum amount in circulation at any one time as many of these coins would have been needed by smaller tradesmen. Other than this there were various issues of copper farthings which were problematic because they could be counterfeited easily. As a result there was probably much less small change available for most of the eighteenth century than there had been in the 1670s and 1680s when Locke, Law and Petty identified it as a major problem. Table 7.1. Forms of circulating currency in the eighteenth-century economy

1694 1695– 1700–1750 1700–1750 1701 1719–1725 1700 1725–1775 1729–1754 1752

1760–1775 1775–1810 1775

Type of currency

Amount

Estimated number of negotiable units

Bank of England Bills of £5 and 10 Exchequer Bills of £5 Minted Gold Coins (Guineas) Minted Silver Coins 1/2p. Copper Farthings 1/2p. Copper Farthings 10s. Lottery Tickets

£1,200,000

180,000

£167,000 £17,000,000

33,400 15,500,000

£1,000,000 £137,000 £30,289 £25,000

10-20,000,000 65,760,000 14,538,720 50,000

£173,000

83,040,000

1/2p. Copper Farthings Bank of England Notes in Circulation (above £5 in value) 1/2p. Copper Farthings Gold Coins

1752–1801 1785

Minted Silver Coins 1/2p. Copper Farthings

1779

County Bank Note Issue

1787–1797

Private Trade Tokens manufactured in Birmingham Boulton’s Copper Cartwheel’ Farthings 5s. Silver Dollar Tokens County Bank Note Issue County Bank Notes Issued of £2 or Less

1799 1804–1815 1808–1809 1808–1809

£4,750,359 £46,454 £20,000,000 (estimated circulation) £143,313 £322,000 (estimated to still be in circulation) £57,000 in notes of less than £5

22,297920

2,866,260

154,560, 000 c.11,550

£122,000

29,400,000

£679,311 £4,457,649 £14,618,350 £6,247,165

256,524,240 10,300,000 c.3–6,000,000

160

Cash, wages and makeshifts

Thus, in the 1750s, when the population was about 6,000,000 (1,263,158 households) and about fifty percent of these were gaining income from wages or other small payments, a weekly cash wage of 7s. would have required £221,053 to circulate weekly. This was greater than the amount of farthings in circulation, and although it is impossible to know how many silver coins there were, from 1725 less that £4000 was minted in most years. As a result it was reported that most shillings and sixpencees were little more than worn blank disks, while half crowns continued to be hoarded. In addition, since the Mint made no provision for distributing farthings, local gluts tended to occur causing shortages elsewhere. In 1754 the butchers, bakers and grocers of London petitioned the government to withdraw the copper coinage because they were burdened with £50 to £500 each which they could not get rid of. Thus, shortages of coin for wages were more acute than ever by the mid-eighteenth century.28 Indeed, as Table 7.1 suggests, it was not until the recoinage of 1786 and, more particularly, the crisis of the Napoleonic Wars (when the need to finance heavy taxation coincided with the suspension of gold payments in 1797) that small change began to be produced in significant quantities.29 In addition, as Peter Mathias has shown, private trade tokens began to be issued again in large numbers after 1780.30 Supplementary evidence supports the view that in a national sense at least the coinage crisis was easing. Poor relief bills in most regions begin their upward spiral in the 1780s, and while this has often been attributed to inflation, the movement and the fact that it was so uniformly felt might have a rather simpler explanation – that by the later 1780s poor law authorities had for the first time the liquid funds to be able to put generous relief policies in place (see below). Moreover, the 1780s and 1790s mark the development of a widespread system of out-parish relief in the northern industrial counties, whereby parishes of settlement paid relief to paupers who had moved elsewhere to work but subsequently become dependent upon relief in their new parish of residence. Payment frequently took the form of giving small change to agents who then distributed it to the paupers concerned. This practice, and the flexible labour markets that it underpinned, would simply not have been possible where currency shortages were severe and enduring.31 Nor should we forget that whatever was happening to the national supply of coinage, an increasing network of provincial banks was developing from the later eighteenth century, potentially fostering a faster local circulation of coinage and offering a connection between those who had money and those who needed it. In Lancashire and parts of Yorkshire, increasing numbers of attorneys filled the same function, often recirculating tiny amounts of ready cash on behalf of their clients and on their own behalf as they made their fortunes.32 Yet, it would be wrong to overstate the impact of any notional increase in the supply of small coinage. Even by the opening decades of the nineteenth century, the supply of coinage was failing to keep pace with the increasing numbers of people dependent upon wages and the scale of industrial and commercial agricultural production. Progressive increases in

Muldrew and King

161

the size of farms in most counties and a gradual move from outwork or workshop production to centralised urban production placed intense pressure on farmers and entrepreneurs to be able to find the cash for weekly or even daily wages. Their complaints about their inability to achieve this objective, and the complaints of workers who were rarely paid regularly can be found in the firm records, diaries and newspapers of much of industrial northern and midland England, even after 1810. A further problem was that while country banks might facilitate liquidity for short periods, their periodic collapse could decimate local and regional currency circulation systems. Nor should we forget that for much of the eighteenth and well into the nineteenth century, the fastest growing industrial areas were also the remotest, generating logistical problems of moving cash around, whatever its nominal supply. It was for this reason that trade tokens became a regular part of everyday life in industrial Lancashire and the West Riding by 1800. This liquidity problem was made even more acute because the number of wage earners increased inexorably in the eighteenth century. Already for 1377–1381 Christopher Dyer has estimated, from poll tax evidence, that the proportion of people who earned most of their living from wage work must have exceeded a third of the population of England.33 This proportion had risen somewhat by the time of subsidy of 1524–1525, and in Alan Everitt’s estimates, seems to have been roughly the same a century later, although there could be large regional differences.34 By 1688 Gregory King estimated that the numbers of households that had to be paid wages or poor relief (labouring people, out servants, cottagers and paupers) had reached forty-seven percent of the population.35 According to Lindert and Williamson’s revisions of Joseph Massie’s social table of 1759 and Patrick Colquhoun’s occupational headcount of 1801 to 1803, the numbers of labouring families and cottagers as well as those who earned wages through textile work, building and mining, was about the same at mid-century, but had risen to sixty percent or more by the beginning of the nineteenth century.36 Thus, while there is no doubt that the economy became increasingly monetised in the sense of measuring value in terms of price, this was done with money of account and did not mean that actual cash changed hands. Clearly, it is better to talk of a price economy not a money economy. The impact of this situation on the form and meaning of wages was profound. In buying and selling, both within towns and between town and country, cashless exchange was the norm. In the context of the manorial economy, while servants and day workers were paid money wages and tenants paid rents, very often these values could cancel each other out, or could be paid in kind worth a similar amount in monetary terms. The remainder would then be paid in cash, or turned into a smaller debt. In addition to this, many servants continued to have their wages supplemented with board and meat, which in the case of servants in husbandry could amount to a sum greater than their wages. Wages could also be supplemented by customary entitlements, such as gleaning or the gathering of firewood.

162

Cash, wages and makeshifts

The ‘meaning’ of the wage is thus influenced by aspects of liquidity in ways that historians have thus far failed to really appreciate. Such means of making bargains and transactions without cash payment can be found in many seventeenth-century farm accounts and memorandum books. Probably the most well known examples are the accounts of the early-seventeenth-century yeoman Robert Loader. He made meticulous efforts to try and calculate the expenses and profits of his farm. He worked out very carefully how much each of his servants cost him in terms of food provided, as well as how much work they did for their pay. As both Ann Kussmaul and Donald Woodward have noted, Loader estimated that each servant cost him about £10 5s. a year in food and drink compared to wages of between £3 and 15s. 6d., and he constantly complained that it would be cheaper to keep fewer ‘unruly’ servants and instead rely on wage labourers.37 When writing up his costs and expenses for 1613 he calculated that if he only kept one maidservant he would save £5, based on his calculations for 1612 when he hired his servants at board wages. Despite this he only made one further experiment hiring for board wages – with his carter William Weston in 1617. A careful examination of his accounting makes clear that the reason for this was that he simply did not have access to enough ready money to pay regular cash wages. When he negotiated with Weston for his board wages he agreed to pay £11 in money together with four bushels of wheat, three weeks board at harvest, the keeping of a hog by Loader and shorter days in the winter, which Loader reckoned to be worth £13 9s. 4d. He termed this ‘exceding great wages’, and indeed it was little different than what a carter usually cost him with board provided.38 Even the wages he assessed in cash were often paid with wheat, or reckoned against other services he provided for his servants. In this way bargains could be made and wages could be cancelled in a very simple way against food he produced, rent he was owed, or the use of his land. Loader’s accounts make it clear that bargains were entered into at market rates between him and his servants, and that entitlements in kind had a monetary value agreed through negotiation. This was not simply barter or payment in kind based on feudal obligation, but a way of engaging in market bargaining with limited use of cash. In this context it is interesting to note that Loader actually wrote about ‘spending’ his wheat or hay to pay wages in the same way as spending actual money.39 It was probably for such reasons that whenever Loader complained about the cost of servants, he first proposed putting forth his land to tillage or rent before suggesting that he should keep his servants at board wages.40 Servants were also often allowed to put animals at pasture on their master’s lands. Like Loader, the Yorkshire farmer Henry Best paid one servant £6 together with barley, oats, oatmeal, a coat and straw, but to another he paid £5 in money and wintered ten of his sheep, forgave his house rent for a year and paid for the cost of a cow for the summer.41 The Catholic estate owner Richard Cholmeley of Brandsby in North Yorkshire also had quite complicated bargaining arrangements with his servants. In his memorandum book from the first quarter of the seventeenth century,

Muldrew and King

163

he sometimes noted wages that he could not pay which were in arrears and which were converted to loans that he owed to his servants. On another occasion when one maidservant left his service he wrote that he owed her £4 10s. for two years and four months’ wages, as well as another 40s., which she had ‘lent’ him, and 33s., which she had disbursed on Cholmeley’s wife’s linen. He owed a neighbouring yeoman’s wife £4 10s., which another maid servant had borrowed from her to pay for her wages. He also recorded owing her £10 which he had ‘borrowed’.42 There are also a number of instances in his memorandum book where it is clear that he paid the monetary portion of the wages he owed after receiving larger cash payments for something he had sold.43 In addition he commonly paid a lot of wages after his rents were paid in.44 This type of exchange could be done with day labourers working on a farm as well. Cholmeley paid some labourers to carry wood with corn, beer, ale and meat as well as cash.45 But it was more common with day labourers, if the cash was not available to pay their wages, to turn them into a debt which the employer owed to the labourer. This debt could then be cancelled against a debt the labour might incur to the employer, usually for grain or some other agricultural produce. This process can be seen very clearly in the double entry accounts of Nathaniel Brewer a farmer of Over Stowey in Somerset. In 1713, for instance, he listed debts due to him from a labourer for various sales of peas, barley and wheat worth about £3 15s. On the debit side he listed debts he owed the man for felling timber and sawing it.46 This practice remained common throughout the eighteenth century, and as Mick Reed has argued well into the nineteenth century.47 Such debts had the advantage that they could be transferred to third parties making the bargain more flexible, as in an instance where Richard Cholmeley paid a tailor for some work done for one of his servants as part of the wages he owed her.48 Eighteenth- and early-nineteenth-century Lancashire, the locus of some of the most substantial notional demands for ready cash to pay agricultural and industrial labour, provides some of the most compelling evidence for these practices. The number of extant account books from householders, doctors and entrepreneurs numbers several hundred. One of the best, that of Rowland Park of Kirkby, records a long history of coping with lack of cash. His devices were innovative. For instance, he wrote off servant wages against medical treatment from the Fylde doctor Richard Loxham, in turn directing the bills of the doctor to the wholesalers who purchased his crops and animals on a six monthly basis. Park even paid his poor law bills in kind or through deferred notes of hand.49 Wages for non-agricultural labour were more difficult to treat in this manner. They could not be cancelled against food produced by a farmer and, as Donald Woodward has shown for building workers in the north, it was even more expensive for industrial employers to provide beer and food than it was for farmers since they had to purchase it.50 In some instances the cash proportion of spinners’ or weavers’ wages might be cancelled against purchases of cloth, but the purchase of cloth and other

164

Cash, wages and makeshifts

necessities was not nearly as common as food. Thus Locke argued that day labourers, those employed in building work in towns, spinners, weavers, stocking knitters or any other industrial workers should ideally be paid every week in cash. But although an analysis of amounts of cash contained in a sample of probate inventories shows that on average more cash was held by residents of towns than rural dwellers, there was still an insufficient amount in circulation to pay industrial wages on a weekly basis. By the end of the seventeenth century most cash was hoarded by tradesmen for use in long distance trade, and comparatively little was available in small change to pay wage earners.51 Because of this wage earners had to sell their labour on credit and wait for their wages to be paid in arrears. In one case from the London Mayor’s Court for 1689, a bottlemaker was forced to borrow £10 to pay his workmen’s wages, which they demanded that night.52 In 1729 it was reported that keelmen involved in the coal industry in County Durham had to live on `trust’ during the winter while waiting for their wages to be paid.53 A shopkeeper who sued a brickmaker in the Birmingham court of requests in the mid-eighteenth century claimed that the latter’s master had told him that during the winter season the defendant ‘can get only clay, consequently his wages are small; trust him with what he wants, and I will see you paid in the summer’. Here the brickmaker’s wages would have simply been turned into a debt owed by the master to the shopkeeper eliminating the need for small change. Unfortunately, in this case, the master broke and fled the city leaving the brickmaker unpaid and liable for the debt to the shopkeeper.54 The experience of ‘putting-out’ entrepreneurs and their workers in the eighteenth- and early-nineteenth-century Lancashire and Yorkshire textile industries provides a useful case study. Despite the increase in the amount of token coinage in the 1790s such entrepreneurs complained constantly about having insufficient liquidity to pay workers for their pieces at the end of a production week and resented any drain on their reserves of ready cash.55 By the late eighteenth century, many of the putting-out concerns in the Rossendale valley and the Manchester hinterland had entered into formal agreements with the trustees of Turnpike Trust companies to purchase large amounts of small change, offering a small premium in the process. Such agreements are to be found in most of the Lancashire Turnpike Trust cash account books. Yet, such innovative schemes notwithstanding, it is generally accepted that putting-out merchants in both Lancashire and the West Riding were usually at least two weeks in arrears with wages. This was a great improvement on the situation fifty to sixty years earlier, but the evidence of workers from a Royal Commission of 1806 suggests that workers accepted, albeit with frequent hostility, that their wages would not be paid on time. The amazing preponderance of small irregular payments to men and women in Yorkshire and Lancashire poor law accounts may well be a reflection of the need for workers to tide themselves over irregular wage periods (and a reflection too of the role of putting-out entrepreneurs in local poor law administration), as might the

Muldrew and King

165

now convincing evidence of a cooperative culture in much of the north and north-west in the eighteenth and early nineteenth centuries.56 Currency shortages in manufacturing areas were also a main incentive for employers to initiate truck systems, thus short-circuiting local credit networks. David Whitehead of Waterfoot Mill, in Rossendale, Lancashire, for instance, was obliged to buy in small change from Manchester to pay his workers because his own was absorbed by his position as banker to the poor law. But even this was not enough, and he opened a mill shop which offered subsidised food and household goods on credit to employees and others in the locality. The Waterfoot Mill Shop account book, running from 1812 to 1826, records debts only at year end when the shop formally balanced accounts obviating the need for cash over very long periods indeed.57 Such practices, though, were not a new means of dealing with the shortage of cash.58 As early as the late seventeenth century John Locke complained that many manufacturers, especially clothiers in the wool trade, trucked commodities for work which, such as they are, good or bad, the Workman must take at his Master’s Rate, or sit still and starve: Whil’st by this means, this new sort of Ingrossers or Forestallers, having the feeding and supplying this numerous Body of Workmen out of their Warehouses, (for they have now Magazines of all sorts of Wares) set the Price upon the poor landholder [who] must sell it to these Ingrossers, on their own terms of Time and Rate; and allow it to their own Day-Labourers under the true market price.59

Workers’ penny clubs in alehouses also played an important role in alleviating liquidity constraints by allowing masters to pay workers less regularly in gold coins which would be used to pay alehouse scores.60 The alehouse keepers could then provide small change in the form of the farthings or worn shillings which middling tradesmen did not like. One case recorded in the law notes of Sir Dudley Ryder concerned an alleged quarrel in a public house where the plaintiff, a master carpenter, said he normally paid his workmen. In this case there was a dispute over 16s. wages demanded by a journeyman, which the plaintiff could not pay because he had no cash. He offered to go home to get it but the journeyman would not trust him. In this case the landlord offered to pay, but the master did not want to become indebted to him and a quarrel ensued.61 William Hutton recorded one case where a master bucklemaker paid two journeymen together with one gold coin ‘when reckoning with his people on Saturday night’ because he lacked silver.62 He also recorded other cases of journeymen spending their wages in penny clubs chalking up alehouse scores, which they were unable to pay. But this function of the alehouse was obviously problematic for moralists because it meant that wages were spent on ale before a worker’s household. Hutton continually complained about Birmingham workers’ spending on ale, even thought they lived on the edge of want.63

166

Cash, wages and makeshifts

At this time the Navy was undoubtedly the largest employer which had to pay in cash. Its problems in finding enough of it might be seen as a microcosm of the economy in general. In the early years of the Restoration wages were often years in arrears, leading to the disaster of the Second Dutch War (memorably described by Samuel Pepys) when no money could be found to pay people to man English ships.64 Such problems continued through the eighteenth century. A system of paying credit by ticket was established, but the credit of the Navy was so bad then they could only be cashed by sailors at a discount, effectively reducing their wages.65 Mariners’ wages in general were an even greater problem than other wages because the transient nature of sailors’ lives meant they were even less trusted with credit than other poor people. It might be years before they returned to port, and often, of course, they would not return at all. Because of this, mariners were the one group of wage-earning workers who sued in large numbers for wages.66 We return to this theme later. For now it is clear that lack of liquidity could influence the level, meaning, form and prevalence of the wage in very subtle ways. But lack of cash and irregular wage payments could also have important indirect effects on the meaning of the wage by placing emphasis on the alternative coping strategies of labouring households, particularly the relationship between work and communal welfare.

Wages in the economy of makeshifts In England, increasingly in the eighteenth century the most significant plank of this economy of makeshifts is often seen by welfare historians to be the income offered by the poor law. National relief expenditure bills rose from about £150,000 in the 1650s to £400,000 in 1700, to almost £2,000,000 by the mid 1780s and over £4,000,000 by 1803.67 This was paid either directly in the form of allowances (in cash and kind) or indirectly through the interaction of local relief policies and the structures of local labour markets. Poor relief is thus a potentially crucial influence on the form and level of wages. Yet it was not the only element of the economy of makeshifts which could have an impact in these terms. Gleaning, embezzlement, exploitation of wasteland, crime, credit, drawing upon kin and other networks, petty production and remarriage are all implicated in the daily and yearly business of making ends meet when ordinary people deploy their own words before the vestry or in letters to the overseer.68 There have been few studies of alternative earning avenues or their relationship to the payment of poor relief and wages, but it is nonetheless important to investigate these variables further as a precursor to dealing with the communal welfare system. Thus, one element in the process of making ends meet was the assumption of rights to perquisites by workers throughout the period dealt with here. These could significantly augment the notional ‘earnings’ of industrial workers, in particular, and proved a persistent thorn in the side of

Muldrew and King

167

putting-out entrepreneurs. Equally, perquisites could substitute for wages rather than simply augmenting their level. The diary of David Whitehead from Rossendale in Lancashire provides a unique insight into the most lucrative of all perquisites, embezzlement. After a series of failed apprenticeships between 1805 and 1816, Whitehead agreed to act as a travelling salesman for a local merchant. Two things made a particular impression on him. First, how little cash there was in the bustling Rossendale valley, an area at the heart of the English industrial revolution; second, the fact that the vast majority of those who wanted to ‘buy’ his goods did not want to pay in cash. Rather, they wanted to barter for the goods Whitehead had to sell with raw materials embezzled from putting-out merchants. These raw materials were usually raw or spun cotton, but might also include wool or made-up garments. Such experiences prompted Whitehead to give up his job as a salesman, and constitute the most visible testimony to the points we have been trying to make here.69 This said, perquisites were of limited use for some day-to-day transactions. To make life bearable, sophisticated credit, borrowing and reckoning systems also developed around industrial production and wage earning.70 We return to this aspect of the economy of makeshifts later in the chapter. Meanwhile, a further strand of the economy of makeshifts was the variation of household residential arrangements. A typical response to inadequate or irregular wages in the domestic economy might involve bringing in unpaid family labour to replace ancilliary workers hired in the labour market. Old people and children or grandchildren were important components of this unpaid labour pool.71 For factory workers and other families where earning opportunities were mainly outside the domestic sphere, giving a home to kin or friends might free up family labour from domestic chores or reduce the opportunity costs for women wage earners by providing childcare. This is not a new observation, as the work of Anderson has shown, but it is important to acknowledge that varying household arrangements could have an important impact on the level and symbolism of the individual and family wage. Of course, coping with liquidity problems in this manner rather than turning to, say, neighbourhood lending and borrowing networks, also had the added advantage of not incurring debts of obligation to other local families which might then have to be repaid.72 Meanwhile, it is important to note that even if they did not live together, those wage earners most vulnerable to irregular cash wages often crowded together in residential terms as a means of maximising credit opportunities, creating lending and borrowing networks and pooling buying power. In the parish of Calverley, the bulk of eighteenthcentury wage earners can be found crowded into low cost housing on common or moorland.73 Housing surveys and lease data for south west Lancashire provide more evidence. A survey of Birkdale in 1815 highlighted four terraced cottages standing on common land, and gave details of the occupants. A mixture of poor widows and families overburdened with children, they lived rent free in these town cottages, but got no other allowances. The overseer clearly envisaged a self-supporting and distinct

168

Cash, wages and makeshifts

community of poor wage earning people. Some confirmation of this contention can be found in an 1816 lease between Charles Blundell and two local gentlemen who wanted to use money raised by a public subscription to build a cottage, on the same common land, for the widow and children of a local sailor who had perished at sea. The fact that a public subscription was held is an important indicator of other welfare avenues, but more significant is that the cottage was to be built right next door to housing for other poor wage earning people.74 Contemporaries clearly attached some significance to proximate residence and the interlocking welfare obligations that they felt might ensue. In addition, both shopkeepers and employers benefited from such spatial proximity – the former could track outstanding credit and reputation much more easily in such areas, while the latter experienced reduced opportunity costs of finding outworkers. Such ‘opportunities’ probably also reduced the pressure on entrepreneurs to find the cash component of wages and encouraged payment in kind. Charitable provision was another potential way of coping with irregular wages and may also have had an impact on their form and level.75 Of course, the term ‘charity’ masks many forms of giving: formal charitable endowments; the charitable activity of nonconformist groups; occasional dispensation of money or gifts by private individuals; and irregular disbursements by charitable funds which stemmed from collections by local elites to tackle the poverty occasioned by factors such as trade depression. If we tally the resources offered by ‘charity’ they were very substantial indeed throughout the period covered by this chapter. Even by 1800, when many commentators suggest the charitable imperative was dwindling, the potential income from formal charitable benefactions alone may have matched or exceeded the sum paid out of the parish rates for poor relief.76 It is true that much charity capital was absorbed by London, and that charity might also be directed in such a way as to offer little real benefit to those who were struggling to cope with meagre or irregular wages.77 This said, many other charities gave out substantial cash sums and made little or no distinction between workers receiving wages and the old, sick or orphaned who might or might not be eligible for poor relief. Indeed, some charities explicitly made payouts to wage earners. An 1811 Dole Book for Halifax records the family circumstances, employment status, wages and other details of those who were given food paid for by a collection in that year. The 228 names on the list represented a more substantial subsection of the population than the 149 on the relief lists for that year, and while the quantifiable benefit of the dole may have been limited the fact that the dole was given at all is testimony to the fact that liquidity problems for labourers became particularly severe during trade downturns, when the meaning of the wage may also have changed.78 In the same manner, personal charity could be a significant boost to wage earners. The incredibly detailed seventeenth-century records of the charitable activities of the Flemming family of Rydal are one example. In 1686 alone, the family gave £98 to the poor households in south Westmorland, far

Muldrew and King

169

outstripping sums raised through local taxation.79 In turn, where recipients of charity were also working, ‘the wage’ might actually be a residual payment, hardly deserving of the attention that historians have paid to it. Employers, aware of this fact, might thus have been under less pressure to pay market wages or wages in cash form than would otherwise have been the case. For those on the outside of the charity processes other welfare avenues also beckoned. While gleaning and foraging on the wastes and commons may have made a substantial contribution to monthly or yearly welfare, such ‘rights’ had been under pressure for many years by the middle of the eighteenth century.80 Yet in most northern counties farm or industrial land was located within the midst of large tracts of waste and common, where customary access to ‘communal’ resources had considerable longevity. In Rossendale, Cumberland, Westmorland and Yorkshire, labouring families could still reasonably expect to be able to cut turf and take other resources from open land even by the later eighteenth century. Intercommunity disputes over the rights of access to wasteland in Yorkshire provides ample testimony to the perceived value of these communal resources by wage earners.81 Certainly, for some wage earners at least these potential avenues might be one way to avoid the periodic large bills (for fuel, for instance) that irregular cash wages made it difficult to budget for. Continuing access to such alternative avenues was almost certainly part and parcel of the continuation of a customary wage for women.82 This said, the supply of makeshift resources came to be outpaced by demand during the eighteenth century. This was particularly true for the southeast and midlands, but was also true generally. It is no surprise, then, to find that commentators from Snell and Wall to Horrell and Humphries have increasingly emphasised that poor relief was a vital and ever more important plank of the economy of makeshifts and increasingly became used as a sort of communal wage ‘supplement’. People combined both income streams and, indeed, an expectation that they would lay at the heart of poor law legislation from its very inception in the late sixteenth century. The nature of this relationship is a key influence on the way that we should read ‘the wage’. In one sense the relationship is easy to specify. For instance, under the Speenhamland system (which periodically from the 1790s to the 1820s linked the level of relief to the price of bread and the size of rural families) farmers were encouraged to pay low and sticky wages in the knowledge that deficits in the household income of rural labouring families would be made up with community resources marshalled via the parish rate. But, the Speenhamland system was just one way in which the poor law might have a very direct impact on the level, form and significance of wage payments. In much of the north, for instance, practices such as paying small pensions to those who sought work outside the parish or paying small pensions at slack times of the industrial year, acted in effect as a labour market subsidy and made industrial wages less responsive to supply and demand than might have otherwise been the case.

170

Cash, wages and makeshifts

The need for employers to keep the core of the local population (and hence a potential labour force in remote areas) in place through seasonal fluctuations, periods of falling piece and time rates, and episodes of life-cycle crisis, allied with the fact that millowners and their relatives often owned the debts run up by a constant tendency for the poor law to overspend, meant that there was an inevitable linkage between labour markets and the poor law for those at the prime of their working age. In turn, there is considerable evidence from vestry minutes and other poor law records that ratepayers and overseers saw efforts to help oneself through work as a gateway to the relief process and continued to insist on work as well as welfare even in the extremities of old age.84 There were also more subtle, long lived and complex relationships between the poor law system and labour markets which shaped the function and symbolism of ‘the wage’. Overseers could engage with the credit system. Evidence from pauper letters in Essex, Westmorland and Lancashire indicates clearly that pleas from paupers for overseers in their parish of settlement to meet debts increased during the eighteenth and early nineteenth centuries. Vestry minutes suggest that many such pleas were met, effectively bolstering otherwise fragile local credit networks and heading off demands for cash wages on a regular basis. Meanwhile, the poor law might also meet extraordinary payments such as rent, which could prove both a burden to working families and a problem for landlords and local economies if it remained unpaid. In places like nineteenth-century Lund, Lancashire, up to one half of all poor law resources in some years were expended on payment of rents.85 The vestry of Garstang consistently struggled with the burden of rent payments, regularly recording resolutions to the effect that no more rent payments would be authorised. This had little effect, as rents continued to be paid.86 In turn, if the poor law increasingly came to remove the need to save surplus wages for extraordinary items such as rent (and also clothing and medical relief), so both the level and the form of the wage could remain substantially static in both agricultural and industrial localities. Moreover, we should not forget that the poor law was in its own right a significant employer and payer of wages, ranging from the workhouse master and mistress, to suppliers of goods and services. Relief and payment of wages might even overlap where the goods and services represented by increasing payment in kind in most places during the later eighteenth century were supplied to poor people by other poor people.87 Overseers were also very likely to support self-employment. In northern communities in particular they can be seen systematically paying for items to support work, rather than simply paying to support wages. This included expenditure on things like coals for blacksmiths, looms, potatoes for seeding and cloth for selling. In terms of overall poor law budgets, these sorts of expenditure were usually small but the effect may have been disproportionate. In the Lancashire cotton spinning parish of Longton, for instance, the poor law authorities entered into agreement with a local merchant partnership that the parish would pay loom rents for local textile workers, keeping a whole range of people off relief who might otherwise

Muldrew and King

171

have been dependent.88 Similarly on 24 March 1816 the poor law authorities employed a loom master to repair the looms of local people who might otherwise be obliged to interrupt work and claim poor relief. The overseers also lent, outside the framework of the poor law accounts, money to buy looms. In short, the poor law could subtly or directly interlink with the labour market and the wider economy of makeshifts. In so doing it might eliminate the need for cash to circulate to pay for expenditure such as rent. The effect was also to redirect cash flows from the wider community in the form of rates, to supplement wages and other earnings which otherwise would have been circulated through market forces, with the inevitable bottlenecks that this would have implied. While more work needs to be done on this crucial relationship, it seems to us clear that it is impossible to view the level, form or symbolism of wages in isolation from either the availability of cash or the nature of the economy of makeshifts as much historiographical literature has done thus far. Nor should we ignore the fact that dealing with the lack of cash and the complexity of family and household earnings meant that credit rather than weekly wages lay at the heart of the family economy.

Wages in a credit economy One piece of evidence for the importance of such credit is the fact that bread continued to be sold in the form of penny loaves, where the size of a loaf changed according to the changing market price of grain, but was always worth a penny thus making it easy for bakers to keep track of what poor families owed for their bread.89 Such reliance on sales credit can be seen in the suits brought before the Bristol Court of Conscience, which was established in 1689 to process a large number of suits in a short time with little expense. Although the limit on cases which could be brought before the court was 40s., most were for less than a pound, and thus show what sort of debts poorer households litigated over.90 Many suits were for debts which the poor owed for food and other necessities, and many were for debts they were owed for small services which were part of the economy of makeshifts. Small tradesmen sued for debts as varied as meat and drink, rent for lodging, butter and eggs, a hat, coffee, bread, poles, shoes, a pistol, and much beer and cider. In 1692, where the goods sold were listed, food and drink accounted for twenty-nine percent of all sales. There were also many suits brought by tailors and cordwainers, but only a handful for unpaid wages. Debts owed for work and specific small services were always more common than wages. These included things as varied as nursing, washing, the sale of milk, schooling, payment for gate keeping, carriage of hay, hauling stones, fees for teaching a deaf child, and dressing wounds.91 Rent for lodging was common as well.92 In the Bath Court of Requests (another name for a court of conscience) studied by Margot Finn, in 1829 and 1839 the majority of

172

Cash, wages and makeshifts

suits brought to court concerned debts for food and drink, while by 1839 more than twice as many suits were brought against labourers than any other occupational listing.93 Such observations fit well with the recollections of a late nineteenth-century observer on his childhood in Yorkshire, where credit was the mainstay of life and the wage was something that figured as a residual issue to be dealt with at the end of a week.94 Even as late as 1900 many wage earners had to purchase goods on credit to survive until they had done sufficient work to receive payment.95 As noted previously, wages could often go unpaid for years on end, and even after substantial periods of contractual work the value of wages could be disputed depending on their economic context. A case brought to King’s Bench by a London mantua maker in 1754 concerning her proper wages shows how complicated the negotiation for a wage could be, and also how many people might be involved in the negotiation. The case arose over an account concerning payment for 160 weeks and two days work between 1749 and 1754. The defendant, another female mantua maker who employed her, admitted that she had paid the plaintiff £48 17s. 3d. after a rate of 6s. a week, while the plaintiff claimed her work was worth 8s. Various other master mantua makers as well as servants were called as witnesses to determine what different work was worth. Wages ranged between 6s. to 10s. 6d. a week depending on ability to plait. Seamstresses earned less, and a woman could earn more if she could do multiple tasks. Wages were also affected by the worker’s reputation for the quality of her work. One master mantua maker who kept nineteen to twenty servants also claimed she provided breakfast, and paid her workers more in the mornings, although she did not explain why. A male master claimed that the common wages for a journeywoman were 6s. a week with 2d. a day for breakfast, while another claimed he did not give so much wages to men as woman mantua makers ‘because we don’t want such extraordinary hand’. It was also established that male workers who were skilled at plaiting earned 15s. a week because it was claimed they could do twice the work, which presumably meant they worked twice as long if they were not as skilled. In the end the court awarded wages of 7s. a week, which it decided was the ‘common’ payment in the trade for the plaintiff ’s skill, which had to be accepted when no agreement could be reached between the bargaining parties.96 William Hutton also claimed that in Birmingham wages fluctuated with fashion, and that journeymen would earn more when something new was in high demand, and less when it became more widely produced. As a result the fluctuation of wages was ‘a constant source of wrangling between master and man.’97 It is thus unsurprising that poor households would run up debts to the maximum extent possible to optimise their limited ability to make ends meet.98 Since, as a proverb of the time had it, ‘Sue a Beggar, &c. and they have nothing to lose, their punishment will ne’r make you satisfaction’ many must have felt they had little to lose by being in debt.99 According to William Hutton, the judge and chronicler of the Birmingham Court of Requests, Birmingham abounded with cases of wives whose husbands had

Muldrew and King

173

left them because of debts. Since these poor women with children to support could not earn enough through their own labour they had to survive by using, every species of finesse …to overreach those who trade in the necessities of life. She subsists by a little ready money and a little credit. The money is temporary, the credit eternal. Her depredations are chalked upon every huckster’s door in the neighbourhood.

If a shopkeeper tried to sue her, however, a separated or abandoned wife could plead that in law she was still married and could not be sued for debts which were legally her husband’s responsibility under coverture.100 In this way, exploiting credit which might not be repaid was one essential part of the strategies of survival of poorer households, and these debts very often exceeded income by wages or other makeshift work which was unpredictable.101 As Hutton put it on another occasion, a debt may be trusted with him who has, but to trust it with him who has not, is like trusting it to the bottom of sea. In the commercial world, a rich knave is preferable to a poor honest man; one pays in money, the other in promises. 102

He also complained that even though skilled artisans in Birmingham could earn 10s. to two guineas a week wages many spent this and had little left but promises for their landlords, a fact that has its analogue in the increasing payment of rents by poor law authorities in the eighteenth century. In this case Hutton argued, their poverty was their security as they had little to distrain and the expense of ejectment or imprisonment was too great. Modern historians have been more cautious, noting increasing distraint in most areas by the 1760s. Pauper letters also suggest that the threat of distraint loomed large. Nonetheless, because the number of small houses built to be rented was great, landlords had to rent to those who might not pay. It was the same for shopkeepers: competition forced them to sell on credit if they wanted business. In his history of the parish of Myddle in Shropshire, Richard Gough singled out poor, honest, households for praise. However, this was the judgement of someone successful, and for many poor households bad fortune such as injury, death, unemployment or an unpaid debt could make their efforts worthless in a stroke.103 As a result manipulating credit, customary entitlements, and charity must have been a more realistic option than honesty and forbearance, especially when need was pressing.

Conclusion A local economic system inhibited by lack of cash; a labour market which was inextricably tied up within a complex economy of makeshifts; and a remuneration system pervaded by customary entitlements and distorted

174

Cash, wages and makeshifts

by the existence of complex credit networks, created many problems for the entrepreneur. As we know from the Industrial Revolution literature, it was difficult in the 1770s to recruit industrial labour, to improve its productivity and output and to make effort a contractual obligation. It is no surprise, then, to see attempts to rationalise payment towards greater, but less negotiable wages on the part of masters from an early date. One of the first to attempt to do this was Ambrose Crowley, as can be seen in the law book of his late seventeenth-century ironworks. Crowley has become famous for attempting to implement a ‘modern’ type of factory discipline and accountable organisation at a very early date.104 The organisation of his works, however, was still very contemporary in that labour was negotiated on an individual basis, and the law book goes into great detail on how the value of different types of work were to be reckoned on a weekly basis.105 In order to reduce the complexity of such reckonings and to make them more amenable to central accounting he attempted to forbid lending money or equipment to his workmen, which he claimed made them, ‘negligent, extravagant and delayed their reckonings’ and which ‘brought disorder upon the Cashier, by answering the workmen’s solicitations in lending and correcting payments’.106 Crowley realised that to prevent this he would have to pay regular wages, and part of his system of reckoning was to make deductions for equipment and his many fines up front to reduce the amount paid every week. But even so the supply of money was not enough for an organisation of such scale, and long before the advent of county banks he claimed he printed his own current bills ‘to the end that they might be valued in all places better than money’.107 In addition to prevent borrowing because of poverty he set up his own poor relief system.108 Peter Linebaugh has also argued that the ‘right’ of workers in the Royal Navy dockyards at Deptford to take chips of wood left over from the carpentry in the yard, came under attack from the administrators of the dockyards who attempted to replace the right to chips with an increase in wages, so as to make the operation more ‘accountable’ and cost effective as the importance of keeping proper accounts grew. But, the Yard was one of the largest employers in the country, employing 900 men and boys in peacetime and 1,200 during war, and it was no easier for the Navy to find cash to pay dockyard workers, when sailors were not being paid, so it is not surprising that workers resisted these attempts.109 The subject of how such rationalisation was achieved, in the face all the problems outlined here, is a crucial aspect of factory organisation, which obviously needs to be addressed. The importance of this question is further emphasised if we consider that the concept of the wage which has been criticised here is as much a product of a certain form of industrial organisation in which an efficient money supply is to a large extent taken for granted. In this sense the ‘modern’ wage in is some measure ideological in that, in conception at least, it is supposed to be a measurement of the great majority of income meant to replace the negotiated survival economy of makeshifts, outdoor relief and credit manipulation.110

Muldrew and King

175

Notes 1. Michael Sonenscher, ‘Work and Wages in Paris in the Eighteenth Century’, in Maxine Berg, Pat Hudson and Michael Sonenscher (eds), Manufacture in Town and Country Before the Industrial Revolution (Cambridge, 1984), p.147. 2. Leonard Schwarz, ‘The Formation of the Wage: Some Problems’, in Peter Scholliers (ed.), Real Wages in nineteenth and twentieth Century Europe: Historical and Comparative Perspectives (Oxford, 1989), pp. 21–39. 3 Peter Linebaugh, The London Hanged: Crime and Civil Society in the Eighteenth Century (London, 1991), pp. 371–401; Donald Woodward, Men at Work: Labourers and Building Craftsmen in the Towns of Northern England, 1450–1750 (Cambridge, 1995), ch. 5. 4. Although in some places the wage might be the centrepiece of family welfare; cf. Anegélique Janssens, Family and social change: the household as a process in an industrializing community (Cambridge, 1993). 5. J. Innes, ‘The “mixed economy of welfare” in early modern England’, in Martin Daunton (ed.), Charity, self-interest and welfare in the English past (London, 1996), pp.139–180; T. Sokoll, ‘Old age in poverty: the record of Essex pauper letters 1780–1834’, in T.Hitchcock, P. King and P. Sharpe (eds), Chronicling poverty: The voices and strategies of the English poor 1640–1840 (Basingstoke, 1997), pp. 127–54. 6. This meant, in comparison to the Parisian artisans studied by Sonenscher, that wage payments in England were more bound up with a contractual market economy. The right to the necessities of life provided by the natural law of self-preservation was supplied by access to credit rather than being bound up with a set of rights associated with specific trade organisations. Sonenscher, ‘Work and Wages’, p.156. 7. Tim Stretton, ‘Women, custom and equity in the Court of Requests’, in Garthine Walker and Jenny Kermode (eds), Women, crime and the courts in early modern England (London, 1994), pp. 170–89; J. M. Neeson, Commoners: Common Right, Enclosure and Social Change in England, 1700–1820 (Cambridge, 1993), ch. 6. 8. Oxford English Dictionary xix, p.803. 9. Richard Britnell, The Commercialization of English Society 1000–1500 (Cambridge, 1993), p. 113; Maurice Keen, English Society in the Late Middle Ages 1348–1500 (Harmondsworth, 1990), p.137. 10. Ann Kussmaul, Servants in Husbandry in Early Modern England (Cambridge, 1981), pp. 5–8. 11. John Locke, Two Treatises of Government, edited by Peter Laslett, (New York, 1963), 2, chs. 28, 85. 12. Peter Laslett, The world we have lost further explored, (third edition, London, 1983). 13. John Law, Money and Trade Considered with a Proposal for Supplying the Nation with Money (Edinburgh, 1705), pp. 13–4, 98–9; Adam Smith, An Inquiry into the Nature and Causes of the Wealth of Nations, 2 vols., eds. R. H. Campbell and A. S. Skinner (Oxford, 1976) 1, ch.8. 14. Statutes of the Realm, 1 p. 307. 15. Reference from Jane Whittle. Also, L. Botelho, Churchwardens’ accounts of Cratfield, 1640–1660 (Woodbridge, 1999). 16. G. E. Fussell, (ed.), Robert Loader’s Farm Accounts 1610–1620, Camden Society, 3rd ser., LIII (1936), i.e., pp. 100, 146, 152, 154, 166; Farming and Account Books of Henry Best, Surtees Society, 33 (1857), p.154. 17. Norfolk Record Office, KL/C39, 105–107. 18. There are over 250 probate accounts for labourers in the Hampshire Record Office, for instance. 19. Richard Britnell, The Commercialization, pp.29–47. 20. Ibid., pp.47–8, 104–15, 113–5. 21. J. R. Wordie, ‘Deflationary factors in the Tudor Price Rise’, Past and Present, 154 (1997), pp. 49–61. 22. Craig Muldrew, The Economy of Obligation: the Culture of Credit and Social Relations in Early Modern England (London, 1998), pp. 99–103.

176

Cash, wages and makeshifts

23. Alan Macfarlane (ed.), The Diary of Ralph Josselin 1616–1683 (Oxford, 1976), pp. 102, 110–11, 115, 118, 188, 221. 24. Barry Supple, Commercial Crisis and Change in England 1600–1642 (Cambridge, 1959), pp.14–8, 54, 85, 131, 166–7, 173ff. 25. Muldrew, Economy of Obligation, p. 102; D. W. Jones, War and Economy in the Age of William III and Marlborough (Oxford, 1988), pp. 15–16. 26. Sir John Craig, The Mint, A History of the London Mint from A.D. 287 to 1948 (Cambridge, 1953), pp. 219–22, 247–51, 253; A. E. Feavearyear, The Pound Sterling, A History of English Money (Oxford, 1931), pp. 151–8, 169–72. 27. Craig, Mint, pp. 179, 220–21, 246, 250–51, 261–6; Feavearyear, Pound Sterling, pp. 158–60, 169–70; J. R. S. Whiting, Trade Tokens; A Social and Economic History (Newton Abbot, 1971), pp. 22–3; Sir John Clapham, The Bank of England, I–II (Cambridge, 1970) 1, pp. 22, 39–40, 140–2, 161–3; L. S Pressnell, County Banking in the Industrial Revolution (Oxford, 1956), pp. 145–7. 28. On the shortage of cash in Yorkshire during this period see John Styles, ‘ “Our Traitorous Moneymakers”: The Yorkshire Coiners and the Law, 1760–83’, in John Brewer and John Styles (eds), An Ungovernable People, The English and Their Law in the Seventeenth and Eighteenth Centuries (New Brunswick, N.J., 1980), pp. 172–249; Craig, The Mint, p.251. The population estimate is from E. A. Wrigley and R. S. Schofield, The Population History of England (Cambridge, 1989), p. 533. A household multiplier of 4.75 was used. 29. Craig, The Mint, pp. 260–7. 30. Peter Mathias, ‘The People’s Money in the Eighteenth Century: the Royal Mint, Trade Tokens and the Economy’, in Peter Mathias (ed.), The Transformation of England (London, 1979), pp. 197ff. 31. For a particularly good example of the practice of out-parish relief, see Rawtenstall Library (RL), The poor law accounts of Cowpe, Lemches and Newhallhey. 32. There is an extensive literature on these issues. See B. Anderson, ‘Provincial aspects of the financial revolution in the eighteenth century’, Business History, 12 (1969), pp. 1–22, L. S. Pressnell, Country banking in the Industrial Revolution (Oxford, 1956) and Pat Hudson, The genesis of industrial capital: A study of the West Riding wool textile industry 1750–1850 (Cambridge, 1986). 33. Christopher Dyer, Standards of living in the Later Middle Ages: Social Change in England c.1200–1520 (Cambridge, 1989), pp. 213–4. 34. Alan Everitt, ‘Farm Labourers’ in Joan Thirsk (ed.), The Agrarian History of England and Wales 1500–1640, 4 (Cambridge, 1967), pp. 163–4; James Sharpe, Social History of Early Modern England (London, 1987), pp. 211–2. 35. Laslett, The World We Have Lost, pp. 36–7. 36. Peter H. Lindert and Jeffrey G. Williamson, ‘Revising England’s Social Tables 1688–1812’, Explorations in Economic History, 19 (1982), pp. 385–407. 37. Donald Woodward, ‘The Means of Payment and Hours of Work in Early Modern England’, in Carol S. Leonard and B. N. Mironov (eds), Hours of Work and Means of Payment: The Evolution of Conventions in Pre-Industrial Europe, Proceedings of the Eleventh International Economic History Congress (Milan, 1994), p. 17; Kussmaul, Servants in Husbandry, p. 40. 38. Fussell, Loader’s Farm Accounts, pp. 72, 90, 107–8, 137. 39. Ibid., pp. 2–3, 20–1, 74. 40. Ibid., p. 90. 41. Farming and Account Books of Henry Best, p.154. For other examples of masters keeping livestock for servants, see Kussmaul, Servants in Husbandry, p. 39. 42. Memorandum Book of Richard Cholmeley of Brandsby, 1602–1623, North Yorkshire County Record Office Publications, 44 (1988), pp. 53, 66. He also forgave rent in lieu of wages: ibid., p.63. 43. Ibid., pp. 80, 84. 44. Ibid., pp. 76–7. 45. Ibid., p. 62.

Muldrew and King

177

46. Somerset Record Office, DD.DR 70 Part 1. 47. This can be seen in the wage book of the Somerset farmer Francis Hamilton from 1802, and in a number of farm accounts from eighteenth- and early-nineteenth-century Ulster. Somerset Record Office, DD/FS 7/4; Vivienne Pollock, ‘Contract and Consumption: Labour Agreements and the Use of Money in Eighteenth-Century Rural Ulster’, Agricultural History Review, 43 (1995), pp. 19–34; M. Reed, ‘ ‘Gnawing it Out’: A New Look at Economic Relations in Nineteenth-Century Rural England’, Rural History, I, (1990), pp. 83–4, 91. 48. Memorandum Book of Richard Cholmeley, p. 186. 49. Lancashire Record Office (hereafter LRO), DDX 115/91,’Memoranda book’ and DDPr 25/6, ‘Doctor’s account book’. The latter document is particularly interesting, testifying to a remarkable bartering economy in West Lancashire in which the doctor was paid in labour, cloth, beans, etc. for his services. 50. Woodward, Men at Work, pp. 142–59. 51. Craig Muldrew, ‘ “Hard Food for Midas”: Cash and its Social Value in Early Modern England’, Past and Present, 170 (2001), pp. 78–120. 52. London Corporation Record Office, Mayor’s Court Equity Proceedings, Bils and Answers; 248E, Box 75. 53. J. M. Fewster, ‘The keelmen of Tyneside in the eighteenth century, Part I’, Durham University Journal, New Series, 19 (1957), p. 27. 54. Hutton, The Court of Requests (Edinburgh 1840), pp. 30–1. 55. See RL, ‘Whitehead collection’. 56. On these issues see J.Smail, Merchants, markets and manufacturers: the English wool textile industry in the eighteenth century (London, 1999) and S. A. King, Poverty and welfare in England 1700–1850: a regional perspective (Manchester, 2000). 57. RL, RC 355,’Whitehead collection’. 58. George W. Hilton, The Truck System, Including a History of the British truck Acts, 1465–1960 (Cambridge, 1960).For the relationship of truck to shortages of cash, and worker’s credit, see esp. pp. 47ff. 59. John Locke, Some Considerations of the Consequences of the Lowering of Interest, and Raising the Value of Money (London 1698), repr. in Patrick Hyde Kelly (ed.), Locke on Money, 2 vols. (Oxford, 1991), 1, pp. 237. Both William Petty and John Law also claimed that the lack of cash in the economy was perhaps the major impediment to economic growth, as the amount of wages which could be paid limited the amount of work labourers would do; William Petty, Treatise of Taxes and Contributions (London, 1662), reprinted in C. H. Hull (ed.), Economic Writings (Cambridge, 1899), 1, p. 36; J. Law, Money and Trade Considered, with a Proposal for Supplying the Nation with Money (Edinburgh, 1705), pp. 13–4, 17, 98–117. 60. Craig The Mint, p. 247; Peter Clark, British Clubs and Societies, 1580–1800: The Origins of an Associational World (Oxford, 2000), pp. 39–40, 129–30; Joan Thirsk and J.P. Cooper, Seventeenth-Century Economic Documents (Oxford, 1972), pp. 97–9; Leonard Schwarz, London in the Age of Industrialisation: Entrepreneurs, Labour Force and Living Conditions, 1700–1850 (Cambridge, 1993), pp. 117–21. 61. Lincoln’s Inn Library, Dudley Ryder’ Law Notes, 1754–56; Volume 12, pp.17–18. 62. Hutton, Court of Requests, p. 24. 63. Ibid., p. 40. 64. Robert Latham and William Mathews (eds), The Diary of Samuel Pepys, (London, 1970–1983), 4 , pp. 144, 149. 65. Nicholas Rodger, The Wooden World. An anatomy of the Georgian navy (London 1986). Mariners were also often given goods being shipped in part payment for wages, as in the case of a sailor employed on a vessel shipping coal from Newcastle to King’s Lynn who was given coal to sell as part wages. Norfolk Record Office, KL/C25/18, 08/17/53. 66. For a discussion of litigation see below; George F. Steckley, ‘Litigious Mariners: Wage Cases in the Seventeenth-Century Admiralty Court’, Historical Journal, 42 (1999), pp. 315–45. 67. Paul Slack, Poverty and policy in Tudor and Stuart England (London, 1985), pp. 171–2; John Rule, Albion’s People: English Society 1714–1815 (London, 1992), p. 129.

178

Cash, wages and makeshifts

68. On the words and lives of the poor, see the contributions to T. Hitchcock, P. King and P. Sharpe (eds), Chronicling poverty (Basingstoke, 1997). 69. RL, RC 352. Raw,’The diary of David Whitehead’. A full transcript of this diary has been kindly donated to Oxford Brookes University library by the librarian at Rawstenstall. See also R. J. Soderlund, ‘Intended as a terror to the idle and the profligate: embezzlement and the origins of policing in the Yorkshire worsted industry 1750–1777’, Journal of Social History, 31 (1998), pp. 647–70. 70. Muldrew, Economy of Obligation, pp. 107–9. 71. For excellent material on this strategy see the narratives and vestry decision contained in LRO DDX 325, ‘Garstang vestry minutes’. 72. Michael Anderson, Family structure in nineteenth century Lancashire (Cambridge, 1972). 73. S. A. King, ‘Dying with style: infant death and its context in a rural industrial community’, Social History of Medicine, 10 (1999), pp. 3–24.. 74. See LRO DDIn/45/14, ‘Report on the town cottages of Birkdale, 1815’, and LRO DDIn/46/36, ‘Lease, 1816’. 75. For the most recent survey see H. Cunningham and J. Innes (eds), Charity, philanthropy and reform from the 1690s to 1850 (Basingstoke, 1998). 76. In the larger urban areas and distant rural counties such as Westmorland and Cumberland charitable resources continued to grow strongly. In the Westmorland communities of Undermilkbeck and Applethwaite, for instance, Thomas Dixon left 20s. per annum to the poor in 1730, James Sattherwaite left the interest on £100 to the poor in 1785, and Margaret Williams left the interest on £67 to the poor in 1789. If we combine these with existing historical legacies, we can see that the capital for these two townships amounted to over £1500 by the later eighteenth century. At 5 percent interest, this would eclipse the formal poor relief bill for the townships at this date. See S. A. King, Poverty. 77. Bible charities, clothing charities, bread charities, and funds devoted to religious or educational purposes offered little immediate aid for cyclical poverty. D. Andrew, Philanthropy and police: London charity in the eighteenth century (New Jersey, 1989). 78. This dole book is reproduced in A. Gilson, Halifax past and present (Halifax, 1892). The tendency to donate to irregular collections for the poor can be seen in most other northwestern communities of all sizes. See, for instance, Cumbria Record Office (CRO), WD/D/D6/68, ‘Donations to the poor in Foulshaw’. 79. See CRO WD/Ry/37/1, ‘List of charitable gifts’; CRO WD/Ry/18/81, ‘Receipts and disbursements left by Jno Flemming 1664’; CRO WPR/62/W1 and W2, ‘Overseer accounts’ and CRO Wd/Te/24, ‘Poor law accounts’. 80. Sara Horrell and Jane Humphries, ‘Old questions, new data and alternative perspectives: families’ living standards in the industrial revolution’, Journal of Economic History, 52 (1992), pp. 849–80, and Jane Humphries, ‘Enclosure, common rights and women: the proletarianisation of families in the late eighteenth and early-nineteenth centuries’, Journal of Economic History, 50 (1990), pp. 117–42. Also P. King, ‘Customary right and women’s earnings: the importance of gleaning to the rural labouring poor 1750–1850’, Economic History Review, 44 (1991), pp. 461–476. 81. See Yorkshire Archaeological Society (hereafter YAS) DD11/11/38, ‘Enclosure data’, and YAS DD12/I/11/15–29, ‘Depositions’. 82. See P. Lane, ‘Work on the margins: poor women and the informal economy of eighteenth and early nineteenth century Leicestershire’, Midland History, 22 (1997), pp. 85–99. 83. For a good example, see S. Peyton, Kettering vestry minutes, 1797–1853 (Northampton, 1933). 84. See King, Poverty. 85. The accounts are in private hands. A photocopy can be consulted at Oxford Brookes University library. 86. LRO DDX 325, ‘Garstang vestry minutes’. 87. For good examples, see LRO DDIn/63/37, ‘Accounts of Birkdale township’.

Muldrew and King

179

88. LRO DDHe/83/84, ‘Cottage and loom rents’. In 1820 the overseers spent £16 hiring looms for twenty-seven people. It is also perhaps important to note that the poor law here had incurred considerable debts to the local manufacturers as a result of overspending and the borrowing to cover this extra expenditure. 89. 5+6 Edward VI ch.14; The Assize of Bread, (London, 1636) 90. Bristol Record Office, Z33, 04413, 04753, 04414. 91. Ibid., Z33 04754. The court records do not list the nature of every suit. In a sample of 257 cases from the King’s Lynn Guildhall Court from 1652 to 1654 only four concerned wages. Norfolk Record Office, KL/C25/17–18. 92. Almost fifty years later, in 1739, there were only four cases dealing with wages compared to over forty for goods sold. Ibid., Z33 14413. 93. Margot Finn, ‘Debt and Credit in Bath’s Court of Requests’, Urban History, 21 (1994), pp. 220–2. 94. J. Lawson, Letters to the young on progress in Pudsey (Chichester, 1972 reprint). 95. G. R. Rubin, ‘The County Courts and the Tally Trade, 1846–1914’, in G. R. Rubin and David Sugarman (eds), Law, Economy and Society 1750–1914. Essays in the History of English Law, (Abingdon, 1984), p. 342. 96. Lincoln’s Inn Library, Dudley Ryder’ Law Notes, 1754–56: Volume 13, pp.17–19; Hutton, Court of Requests, p. 52. 97. Hutton, Court of Requests, p. 52. 98. John Locke, Some Considerations of the Consequences, pp. 236–7. Also see, Charles Davenant, ‘A memorial concerning credit and the means and methods by which it may be restored’, (1696), printed in G. Heberton Evans Jr. (ed.), Two manuscripts by Charles Davenant (Baltimore, 1943), pp. 72, 97, 102; William Petty, A Treatise of Taxes, pp. 34–6. 99. England’s great happiness, or a dialogue between Content and Complaint (London, 1677), reprinted in J.P. McCulloch, A select collection of early English tracts on commerce (Cambridge, 1954), p. 18. 100. Hutton, Court of Requests, p. 47. 101. For an estimate of the amount of desperate debts which might have been forgiven, see Muldrew, Economy of Obligation, pp. 177–80, 304–7. 102. Hutton, Court of Requests, p. 42. 103. Richard Gough, The history of Myddle, edited by D. Hey, (Harmondsworth, 1981), pp. 101, 145, 316. 104. M. W. Flinn (ed.), The Law Book of the Crowley Ironworks, Surtees Society, CLXVII (1952), pp.xix, xxv–xxvi. 105. Ibid., pp. 133–9. 106. Ibid., pp. 66, 71, 128, 139–42. 107. Ibid., pp. 50–6 108. Ibid., pp. 154ff. 109. Linebaugh, London Hanged, pp. 371–401. 110. For an excellent discussion of the problems this entailed at a bureaucratic level, see Roger Davidson, Whitehall and the Labour Problem in Late-Victorian and Edwardian Britain (London, 1985), chs. 4–6.

AGE, GENDER AND WAGES

CHAPTER 8

GENDERED WAGE SYSTEMS AND INDUSTRIALISATION IN FINLAND IN THE LATE NINETEENTH AND EARLY TWENTIETH CENTURIES*

Sakari Heikkinen

Do wages have sex? Are the wages of women different from those of men? Does woman’s place in the world of wage labour differ from that of man? Historians’ answer to these questions is affirmative. Yes, women’s wages have been – and still are – in general lower than men’s wages. Yes, women have in general participated less in wage labour than have men.1 But do wage systems have sex too? Do the systems of definition and measurement of wages (e.g., time or piece rates) themselves have gender characteristics? In this article I try to shed a little northern light on this subject that is less studied than wage gaps and participation-rate differences, by examining wage systems in Finland in the late nineteenth and early twentieth centuries. In the middle of the nineteenth century, Finland – a latecomer compared with the economic core of Europe – was only at the beginning of her industrialisation process. Although the country was still quite agrarian in the 1910s, she had gone through a considerable economic change during the preceding fifty years.2 Along with growing new industries and an emergent market economy, wage labour expanded. The proportion of wage labourers in the economically active population grew from fortythree percent in 1850 to fifty-two percent in 1910.3 How industrialisation changed the role played by gender in wage formation is the question I am discussing on the basis of the Finnish data. My focus will be on the wage systems within manufacturing, on which there is relatively good material

184

Gendered wage systems

in the ‘labour statistical studies’ carried out at the beginning of the twentieth century.4 The Finnish evidence scrutinised here implies that the role played by gender in the making of wage systems was complex. The notable female–male wage-system disparities found should be analysed, I argue, with a view to the persistent female–male wage gaps, and to occupational segregation. Therefore this chapter will first examine the proportions of women and men among the strata of wage labourers and the degree of occupational segregation in manufacturing industries. Secondly it will discuss female–male wage differentials; in explaining these, the differences in work careers of women and men are explored. These sections pave the way to the main theme of this article: the gender characteristics of the wage systems in Finnish manufacturing industries around 1900. In the concluding section tentative explanations for female–male wage-system differences are discussed.

Women’s and men’s jobs In the nineteenth and early twentieth centuries the majority of Finnish wage workers were agricultural labourers. According to my estimates the proportion of agricultural workers in the total sum of wage labourers was sixty-eight percent in 1850, sixty-three percent in 1880 and fifty-four percent in 1910. Within agricultural workers there were two basic groups: annually hired farm servants on the one hand, and day labourers on the other. Farm servants made up almost two thirds of agricultural labourers in 1850, a little more than half in 1890 but only a quarter in 1910.5 According to the 1910 population statistics, thirty-nine percent of agricultural workers were women. There were, however, notably more women among the annually hired farm servants (sixty-two percent) than among the day labourers (thirty-five percent).6 Inconsistencies and deficiencies in the population statistics do not allow us to measure precisely changes in the proportions of male and female workers in agriculture from 1850. This much, however, can be said: the proportion of women within all farm servants rose after 1890, when it was fifty-three percent (the figure was equal in 1880).7 The rapid growth in the numbers of female farmhands can be explained by the growth in dairy farming, because livestock tending and dairying was predominantly women’s work.8 The principle ‘man to the plough, wife to the cow’,9 applied also to Finland in the nineteenth and early-twentieth centuries. In sum we may roughly estimate that about forty to forty-five percent of agricultural wage labourers were women during the latter half of the nineteenth century and the early twentieth century. In manufacturing industries the proportion of women wage workers was smaller than in agriculture, implying that industrialisation did not necessarily lead to an increase in the female participation rate – a conclusion that has been made for British and French industrialisation.10 About

Heikkinen

185

every fourth manufacturing worker was a woman during the period from the 1850s to 1910s. The proportion fluctuated somewhat: it was twentyfour percent in 1855, nineteen percent in 1885 and grew again to twenty-nine percent in 1913.11 These variations were the result of two phenomena: changes in the structure of the manufacturing sector on the one hand, and changes in the proportion of women in the labour force of the different branches on the other. There were male-dominated manufacturing industries and female-dominated industries, but the gender character of an industry could also change – for instance because of changes in production techniques. Textile manufacture was the biggest sector with a female majority in the labour force: between 1845 and 1875 this fluctuated between fiftyfour and fifty-nine percent, and subsequently grew to seventy-four percent in 1913. Match manufacture was a female-dominated industry, women formed fifty-seven to sixty-five percent of its labour force between 1875 and 1913, but this was a minor branch compared with the textile industry. Tobacco manufacture, again, is an example of an industry with a changing male–female composition in its labour force: the proportion of women varied between twenty and thirty-nine percent in 1845 to1875, jumped to seventy-one percent in 1885, and rose again to eighty-six percent in 1913. This rapid increase may be attributed to changes in production techniques and structure, with the production of cigarettes, manufactured by women, expanding. Male-dominated sectors within manufacturing included first of all the metal industry, where women comprised no more than two percent of the workforce. The proportion was a little higher in the sawmill industry, fluctuating at around ten percent. The manufacture of leather and glass, among others, was largely male-dominated. ‘Intermediate’ branches were not totally dominated by male workers, although women accounted for less than half the workforce. The most important of these was the pulp and paper industry, where between 1885 and 1913 over one third of the workers were women. Until the 1870s, that is before the expansion of the new wood-processing paper industry, the proportion of women was smaller, varying between twelve and twenty-four percent between 1845 and 1875.12 The distinction between women’s and men’s jobs in manufacturing can be analysed in more detail on the basis of so called ‘labour statistical studies’, carried out at the beginning of the twentieth century. These studies examine eight manufacturing industries that together comprised two-thirds of the total labour force of the manufacturing sector.13 Labour statistics show that manufacturing occupations were clearly segmented into female and male jobs. The degree of segregation is shown in Table 8.1 by using the ‘index of dissimilarity’ (D), which puts its values between zero (no segregation) and one hundred (total segregation). The value of D indicates how great a percentage of either women or men would need to change occupations to make the occupational distributions of women and men equal.14 The index of dissimilarity was highest in the textile industry (82.3) and lowest in the sawmill industry (50.8). The index across all 387

186

Gendered wage systems

occupations in these eight branches reached 75.4, meaning that female and male jobs were very strongly segregated. When calculated across all 129 manufacturing industries in 1910, the dissimilarity index is lower (at 55.7), but this is the result of the more aggregate way of computing. The value is quite near that of 60.8 that was calculated by Claudia Goldin for twenty manufacturing industries in the United States in 1890.15 Table 8.1. Female–male segregation of occupations and manufacturing industries, 1902–191016

Tobacco manufacture (1902) Textile manufacture (1902) Bakeries (1903) Printing industry (1905) Machinery manufacture (1907) Pulp and paper manufacture (1909) Glass manufacture (1910) Sawmill industry (1910) All occupations in industries above

Number of occupations

Index of dissimilarity (D)a

28 127 17 24 35 66 30 60 387

61.1 82.3 53.3 55.1 78.1 77.1 79.3 50.8 75.4

Number of industries All manufacturing industries in 1910 a

129

55.7

Index of dissimilarity is D = ∑ fi – mi /2, i where fi = the percentage of female workers in occupation/industry i Mi = the percentage of male workers in occupation/industry i

(The distribution of female workers in machinery manufacture [only 1.4 percent of the labour force] is partly estimated)

Of course there were women working outside agriculture and manufacturing, but the information concerning them is quite fragmentary. Outside agriculture, domestic service was the most common area of female employment until the turn of the twentieth century, when manufacturing took its place. The clear majority of all domestic servants were women: seventyfour percent in 1880, seventy-nine percent in 1890 and ninety-three percent in 1910.17 Also, shop workers attendants were mostly women. According to surveys of two cities in 1914 to 1915, the female proportion of shop assistants reached eighty percent.18 House construction, again, was predominantly men’s work, but there were female workers too. According to the patchy data available on construction workers between 1886 and 1908, women made up a little over ten percent of the labour force. There were very few women in skilled occupations, such as painters and glaziers, so that most of them were classified as ‘unspecified’, common building workers, who made up the bulk

Heikkinen

187

of the construction labour force.19 The fact that women were also doing heavy building work, such as carrying bricks, should warn us against seeking causes for female–male occupational segregation only in physiological characteristics, such as women’s allegedly ‘nimble fingers’. Men and woman took care of quite different tasks in a traditional agricultural society. Industrialisation changed the established division of labour by bringing in new occupations but, as elsewhere, these did not abolish the division into female and male jobs, since women and men were very dissimilarly distributed in the newer as well as in the older manufacturing occupations.

Female – male wage differentials One of the most enduring facts of wage history is the existence of notable and persistent wage differentials between men and women. How these differentials should be explained causes much debate.20 What was the role of supply and demand on the one hand, and the role of tradition, custom, cultural norms – or whatever name one may give to non-economic factors – on the other hand in determining the female – male wage gap? How did industrialisation and the evolving capitalist market economy change the role of custom in the formation of the wage, and what effect did this have on the wage differential? It is obviously more difficult to answer these questions than it is to gauge the extent of the wage gap itself. Whatever the cause, Finnish wage data show strong continuities in female – male wage differentials. In agriculture the gap seems to have been stable for a very long time. Between 1700 and 1740 the wage of annually hired female farms servants was, on average, sixty-seven percent of the male wage (in the provinces of Satakunta and Northern Ostrobothnia), and between 1907 and 1913 it reached sixty-five percent. Furthermore, the female – male pay ratio calculated from the wage series for agricultural day labourers for 1891 and 1913 was at about sixty-seven percent. We thus may conclude that ‘traditionally’ – at least in the eighteenth and nineteenth centuries – women’s wages in Finnish agriculture were twothirds of men’s wages.21 According to the evidence reported in Table 8.2, the female – male wage differential among unskilled urban ‘outdoor’ workers – e.g. workers in the building trade, in ports and in various casual jobs – seems to have been, on average, somewhat larger than the wage gap among agricultural workers. Women’s wages in the town of Vaasa were about two-thirds of male wages in 1851 to 1880, i.e. the same as in agriculture, but the 1861 to 1880 figures for Oulu town, as well as those for Helsinki for 1889 – 1910, show lower percentages of around fifty-five. It is also interesting to note that both the Oulu and Helsinki figures indicate a decline in women’s relative wages. Far-reaching conclusions, however, should not be drawn from these figures since the data are quite limited.

188

Gendered wage systems

Table 8.2. The ratio of female urban outdoor workers’ wages to male workers’ wages, 1851–1910, (percentages).22

1851–1860 1861–1870 1871–1880 1889–1900 1901–1910

Vaasa Summer

Oulu Summer

Helsinki Whole year

63.5 68.4 70.3 n.a. n.a.

n.a. 59.6 55.4 n.a. n.a.

n.a. n.a. n.a. 54.8 52.8

(The percentages were calculated from wages paid to outdoor workers hired by the towns. The figures are averages of yearly female:male pay ratios)

The female–male wage gap among manufacturing workers during the first decade of the twentieth century can be calculated quite reliably from the data provided by the labour statistical studies mentioned above. The first feature to note in Table 8.3 is the substantial variation in female–male wage differentials between different manufacturing industries. The table also shows quite a strong correlation between the female-male composition of the labour force and the size of wage gap: the higher the proportion of women, the smaller the wage gap. Manufacturing industries were thus divided into two: male-dominated industries with large wage gaps on the one hand, and female-dominated industries with smaller wage gaps on the other.23 If intra-industry differentials are weighted according to the number of workers in the branch (men and women in 1910), the average female–male wage ratio of manufacturing turns out to be fifty-five percent.24 The percentage is thus about the same as in urban outdoor works (see Table 8.2), and over ten percentage points lower than in agriculture. It is interesting that the female–male wage ratio in Finland was almost exactly the same as in the United States, where female manufacturing wages were 55.4 percent of male wages in 1900 and 55.6 percent in 1905.25 Table 8.3. The ratio of female manufacturing workers’ wages to male workers’ wages, 1902–1910 (percentages)26

Tobacco manufacture (1902) Textile manufacture (1902) Printing industry (1905) Machinery manufacture (1907) Pulp and paper manufacture (1909) Glass manufacture (1910) Sawmill industry (1913)

Women’s wages/ men’s wages

Female workers in the labour force

66.2 62.8 50.1 44.1 50.3 38.8 56.7

81.4 70.1 32.3 1.4 29.2 12.5 12.6

(The best of the available wage series were used for calculating the wage ratio. Weekly wages were used for the tobacco and textile industries, hourly wages for machinery manufacturing and sawmill industry, and annual wages for other branches)

Heikkinen

189

The female–male wage differentials for manufacturing workers examined above are not, of course, ‘pure’ sex differentials, because the average wages of women and men were also influenced by differences in age, experience and skill. Table 8.4 and Figure 8.1 show that both men’s and women’s wages were highest in the age group thirty to forty-four, but they also reveal that women’s age-earnings profile was much flatter than that of men. Female and male workers started at about the same wage level when they were younger than eighteen, but at the age of eighteen to twentynine there was already a wide wage gap between them, and this became even wider at thirty to forty-four years. This wage differential then prevailed into older age groups, too. Table 8.4. Wages of manufacturing workers by sex, age and branch, 1902–1910 (average male wage in the branch = 100)27 –17 Tobacco manufacture (1902) Textile manufacture (1902) Printing industry (1904) Machinery manufacture (1907) Pulp and paper manufacture (1909) Glass manufacture (1910) Weighted averages

Men Women Men Women Men Women Men Men Women Men Women Men Women Women /Men

52.1 49.2 45.3 43.0 24.5 24.9 40.8 47.5 42.7 27.1 (33.1) 39.8 38.8 0.974

18–29 92.4 65.9 89.7 63.6 85.2 46.9 83.3

30–44 119.4 69.8 118.1 69.8 149.0 85.9 115.2

91.7 109.5 52.0 50.9 85.5 119.6 39.5 (39.3) 87.3 115.4 60.5 65.0 0.693 0.563

45–59 113.9 69.0 116.0 66.9 138.6 (71.2) 113.4

60+ (126.0) (60.7) 105.8 61.1 (97.4) – 93.0

102.2 85.7 46.7 42.2 130.8 95.2 (38.0) (39.2) 111.1 93.8 60.7 54.8 0.546 0.574

(The figures in parentheses represent less than twenty workers. The weighted averages of relative male and female wages were calculated from branch averages by using the number of male or female workers of the relevant age group in the different branches as weights. The women/men ratio on the bottom line is the ratio of male and female weighted averages)

The difference in the mean age of female and male manufacturing workers – women were on average a couple of years younger than men – explains some of the female-male wage differential, but it does not help to explain the gaps within the age groups. It is not age as such, but the experience and skills acquired that count. There was a notable difference in the average length of the careers of men and women: the proportion of experienced workers was notably higher among male than female workers in each and every manufacturing industry examined by earlytwentieth-century labour statisticians (see Table 8.5).

190

Gendered wage systems

Figure 8.1. Average wages of female and male manufacturing workers by age, 1902–1910

Averga male wage = 100

120 100 80 60 40 20 0 –17 Men

18–30

30–44

45–59

60–

Women

Table 8.5. The age and work experience of manufacturing workers, 1902–191028 Mean age Men

Tobacco manufacture (1902) Textile manufacture (1902) Printing industry (1905) Machinery manufacture (1907) Pulp and paper manufacture (1909) Glass manufacture (1910) Sawmill industry (1910)

31.0 33.4 25.5 33.1 34.9 30.4 32.6

Work experience, years in the branch

Women Men

27.4 26.4 21.6 29.2 29.5 28.2 30.9

12.2 12.9 11.5 16.1 9.0 12.8 10.7

Women Women/ Men (=FEMEXP) 9.2 8.6 4.8 5.0 6.4 4.7 5.6

0.756 0.663 0.416 0.311 0.707 0.369 0.526

Labour statistics do not, unfortunately, contain individual-based data that would allow wages to be correlated with work experience. We have therefore to content ourselves with a much coarser regression based on branch averages. Women’s relative wage (FEMWAGE), i.e. female/male wage ratio, was used as a dependent variable, and the indicator measuring the relative length of the female worker’s career (FEMEXP, see Table 8.5) as an independent variable. The estimated relationship was FEMWAGE = 0.279 + 0.450 FEMEXP (R2 = 0.678; significant at 5 percent level). This yields the predicted value of FEMWAGE at 0.729 (i.e., female wages = 72.9 percent of male wages), when the explanatory variable FEMEXP is given the value of 1.0, meaning that there is no ‘experience gap’ between female and male workers.

Heikkinen

191

The estimation above is, of course, a very rough one; it is based on scanty and all too aggregated data. Yet, it seems reasonable to conclude that the ‘pure sex differential’ in the wages of female and male manufacturing workers was much smaller than the crude percentage (female:male wage ratio of fifty-five percent) calculated from Table 8.3 indicates. On the other hand, however, although age and length of job tenure explain some of the wage gap between female and male manufacturing workers, a notable differential – about thirty percent of male wages – is still left to be explained. It is interesting that this ‘residual’ female:male wage differential in manufacturing is about the same size as the ‘traditional’ wage gap in agriculture. The remaining wage gap can again be interpreted in the competitive framework, resulting from the differences in productivity, or it can be explained as a result of discrimination – or both factors may be partial causes.29 The limited data used here, however, do not give the means for weighing these alternative explanations.

Gender characteristics of wage systems The existence of distinct wage differentials between female and male workers is, of course, not surprising: it only confirms generalisations made in the literature on the history of wage labour. Less obvious, however, is the existence of female–male differences within wage systems. Whether there existed gender-based disparities in wage systems and the nature of these possible disparities, is examined in the following section. Wage systems can be classified into two basic categories: systems in which wages are measured by time worked, and those in which they are measured by the result of the work. Within these two categories there are different subclasses. Wages can be determined by the year, month, week, day or hour in the payment-by-time systems, or in the case of payment by results by straightforward piecework pay, or as payment for a more complicated task. In the latter case, they can be paid to an individual worker or to a group of workers. Furthermore, combined systems can be created by using both working time and results as the basis of payment.30 Time wages were dominant in Finnish agriculture in the nineteenth and early twentieth centuries. Farm servants were normally hired for a year and for a reward that was agreed when the contract was made and the master paid the recruitment fee to the servant. This usually occurred in autumn. The wages for the agricultural labourers who were hired for a shorter period, ‘day labourers’, were determined as a daily rate, and because the length of the working day varied within the year, the daily wage also varied. In particular the summer day rate was higher than that of winter, to take account of the much greater length of the former. Combined daily and piece rates were also used, especially during the busy harvest season. It could be agreed, for example, that for a daily wage the day labourer had to cut fifteen shocks of rye, and for extra shocks he then received

192

Gendered wage systems

extra pay.31 The time rate, however, was the dominant form, both for female and male agricultural workers. Both time and piece rates were used for urban ‘outdoor’ work. Daily wages seem to have been more common from the 1820s to the 1870s, and were used for carpenters and other labourers in house building. Dockers and street pavers, however, worked for piece-rate wages, and towards the end of the nineteenth century piece rates gained ground. Payment by results was used for bricklayers and stoneworkers, first in the form of subcontracting, then as contracts taken out by working gangs, and further as individual workers’ piece-rate contracts. This development was resisted by the fledgling trade unions, but they did not manage to turn the tide. In terms of time, daily rates were changed into hourly rates at the turn of the twentieth century.32 Piece rates were clearly dominant in railway construction, two-thirds of the work being paid this way and one-third by daily rate from 1890 to 1913.33 Both wage systems were used in the growing forestry sector, too. According to a study conducted in 1910, piece rates were applied to timber felling, where working-gang contracts were common, whereas daily rates were the rule in timber floating. Time wages were dominant also in loading sawn goods onto ships (79 percent against 21 percent of workers).34 There is not much to say about female–male differences in the wage systems of construction and forestry works, since these sectors were very much male-dominated. In manufacturing, however, the gender aspect is relevant because here women made up a significant part of the labour force. Because of the relatively good information on wage systems in the labour statistics studies of the early twentieth century, manufacturing is a particularly rewarding area of research. Five of these studies show the percentages of workers paid at piece or time rates. The industries concerned are presented in Table 8.6, ranked in descending order, according to the proportions on a piece-rate system. The table shows, first, that there was great variation between these five industries in this respect, and second, that the piece-rate system was most common in industries where the majority of workers were women. Table 8.6. Wage systems in manufacturing industries 1902–1910 (percentages)35 Workers in different wage systems Piece rate

Tobacco manufacture (1902) Textile manufacture (1902) Glass manufacture (1910) Pulp and paper manufacture (1909) Printing industry (1905)

72.2 57.1 30.1 22.8 20.3

Time rate

27.3 42.3 64.7 52.4 79.3

Piece and time rates 0.5 0.6 5.2 24.8 0.4

All Female workers workers in the labour force 100.0 100.0 100.0 100.0 100.0

81.4 70.1 12.5 29.2 32.3

Heikkinen

193

For three manufacturing industries – tobacco, textiles and printing – we can take a closer look at female–male differences in wage systems. Table 8.7 shows a clear gender bias in wage systems of these industries. Piece rates were the rule among female workers in tobacco and textile manufacturing, which were women-dominated, whereas the majority of male workers received time-rate wages. In the printing industry, where male workers made up the majority of the labour force, time wages were dominant among both men and women, but even there they were more common among men than women. Table 8.7. Male and female wage systems in the tobacco, textile and printing industries 1902–1905 (percentages)36 Piece rate

Tobacco manufacture (1902) Textile manufacture (1902) Printing industry (1905)

Men Women Men Women Men Women

28.7 81.9 20.0 72.9 18.1 24.9

Time rate

70.3 17.7 78.7 26.8 81.6 74.5

Piece and time rates 1.0 0.4 1.2 0.4 0.3 0.7

All workers

100.0 100.0 100.0 100.0 100.0 100.0

For these three industries there is also information on wage systems in different occupations. In Figure 8.2 the percentage of workers receiving piece-rate wages is plotted against the proportion of women in the labour force. The data refer to sixty-five occupations (with at least thirty workers) in the tobacco, textile and printing industries. Figure 8.2 also shows branch-level observations concerning these three industries plus two branches (pulp and paper, glass), for which there is no occupational data, as well as the quadratic trends for both series. An examination of sixty-five different occupations within the tobacco, textile and printing industries gives a fairly good (R2=0.504) and statistically significant (at the 1% level) correlation between the piece-rate proportion and the female-labour-force proportion. Furthermore, the quadratic regression clearly gives a better fit than the linear regression, although the U-shape of the curve is rather slight. Results from the industry-level quadratic regression form a more distinct lopsided U-shape.37 This regression is based on only five aggregated observations, so no far-reaching conclusions can be drawn from it. Nevertheless, it hints that if we had occupational data on the two industries (glass, pulp and paper) not included in the occupational regression, the U-shape would be clearer than it is now with occupation-level data available for only three industries.38 To sum up, these regressions strongly indicate that the greater the proportion of the labour force formed by women, the more common was the piece rate; yet, they also suggest – although less strongly – that a clear male-dominance in the labour force tended to raise the percentage of

194

Gendered wage systems

piece-rate workers. Similarly, in US manufacturing Goldin’s examination of forty-eight manufacturing industries in 1890 show a strong correlation between the piece-rate percentage and female-labour-force percentage. Her data, however, do not show the U-shape pattern suggested by the Finnish data, i.e. in the 1890 US data the piece-rate percentage does not increase when the male-labour-force percentage approaches 100 percent.39 The female:male differences in wage systems shown by the statistical exercises above could plausibly be interpreted as an indicator of occupational segregation. This interpretation gains support from a closer examination of those occupations, which were neither male- nor female-dominated. From Figure 8.2 we see that eight of a total sixty-five occupations in three industries belonged to this intermediate category (the proportion of women at 33 to 67 percent).40 Within these eight occupations the gender bias found on industry and occupation levels disappears. In three occupations the piece rate was more common among women than men, in three cases the contrary was true and in two occupations the percentage of piece-rate workers was equal among women and men. The average piece-rate percentage for women (13.1 percent) was lower than for men (16.3 percent), whereas in general, as we have seen, the order of magnitude was reversed. Furthermore, both percentages were notably lower than the average percentage for all sixty-five occupations (37.3 percent). 100

Piece rate, %

67

R 2 = 0,990

33 R 2 = 0,504

0 0

33

67

100

Women, % Note: Big symbols refer to the branches, small to the occupations. Tobacco, textile, printing

Pulp and paper, glass

Figure 8.2. The correlation between piece rate frequency and the proportion of women in the labour force, by manufacturing industries and occupations, 1902–1905

Heikkinen

195

Custom or market? We have seen above that gender played a role in the world of wage labour, not only in the sense that women and men worked in different occupations and earned unequal wages but also in the sense that wage systems were not ‘gender neutral’. At least the evidence from Finnish manufacturing industries in the early twentieth century shows a significant asymmetry in the wage systems of female and male manufacturing workers. Women were in general paid by piece rate more often than men, but the percentage of piece-rate workers – according to the scanty data available – increased again when the proportion of men in an occupation approached 100 percent. Within nonsegregated occupations with approximately equal proportions of women and men, the percentage of piece-rate workers was low and there was no significant difference between sexes. Thus, not only occupations but also wage systems were segregated. How could the gender-based asymmetry of wage systems be explained? Is the female–male disparity in wage systems somehow connected to the other two aspects of female–male differences discussed above, i.e. wage gaps and occupational segregation? Because these questions are too big to be answered conclusively on the basis of the limited Finnish data used, the suggestions given in the following paragraphs should rather be seen as hypotheses for further research. The ‘gender-wage debate’ on explaining large and persistent female– male wage gaps has been to a great extent organised around the polarity of ‘custom’ and ‘market’.41 According to the ‘customary wage’ interpretation, wage gaps existed because of tradition: women received lower wages than men because that was the established custom. According to the ‘market wage school’, i.e., neoclassical economic theory, wage gaps should be interpreted as stemming from productivity differences. A customary-wage explanation implies the existence of wage discrimination against women, whereas a market-wage explanation does not of course give support to a discrimination thesis – that is, not in a competitive labour market. Several recent studies have convincingly argued that wage discrimination per se was not the cause of large female–male wage gaps during the early stages of industrialisation. When comparing female and male wages in the same occupations and taking into account the differences in the productive attributes of women and men, the discrimination hypothesis does not receive support.42 The rejection of the wage discrimination hypothesis does not, however, imply that there was no discrimination at all and that all differences between female and male wages were market-based. One form of ‘nonwage’ discrimination was occupational segregation, which – as we have seen – was very distinct in Finnish manufacturing industries in the early twentieth century. The ‘occupational crowding’ resulting from segregation would press female wages lower than they would be in a nonsegregated situation.43 Would it be right to say that here ‘custom’ walks into the house of market wage theory through the back door, since

196

Gendered wage systems

occupational segregation can hardly be explained purely in terms of neoclassical economics? Furthermore, the female–male wage-system disparity described above can be interpreted in the alternative frameworks of ‘custom’ or ‘market’. Industrialisation and the development of market economy furthered more exact measurement of work performance: daily wage rates, for instance, were changed into hourly rates. And a piece-rate system, one could argue, was best suited for a competitive market economy, since wages were measured according to the work actually done. Why, then, were women’s and men’s wage systems not symmetrical in manufacturing industries? Skill differences offer a possible market-based explanation to the puzzle. If piece rates were more easily applied in low-skill jobs than in high-skill jobs, then women labourers – in general less skilled than men – would be paid more often by piece rates than men. A simple ‘cultural’ explanation, again, could be derived from the supposed unpopularity of piece rates. If men had better resources – social and political – with which to oppose the introduction of piece-rates systems than were available to women, then a greater percentage of male workers would be paid by time. This, however, may be too simple an explanation, since piece rates were also common in male-dominated occupations. A further solution to the puzzle may be sought in neither custom nor market as such, but in their contradiction. We could assume that the traditional wage structure of agriculture with its stable female–male wage gaps formed the starting point for the modern wage systems of an industrialising economy. This does not seems as an implausible assumption in the case of Finland, where female agricultural workers wages stayed for two centuries in an almost unchanging proportion (about two-thirds) to men’s wages. Since these were time rates – annual or daily – paid for quite segregated jobs, we may presume that women could be paid lower time rates also in manufacturing simply on the basis of ‘tradition’. Different time rates for female and male workers did not visibly contradict the market logic of productivity-based wage differentials since occupations in manufacturing were clearly sex segregated. Different piece rates for women and men, however, would appear to contradict market rationality. It is, thus, no wonder that sex-neutral piece rates were the rule during the British industrial revolution.44 The hypothesis sketched above offers a possible explanation for the Ushaped piece-rate curve (see Figure 8.2) that indicated that piece rates were least common in integrated occupations. If women and men worked in the same occupations and if they could not – because of the ‘market’ – be paid by different piece rates, they should be paid by time rates in order to maintain the ‘customary’ female–male wage gap. Thus, the female–male disparity in wage systems, made possible by occupational segregation, could be interpreted as a ‘solution’ to the contradiction of custom and market. Whether this hypothesis stands the test of more extensive evidence remains to be seen.

Heikkinen

197

Notes * I wish to thank Carmen Sarasua, Peter Scholliers, Leonard Schwarz and two anonymous referees for their insightful comments, and Leonard also for checking and improving my English. This article is based on research supported by Finland’s Academy (the project Markets, Institutions, Actors: A Comparative Analysis of the Finnish Economic Crises in the twentieth Century). 1. In the mature industrial economies of the twentieth century, female–male wage gaps have narrowed, and participation rates of women and men have converged (Glen Cain, ‘The Economic Analysis of Labor Market Discrimination: A Survey’, in Orley Ashenfelter and Richard Layard (eds), Handbook of Labor Economics, 1 (Amsterdam, 1986), pp. 693–785; Mark R. Killingsworth and James J. Heckman, ‘Female Labor Supply: A Survey’, in Ashenfelter and Layard, Handbook, I, pp.103–204; John Pencavel, ‘Labor Supply of Men: A Survey’, in Ashenfelter and Layard, Handbook, 1, pp. 3–102. The development, however, has not been linear throughout the earlier phases of industrialisation as Goldin has suggested in her seminal book on female employment and wages in the USA, Claudia Goldin, Understanding the Gender Gap. An Economic History of American Women (New York and Oxford, 1990). 2. According to the official population statistics, eighty-two percent of Finland’s population (1.6 million in 1850, 3.1 million in 1913) was classified as agrarian in 1850, and sixty-seven percent in 1910. Judged on the basis of production figures, Finland was somewhat more of an industrialised economy than one might conclude from the population statistics. The proportion formed by primary production had diminished from sixty-one percent in 1860 – 1864, to forty-four percent in 1909 – 1913, that of manufacturing had more than doubled to nineteen percent, and the service-sector share had grown to twenty-seven percent. It should be noted that the proportion of ‘pure’ agriculture had dropped to below thirty percent in 1910. The high remaining proportion of forestry production can largely be explained by the growth of export-oriented woodworking industries which were the fastest growing of the manufacturing industries; see Sakari Heikkinen, Labour and the Market. Workers, Wages and Living Standards in Finland, 1850–1913 (Helsinki, 1997), p. 36; Riitta Hjerppe, The Finnish Economy 1860–1985: Growth and Structural Change, (Helsinki,1989), pp. 215–29. 3. Heikkinen Labour, p. 42. 4. The eight studies I am utilising (see note 16) contain very rich information on workers and wages in manufacturing industries. The studies, however, have their deficiencies, which limit my analyses. First of all, they are cross-sectional studies, thus they give no longitudinal data, on the basis of which changes in wages and wage systems over time could be examined. Secondly, the data of the ‘labour statistics’ is classified thus making individual-level analysis impossible. Thirdly, these studies contain information only on workers and wages, not on production, therefore there is no possibility to compare, for instance, the relative productivity of male and female labourers. 5. Heikkinen, Labour, p. 195. 6. Official Statistics of Finland (Helsinki) VI, 47. 7. Official Statistics of Finland VI, 11, 22. 8. Matti Peltonen, Talolliset ja torpparit: Vuosisadan vaihteen maatalouskysymys Suomessa (Helsinki, 1992), p. 217. Fieldwork, again, was virtually a male monopoly, with the exception of harvesting, in which women took part almost as much as men. Land clearance and construction, as well as maintenance, also were men’s work. It should also be remembered that female farm servants were not ‘pure’ agricultural workers but functioned at the same time as domestic servants. 9. Ivy Pinchbeck, Women Workers and the Industrial Revolution 1750–1850 (1930; reprint, London, 1969), p. 37. 10. Sara Horrell and Jane Humphries, ‘Women’s labour force participation and the transition to the male-breadwinner family, 1790–1865’, Economic History Review, 48 (1995), pp. 89–117.

198

Gendered wage systems

11. Riitta Hjerppe and Per Schybersgon, Kvinnoarbete i Finlands industri ca. 1850–1913 (Helsingfors, 1977), p. 4. Manufacturing statistics have been compiled according to consistent principles since 1885; the data before 1885 are not perfectly comparable and are based to some degree on estimations of the authors. 12. Hjerppe and Schybergson, Kvinnoarbete; Official Statistics of Finland XVIII, pp.1–30. 13. The proportion is calculated on the basis of 1910 manufacturing statistics, Official Statistics of Finland XVIII. 14. See Goldin, Understanding the Gender Gap, p.75. 15. Goldin, Understanding the Gender Gap, pp. 75–81. 16. G. R Snellman, Undersökning av tobaksindustrin i Finland (Helsinki, 1903); G.R. Snellman, Undersökning av textilindustrin i Finland (Helsinki, 1904); G.R. Snellman, Undersökning af bagareyrket i Finland (Helsinki, 1905); G.R. Snellman, Undersökning av tryckeriindustrin i Finland (Helsinki, 1907); G.R. Snellman, Undersökning av mekaniska verkstäder i Finland (Helsinki, 1911); Vera Hjelt, Undersökning av yrkesarbetares levnadsvillkor i Finland (Helsinki, 1911); G.R. Snellman, Undersökning av pappersindustrin i Finland (Helsinki, 1912); G.R. Snellman, Undersökning angående glassindustrin i Finland (Helsinki, 1913); G.R. Snellman, Undersökning angående sågindustrin i Finland (Helsinki, 1914). 17. Kaarina Vattula, ‘Palvelustytöstä konttoristiin – naisten työhönosallistuminen 1880–1940’, in Yrjö Kaukiainen (ed.), När samhället förändras – Kun yhteiskunta muuttuu (Helsinki, 1981), pp. 63–90. 18. Marjaliisa Hentilä, Keikkavaaka ja koussikka. Kaupan työ ja tekijät 1800–luvulta itsepalveleuaikaan (Helsinki, 1999), p. 491. The survey concerns cities of Helsinki and Viipuri, which were first and fourth largest respectively. 19. Official Statistics of Finland XVIII, 24, 25, Official Statistics of Finland XVIII, 3–23. 20. See Joyce Burnette, ‘An investigation of the female–male wage gap during the industrial revolution in Britain’, Economic History Review, 50 (1997), pp. 257–281; Leonard Schwarz, ‘The Formation of the Wage: Some Problems’, in Peter Scholliers (ed.), Real Wages in nineteenth and twentieth Century Europe: Historical and Comparative Perspectives (Oxford, 1989), pp. 27–8. 21. Heikkinen, Labour, p. 84. 22. Heikinen, Labour, p.86. 23. The female–male wage gap was somewhat narrower in the Finnish than in the British textile and metal industries, where the female–male wage ratio was 54.9 percent in textile and 37.3 percent in metal industry in 1906 (Edward Hunt, British Labour History 1815–1914 [London, 1981], p. 104). It is interesting that, despite the difference in the size of wage gaps, the ranking of these industries was the same and that relatively speaking the pay ratios were about the same: the textile/machinery ratio was 1.42 in Finland and the textile/metal ratio 1.47 in Britain. 24. Official Statistics of Finland XVIII, Manufacturing statistics. 25. Goldin, Understanding the Gender Gap, p. 60. 26. Labour Statistics, 1–4, 12, 14, 15, except Sosiaalinen Aikakauskirja and Manufacturing statistics (1913) for sawmills. 27. Source: Labour Statistics, 1–4, 12, 14, 15. 28. Source: Labour Statistics, 1–4, 12, 14, 15. 29. On the limits of ‘residual’ analysis, see Donald Cox and John V. Nye, ‘Male–Female Wage Discrimination in Nineteenth-Century France’, The Journal of Economic History, 49 (1989), p. 904; on discrimination in general, see, e.g., Cain, ‘Economic Analysis’ of Labor Market Discrimination’, and Goldin, Understanding the Gender Gap, pp. 83–118. 30. For a 1920s catalogue on wage systems, see von Zwiedineck–Südenhorst, ‘Lohntheorie und Lohnpolitik’, in Handwörtebuch der Staatswissenschaften, 4th edition., 6 (Jena, 1925), pp. 396–426; for a more modern view, G. H. Webb, ‘Payment by Results Systems’, in Angela M. Bowley (ed.) Handbook of Salary and Wage Systems, 2nd. edition (Aldershot, 1982), pp. 285–96; and Anne G. Shaw and Pirie D. Shaw, ‘Payment by Time Systems’, in Bowley, Handbook of Salary, pp. 297–308.

Heikkinen

199

31. A. M. Soininen Soininen, Maataloustyöväen palkkakehitys 1800–luvun lopussa ja 1900–luvun alussa: Ajanjakso 1878–1913 (Helsinki, 1981), p. 7. 32. Eemil Hynninen, Ulkotyöväen palkkasuhteen kehitys Suomen suurimmissa kaupungeissa 19. vuosisadalla ennen elinkeinovapautta (Helsinki, 1924), pp.68–89; Aarne Mattila, Työmarkkinasuhteiden murros Suomessa: Tutkimus työntekijäin pyrkimyksistä osallistua työehtojen määräämiseen 1880–luvulta vuoden 1905 suurlakkoon, (Helsinki, 1969), pp.113–5; Eino Kuusi, Talvityöttömyys, sen esiintyminen, syyt ja ehkäisytoimet Suomen suurimmissa kaupungiessa (Tampere, 1914), pp. 98–108. 33. Official Statistics of Finland XX, Railway statistics. However, there seems to have been a slight drift towards daily wages, which accounted for 67.5 percent of working days in 1890 to 1899, 65.7 percent in 1900–1909, and 64.7 percent in 1910 to 1913. 34. Labour Statistics 16, pp. 188, 216, 224. In loading work, hourly wages were somewhat more common than daily wages. 35. Source: Labour Statistics, 1, 2, 4, 14, 15. 36. Source: Labour Statistics, 1, 2, 4. 37. The regression is statistically significant at the 1% level; R2 = 0.990. 38. In the case of the three industries (tobacco, textile and printing) occupation-level regression gives a slight U-shape whereas industry-level regression gives a straight line. If the same applies for the two other industries (glass, paper and pulp) the occupation-level regression of the five industries would give an even more distinct U-shape than the industry-level regression displayed in Figure 8.2. 39. Goldin, Understanding the Gender Gap, pp. 78–9. 40. The small number of these occupations (12 percent against the 33 percent expectation value of equal female–male distribution) is an indicator of occupational segregation. 41. See, e.g., Janet Greenlees, ‘Equal Pay for Equal Work?: A New Look at Gender and Wages in the Lancashire Cotton Industry, 1790–1855’, in Margaret Walsh (ed.) Working Out Gender. Perspectives from labour history (Aldershot, 1999), and Burnette, ‘Investigation’. 42. Cox and Nye, ‘Male–Female Wage Discrimination’; Burnette ’Investigation’; Greenlees ‘Equal Pay’. 43. Burnette ‘Investigation’, pp.261–2; Cox and Nye, ‘Male–Female Wage Discrimination’, p. 915. 44. Burnette, ‘Investigation’; Greenlees, ‘Equal Pay’.

CHAPTER 9

ENGENDERING THE EXPERIENCE OF WAGES: THE EVOLUTION OF THE PIECEWORK

SPANISH TOBACCO MONOPOLY, 1800–1930S

SYSTEM AT THE

Lina Gálvez-Muñoz

Gender and wages have mainly been linked in the literature through the study of earnings differentials and discrimination and, to a lesser extent, through the employer’s choice of a particular system of remuneration. However, wage discrimination – which occurs when the female-male wage ratio does not match the productivity ratio – is only one potential kind of discrimination, and not even the most important for explaining the potential of gender as a useful category for historical analysis. The study of the relationship between the employer’s choice of a remuneration system and gender shows how the price of labour was only one variable in explaining employers’ hiring patterns. This is consistent with recent historical and theoretical work showing that the method of pay is in fact the glue in the labour market, affecting and itself affected by the choices and strategies of suppliers and demanders of labour time. Payment schemes are not identical and their effects are not neutral.1 Gender helps to explain how the operations and dynamics of the labour market were not independent of the payment schemes. This chapter assumes that employers’ choices of a particular organisation of production and, as a matter of fact, towards a particular system of remuneration were not indifferent to gender. Within the context of a specific organisation of production that comprises the choice of a technique, the recruitment, maintenance, monitoring of workers – including its relation with unions if they exist – and the reward of workers, an analysis of the

202

Engendering the experience of wages

payment scheme adopted constitutes an excellent area for testing the use of gender as an analytical tool.2 Through the study of the evolution of the piecework system at the Spanish tobacco monopoly, this chapter sets out to test the potential of gender in explaining employers’ choices towards a specific remuneration system and a specific organisation of production. In addition, this chapter shows how earnings differentials were just one variable in explaining employment hiring patterns by gender. Gender discrimination in the field of earnings during the industrial revolution has caught the attention of economic historians, both because of its magnitude and because of its potential to explain the extent to which there was a functioning labour market. In fact, the earning differential between men and women was so large, with women earning between one-third to two-thirds of the male wage, that many historians have found it impossible to be satisfied with the neoclassical assumption that wages must equal the marginal product of labour and that the gender gap must therefore be explained through productivity differences.3 The work of J. Burnette is relevant here, in particular her suggestion that female wages during the British industrial revolution followed Gary Becker’s view that wage discrimination would be more compatible with monopolistic than with competitive markets.4 Burnette has sought to demonstrate that women workers in more competitive industries such as agriculture and most cottage industries received market wages based on their marginal productivity, not simply a customary women’s wage; this would explain why a crowding out phenomenon did not occur in such industries. Discrimination did exist but took other forms, particularly through institutions such as unions and professional organisations that blocked women’s access to certain occupations and limited their ability to acquire skills.5 Such behaviour seemed to have increased during the Second Industrial Revolution, when technological change was accompanied by ‘horizontal’ segregation by gender (i.e., men and women were concentrated in different industries).6 This makes it more difficult to explain gender division within the labour market in relation to earning differentials, since occupational crowding by industry was increasing as well as the differentiation of tasks within the same industry or even the same company. In addition, the model of the male breadwinner, although neither new nor universal, was probably being used more explicitly to help employers and institutions justify, among other things, the payment of lower wages to women than to men.7 In fact, gender differences went beyond earnings differentials and institutional constraints. The gendered division of labour within the family meant that men and women were different types of workers, especially in their allocation of time, including the female demand for flexibility in order to attend to household duties, as well as in regard to investment in human capital. The different effects of marriage and family on female, as opposed to male labour supply make it self-evident that if gender is to be included as an analytical tool and if it is to progress beyond a simple differentiation of male and female patterns, then the family needs to be placed at the same level of analysis as the market. Differences

Gálvez-Muñoz

203

between male and female workers, such as degrees of flexibility, union pressures, or specific skills that were socially constructed and not always considered as skills, guided employers’ choices.8 Among the different types of remuneration systems it is probably piecework, with its long historic attachment to domestic and subsequently to female labour, that can most easily be linked with gender. However, in neoclassical economic theory the choice of a piecework system by an employer is not normally analysed in relationship to gender but instead to factors such as the ownership structure of the company, firm size, technology, market structure and institutional constraints. Piece rates, bonus schemes or any type of result-based reward system are basically treated as the most obvious way to raise worker effort – as long as this output can be measured. A system based on results is simply considered the most obvious one to adopt, unless employers have considerable difficulty in assessing the marginal product.9 In business and management literature a piecework reward system is similarly presented as bearing the simplest relationship between effort and reward. Three main groups of payment systems must be distinguished: by results, by time and by company-wide bonus schemes. Payment by results presumes that if money can be used to induce effort, then more money will result in more effort. In essence, the result of the effort must be clearly seen. The method of setting the standard of output is usually settled by a simple piecework bargain.10 Recent historiography, however, has described a more complex picture. By showing the diversity that existed between different systems of remuneration it has demonstrated that the labour market did not operate in a quasi-perfect manner and that wage changes were therefore not solely a consequence of changes in the supply and demand for labour. There has been a tendency to explain this by reference to custom, especially after E. Hobsbawm in 1960 stressed the importance of customary expectations in wage determination before at least the mid-nineteenth century.11 Workers expected remuneration appropriate to their status as well as to their need to support an expected quality of life. Piece rates could remain unchanged over very long periods. J. Rule has supported this, stating that as late as the second quarter of the nineteenth century wages cannot be related to any simple concept of the sale of labour power in a market.12 Nevertheless, Rule sees payment by results becoming managerial orthodoxy from the 1830s, moving outside its traditional arena within the domestic system. This development took place during a period of relatively high unemployment, and was very much linked to the intensification of labour, with a new stress on time discipline and accommodation to the pace of machinery. The development was not, of course, uniform, and in some industries and some regions older traditions survived into the twentieth century.13 Custom seemed to operate more in some markets than in others where market mechanisms were more efficient in establishing wages.14 If gender is introduced, the analysis becomes more complicated. Rule has stated that outside agriculture the starting point for any discussion of wage forms is the distinction between those who sold simple labour power

204

Engendering the experience of wages

and those who sold skilled labour power. But one of the potentials of gender as an analytical tool is to call for a rethinking of traditional categories of analysis within the different social sciences. Despite the historiography, the notion of skill was not only technical. Skills were socially constructed and gender contaminated. Only if the social construction of skills is taken into account will it be possible to understand the relationship between gender and an employer’s choice of a particular remuneration system. By not doing so, the traditional historiography has been in danger of considering the same job skilled or unskilled depending on the gender involved: there has been a tendency to place women in the same statistical category as children, disregarding their skills.15 In fact, neither market nor custom are sufficient if gender is included in the picture. The gender division of labour was also an effective tool for determining wages and the type of remuneration imposed or selected, and it shows how collusion could exist among employers and employees in relation to a specific organisation of production.16 In addition to variables such as local labour market conditions or institutional constraints, the choice of male or female workers when a piecework system was preferred did not occur randomly. For instance, neoclassical economic theory considers that the piece rate for a given job is likely to be higher than the time rate of pay for the same job. Two reasons are given: payment by the piece will induce greater effort and this effort must be paid for if workers are to consent to be paid by the piece; and payment by the piece gives rise to a more variable income than paying by time, and this extra variability should give rise to a compensating differential. Again, a gender approach calls for a redefinition of certain assumptions. Both these reasons assume that the main advantage of a piecework remuneration system is to increase workers’ output. But the advantages of the piecework system could be of a different nature if the imperfections of the market derived from the gendered division of labour are accepted. In that assessment, variability is automatically considered as negative for labour, in part because it has been analysed through male categories, measuring the labour supply curve as a trade off between wages and leisure. But if instead it is analysed through a gender perspective, including the homework in the labour supply curve, variability is not necessarily seen as negative for the worker, since the opportunity cost will have been different for men and women. No compensating differential will be needed, and a female workforce will be preferable if it is available and there are no institutional constraints towards hiring it. A flexible labour commitment could generate collusion between employers’ and workers’ needs, especially among a female workforce of married women who needed a flexible use of time to cope with household duties. During the management of the Compañía Arrendataria de Tabacos (or CAT) seventy-four percent of cigarreras were married women, which makes sense taking into consideration the high unemployment rates among working-class males and the secure incomes received by the cigarreras working for the Tobacco monopoly.17 In fact, there were no differences

Gálvez-Muñoz

205

in the working life of married and single women: forty-five years for married and forty-seven years for single women. The long average working life of the cigarreras during the CAT management is related to their entry age and to the absence of retirement, until it was introduced in the 1940s, which ensured they kept going to the factory – though employed in auxiliary tasks – until, in most cases, the onset of terminal illness or death.18 In addition to this, the hiring of a female workforce could compensate for any increases in the cost of labour implied by piecework because it was acceptable to pay lower wages to women, as the latter were not considered to be the family’s breadwinners. Obviously, this is related to the gendered division of labour operating in the market and the necessity of including non-market work, i.e. household or other work developed from within the family, in order to analyse the different behaviour and the different considerations of men and women in labour markets.19 The prioritisation of men as breadwinners, the flexibility needed by women in order to make compatible ‘their’ domestic duties with their market commitments and the institutional barriers built by labour laws or unions have been essential in the different considerations that men and women have historically been given by employers. Only if the family is positioned at the same level as the market can the process of engendering the experience of wages make sense. This chapter concentrates on the Spanish tobacco industry from the establishment of the fiscal monopoly in the seventeenth century to the mechanisation of production in the 1930s. It will focus on two periods. The first is the earlier nineteenth century, when changes in consumption patterns brought about important changes in the production process that ended up with the substitution of a male by a female workforce. The second period will concentrate mainly on the formative decades at the end of the nineteenth century when significant changes in the piecework system and mechanisation took place.

The Spanish fiscal tobacco monopoly Tobacco consumption appeared in Western civilisation at the time of the colonisation of America, but was already strongly present in the Amerindian cultures. From an early stage governments realised the possibilities of such a product and started to tax it or to control its trade, production and consumption under a monopolistic regime. From the beginning of seventeenth century all Spanish tobacco production was concentrated in Seville under a centralised manufacturing system.20 In 1637, the State created a fiscal monopoly, being the first nation to do so.21 Initially run by individual entrepreneurs, in the eighteenth century the Treasury took direct control over the monopoly’s management, and in 1887 the state decided to lease the management of the tobacco monopoly to a private company, Compañía Arrendataria de Tabacos (CAT).

206

Engendering the experience of wages

During the seventeenth and the first half of the eighteenth centuries Spaniards consumed tobacco predominantly as snuff, made in its various forms in the Seville factory (exquisito, fino, palillos, grosso, negro, cucarachero); it was also imported from Cuba, in which case it was known as Polvomonte of the Indies.22 The number of workers at the Seville factory increased from around fifty in 1647 to some 1,200 a century later. In this period, all the factory workers were men, a requirement for snuff production since physical strength was essential for moving the millstones and the tobacco bundles. Although representing a small proportion of production, cigars were produced in the factory from the seventeenth century, though mainly in Cádiz, where a subsidiary factory of Seville was established. In Seville, cigars were made by men in order to avoid mixing the sexes in the same building, in Cádiz the work was done by women. In both places, the cigarmakers were remunerated under a piecework system that contrasted with snuff tobacco workers, who were paid fixed daily wages. While cigar production was entirely manual, snuff tobacco production underwent continuous technical improvements.23 During the second half of the eighteenth century tobacco consumption shifted from snuff to smoking. In 1740, snuff constituted 67.4 percent of the consumption provided by the monopoly, whilst consumption of leaf, cigars and branch represented 32.6 percent. By 1798, however, these proportions had been reversed, with snuff having fallen to 37.8 percent whilst leaf, cigar and branch consumption had risen to 62.2 percent.24 For Rodríguez Gordillo the transition had already occurred by 1780, when the consumption of cigars came to exceed that of snuff. The increase in cigar consumption was possible thanks to an increase in production derived from a directive of 1761, which regulated previous advances in production methods and improved administrative mechanisms for ensuring that the ever more numerous workforce carried out its operational duties. But this directive also limited the number of cigarmakers in the factory of Seville to five hundred. With the increasing consumption of cigars, this number was insufficient. Rodríguez Gordillo has interpreted this measure as a precaution by the Treasury to avoid social unrest in the factory. Since national production was dependent on that factory and its subsidiary in Cádiz, a high concentration of workers under the same roof was avoided.25 However, Gálvez, although accepting the avoidance of social unrest as an explanatory variable, has inserted that measure into a more global strategy related to the opening of new factories for smoking tobacco during subsequent decades and with the shift from a male to a female workforce.26 The substitution of male by female cigarmakers is related to changes in consumption as well as to the fiscal character of the Spanish tobacco industry. The ownership and market structure of a company are of course vital factors in the determination of its strategy. In this case, the monopoly situation, the double role of the tobacco monopoly – to collect taxes and to provide tobacco – as well as its public responsibility are essential in order to understand the strategies pursued by the Treasury and afterwards by

Gálvez-Muñoz

207

CAT.27 Snuff tobacco, the manufacture of which was technologically complex, was increasingly pushed out by a product, the cigar – but soon also the cigarette – that was completely hand made. While in countries where the market was not organised under a fiscal monopoly regime the production of cigars was carried out in small workshops, in monopoly countries production was centralised. Being a fully manual task the advantages of centralised production were not evident, while the disadvantages in terms of flexibility and production costs were more obvious. In fact, the Spanish tobacco industry adds complexity to the old debate about the triumph of the factory system and the spread of the piece-rate system within the factory. For Rule, discussing the English cotton mills during the earlier nineteenth century, piece rates and time discipline combined to intensify labour. In 1833, around half of all cotton mill employees were on piece rates. In the second quarter of the nineteenth century the discovery of the benefits of payment by results were propagated so widely as to be regarded as an innovation of major significance, evidently ignoring the fact that payment by results had inevitably been the common system of payment under the domestic system. What was significant, however, was the deliberate adoption of piece rate systems in the factory where the possibility of checking in and out would seem to make time wages more logical. However if the origins of the triumph of the factory system are moved beyond motivations of technology and labour control, it could nevertheless be the case that a pure piecework system could appear within a factory as more ‘logical’ – despite the difficulties of using that word in historical research – than remuneration based on time.28 In line with the heterogeneity highlighted by authors such as Berg, Williamson, Clark and Jones who considered it an exaggeration to present control and efficiency as opposite alternatives, this analysis shows that other factors exogenous to the production system could explain the implementation of the factory system.29 The Spanish tobacco industry adds complexity to that debate since the main variable in explaining the adoption of the factory system was external to changes in the production system. The tobacco fiscal monopoly had a double role: to provide products, and to collect taxes. Centralisation of production was the best way to control a tax. It has been argued that the main aim of centralised tobacco production was the control of quality, as was the case with other eighteenth-century manufactures dependent upon the Crown. In the case of tobacco the aim was mainly to avoid smuggling. In fact, the only trace of industrial discipline that was maintained in the Spanish tobacco factories was the register. All workers entering and leaving the factories were registered. Apart from that, the work was done in groups and done manually, while the workers needed to bring their tools with them. Women could bring their babies to the factories, where they stayed in cradles provided by the management while a flexible use of time – especially entry time – was pursued. Since the Treasury needed production to be centralised for fiscal reasons, the production system needed to be as efficient as possible within

208

Engendering the experience of wages

the bounds set by centralisation. The capital/work ratio, which had applied in the case of snuff workshops that were more mechanised, thus decreased with the change in consumption towards smoking tobacco. Throughout the seventeenth and eighteenth centuries the Seville factory produced snuff, the manufacture of which permitted the use of different types of mills, thereby producing a capital-intensive production system. In the nineteenth century, on the other hand, the factories produced cigars and cigarettes, products which, given the prevailing levels of technology, required labour-intensive production methods. These products could not be manufactured mechanically until the 1880s because, with the exception of the shredder, suitable machines were not available on the market.30 It is in this context that it is necessary to understand the limits in the number of male cigar makers in the factory of Seville that were laid down by the 1761 directive. A more labour-intensive mode of production would naturally be cheaper with a female labour force. The flexibility of the piecework system suited both the cigarreras and the factory management, since it was easy to measure output, and when output can be measured, payment by result – when possible – remains the most direct option for linking effort to remuneration.31 The opening of new factories with female cigar makers and the substitution of male cigar makers by female cigar makers in the Seville factory are the outcome of that strategy. It was only in Seville that the drama of this was realised, since Seville was the only factory where men had manufactured cigars during the eighteenth century. In fact, despite the Treasury’s preference for female cigarette makers – manifest in the labour force of the new factories – it was difficult to lay off staff in Seville because it was a royal factory dependent upon the Treasury as well as being the largest industrial employer in the town. The reorganisation of the Seville factory did not take place until the mass dismissals during the shut down of the factory because of the Napoleonic Wars in 1808, which affected as many as seven hundred male cigarreros. When the Seville factory reopened in December 1812 the first female workers entered the factory. Some male workers were re-employed in order to teach them the trade and to reduce clandestine cigarette production. So during almost two decades, a mixed population of cigar makers worked in the tobacco factory of Seville. But when the triumph of smoking tobacco over snuff tobacco had become clear and the new factories in other Spanish cities were fully operational, a large number of men were laid off in 1830, provoking a revolt. The Treasury justified the layoff by appealing to the quality of the handmade cigars and cigarettes and arguing that the products made by male workers were of lower quality than those made by women. The cigarerros argued that they received worse raw material than the women and that, because they had to sustain their families, they worked more hastily, the piecework system forcing them to produce a large number of cigarettes in order to generate a daily wage sufficient for this purpose.32 Excessive speed of production has always been considered to be one of the main problems of piece-rate systems, especially in capital-intensive sectors where this can result in

Gálvez-Muñoz

209

damaged machinery. Therefore in a capital-intensive environment piece rates are likely to give the wrong incentives.33 Men were only re-admitted if they agreed to earn the same piece rates as women, an insult almost guaranteed to produce a revolt, and which shows how important it was to keep down labour costs in such a labour-intensive production system.34 From this period onwards the number of male workers began to fall, since no other cigarrero was admitted in any Spanish tobacco factory, whilst the number of female workers continued to increase, reaching a figure in excess of 6,500 in 1887, just in Seville.

The gendered division of labour and pay systems When the cigarreras entered into the tobacco factories in the nineteenth century, a gendered division of labour that implied different types of work and different remuneration systems was established. The gendered division of labour was well defined in the tobacco factories. At the end of the nineteenth century, tabaqueros – male workers – were in charge of the supervision of the factory building, the services inside the plant, the maintenance of engines and machinery and the repair and maintenance of the factories themselves. Work in the production workshops was only carried out by cigarreras. There were different types of workshops related to the different products, or to the different stages in the production chain. The cigarreras all had skills related to specific stages of the manufacturing process that were also linked with their different abilities and ages. However, the organisation of production was the same in all workshops. All the cigarreras, except for the porteras and maestras, worked in groups and by the piece, which gave them a great deal of freedom in their use of time, certainly much more so than the tabaqueros who had individualised jobs and were paid a fixed daily wage. Between six to ten cigarreras worked around a table called rancho, directed by an ama de rancho who was charged with the distribution of tobacco, tasks and wages, and the maintenance of order and cleanliness on the rancho. The ama de rancho paid each worker, depending on individual output. A number of ranchos made a taller, a workshop, directed by the maestra who was the major authority in the workshop, having sole responsibility for the supervision of production and discipline within the workshop, on which she had to report to the head of the factory. The oldest maestras acted as porteras, stationed at the entrance of the workshop to register all comings and goings and to prevent the theft of tobacco by inspecting the workers at the factory gates. The gendered division of labour and the differences in the remuneration system were maintained during the mechanisation process, although some changes were introduced even before the arrival of machines: for instance, a technical division of labour began to be implemented inside the workshops. The latter was one of the modernising goals of CAT, realised with the increased mechanisation of work. Until then, the only

210

Engendering the experience of wages

specialisation that had existed had been between different workshops. The best workers concentrated on the manufacture of cigars, especially those of a superior quality. At the other end of the scale were the stemmers, performing the least skilled tasks. The oldest cigarreras were sent to these workshops, giving them the opportunity of maintaining some income, a pattern that continued throughout the period studied here. The ageing workforce was tolerated by the cigarreras because this was a family occupation, with only the daughters of the cigarreras being given a job.35 An important consequence of the mechanisation of the production process for the gendered division of labour during the earlier twentieth century was a dramatic change in the composition of the labour force with respect to gender. For instance, in the Tobacco Factory of Seville; in 1896 the tobacco factory employed 5,238 cigarreras and ninety-nine tabaqueros, but by 1944 the number of cigarerras had dropped to 365 while the number of tabaqueros had more than doubled to 191. This was caused by the cessation of new admissions of cigarreras except on two occasions: the establishment of semi-mechanical cigar workshops (1909–1912) and the mechanisation of the cigarette workshops (mainly during the first half of the 1920s). Cigarreras continued to be employed in production workshops and tabaqueros continued to be used in everything related to the maintenance of machines and the factory itself. The differences between cigarreras and tabaqueros were also evident in their different remuneration systems. Within the techno-labour system of the factory, male workers were paid by a fixed wage, while the female workers were paid by piece rates.36 The exceptions were relatively unimportant.37 With such a concentration of workers doing a manual task, a female working population under a piecework system was greatly preferable to a less flexible, theoretically family-breadwinning and generally more unionised male one. In Spain, workers were kept together in factories to enable the monopoly more easily to control smuggling, robbery and fraud. Since for fiscal reasons production needed to be concentrated, the female workforce under a piecework system combated the disadvantages of a centralised production system in terms of lower production costs and greater flexibility, as was the norm in small workshops. The Treasury increased the number of factories in order to meet growing demand and the number of workers grew because it had no choice but to do so, given the lack of technological progress in the production of smoking tobacco until the end of the nineteenth century. The Treasury invested in machinery wherever possible, but the lack of public funds delayed the mechanisation process, as well as the subsequent maintenance and replacement of the machines.38 Although various cigarette rolling machines were invented, it was not until 1881 that the Bonsack system was patented. The efficiency of this system forced the world tobacco industry to mechanise cigarette production. Adopting the mechanical production system implied important financial resources, more commercial management, and a high political risk since the cigarreras numbered over 30,000. But the Treasury did not have a

Gálvez-Muñoz

211

choice, as the disparity between demand for cigarettes and production by the monopoly had been increasing from the beginning of the 1870s, as indicated by the rise in contraband activity.39 The large investments that were required were one reason for seeking sources of finance other than the State budget, but as García de Torres had remarked in 1884, what was considered more difficult than finance, were the political problems involved in re-equipping the factories. In fact, the major problems lay more in the difficulties of managing ten factories with a large number of workers and in the relations with the Treasury, than in the lack of finance. Nevertheless, one of the main objectives of the lease was to increase tobacco revenues for a Treasury in continuous deficit. The other main objective was related with that increase, since it was expected that the Lease Company would modernise the industry in order to compete more effectively against smuggling, the only real competing force within a monopoly.

CAT’s management: continuity and change in the piecework system By 1887, the starting date of CAT’s management, the tobacco industry turned over to the Treasury twelve percent of state revenues,40 and the employees in this industry, more than 30,000 in number, represented three percent of the Spanish industrial population.41 There were ten factories in Spain then, most of them created during the nineteenth century as a result of expansion and the changes in consumption patterns. The Seville factory, the oldest and most important factory within the state tobacco monopoly, had around 7,000 workers. However by this time the cost of sustaining the growth of production had become very high. The Treasury was unable to invest in modernising the factories because the State was interested only in fiscal receipts and in avoiding the protests that modernisation would cause. Instead, it decided to lease the management of the tobacco monopoly to a private company. Although the changes occurring in the international tobacco industry after the invention of the Bonsack machine were important in this decision, the main reason for the lease of the tobacco monopoly remains a fiscal one. The public management of the monopoly did not maximise profits, and it had created an unbalanced relationship between the demand and supply of tobacco. The Spanish Liberal government therefore announced a competition for the lease of the management to a private company. CAT was created to apply for this lease, its main shareholder being the Bank of Spain, which was then a private bank, although with a monopoly on the issue of money from 1874.42 CAT obtained the contract in 1887, with several renewals (1892, 1896, 1900, 1909 and 1921), until its replacement by the Tabacalera S.A. in 1945, a company created by CAT shareholders. Private management of the tobacco monopoly was indeed more efficient than direct administration, making a greater profit from the monopoly.

212

Engendering the experience of wages

The previous public management had been impeded by an entrenched bureaucracy and by the Spanish Treasury’s chronic lack of capital, so it was expected that the new company would undertake the capital investment necessary for large-scale production. However, this was not a straightforward task. The legacy of State management had created a disequilibrium between the demand and supply of tobacco, and it had also left a fully manual productive structure dependent upon a skilled female labour force. Since the tobacco monopoly was not abolished, the Treasury ministers sought to increase its profits. They were aware of the factories’ deficiencies and they also had the necessary strategies to improve production efficiency and increase profits from the tobacco industry. Practical application of the improvement measures was complicated, however, because of budgetary restrictions and the inflexible administrative structure of the tobacco monopoly. The large number of interested parties generated by the monopoly, from leaf suppliers to workers and tobacconists, no doubt also placed numerous obstacles in the way of reform. Local powers in the cities in which the factories were located also hindered reform. The response of the company was a slow adaptation of its production and marketing to mechanised production techniques. As a consequence, the company needed to maintain a flexible mode of production and a flexible labour force, but since it continued to be a fiscal monopoly, production needed to be concentrated in its factories. However, the technological innovations available from the 1880s implied advantages in economies of scale that harmonised with a centralised system of production. Making radical changes to an existing techno-labour system is of course very difficult and would not normally be pursued, especially if a company is enjoying a monopoly situation. In such a case, the institutional constraints derived from managing a fiscal monopoly with public responsibility need to be considered. Lipartito has found similar strategic behaviour in the shift from educated ladies to mechanical switching devices in the Bell Telephone Company, a sprawling monopoly composed of dozens of regional firms that dominated telephony in the US from 1880 to 1894. For Lipartito, locating technology in its various contexts makes change far more problematic and far less deterministic than it is in either Marxist or neoclassical economic models. Managers and engineers generally lack total control over innovation. They proceed with imperfect knowledge, and they concentrate resources on making incremental improvements to existing technology. They build strong systems that follow historically determined patterns of change derived from previous events and choices. Often, the interests of workers, consumers and politicians coalesce around these evolving systems. Such vested interests reinforce the tendency of technology to follow its existing trajectory. Subject to these powerful forces for inertia, techno-labour systems respond slowly to change, as was indeed the case with CAT.

Gálvez-Muñoz

213

CAT’s constraints on rapid mechanisation CAT’s management was seriously constrained by institutional, technical and organisational factors inherited from the past, or imposed in the different leasing contracts, or by the Reales Ordenes dictated by the Treasury from 1887. It could not act completely as a private company, even if legally it was one, because of the restrictions imposed by the Treasury (the titular owner of the monopoly). In addition the fact that the main shareholder was the Bank of Spain made the commercial management of the company more difficult, since the latter was prohibited from diversifying and bargaining with other banks over increasing its capital. Furthermore, it was forbidden to modify its statutes. In fact, neither the Treasury nor the Bank of Spain trusted the board of CAT, and took steps to limit its autonomy in decision making. The first CAT managers wished to introduce modern business practices in order to control and coordinate the different parts of the business. These ranged from the acquisition of raw materials to retailing, but from the outset they encountered a great number of restrictions on their freedom to do this, since the lease contract established rigid norms limiting the new company’s mercantile activities. Furthermore, the managers of CAT did not help matters by failing to take steps toward diluting their heavy and rigid bureaucratic inheritance. In fact, they increased it. They were particularly anxious to control every single decision. To some extent this was forced on them, as every single item of capital expenditure, even those involving only small sums, had to be approved by the Board and subsequently by the Treasury. This extreme centralisation was in part determined by the nature of the contractual relationship between the two bodies; depending on the nature of the expenditure, this was charged to the company or to the Treasury, and thus both the bodies needed to be in constant contact. In addition to centralising virtually all decision making, the new management wished to control the power and autonomy hitherto enjoyed by the factories so as to introduce economies of scale as well as to obtain better quality and greater uniformity in production. The underlying objective was to win the market from the smugglers (which the public management had clearly failed to do) and, in doing so, to increase profit. They therefore established a strong control structure in the company in order to improve their position when introducing reforms, but of course this entailed limiting the company’s flexibility and increasing its bureaucracy. One area in which the Board of the company tried hard to increase its control was in the factories’ workshops. In order to increase control over the workers, new rules were designed, and an inspection service was created to monitor this implementation. However, the new rules were not always followed, especially by the cigarreras, and CAT tolerated this. In the first place, these women – skilled workers under a piecework system that was still basically manual – controlled the production process, and secondly, their large number put them in a position to make a public issue

214

Engendering the experience of wages

out of any conflict generated by CAT’s attempts to change production methods. The management’s aim was to intensify labour effort, so combining piecework with time control. While CAT did find a piecework remuneration system already in operation when it took over the management of the tobacco monopoly, it was far from controlling the cigarreras’ working time. The cigarreras based their commitment to the factory and to their household on a flexible use of time and they were still too powerful to be made to surrender that flexibility. In practice, every facet of CAT corporate behaviour was regulated through contracts signed with the State; the company faced controls and limitations established by the Government, especially the Treasury, on all its activities. Yet no issues were as dynamic and as prone to conflict as those connected with labour relations. Because of the important socioeconomic consequences and the effects on company management, this kind of intervention had been subject to conflicts: it thus had to be more explicit. The matter was, in fact, highly complex. Industrial relations between CAT and its workers were negotiated in many different locations: the Treasury and ‘Government’ ministries; the office of the President of the Bank of Spain; the President of CAT’s office; the office of the Government representative in the provinces in which the factories were located; and finally the offices of the factories’ directors, as well as the workshops – more so than in the union’s headquarters. The cigarreras’ opposition to CAT was not predominantly over the new working discipline, but rather over the modernisation plans affecting the production process. The mechanisation of the workshops implied a change in labour routines, and the loss of control over the production process that had previously been exercised by the workers when production had been fully manual. For the cigarreras this loss of control could imply the loss of their labour flexibility that was the basis of their commitment to the factory, especially because most of them – an average of seventy-four percent in the factory of Seville during the CAT management – were married women who had to combine factory work with their household tasks. Indeed, the modernisation CAT wanted to implement implied standardisation, mechanisation, and a different use of time (punctuality and assistance) which were all directly at odds with the cigarreras’ traditional practices. These practices could be considered typical of manual workers under a system of piecework remuneration, though Clark has found evidence of workers under a piecework system who were in fact subject to a strong discipline – such as dismissal or fines – with regard to their use of working time.43 To reform the workshops and to increase productivity, the company needed to economise on inputs. In order to attain this two steps were necessary: first to rationalise production through the division of labour, and secondly, to avoid labour unrest. This last step was necessary for a private company managing a public monopoly with complex agency problems, added to constant public and social scrutiny. Taking into consideration the cigarreras’ power inside and outside the factories, CAT

Gálvez-Muñoz

215

bargained for this collaboration through incentives rather than through repression. These incentives were aimed at increasing productivity and improving quality. CAT’s choice of strategy is related to the company’s public responsibility and the socio-political impact on public opinion that might be caused by labour conflicts in tobacco factories. If social order were maintained, the government would not be called upon to reconsider the private management of the tobacco monopoly. These operational constraints forced CAT to look at original means of implementing the reforms needed to modernise the factories, whilst avoiding confrontation with the workforce. An essential link between the rationalisation of the production process and CAT labour policy was the need to reduce the size of the workforce in order to replace workers by machines. When CAT began to manage the monopoly it became the biggest industrial employer in the country. In July 1887 there were 31,834 cigarreras while the company’s management considered that only 26,512 were needed in order to supply the market and to prepare the workshops for their mechanisation. However, the cigarreras did not only play an important socio-economic role, they also played a political role at the central and especially at the local level. They could do this partly because of their number, and partly because the profession of cigarrera was part of the family economy in which these women were the breadwinners for family networks.44 On the other hand they were skilled manual workers remunerated through a piecework system and, consequently, it was difficult to ‘recycle’ these women for mechanical work. The government was already aware that labour reform was needed in the factories, and in the 1887 leasing contract it permitted the company to reduce the labour force by twenty-five percent. Even though CAT did not make use of this opportunity, there was unrest among the cigarreras. Usually, as with the case of new rules that had not been expected by the cigarreras, this unrest was treated with indulgence because the company’s main concern in these early years was to balance production and consumption. At that time, this implied an increase in production, especially of cigarettes. The cigarreras’ unrest tended to be expressed in brutality done to the workshops’ material and to the managers, and tended to have very damaging consequences on CAT’s plans for reaching a balance in the market. Normally, these revolts provoked a fall in production with a subsequent supply shortage, and sometimes (as happened with the conflicts and the closure of the Cádiz and Alicante factories in 1888 to 1889) a decrease in the output of products most in demand among the consumers. In order to maintain discipline the factories were closed during these conflicts, but they normally reopened with the mediation of the Government’s provincial delegate, demonstrating the public and socio-political dimension of the tobacco-monopoly management. In any case it was clear that the menace of social unrest weighed heavily on CAT’s labour-management decisions. The factory lockouts were normally short and the reopening usually involved the mediation of local authorities and sometimes even the Spanish

216

Engendering the experience of wages

Crown. Since the cigarreras’ job was at the core of large family networks, their demonstrations inevitably mobilised far more people than the actual factory workers. This workforce was kept because the flexibility provided by the cigarreras was convenient for a slow modernisation process and because of the socio-political weight of the cigarreras in some local economies. In addition, both the company and the cigarreras were monopolistic. The cigarreras could not use their skills outside the tobacco monopoly – unless doing so illegally – but on the contrary, they controlled the apprenticeship system in the factories and could intervene in public opinion, particularly on a local basis, especially because the working alternatives were few and they were the breadwinners in a high proportion of families.45 The legacy of a weak relationship between production and market demand, inherited from the State management, obstructed CAT’s reform plans as much as did the cigarreras’ opposition and the government’s constraints. In fact, the imbalance between tobacco-monopoly production and consumers’ needs not only explained the company’s indulgence towards labour conflicts and the workers’ lack of commitment to the new rules, but it also explained why the permitted reduction of twenty-five percent of the labour force was not followed. In order to increase production CAT needed more hands and some new cigarreras were even admitted in the first years. This issue has to be related to some particularities of the Spanish tobacco market. Many of the factories’ products were not demanded by the customers and were liable to be spoilt by lengthy storage, while cheap cigarettes, that were in demand, were underproduced and consequently often smuggled into the country. In addition, the market value of these products differed from their cost of production. CAT pursued the reduction of production costs mainly through standardisation and an increase in quality in order to compete with smuggling, to increase profits and to balance the supply and demand of tobacco. The mismatch was so important that the main goal of CAT was to eliminate it. CAT tried, by both mechanical and manual means, to produce new goods, that were better targeted at consumers’ tastes and priced more cheaply. In a market where consumers’ habits were based fundamentally on custom, changes had to be introduced slowly and with caution; in fact, only some of the new CAT products were accepted by the market. In addition, there were some particular characteristics of consumption in Spain that were probably very influential in explaining the slow path of mechanisation. First of all, the demand for ground tobacco was low, and ground tobacco was the easier product to mechanise. This was due partly to the disequilibrium of the production during the State’s management which priced ground tobacco and cigarettes equally. Furthermore, squared ground tobacco (picadura al cuadrado) was preferred to picadura de hebra, which was consumed in other countries. As a result, international manufacturers built machines designed for the hebra and not for producing picadura al cuadrado, leading to the use of indigenous machinery which proved to be quite solid, but not very fast or efficient.

Gálvez-Muñoz

217

The imbalance between tobacco-monopoly production and consumers’ needs not only explained the indulgence developed by the company towards labour conflicts and its workers’ lack of commitment to the new rules, but it also helps to explain the slow path followed by the company in reducing personnel and introducing machines. Secondly, Spanish consumers were reluctant to purchase mechanically rolled cigarettes. At the end of the nineteenth century, cigarette consumption in the United States was only two percent of total tobacco consumption, almost three percent in France, eleven percent in Italy and thirty-one percent in Spain.46 French consumption patterns – mainly ground tobacco – and the French government intention to mechanise the tobacco factories using French machinery and French engineers ended up with an early transition to mechanised production methods at the end of nineteenth century. However, the piece-rate system was kept, although piece rates were lower than before since productivity increased with the use of machines.47 In the Italian monopoly, female tobacco workers were also under a piecework system that was kept longer than in other places in part because of consumption patterns. At the beginning of the twentieth century the Italian market saw the expansion of cigar consumption that were mainly handmade. That also explained why in Spain the size of the female workforce decreased, while it increased in Italy.48 In the United States, salaries were on a time basis, and wages were fixed per hour.49 The consumption of factory-rolled cigarettes in Spain was widespread before the arrival of the mechanically rolled cigarette.50 However, in other countries the cigarette was a new product, while in Spain it had to fight against the old established handrolled cigarette.51 The Spanish consumer did not easily accept mechanically rolled cigarettes, and the quality of manually-produced cigars, in particular, far surpassed that of the mechanised ones.52 However, thanks to the success of CAT’s efforts to balance the market by multiplying the products demanded by consumers, and by rationalising the workshops with the division of labour, production was balanced with consumption as early as February 1891. Two factories were opened, one in Logroño and another in Valencia, where the mechanical workshops were concentrated. It was only then that CAT decided not to reduce the labour force abruptly by twenty-five percent, but rather to stop hiring new workers. They were able to take advantage of the flexibility provided by the piecework system and by the female workers. This workforce was now reduced, but more slowly and without mass dismissals, which helped the company to avoid any questioning of its management of the tobacco monopoly. Indeed, CAT was to hire new cigarreras only twenty years later to fill the new mechanical workshops. An abrupt and drastic reduction of personnel would have exposed the local labour markets and would have been a measure against the State, even though a drastic replacement of personnel by machines would have furthered the objective of the lease (i.e., to increase profits and tobacco revenues). CAT continued to manage the monopoly because the constraints its management faced retarded the

218

Engendering the experience of wages

reforms but it did not altogether stop them; the management of the tobacco monopoly remained profitable for the company and its shareholders, while it found an alternative way to fully mechanise the production process. The three constraints (institutional, labour and market) interacted to slow down the modernisation of the tobacco monopoly.

CAT’s responses and the evolution of the piecework system Alongside these three constraint there are at least three other reasons that explain why CAT did not pursue radical mechanisation, and why the management chose not to reduce the workforce by twenty-five percent. The first one is the monopoly. It is only under a monopoly regime that a smooth transition makes sense.53 Furthermore, a total mechanisation of the factories would have required a large capital. Thirdly, the industry’s size made it even less possible to invest in technology on a massive scale. CAT did not reduce the labour force immediately, though the maintenance of the workforce cannot anymore be considered as charity.54 Using the flexibility provided by a female labour force, CAT reduced its labour force slowly, avoiding social disorder, and using piecework to promote its workers’ attendance or absence depending on the production needs of the moment. This alternative strategy showed how the company used the characteristics of a female workforce to enable it to survive the different constraints. That the company sought to use the cigarreras’ flexibility to further its own objectives can be demonstrated by comparing two very different periods – the later 1880s and the First World War. During CAT’s early years the company allowed the cigarreras to be absent: absenteeism helped to reduce the workforce and to rationalise the workshops without provoking social unrest (even if non-attendance for eight consecutive days was a formal reason to be fired). In fact, this rule was never applied as CAT’s daily record of the cigarreras’ attendance to the factories demonstrates – this attendance log was not in place before the new management arrived.55 Flexible production, in order to settle the mismatch between supply and demand inherited from the State management, required a flexible labour force. Therefore, during the first decades of CAT management, before the introduction of semi-mechanical and mechanical workshops, the firm implemented a labour-hoarding strategy.56 The maintenance of more cigarreras than were required, helped to lower the hiring and firing costs. Due to the cigarreras’ culture of absenteeism, or at least of irregular work, it was necessary to have fifteen to twenty percent more workers in the workshops in order to maintain production. Piecework, the specific qualifications of the cigarreras and the existence of highly substitutable workshops eased internal labour mobility, mitigating the effects of absenteeism on production. This system also permitted the coexistence of absenteeism and mechanical workshops where the investment in technology required a continuous flow of personnel. Therefore, this firm’s inaction constituted a conscious policy of avoiding social protest by transferring all

Gálvez-Muñoz

219

responsibility to the worker. This gradual transition to an industrial organisation of production would have been impossible with a male labour force that was not flexible and that was committed to labour unions, which at that time were fighting to avoid piecework and to gain a minimum wage. On the other hand, under different conditions, CAT also took advantage of the cigarreras’ flexibility. During the First World War CAT adopted a completely different policy. By that time the policy of natural wastage had led to a large reduction in the workforce. In addition, most of the tobacco machinery used in Spain was imported, and these imports collapsed during the war. Furthermore, CAT also faced problems with the supply of tobacco. The result was insufficient production that could only be increased by employing the workforce full time. As a consequence, in 1917 the company promoted the cigarreras’ attendance to the factories by introducing changes to the piecework system. The end of a pure piecework system in the Spanish tobacco industry in order to promote workers’ attendance could be compared with the measures introduced by Ford to avoid turnover among his workers. Although the piecework system continued to be the norm under Taylorism, one of its negative effects for the employers was an increasing turnover. In fact, Taylor took the supply of labour for granted, and was mainly interested in increasing the workers’ efficiency. It was Ford who transformed the way of keeping labour.57 An excess or shortage of labour is of course crucial in explaining employers’ choices and general trends.58 It was in a period with clear shortages of production that CAT decided to provide a fixed daily subsidy to those who turned up to work. The beginning of the subsidies for attendance in 1917 introduced a two-part (fixed and variable) model of payment. At the same time, these subsidies acted as an incentive to job attendance, so that the opportunity cost of absenteeism for the household economy was now higher, especially during the war years when Spain suffered high inflation and a subsistence crisis.59 As a matter of fact, the lowest absenteeism rates correspond to this moment. However, the continuous changes at the beginning of the implementation of that mixed model of payment demonstrate the adaptation period. As can be observed in Table 9.1 the fixed amount advanced from being less than twenty percent in 1918 to more than seventy percent by 1933. Table 9.1. The evolution of piecework system Year Before 1917 1917 1918 1919 1930 1933

Daily fixed amount in PTAs

Piece rates on total wages, percentage

0 0.25 0.5 1 2.5 5

100 89 81 69 47 30

Source: Gálvez, Compañía Arrendataria de Tabacos.

220

Engendering the experience of wages

During the entire period under discussion the company consistently sought to maintain the social order, while it used incentives to ensure that the number of cigarreras employed fluctuated in line with market changes and production plans. In fact, in the 1930s the factory again encouraged the cigarreras’ non-attendance by conceding sick leave regardless of whether they were actually ill or not. This time, the reason was mainly related to social needs rather than to the needs of production. The company was expecting the government to pass a compulsory retirement law for tobacco workers in order to get rid of the excess of labour that, for political reasons, could not be fired. In fact, if during the First World War the company suffered a shortage of production and of labour, the introduction of apprenticeships in the 1920s and the mechanisation of the majority of the workshops resulted again in an excess of workers.60 Until this time, the manual workshops with skilled workers had been maintained to allow for any shift in consumer preference toward manually-produced cigars and cigarettes; accordingly, some new workers could be trained in traditional manual techniques. However, the definitive acceptance of machine-made cigars and cigarettes placed the manual workshops in a residual position. The average daily wage of a cigar maker in the 1930s amounted to 7.17 pesetas. Sick leave implied a reduction to 5 pesetas per day. Since workers could not be easily dismissed, absence was a double advantage for management, both in lowering average wages and in controlling excess production of tobacco, which would otherwise have had to be stored and possibly thrown away. Moreover, the remaining cigarreras in the manual workshops were very old (most of them over sixty-five). Doctors gave uncontrolled leave of absence for illness: this gave the cigarreras money and time. In fact, when Nicot, the Seville cigarreras and tabaqueros union, in 1933 denounced the doctors for negligence in granting leave of absence, only two women signed the protest and followed the Union’s guidelines, even though the cigarreras constituted eighty percent of the union’s members. They defended their family economy strategies against union strategies, and the flexible use of their time was essential in order to make both household and factory work compatible. Therefore, in this period, absenteeism could be understood as a sort of part-time job necessary for the running of the household economy, because most of them were married women. At the same time, the company was moving towards a fixed daily wage since production was fully mechanised. The only manual workshops that remained were the ones in which the old manual cigarreras waiting for retirement were concentrated.

Conclusion Contrary to more static models developed within economic theory and management literature, the historiography has shown a more complex picture of the experience of wages. The gendering of this experience

Gálvez-Muñoz

221

increases the diversity since it shows that neither market nor custom is sufficient in itself to explain employers’ choices of a particular system of remuneration. Since the organisation of production was not indifferent to gender, the inclusion of gender as a principal analytical tool helps to explain the diversity of the different forms of remuneration, as well as their evolution. The analysis of piecework at the Spanish tobacco monopoly shows that, in addition to market structure, local labour markets, technology and institutional constraints, gender can be crucial for a full understanding of the reasons for the adoption of a particular system of remuneration. Two main strategies pursued by the Spanish tobacco industry illustrate this. On the one hand, changes in tobacco consumption patterns at the end of the eighteenth and the beginning of the nineteenth centuries brought about important changes in the production processes within the Spanish tobacco monopoly. Their factories moved from being capital intensive to labour intensive. While in other countries, where the market was not organised under a monopoly, production was concentrated in small workshops, in Spain thousands of workers were concentrated in factories producing hand-made cigars and cigarettes. Since the concentration of production had a fiscal origin that was not in principle related to any advantages in the production process, a female workforce under a piecework system brought flexibility to the factories. Piecework was maintained after the privatisation of the tobacco monopoly management to CAT in 1887. At that time, the mechanisation of cigarette production had started across the world and CAT was therefore expected to modernise the Spanish tobacco industry by introducing machines into the factories where over 30,000 workers were employed, thereby converting itself into the largest modern industrial employer in the country. The company faced strong institutional, labour and market constraints that made the industrial transition slow and sensitive. In parallel with the mechanisation of the workshop the piecework system was slowly substituted by a mixed payment system in the interwar decades and later by a fixed wage based on a standard eight-hour working day, maintaining a bonus system as a heritage of the former piecework system. During these decades machines were persistently substituted for labour, but the flexibility derived from a female workforce under a piecework system helped the company to make a politically safe industrial transition as was needed by a company managing a public monopoly, always under tight social and political scrutiny.

Notes 1. M. Huberman, Escape from the market. Negotiating work in Lanchashire (Cambridge, 1996), p. 63. 2. For the different levels at which gender could work as an analytical tool in the social sciences, see J. Scott, ‘Gender: A Useful Category of Historical Analysis’, in J. Scott (ed.), Gender and the Politics of History (New York, 1988); L. Gálvez ‘Género, Historia y Empresa’, Leviatán, 81 (2000), pp. 131–168.

222

Engendering the experience of wages

3. M. Berg, ‘What difference did women’s work make to the Industrial Revolution?’, History Workshop Journal, 35 (1993), pp.22–44; J. Humphries, ‘Lurking in the wings… Women in the Historiography of the Industrial Revolution’, Business and Economic History, 20 (1991) pp.32–44; E. Jordan, ‘The exclusion of women from industry in nineteenth-century Britain’, Comparative Studies in Society and History, 31:2 (1989), pp. 273–96. 4. G. Becker, Economics of Discrimination (Chicago,1971), p. 44 establishes that in competitive markets, firms that do not discriminate have lower costs and tend to expand at the expense of discriminating firms. 5. J. Burnette, ‘An investigation of the female-male wage gap during the industrial revolution in Britain’, Economic History Review, 50:2 (1997), p. 279. 6. Jordan, ‘The exclusion of women’. 7. A. Janssens, ‘The rise and decline of the male breadwinner family? An overview of the debate’, International Journal of Social History, 42 (1997), pp. 1–23; C. Creighton, ‘The Rise of the Male Breadwinner Family: A Reappraisal’, Comparative Studies in Society and History, 38 (1996), pp. 310–337. 8. L. Gálvez, ‘Género y Cambio Tecnológico: Rentabilidad Económica y Rentabilidad Política de la Gestión Privada del Monopolio de Tabacos, 1887–1945’, Revista de Historia Económica, 1 (2000), pp. 11–45, p.17. 9. S. W. Polachek and W. S. Siebert, The Economics of Earnings (Cambridge, 1992), p. 250. 10. From the conventions established in the early days of work study (by Bedaux and his associates) the point of intersection between basic rate and work done is presumed to be that at which in one hour, sixty minutes of work were carried out by a competent and conscientious worker who was not motivated by an incentive (A. M. Bowey, Handbook of Salary and Wage systems (Aldershot, 1982), pp. 280–5.) 11. E. J. Hobsbawm, ‘Custom, wages and work-load in nineteenth-century England’, in A. Briggs and J. Saville (eds), Essays in Labour History in Honour of G.D.H. Cole (London, 1960). 12. J. Rule, The Labouring Classes in Early Industrial England 1750–1850 (London, 1986), p. 119. 13. The combination of that system with the domestic system ended up as the sweating system. The piecework system continued to be the norm under Taylorism, though one of its negative effects for employers was an increasing turnover. Another common type of remuneration was that of surveillance subcontractors. For Rule, The Labouring Classes, p.108, subcontracting was widespread and in many cases, for example in mining, increased rather than decreased in extent over the nineteenth century, replacing earlier forms of direct hiring. 14. It occurred, for instance, in labour markets where the employers sought to reduce the labour turnover of skilled workers as in Lancashire (Huberman, Escape from the market, pp. 151–2). 15. North American male cigar makers were skilled workers; they have been considered by historians as members of the labour aristocracy: P. A. Cooper, Once a cigar maker: Men, Women and Work Culture in American Cigar Factories, 1910–1919 (Urbana, 1987); M. Prus, ‘Mechanization and the gender-based division of labour in the U.S. cigar industry’, Cambridge Journal of Economics, 14 (1990), pp. 63–79. Spanish female cigar makers, performing the same job, have been considered as unskilled workers: C. Arenas, Sevilla y el Estado (1892–1923). Una perspectiva local de la formación del capitalismo en España, PhD Universidad de Sevilla (Sevilla, 1995). For a gender analysis of the concept of labour aristocracy and skill, see L. Gálvez, Compañía Arrendataria de Tabacos. Cambio tecnológico y empleo femenino, 1887–1945 (Madrid, 2000). 16. Huberman Escape from the market, p. 64 has also found collusion on both sides of the market in Lancashire in the first decades of the nineteenth century. Although the form of payment did not cover all possible contingencies, and in this sense the contract remained incomplete, payment by piece persisted in Lancashire because over time it consolidated elements of fairness and it induced co-operation on both sides of the market.

Gálvez-Muñoz

223

17. The cigarreras´ husbands in the city of Seville during the management of the Compañía Arrendataria de Tabaco (or CAT) were artisans and skilled workers (twenty-eight percent), semi-skilled workers and tabaqueros (twenty-five percent) and jornaleros, which means they had an irregular labour market commitment (forty-two percent), and others professions (five percent). 18. With the mechanisation process apprenticeship became less important than before and, as a matter of fact, it was unnecessary to train the cigarreras from an early age. Equally the modernisation of the company included the formalisation of the entry age. As a matter of fact, the average entry age of cigarreras evolved from thirteen years before 1888, to sixteen from 1888 to 1915, to twenty from 1916 to 1926, and to twenty-four from 1927 to 1935. 19. Within English history, see the critiques of labour historians such as E. Hobsbawm and E. P. Thompson, in S. Alexander, A. Davin and E. Hostettler, ‘Labouring Women: a Reply to Eric Hobsbawn’, History Workshop, 8 (1979), p. 175; J. Scott, ‘Women in the Making of the English Working Class’, reprinted in J. Scott (ed.), Gender and the Politics of History (New York, 1988). C. Sarasúa, The rise of the wage worker. Peasant families and the organization of work in Modern Spain, PhD, European University Institute (Florence, 1995), pp. 5–6, has criticised the same attitudes in Spanish historiography, which has tended to concentrate on analysing the existing hierarchy of the production structure of the pre-industrial domestic unit. 20. Concentration of production increased with the building of a new factory, opened in 1758 and especially designed for the production of snuff tobacco. J. M. Rodríguez Gordillo (‘El personal obrero en la Real Fábrica de Tabacos’, in Sevilla y el Tabaco [Sevilla, 1984], pp. 68–75) maintains that this factory represented the largest civil building in Spain and the largest tobacco factory in the world. It brought together onto a single site machine warehouses, stables, production and repair workshops, storerooms and rooms for auxiliary activity. Previously, these had been in scattered locations. However, authors such as J. M. Valdaliso and S. López (Historia Económica de la Empresa [Barcelona, 2000], pp. 87–90) have difficulties in considering the Royal manufacturers as factories since they were not profit centres. 21. Although the first big European state establishing a tobacco fiscal monopoly was Spain in 1637, the first fiscal tobacco monopoly was established in Mantua in 1634. 22. Three main forms of tobacco consumption passed to Europe: chewing, smoking and snuffing, and from Europe they were passed everywhere through colonial expansion (J. Goodman, Tobacco in History: the Cultures of Dependence (London, 1993). 23. Apart from an increase in the number of workers the equipment pool also grew in the period 1647 to 1750: the figure of three mills in the former year rose to one hundred in the latter, and the number of horses rose from five to 257. Animal traction mills replaced manual mortars, whose number fell from twenty-six to ten between 1647 and 1700. There was not only an increase in the quantity of production tools; their technical efficiency also increased as a result of the improvement of the mills, which initially had one stone and later two or even more, moved by an animal. The sieving process evolved in a similar fashion, so that towards the end of the period a single operator could move several cases of sieves; F. Comín and P. Martín-Aceña, Tabacalera y el estanco del tabaco en España 1636–1998 (Madrid, 1999), p. 57. 24. Comín and Martín-Aceña, Tabacalera y el estanco del tabaco, p.62. 25. J. M.Rodríguez Gordillo, ‘El tabaco: del uso medicinal a la industrialización’ in J. Fernández Pérez and I. González Tascón, La Agricultura Viajera. Cultivo y manufacturas de plantas industriales y alimentarias en España y la América virreinal (Madrid,1990). 26. Galvez, Compañía Arrendataria de Tabacos. 27. In early studies, such as J.R.Commons, ‘American Shoemakers, 1648–1826: A Sketch of Industrial Development’, Quartely Journal of Economics (1909) – cited in H. Gospel, ‘Managerial Structures and Strategies: An Introduction’ in H. Gospel and C. Littler (eds) Managerial Strategies and Industrial Relations. An Historical and Comparative Study (Aldershot, 1983), p. 3, the firm’s ownership structure and its market position were

224

28.

29.

30.

31.

32.

33. 34.

35.

36.

Engendering the experience of wages already considered essential when explaining the different production systems and, as a matter of fact, the different remuneration systems chosen by employers. For D. Landes (The Unbound Prometheus. Technological change and Industrial Development in Western Europe from 1750 to the Present (Cambridge, 1969), pp. 56–58), the factory system triumphed because it was the most efficient way to produce with the new technologies. For S. Marglin, (‘What do bosses do? The origins and functions of hierarchy in capitalist production’, Review of Radical Political Economics, 6:2 (1974), pp. 60–112), the answer was not technical efficiency but the accumulation of capital and a better control of the workforce. M. Berg, ‘On the origins of capitalist hierarchy’, in B. Gustafsson (ed.), Power and Economic Institutions, Reinterpretations in Economic History (Aldershot, 1991) pp. 182–3, considered it an exaggeration to present control and efficiency as opposite alternatives. O. Williamson, ‘The organization of work: A comparative institutional assessment’, Journal of Economic Behaviour and Organization, 1 (1980), pp. 5–38, agrees with Marglin that factors other than technological help to explain the triumph of the factory system, but he insists in the superiority in organisation of the factory system and the savings in transaction costs. S. R. H Jones, ‘The organization of work: a historical dimension’, Journal of Economic Behaviour and Organization, 3 (1982), pp. 122–24, admits a saving in transaction costs but insists that this was not the only consideration for employers. Other considerations, such as the cheaper labour force in domestic industry were important. The choice was not free because there is not a unique and perfect labour market. For Jones the advantages of the factory system were mainly related to savings in inputs, especially in the labour force through the use of machines. Until cigar and cigarette production became mechanised at the end of the nineteenth century, the only technical innovations were the use of the leaf (capillo) as an outer layer for cigars, and the mechanisation of shag production. The difficulties in measuring the link between effort and reward appear mainly in three situations: when the job entails team production so that the output of individual workers is not discernible, when there are long delays in observing output, or when the job entails numerous and varied tasks so that creating an output index is difficult. In these instances, because worker output is difficult to measure, employees have an incentive to shirk. Since it is in both the workers’ and the firm’s interest to maximise each worker’s productivity, both parties will benefit from creating an environment with disincentives for shirking. Creating such an environment entails augmenting the human capital model to allow for a lifecycle contract, defining a wage payment scheme between workers and firms. (Polachek and Siebert, The Economics of Earnings, p. 250). For the beginning of the twentieth century L. Gálvez ‘Breadwinning patterns and family exogenous factors: Workers at The Tobacco Factory of Seville during the industrialization process (1887–1945)’, International Review of Social History, 42, supplement (1997), pp. 87–128, has demonstrated that seventy-two percent of female tobacco workers at the factory of Seville were breadwinners. This calculation cannot be carried out for the 1830s. Polachek and Siebert, The Economics of Earnings, p. 257. Gálvez, Compañía Arrendataria de Tabacos. Rule (The Labouring Classes. pp. 117–8) says that nominal rates under the old putting-out system were hardly ever subjected to direct cutting which was to become their continual fate under the ‘sweating’ practices of the ‘new domestic system’ in the second quarter of the nineteenth century. Here cutting could be more extreme because it involved the competitive exploitation of female labour. No gap between man and man was as large as that between men and women. Rule, The Labouring Classes, p. 123, describes how Liverpool shipwrights resisted a piece-rate system which, although it would have allowed larger earnings to the young and fit, would have involved considerable unemployment among the older shipwrights. ‘Techno-labour’ was used for the first time by K. Lipartito, ‘When Women Were Switches: Technology, Work and Gender in the Telephone Industry, 1890–1920’, American Historical Review, 99:4 (1994), pp. 1075–1111. For this author, socially constructed technical systems include a socially constructed labour force, which is why

Gálvez-Muñoz

37.

38. 39. 40.

41. 42.

43. 44. 45. 46.

47.

48. 49.

225

it might be best to speak not of technical systems but of techno-labour systems. Employers use different categories of workers with different experiences and skills to operate different systems. Distinctions and divisions among workers, whether wholly rational or not, tend to become deeply embedded in system structures, gender being an essential component (Lipartito, ‘When Women Were Switches’, pp. 1089–90). These few exceptions do, however, confirm how important the specific task was in order to explain the remuneration system adopted, apparently beyond gender differences. The mechanics, male workers, were under a piecework system since they were with the cigarreras, dealing directly with production output, while the maestras, female workers with supervisory tasks were paid a fixed wage. But in relation to the total workforce these groups were not large. J. García de Torres, Las rentas estancadas. Apuntes históricos, observaciones y datos estadísticos (Madrid, 1884). Comín and Martín-Aceña, Tabacalera y el estanco del tabaco. F. Comín, ‘El sector público’ in A. Carreras (ed.), Estadísticas históricas de España, siglos XIX y XX (Madrid, 1989) p. 404, estimates the percentage that the tobacco monopoly turned over to ordinary treasury incomes was thirteen percent in 1850, twelve percent in 1900, seven percent in 1935 and two percent in 1970). In 1887, CAT had to pay an annual fee of 90 million PTAs. Several changes in the contract established a proportional fee to be paid to the Treasury. During CAT period this varied from ninety to ninetyfive percent of tobacco monopoly incomes. L. Alonso, ‘La modernización de la industria del tabaco en España, 1800–1935’, Programa de Historia Económica, Fundación Empresa Pública (Madrid, 1993) WP 9304. The other CAT shareholders were banks and businessmen already implicated in other tobacco businesses as raw tobacco suppliers, the Marqués del Campo, Banco de Castilla, Banco Hispano Colonial and Crédito Mobiliario Español, as well as the bankers representing the Rothschild interests in Spain (Weisweiller & Bauer, and Urquijo & Co). G. Clarck, ‘Factory Discipline’, Journal of Economic History, 54 (1994), pp. 128–163, p. 132. Gálvez, ‘Breadwinning patterns’. For an optimistic and a pessimistic estimation, see Gálvez, ‘Breadwinning patterns’, pp. 101–113. For Spain and Italy the data correspond to 1894 to 95, for France to 1892 (E. Delgado, Consideraciones sobre la Ley de 30 de agosto de 1896 renovando el contrato a la Compañía Arrendataria (Madrid, 1897), and for the United States to 1900. C. Tate, Cigarette wars. The triumph of the ‘Little White Slaver’ (New York, 1999). When the French production monopoly was mechanised in most products at the end of nineteenth century the piece rate was the standard payment system for all the female employees except the maîtresses (as in the Spanish case). That was also so for the cigar makers with an individualised job, as well as for those working on the cigarette rolling machines, thus working at a constant speed. The striving of the factory to amortise the investment on technology and the surveillance of the quality of the product, could be at the base of such choices (M. H. Zylberberg-Hocquard, ‘Les ouvrières d’État (tabacsallumettes) dans les dernières années du XIXe siècle’, Le Mouvement Social, 105 (1978), pp. 95–6). So the reasons were different in the Spanish case compared to the French, in part because the mechanisation path was different, as was the efficiency of both tobacco industries, the French industry being more efficient than the Spanish. For an international comparison of productivity on tax revenue in the tobacco industry, distinguishing especially between monopoly and non-monopoly countries, see A. Madsen, The State as manufacturer and trader (London, 1916). For the conditions of the Italian tobacco monopoly, see E. Benenatti (ed.), Impresa e Lavoro di Stato: La Manifattura tabacchi fra ottocento e novecento (Turin, 1999). In the studies made by Evans in the 1930s for the US Department of Labor in order to study the impact of mechanisation on the cigar industry, especially among the female workforce, the wage comparison among workers using mechanic and hand methods, and the ones among male and female workers, were always on an annual or hourly basis.

226

50.

51.

52.

53.

54.

54. 56.

57.

Engendering the experience of wages Only stemmers were paid by a piece rate. On the contrary, apprentices were also paid low times rates (M. J. Prus, ‘Mechanisation and the gender-based division’, p. 64). Even if working hours in the non-mechanic workshops as late as 1911 did not have a regular working day, ‘Many cigar factory employees do not work all the hours the factory is opoen and the work afforded’, W.P. Evans, ‘Mechanization and Productivity of labor in the Cigar Manufacturing Industry’, Bureau of Labor Statistics, 660 (1938). However, the cigar production within workshops or factories developed negatively compared to cigarette production that started to manufacture mechanically, which became the norm in the United States. The Chicago Daily News printed at the time of the Spanish-American War and in the context of an anti-cigarette movement that ‘Spain might not have reached its present state of deterioration if it had prohibited cigarette manufacturing before it became a national occupation and misfortune’; quoted in C. Tate, Cigarette wars, p. 20. One of the reasons why Spanish consumers preferred hand-made to mechanically-made cigarettes was the bigger weight of the manually-made product. (Delgado Consideraciones sobre la Ley, p. 188). During CAT’s first few years of running the industry important investment in technology took place only in two newly-established factories, in the old factories this affected only the picaduras’ workshops. CAT decided to mechanise the first phase of the production process. Workshops for cigars and cigarettes, where the largest number of cigarreras were concentrated, did not experience this investment impetus. CAT already knew that it had to be very prudent in introducing machines to the workforce. For example, the Luddist-inspired revolt of 1885 in the factory of Seville was still a very recent experience. The reason for this insurrection was the rumour of the arrival of cigarette machines. Gálvez, Compañía Arrendataria de Tabacos, has noticed that their actual arrival took place more than thirty years later. Independent of the lease of the right to a state monopoly by a private company, the fiscal monopoly involves a distortion in the allocation of resources related to the competitive model. In fact, when the management of a monopoly is run by a private company, the subjection of the company to private right does not imply that the company develops a private activity: the main characteristics of the company activity are derived from the existence of the monopoly, which determines its dominant position in the market. Since the industrial transition at the Spanish tobacco monopoly was very smooth it has been considered in the historiography as an anti-capitalist firm and as a mere administrator of a fiscal resource without any industrial or commercial interests. See Arenas, Sevilla y el Estado. Recent studies from the business point of view by Gálvez, Compañía Arrendataria de Tabacos, and Comín and Martín-Aceña, Tabacalera y el estanco del tabaco have shown that, although strong institutional constraints derived from the public dimension of the fiscal monopoly, it was nevertheless managed as a capitalist firm, at least from the lease of the management to a private company in 1887. The Register had continued to exist from the eighteenth century, but a record of daily attendance was not kept. The concept of labour hoarding comprises the company strategy applied during economic recessions, mostly involving skilled workers with specific skills. For economists the concept has its origin in the variability and the uncertainty of modern economies, and in the concept of work as a quasi-fixed factor of production; that implies that managers have an incentive for considering workers as capital when they have taken on part of their training. The first person to sustain the idea of work as a quasi-fixed factor of production was W. Y. Oi, ‘Labor as a quasi-fixed factor’, Journal of Political Economy, 70 (1962), pp. 538–55. The Marxist concept of a surplus army of labour could not be used in this model. The latter implies the existence of an excess labour population which worked as crowbar for capitalist accumulation (H. Braverman, Labor and Monopoly Capital (New York, 1974). Nevertheless, this concept does see the worker as exploited, while the cigarreras profited in terms of time from this company strategy, which was in itself mostly an adaptation to the earlier situation. P. Collier and R. Horowitz, Los Ford. Una Epopeya Americana (Barcelona, 1990).

Gálvez-Muñoz

227

58. S. Pollard has remarked that the intensification of labour, as well as the demographically-caused increase in the supply of labour, were necessary to explain how the demand for labour in the early industrial revolution was met without an increase in its price (S. Pollard, ‘Labour in Great Britain’, in P. Mathias and M. Postan (eds), The Cambridge Economic History of Europe, Vol.VII. The Industrial Economies: Capital, Labour, and Enterprise (Cambridge, 1978), p. 158. 59. Spanish neutrality during the First World War led to an increase in exports and created high inflation in the internal market that brought increasing social conflict. The tobacco workers also mobilised to ask for an increase in piece rates. Instead of that increase the company introduced subsidies for attendance at the factory in order to solve production shortages. 60. In an inspection of the workshops made in the Tobacco factory of Seville on 18 May 1931, the administration made a list of the cigarreras who stayed in the workshops and the cigarreras who were taking any kind of leave. For example, in the mechanical cigarettes workshops, it was calculated that the number of workers needed for production were fifty-four, but there were sixty-five in the list. The eleven surplus cigarreras smoothed the functioning of the workshop when other cigar makers did not attend at their jobs. If the number of assistants exceeded fifty-four, they were employed in other workshops. (Gálvez, Compañía Arrendataria de Tabacos, pp. 234–238).

CHAPTER 10

AGE, GENDER AND THE WAGE IN BRITAIN, 1830–19301 Paul Johnson

Has the labour market in Britain since early industrialisation been characterised by customary and stable wage differentials, or by flexible wages that have reflected the marginal productivity of the worker and which have readily adjusted to changing supply and demand conditions? A classic article from 1955 by Henry Phelps Brown and Sheila Hopkins argued, on the basis of a long-run survey of building workers’ wages, that the pay differential between skilled and unskilled workers had barely changed over the seven centuries up to 1914. The explanation offered was that ‘differentials were largely ruled by custom’ and it required the exogenous shock of the First World War – conscription, rapid price inflation and the ‘dilution’ of skilled trades with unskilled, often female, labour – to break the entrenched earnings norms within occupations.2 Further research has provided indirect support for the idea of customary wage determination. E. H. Hunt has shown that regional wage differentials for workers in the building trades and in agriculture persisted over a hundred-year period during which the geographical and employment structure of the population changed significantly, as did the sectoral composition of the economy.3 Other historians have looked beyond wage data, and have used information on unemployment, indebtedness and nuptiality to argue that the British labour market in the nineteenth century exhibited pronounced and enduring regional diversity.4 This is taken to imply that institutional rigidities, which may have included customary wage setting, restricted the degree to which competitive pressures could generate flexibility in the labour market. The role of customary practices in determining wages may have been even more important in nineteenth-century Britain for women than for

230

Age, gender and wages

men. There is a large historiography relating to women’s work, which overwhelmingly argues that patriarchal institutions, both familial and commercial, structured the system of workplace authority and reward to diminish the return to female labour.5 Customary expectations served to reduce the pay of women workers below that of their male work mates both directly and indirectly. Direct discrimination took the form of lower pay for equal work, as in the Lancashire cotton industry around 1830 when female mule spinners were employed at half the piecework rates of men.6 Indirect discrimination could take the form of job segregation, as in the Prudential Insurance Company where, from 1871, female clerks were formally excluded from the normal promotion ladder open to their male counterparts.7 More common was an informal barrier to higherstatus positions. For example, in the Courtauld silk mill at Halstead, Essex in the 1860s, all but a handful of female workers were confined to lowergrade winding and weaving tasks. Men, however, dominated overseeing, clerical and maintenance work, where earnings were at least twice the female level.8 The belief that wage-setting practices in Britain were dominated by rigidity and custom throughout the nineteenth century and into the twentieth has not been universally accepted. A number of historical studies based on very different methodologies and theoretical perspectives have argued that customary labour practices were no match for the market pressures of nineteenth-century industrialisation. E. P. Thompson famously proposed that the employment pressures of industrial capitalism crowded out customary attitudes to time and work discipline, and introduced a more regulated and monitored workplace environment.9 He also suggested that by the early nineteenth century an older ‘moral economy’ was being subsumed by capitalist rationalisations, although he was cautious about the extent to which this interpretation could be transferred in its application from the specific locus of food riots to the more general analysis of ‘just prices’ and ‘fair wages’.10 Coming from a very different intellectual position, the quantitative economic historian Jeffrey Williamson has argued vigorously that ‘the wage structure exhibited enormous variability across the two centuries following 1710’, and that it did so largely because of market pressures.11 By extending his wage series beyond just the building trades examined by Phelps Brown and Hopkins, Williamson suggests that manual worker skill differentials changed (declined) markedly between the late eighteenth and early twentieth centuries. He also finds that, for the economy as a whole, the wage premium for skill followed a ‘Kuznets curve’ in which skill differentials rose up to 1871 and declined thereafter, which he interprets as a clear indication of a growing skills shortage through to the mid-Victorian period.12 The evidential underpinning of this claim has been challenged on the grounds that the estimates are based on highly unrepresentative data on the earnings of service sector workers.13 Charles Feinstein has argued that Williamson’s inferences about widening skill differentials among manual workers are incorrect, and has concluded that ‘the general

Johnson

231

picture is one of broad stability, most notably in the ratio of skilled to unskilled pay and in the overall distribution of earnings’.14 Nevertheless, Williamson’s subsequent work on labour migration and urban growth provides, he believes, further evidence that the labour market was highly flexible, at least as far as the mobility and remuneration of younger workers was concerned.15 A similarly neoclassical economic analysis has led Joyce Burnette to argue that the lower pay received by women workers in Britain during the late eighteenth and early nineteenth century had a rational, and marketbased, explanation.16 Although females typically earned between one third and two thirds of the male wage in the same employment, Burnette argues that this wage gap can be largely explained by the lower value of work effort (in terms of time, physical strength and skill) supplied by women. Burnette accepts that other forms of labour market discrimination – such as actions by employers or unions to confine women to a small number of low grade jobs – could and did occur in the less competitive sectors of the nineteenth-century economy, but she concludes that there is little empirical evidence to support the idea that female industrial and agricultural wages were set according to customary rather than market pressures. As this brief discussion of the literature on wage setting and wage flexibility illustrates, existing evidence has not enabled historians to resolve their differences and produce an uncontested account of the evolution of the earnings structure of early industrial Britain. The dispute is essentially empirical. If wage differentials for manual workers – differentials based on occupation, skill and gender – were extremely stable over a century of enormous change in the nature of work, the quality of labour and the structure of the economy, then this would point towards some exogenous structuring of the labour market. We may wish to follow Phelps Brown and apply the term ‘custom’ to this exogenous influence, although this is merely a description, not an explanation, of the phenomenon. And if female wages were consistently lower and employment opportunities consistently more restricted than those for men, then this can be taken as strong evidence of gender structuring of the labour market. If, on the other hand, wage differentials changed markedly over time and exhibited considerable long-run flexibility, this is prima facie evidence that the labour market was responsive to changing supply and demand conditions. It is, of course, possible for supply and demand conditions in the labour market to change over time in such a way that pay differentials remain stable, but this is highly unlikely – it is just one outcome from an infinite set of possibilities. A primary reason for this lack of historical consensus is the paucity, and intractibility, of the underlying data. Historical wages data are seldom sufficiently detailed to sustain clear tests of alternative hypotheses. For example, it is often not known whether jobs with the same titles required identical levels of skill and effort; or whether different rates of pay for workers in the same job reflected real differences in the quantity or quality of the individual’s output. A resolution of the debate must await the

232

Age, gender and wages

collection and analysis of a large amount of individual wage data from the surviving wage books of nineteenth century employers. However, not all the existing wage evidence has received the detailed attention it deserves. This chapter takes a further look at the question of whether stability or flexibility has been the primary characteristic of earnings in Britain in the past two centuries, but instead of examining pay differentials by skill and occupation it examines differentials by age. The relationship between earnings and age has received relatively little attention from historians of nineteenth-century Britain, even though it is another of the potential indicators of long-term wage flexibility or rigidity.17 A primary reason for this is the relative scarcity of age-specific earnings data. Although the British government collected and published a large amount of information on earnings and on wage rates in the nineteenth century, and subsequent compilations were produced by Bowley, Wood and Beveridge, very little of this contains information on age-specific earnings (except for a general distinction between ‘juvenile’ and ‘adult’ wages).18 There are, however, three cross-sectional studies of earnings by age from 1833, the late 1880s, and 1929 to 1931, which can be used to evaluate the stability over time of the relationship between age and earnings, and also the stability of the wage gap between males and females of the same age. The first part of this paper discusses these three collections of earnings data, and identifies problems of comparability and reliability. Part two reviews other studies of earnings by age, and itemises the factors that might have changed the relationship between age and earnings over the hundred years covered by these three sources. The third part presents a descriptive analysis of the data, and the final section offers an interpretation of the stability of ageearnings profiles and male-female wage gaps observed in the data.

Earnings data and sources In order to determine whether the relationship between age and earnings changed over the century from the 1830s, and whether male and female patterns changed in a similar way, it is necessary to examine detailed records of the actual earnings of large numbers of workers. Information on the formal wage rates published by employers in local lists, or negotiated between employers and unions, gives no indication of how the actual earnings received by workers varied with respect to age, or any other characteristic. Modern surveys of earnings find the age effect to be substantial: for example, from the 1970s to the 1990s in Britain, the earnings of male manual workers aged between 18 and 20 were barely sixty percent of the earnings of similar workers aged 30 to 39, while workers aged 60 to 64 received eighty-five percent of the wage for 30 to 39 year olds.19 The sources used in this paper enable us, for the first time, to determine whether such large age effects on earnings are a peculiarity of an advanced industrial economy, or whether similar age structuring can be found in the earnings pattern of early industrial society.

Johnson

233

Data for 1833 were collected by Dr James Mitchell, who was appointed by the Royal Commission on the Employment of Children in Factories to conduct an extensive survey of the conditions of employment of workers in a large number of factories throughout the kingdom.20 Mitchell distributed printed forms to factory employers which required them to state, for each individual worker, their age, sex, type of work, weekly wage, ability to read and write, together with time lost during the preceding year due to sickness and whether they were a depositor to a savings bank or subscriber to a benefit society. Little information was forthcoming on sickness or thrift, but details of earnings were obtained from eighty-three cotton factories, sixty-five wool factories, seventy-seven flax factories, twenty-eight silk factories, and ten lace factories. Additional returns were supplied by seven potteries, a dye house, a glassworks and two paper mills, but the small number of workers covered means that these returns have been excluded from the analysis presented here, which relates to 21,483 male and 30,301 female textile workers.21 Mitchell grouped the returns by place and type of factory to produce aggregate tables which report, for males and females separately, the number employed and the average weekly wage for each year of age, for all ages between the lower and upper age thresholds of six and eightyeight. Mitchell emphasised the reliability of the returns, noting that ‘the principle adopted in the construction of the forms of return from the factories, requiring the particulars of the employed one by one, instead of general results, necessarily secured a degree of accuracy which otherwise could not have been expected’. Moreover, he felt that the factories providing the returns ‘are sufficiently numerous to afford a fair section of the whole, and more particularly so as all of them have a general resemblance, so as to support each other’.22 The type of work undertaken by employees whose wages are reported by Mitchell is not listed in the published tables. The subdivision of workers by product – cotton, wool, flax, silk and lace – means that different types of worker with different levels of skill and different rates of wages were combined to produce the average weekly wage for each age. Spinners, weavers, piecers, carders, errand boys and supervisors may all be included in unknown proportions in this data. In fact, as will be discussed further below, the overall shape of the profile of earnings by age is remarkably consistent across the different textile types, although the absolute level of the wage varies by region. Regression analysis has been used to estimate the extent of this regional wage variation in the Mitchell data.23 Table 10.1 presents estimates of the percentage deviation of the regional wage in textiles from the Lancashire wage, and compares these estimates to those derived by Hunt for agricultural workers in the 1860s.24 It can be seen that both the level and the ordering of regional wage differentials are similar across these two independent data sets.

234

Age, gender and wages

Table 10.1. Estimates of percentage deviation of regional wages from level in Lancashire Male textile workers, 1833 Aberdeen Paisley/Renfrew Glasgow/Lanark Leeds/West Riding Derby Norfolk Devon Somerset Gloucestershire Wiltshire

-31 -20 -5* -1* -7* -13 -26 -39 -41 -42

Male agricultural labourers, 1867–70 -27 -13 -15 -1 -13 -17 -30 -31 -28 -28

Note: * indicates estimate is not statistically significant at five percent level Source: Author’s calculation from US. Bureau of Labor; Hunt, ‘Industrialization’, pp.965–6.

Mitchell himself noted the extent of these regional wage variations We should naturally expect, in a country like Great Britain, where the means of intercourse by land and water are so abundant and so excellent, that the whole commercial community would be so united that there would be but one price of labour as there is but one price of goods, but the above tables show how much it is otherwise.25

Mitchell followed the then-fashionable wages fund theory in ascribing this regional wage variation to a corresponding variation in the level of the poor rate. In fact the similarity with Hunt’s entirely independent estimates of the regional wage gap thirty years later suggests that the labour market in nineteenth-century Britain was regionally segmented, with the level of the agricultural wage relative to that in manufacturing similar within each regional labour market, but with significant and sustained differences in average wage levels between regional labour markets. Whereas the wage information for the 1830s was grouped by Mitchell to produce average age-specific wages for each region and textile type, that for the 1880s and 1930s is derived from household surveys, and so is available at the individual level. Data for the 1880s comes from 1,024 British working class household budgets collected by the U.S. Bureau of Labor in the late 1880s, as part of a European-wide study of workingclass living costs.26 The data covers workers in coal, iron and steel, textiles and glass, but individual earnings data can be identified for only seventyseven females, so the analysis here is confined to just over 1000 male workers. Michael Haines has conducted an extensive analysis of this source, and has found that both the income and expenditure data is consistent with standard economic expectations about income levels and expenditure elasticities.27 Sara Horrell and Deborah Oxley have also used this source, and conclude that the earnings recorded in the budgets are

Johnson

235

representative of occupation-specific earnings for full-time male workers, although they note that the survey is biased towards ‘respectable’ twoparent working-class families.28 Data for 1929 to 1931 are derived from a random sample of over 28,000 working-class households in London, collected as part of the New Survey of London Life and Labour.29 This survey was consciously designed as a follow up to Charles Booth’s late Victorian poverty survey of London. Information is available on the earnings, occupation, hours of work, place of work, residence and household structure of over 25,000 male and 10,000 female workers. Although the survey was conducted during the interwar depression, the unemployment rate in London remained fairly low (for males aged twenty-one to sixty-four the survey records a rate of 7.2 percent), so it is unlikely that reported earnings were significantly affected by short-term labour market factors.30 This is an extremely rich source, which allows us to examine not just weekly earnings but also hourly earnings, and thus directly to address the issue of whether female workers received lower weekly earnings because they worked for fewer hours per week. These three sources provide detailed information on the age and actual earnings of British manual workers over a hundred-year period. However, we should recognise at the outset that the use of these data to make comparisons over time is constrained by geographical and sectoral variation. The 1830s data have wide geographical spread, whereas data for the 1930s are confined to London, and the 1880s data lack direct locational identifiers, but are probably biased towards the midlands and the north of Britain. All observations for the 1830s come from textile factories, those for the 1880s also include coal, iron and steel, whereas data for the 1930s reflect the character of the London labour force, with extensive manual employment in both services and manufacturing, but with almost no workers employed in textiles, or in coal, iron or steel. It cannot, therefore, be assumed that any change in earnings differentials by age or gender between the 1830s and the 1930s is the result simply of a decline in the importance of customary wage setting practices. The next section of the paper therefore examines how other economic and social factors may have influenced the structure of age and gender-specific earnings differentials, and how these factors evolved over the century from 1830.

The determinants of earnings differentials Pervasive differences in pay by age and gender may be an entirely rational response by employers to age- and gender-specific variations in the productivity of workers. For instance, handloom weavers accumulated skills through learning-by-doing, so we should expect younger, less experienced, workers to be less productive, and thus to receive lower weekly wages, than their older and more highly skilled peers. And if weaving also required considerable strength and stamina to maintain high quality

236

Age, gender and wages

output over a ten-hour shift (as was true until the widespread adoption of power looms in the early 1840s), then it is likely that, on average, adult male weavers would command higher wages than juveniles or adult women. Economists combine these various worker attributes of education, skill, strength, stamina and aptitude under the umbrella term ‘human capital’, and over the past thirty years have developed a sophisticated set of models to explain how human capital can account for variations in earnings by age and gender.31 Skills are assumed to be created by formal training and education, and to grow with age and/or work experience. Earnings are depressed at younger ages, since young workers receive part of their total employment compensation in the form of human capital. At older ages obsolescence and depreciation of human capital lead to a decline in productivity and a flattening and then decline in earnings. The overall profile of the relation between age and earnings is thus normally hump-shaped. Many additional factors have been found to affect this underlying relationship between age (or experience) and earnings, of which the most obvious is education. In modern studies formal education is found to both raise and steepen the age-earnings profile, with workers who hold higher formal educational qualifications experiencing faster earnings growth in their initial years in the labour market.32 The same seems to be true for nineteenth-century Britain. An analysis of the earnings of child workers aged seven to fourteen years reported in 1843 to the Royal Commission on the Employment of Children in Trades and Manufactures indicates that literacy (a proxy for formal education) raised earnings by eleven percent.33 Estimates of literacy derived from the 1841 census indicate that sixty-seven percent of adult men and fifty-one percent of adult women were literate, so this enhanced literacy of males over females could partially account for a male wage premium in the early nineteenth century.34 However, by the late nineteenth century Britain had achieved practically universal literacy for both males and females, and minimum education standards further improved with extension of compulsory schooling to age fourteen in the 1920s. This should mean that any education-induced pay gap between males and females should have narrowed over the course of the nineteenth century, while the higher general level of education should have raised the age-earnings profile and made it steeper at the younger ages. A second factor that may have affected the general relationship between age and earnings is demography. A number of modern American studies have found that changes in the size of successive birth cohorts has an effect on earnings. Specifically, an increase in the supply of young labour from the large postwar ‘baby boom’ cohorts increased the slope of the earnings profiles across the labour market of the United States by reducing the relative earnings of younger male workers, although this effect was not observed in female earnings.35 The opposite effect could be anticipated for the century from 1830, when the proportion of fifteen to nineteen year olds fell from just over ten to just under nine percent of the population, while the size of the 40 to 49 age group rose from under ten to over twelve percent.36

Johnson

237

A third factor is technological progress. Between the 1830s and the 1930s production processes were transformed by the application of mechanical power – initially provided by steam engines, and subsequently by electric motors. This substantially reduced the physical effort required of workers in a wide range of manufacturing and service industries such as textiles, transport and food processing, and so should have led to a reduction in the wage premium offered to adult male workers on account of their greater average physical strength. The pattern of wage differentials can also be affected by institutional factors. Collective bargaining by workers in unionised firms in modern Britain has significantly raised wage levels above those in matching, nonunionised firms.37 In early-nineteenth-century Britain trade union membership was negligible (it was illegal until 1825), but it rose from the late nineteenth century to reach almost thirty-one percent of eligible male workers and thirteen percent of eligible female workers by 1931.38 The higher level of union membership among male workers should have raised average male wages relative to average female wages. Employment law can affect the structure of earnings by restricting access to certain trades by certain sorts of workers (for instance by outlawing child employment, or by setting minimum qualification standards for doctors). The law can also structure pay rates, for instance by establishing minimum rates or by imposing equal pay for workers of different gender or ethnicity. Minimum wage laws were introduced in 1909 for a limited range of (female dominated) trades; this will have tended to raise the relative wages of younger workers. Finally, the occupational mix of the workforce may affect earnings differentials, since modern data reveal large differences in the shape of the age-earnings profile between occupations.39 This is particularly pronounced between manual and non-manual workers, with non-manual profiles being steeper, and peaking later, than those for manual workers. However, all earners included in our three sources from the 1830s to the 1930s were manual workers, and although there may be differences in earnings profiles between different manual trades (an issue to be returned to in part three below), this historical data will not be affected by the growth from the later-nineteenth century of white-collar employment.40 In aggregate, the joint impact of better education, new technology and minimum wage laws is likely to have been a reduction in the male/female wage differential, although trends in unionisation may have worked in the opposite direction. The expansion of formal education should have steepened the age-earnings profile for younger workers, whereas demographic change and technological progress should have raised the relative earnings of new labour market entrants over the century from 1830, thus making the age-earnings profile flatter. Overall, therefore, there is a wide range of economic and institutional factors that might be expected to alter the ageearnings profiles of men and women over the hundred years from 1830, quite apart from any change deriving from a reduction in the significance of customary wage setting.

238

Age, gender and wages

Earnings differentials by age and gender, 1830–1930 This section begins by examining the relationship between age and earnings for male manual workers in the century from 1830, before moving on to consider equivalent data for women. Absolute earnings increased considerably over this hundred-year period, a result both of long-run increases in productivity and of the considerable amount of price inflation during the First World War. In order to standardise for these factors, earnings data in Figures 10.1 to 10.3 below are displayed relative to the average earnings of workers aged thirty to thirty-four. It is first necessary to examine the textile earnings data for 1833 to see whether there were significant differences in the relationship between age and earnings across different manufacturing areas (differences that would be additional to the variation between regions in the absolute level of earnings noted in Table 10.1). It might be thought, for example, that the relationship between age and earnings in the most modern and mechanised manufacturing areas would be different from that which existed in regions of traditional textile production where customary wage-setting practices may have been more likely to predominate. In fact, as Figure 10.1 shows, the age-earnings profiles for male workers up to age forty-two were very similar in both the (modern) Lancashire cotton industry and in the (traditional) Wiltshire woollen industry where there was considerable worker resistance to mechanisation.41 At higher ages the workers in Wiltshire fared relatively better than those in Lancashire, which suggests that the fast pace of technological change in the more mechanised cotton industry particularly disadvantaged older workers who had not upgraded their skills. This similarity in the shape of male age-earnings profiles is not confined to different branches of the textile industry, or to the 1830s. Figure 10.2 superimposes earnings data for the 1880s and 1930s on the Lancashire cotton data for 1833. Despite a hundred years of massive technological progress, productivity growth, expanded educational opportunity, demographic change, increased unionisation and a shift from piecework to standard hourly or daily wages, there appears to have been practically no change at all in the relationship between age and earnings for male workers from their point of entry into the labour force until their early forties. There was, of course, substantial sectoral change in the British economy over this period, but this can have had little impact on the overall relationship between age and earnings for manual workers, since there was almost no intersectoral variation in age-earnings profiles. Figure 10.2 reports two separate sets of data for the 1880s relating to males over twenty-one in metals and mining (526 workers) and textiles (470 workers). These two lines trace an almost identical path in Figure 10.2. The same is true for the data from interwar London: the earnings profile for all male workers in 1930 (reported in Figure 10.2) is indistinguishable from the separate profiles (not reported here) for men employed in engineering (1,850 workers) and in clothing (556 workers), these being the two sectors of the metropolitan labour market closest to the employment included in the 1880s survey. Before

0

20

40

60

80

100

10

13

16

Figure 10.1. Male earnings in 1833

% of earnings at age 30–34

120

19

22

28

31

Wiltshire wool

25

age

34

37

43

46

49

Lancashire cotton

40

52

55

58

Johnson 239

240

Age, gender and wages

considering the possible explanations for this amazing long-run consistency in the relationship between age and earnings for males, it is important to see whether the same stability holds for female earnings. Unfortunately the 1880s US Bureau of Labor survey does not provide adequate information on female earnings, so the analysis here is confined to a comparison of textile earnings in the 1830s and metropolitan earnings in the 1930s. Figure 10.3 shows that there is less consistency in the female age-earnings profiles over time than for males. In 1833 the earnings profile for females in the Lancashire cotton industry was clearly different to that for women workers in Wiltshire woollens. In Lancashire earnings for juveniles were higher relative to prime age workers, as were earnings for workers in their late thirties and early forties. This perhaps reflects the relatively high demand for female workers during this period of rapid expansion of cotton production. In fact the earnings profile in the ‘traditional’ Wiltshire woollen industry is closer at all ages to that for interwar London than is the profile for the ‘modern’ Lancashire cotton industry. This suggests that the Lancashire data are reflecting the impact of short-term supply and demand forces which, in the long run, have done little to alter the overall shape of the female age-earnings profile. This does not mean that female earnings were unchanged over the century from 1830. Although the pattern of earnings relative to age was fairly stable, the relationship between female and male earnings changed considerably. Figure 10.4 shows female age-specific earnings as a proportion of the equivalent male earnings in the 1830s and 1930s. In both periods male and female earnings were more or less identical for juvenile labour market entrants, but in the 1830s average female earnings fell to about half the male age-specific equivalent by age twenty-two, and thereafter fluctuated around forty to forty-five percent of the male wage (this holds for all textile types and regions). By the 1930s the decline in relative earnings for juvenile workers in London was less steep, and from age twenty-four female earnings fluctuated at or just above sixty percent of the male age-specific equivalent. Joyce Burnette has speculated that lower female earnings in the nineteenth century can be accounted for in part by the lower amount of labour supplied by women in an average working week.42 This proposition can be tested directly for the 1930s. The New Survey of London Life and Labour reveals that, on average, women worked fewer hours per week than men. In the 1930s male employees in their early twenties worked around 47 hours per week, compared with 45.5 hours for females, but by age forty-five the average female working week had fallen to 37 hours, whilst for males there had been no change. The top line in Figure 10.4 takes account of this reduction in female hours of work, and shows what the female age-specific earnings in the 1930s would have been if women had worked the same number of hours per week as men. For women in their mid-twenties, only three or four percentage points of the male/female wage gap can be explained by reference to the length of the working week, but for women in their mid-forties, about half the wage gap is accounted for by the shorter working week of women.

0

20

40

60

80

100

10

12

14

16

18

22

24

26

Lancashire cotton 1833 Textiles 1880s

20

Figure 10.2. Male earnings by age, 1830s, 1880s and 1930s

% of earnings at age 30–34

120

28

30 age

32

34

38

40

42

44

46

48

Metals and mining 1880s London manual workers 1930s

36

50

Johnson 241

0

20

40

60

80

100

Lancashire cotton 1833

Wiltshire wool 1833

age

London manual workers 1930s

10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50 52 54 56 58 60

Figure 10.3. Female earnings by age, 1830s and 1930s

% of earnings at age 30–34

120

242 Age, gender and wages

0

20

40

60

80

100

age Wiltshire wool 1833 London 1930s (adjusted for hours worked)

20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50 52 54 56 58 60

Lancashire cotton 1833 London 1930s (actual)

14 16 18

Figure 10.4. Female earnings as a proportion of male earnings, 1830s and 1930s

percentage of age-specific male earnings

120

Johnson 243

244

Age, gender and wages

Interpretations The three earnings surveys analysed in this paper reveal that there was some variation over the period 1830 to 1930 in the relationship between age and earnings. In the 1830s age-specific relative earnings for older (40+) males in Lancashire cotton textiles were lower than for males in the traditional Wiltshire woollen industry, whereas for older females Lancashire provided higher relative earnings. This is consistent with local labour market conditions: female employment rates in Lancashire were exceptionally high. Furthermore, over time the wage gap between male and female earnings narrowed, and the earnings of men over forty became more stable, both of which are consistent with a technology driven reduction in the wage premium for physical strength and stamina. But far more striking than these changes is the long-run stability in the relationship between age and earnings. This is all the more remarkable because, quite apart from the influence of the factors identified in part two above, the entire basis of wage payment changed over the period studied here. In the early nineteenth-century textile industry manual workers were typically paid by piece rates – that is, according to the volume and quality of the goods they produced: this was true in both the ‘modern’ and the ‘traditional’ textile regions.43 But by the interwar period, time rates – that is, fixed payments by the hour, day or week – had become the norm. It seems implausible that the relative age-specific productivity of manual workers remained more-or-less invariant both between different economic sectors and over a century of unprecedented economic and technological change, yet that is the inference to be drawn from Figures 10.1 to 10.3 from a standard human capital perspective. An alternative interpretation is that the relationship between age and earnings reflects not the underlying age-specific productivity of the workers, but rather the inheritance of some customary expectations about the relative earnings of younger, prime-age and older workers. A determining role for custom is incompatible with standard economic assumptions about wage determination which posit that individuals (or individual firms) maximise, that equilibrium is the norm, and that archaic or redundant wage forms will be rapidly abandoned in favour of remuneration systems that are technically more efficient. But it fits well with the alternative paradigm of organisational ecology, in which it is the fitness of an organisation (firm) to conform to its environment (a set of political, social, technological and cultural factors) which determines its ability to prosper. Whereas economics assumes that behaviour (such as a particular technique of production or a particular form of wage payment) is selected on the basis of technical (profit maximising) efficiency, organisational ecology assumes that it is selected on the basis of its compatibility with its environment.44

Johnson

245

How, then, might customary age-specific wage differentials come to be sustained over the period 1830 to 1930? The basic idea relates to the concept of environmental imprinting, first developed by Stinchcombe. He argued that ‘organizational inventions that can be made at a particular time in history depend upon the social technology available at the time… Then, both because they can function effectively with those organizational forms, and because the forms tend to become institutionalized, the basic structure of the organization tends to remain relatively stable’.45 A system of age-related earnings is exactly the sort of organisational invention that can become imprinted on the environment. In early-modern England, households endured the greatest financial pressure when the male head was aged around thirty-five; thereafter the entry of younger children into the labour market began to ease financial pressures. There was, therefore, a strong incentive both for the individual worker to provide maximum work effort and maximum output around this age, and for society to sanction high wages at this age, in order to reduce the likelihood of familial resort to the Poor Law. Furthermore, craft workers controlled apprenticeship and journeyman training systems to ensure that peak earnings could not be obtained by junior workers, and as Phelps Brown and Hopkins showed, wages for unskilled labourers were closely linked to those received by skilled workers. In theory, traditional wage differentials could have been abandoned by individual employers as they took advantage of changes in the supply of and demand for different types of workers, and changes in age-specific productivity resulting from new technology and mechanisation. However, any individual employer had to evaluate the cost of wage innovation imposed by the environment in which he operated. Two costs in particular can be identified. First, the cost of non-cooperation from fellow employers who did not have access to new technology, and so who could not adopt a different (and technically optimal) set of agespecific wages. Second, the cost of non-cooperation from workers who would not immediately benefit from the new wage system. Even in Lancashire, the most modern part of the British textile industry, production was highly fragmented in the early and mid-nineteenth century, with large numbers of small firms operating at each stage of the manufacturing process. An antagonistic action by one employer, such as abandoning locally agreed wage payment systems, could lead to both commercial and social ostracism by the rest of the employer community. It could also lead to a labour boycott, especially if the new wage system offered less of a premium to more senior workers. Unless there was complete substitutability between workers of all ages, such a labour boycott could bring production to a halt. In these circumstances, the most likely response of any employer who found that technology was making certain age groups more productive, was not to change the

246

Age, gender and wages

inherited age-earnings profile, but rather to change the proportion of workers of different ages. This is, of course, exactly what happened with the mechanisation of cotton spinning – men were replaced in large numbers by child and female workers, but the age-earnings profiles in Lancashire cottons remained much the same as in traditional (non-mechanised) textile areas. From the 1840s, further technical change began to limit the productivity advantage of child workers at the prevailing wage rates, and factory legislation also restricted their hours of work. Yet the age-earnings profile changed little for girls and not at all for boys between the 1830s and the 1930s. A traditional relationship between age and earnings was so deeply imprinted on the social and cultural environment that a century of economic change and technical innovation failed to undermine it. This does not mean that the wage system in Britain from the 1830s to the 1930s was completely unresponsive to economic pressures: excess supply always led to a collapse of the wage, as was famously the case for handloom weavers in Lancashire in the 1840s, and Spitalfields silk weavers in the 1860s. It is also clear for the 1930s that some of the earnings gap between men and women – a feature often ascribed entirely to sociocultural factors – was in fact the result of women working for fewer hours per week. Yet the evidence from the three earnings surveys used in this paper points overwhelmingly to the importance of custom or tradition in setting the relationship between earnings and age. Indeed, the stability of the age-earnings profile for male manual workers shown here for the century from 1830 appears to continue to the 1970s and 1980s, indicating that these customary pressures have been sustained in Britain from the early industrial revolution through to postwar de-industrialisation.46 In the light of this finding, perhaps historians of wages in modern Britain should be more questioning of standard economic assumptions that market competition generates optimal wage rates, and should pay more attention to the mechanisms whereby existing or inherited wage forms become imprinted on the economic environment.

Notes 1. This paper forms part of a larger research project on Laws and Markets in Victorian England, supported by the British Academy. Earlier versions were presented at the London School of Economics and at the Fondation des Treilles, France; I wish to thank participants for their constructive comments. Thanks are also due to Michael Haines, Colgate University, for generously supplying me with a computerised version of the US Bureau of Labor survey, and to Susannah Morris, London School of Economics, for introducing me to the literature on organisational ecology. 2. E. H. Phelps Brown and S. V. Hopkins, ‘Seven centuries of building wages’, Economica, 22 (1955), pp. 195–206. For a more recent evaluation of the extent of customary wages in the early modern period see Donald Woodward, ‘The determination of wage rates in the early modern north of England’, Economic History Review, 47 (1994), pp. 22–43.

Johnson

247

3. E. H. Hunt, Regional Wage Variation in Britain, 1850–1914 (Oxford, 1973); E. H. Hunt, ‘Industrialization and regional inequality: wages in Britain, 1760–1914’, Journal of Economic History, 46 (1986), pp. 935–66 4. Humphrey Southall, ‘The origins of the depressed areas: unemployment, growth, and regional economic structure in Britain before 1914’, Economic History Review, 41 (1988), pp. 236–58; Paul Johnson, ‘Small debts and economic distress in England and Wales, 1857–1913’, Economic History Review, 46 (1993), pp. 65–87; Humphrey Southall and David Gilbert, ‘A good time to wed? : marriage and economic distress in England and Wales, 1839–1914’, Economic History Review, 49 (1996), pp. 35–57. 5. Pam Sharpe, ‘Continuity and change: women’s history and economic history in Britain’, Economic History Review, 48 (1995), pp. 353–69; Maxine Berg, ‘What difference did women’s work make to the Industrial Revolution?’, History Workshop, 35 (1993), pp 22–44; Jane Humphries, ‘“…The most free from objection…”. The sexual division of labour and women’s work in nineteenth-century England’, Journal of Economic History, 47 (1987), pp. 929–49; Sonya Rose, Limited Livelihoods: gender and class in nineteenth-century England (Berkeley, 1992); Deborah Valenze, The First Industrial Woman (Oxford, 1995). 6. Mary Freifeld, ‘Technological change and the ‘self-acting’ mule: a study of skill and the sexual division of labour’, Journal of Social History, 11 (1986), p. 334. 7. Ellen Jordan, ‘The lady clerks at the Prudential: the beginning of vertical segregation by sex in clerical work in nineteenth-century Britain’, Gender and History, 8 (1996), p. 66. 8. Judy Lown, Women and Industrialization: gender at work in nineteenth-century England, (Cambridge, 1990). 9. E. P. Thompson, ‘Time, work-discipline and industrial capitalism’, Past and Present, 38 (1967), pp. 56–97. 10. E. P. Thompson, ‘The moral economy reviewed’ in E. P. Thompson (ed.), Customs in Common (London, 1993). See especially pp. 336–42. 11. J. G. Williamson, ‘The structure of pay in Britain, 1710–1911’, Research in Economic History, 7 (1982), p.22. 12. Jeffrey G. Williamson, Did British Capitalism Breed Inequality? (London, 1985), chapter 3. 13. R. V. Jackson, ‘The structure of pay in nineteenth-century Britain’, Economic History Review, 40 (1987), pp. 561–70; C. H. Feinstein, ‘The rise and fall of the Williamson curve’, Journal of Economic History, 48 (1988), pp. 699–729. 14. Feinstein, ‘Rise and fall’, p. 729. 15. Jeffrey G. Williamson, Coping with City Growth during the British Industrial Revolution (Cambridge, 1990), p. 207. 16. Joyce Burnette, ‘An investigation of the female-male wage gap during the industrial revolution in Britain’, Economic History Review, 50 (1997), pp. 257–81. 17. For a rare example of the analysis of age-specific earnings data, see H. M. Boot, ‘How skilled were Lancashire cotton factory workers in 1833’, Economic History Review, 48 (1995), pp.283–303. 18. See the references in Charles Feinstein, ‘New estimates of average earnings in the United Kingdom, 1880–1913’, Economic History Review, 43 (1990), pp. 595–623. 19. B. Black, ‘Age and earnings’, in M. B. Gregory and A.W.J. Thomson (eds), A Portrait of Pay, 1970–1982: an analysis of the New Earnings Survey (Oxford, 1990), pp. 274–98. 20. Royal Commission on the Employment of Children in Factories (1834), Supplementary Report, Part I (hereafter Mitchell Report). Parliamentary Papers, 1834 XIX. 21. Details of the exact name and location of each textile factory supplying returns are included in the appendix to the report. Mitchell Report, pp. 62–3. 22. Ibid, p. 19. 23. The equation was of the form lnW=AGE + AGE2 + RD where: lnW is the natural log of the weekly wage; AGE is age of worker in years; AGE2 is age squared; RD is a regional dummy variable. The excluded region is Lancashire.

248

Age, gender and wages

24. Hunt, ‘Industrialization’, p. 965–6. 25. Mitchell Report, p.40. 26. US Congress, House of Representatives, Sixth Annual Report of the Commissioner of Labor, 1890, Part III, ‘Cost of Living’ (Washington D.C., 1891); US Congress, House of Representatives, Seventh Annual Report of the Commissioner of Labor, 1891, Part III, ‘Cost of Living’, (Washington D.C., 1892). 27. Michael Haines, ‘Industrial work and the family life cycle’, Research in Economic History, 4 (1979), pp. 289–356. 28. Sara Horrell and Deborah Oxley, ‘Crust or crumb? Intrahousehold resource allocation and male breadwinning in late Victorian Britain’, Economic History Review, 52 (1999), pp. 494–522. 29. H. Llewellyn Smith, New Survey of London Life and Labour, 9 volumes, (London, 1930–35). A computerised version of the household survey has been deposited at the ESRC Data Archive, University of Essex, file number SN3758. 30. For a more detailed examination of the reliability of the NSLLL data, see Dudley Baines and Paul Johnson, ‘Did they jump or were they pushed? The exit of older men from the London labor market, 1929–1931’, Journal of Economic History, 59 (1999), pp. 949–71. 31. The pioneering work was by Jacob Mincer, Schooling, experience and earnings (New York, 1974). 32. Richard Freeman, ‘The effect of demographic factors on age-earnings profiles’, Journal of Human Resources, 14 (1980), pp. 289–318. 33. Data relates to 261 cases which contained the necessary information, drawn from the Royal Commission on the Employment of Children in Trades and Manufactures, Supplementary Reports, Parliamentary Papers, 1843, XIV and XV. The natural logarithm of the weekly wage was regressed on age, work experience and literacy. The estimated equation was lnW = .127AGE + .049EXPERIENCE + .11LITERACY. F=48, R2=.35; all coefficients significant at the 5% level. 34. Michael Sanderson, Education, Economic Change and Society in England, 1780–1870 (London, 1983), p. 17. 35. Finis Welch, ‘Effects of cohort size on earnings: the baby boom babies’ financial bust’, Journal of Political Economy, 87 (1979), S65–S97. 36. Brian Mitchell, International Historical Statistics: Europe 1750–1988 (Basingstoke, 1992), pp. 41–2. 37. P.B. Beaumont and R.I.D. Harris (1990), ‘Collective bargaining and relative wages’ in M. B. Gregory and A. W. J. Thomson (eds), A Portrait of Pay, 1970–1982: an analysis of the New Earnings Survey (Oxford, 1990), pp. 404–29. 38. George Bain and Robert Price, Profiles of Union Growth (Oxford, 1980), p. 37. 39. E. H. Phelps Brown, The Inequality of Pay (Oxford, 1977), p. 268. 40. White-collar jobs are also more likely to be affected by what economists call ‘internal labour markets’, in which formal promotion systems operate to target high pay to loyal long-serving workers. These internal labour markets are most common in very large firms, but according to Hannah few British employees were affected by internal labour markets before the 1930s. See Leslie Hannah, Inventing Retirement (Cambridge, 1986), pp. 133–137. 41. J. de L. Mann, The cloth industry in the West of England from 1640 to 1880 (Oxford, 1971), chs. 5 and 6. 42. Burnette, ‘An investigation’, pp. 268–269. 43. Adrian Randall, ‘Work, culture and resistance to machinery in the West of England woollen industry’ in Pat Hudson (ed.), Regions and Industries (Cambridge, 1989), p.186; Michael Huberman, Escape from the market: negotiating work in Lancashire (Cambridge, 1996), p. 5.

Johnson

249

44. For an introduction to this literature see M. T. Hannan and J. Freeman, Organizational Ecology (Cambridge, Mass., 1989); Elaine Romanelli, ‘The evolution of new organizational forms’, Annual Review of Sociology, 17 (1991), pp. 79–103. 45. A. L. Stinchcombe, ‘Organizations and social structure’ in J.G. March (ed.), Handbook of Organizations (Chicago, 1965), p. 153. 46. Black, ‘Age and earnings’, pp. 280–281.

CHAPTER 11

AT WHAT COST WAS PRE-EMINENCE PURCHASED? CHILD LABOUR AND THE FIRST INDUSTRIAL REVOLUTION

Jane Humphries

Sir, I hardly know whether any argument is necessary to prove that the future hopes of a country must, under God, be laid in the character and condition of its children; however right it may be to attempt, it is almost fruitless to expect, the reformation of its adults, as the sapling has been bent, so will it grow… Now, Sir, whatever may be done or proposed in times to come, we have, I think a right to know the state of our juvenile population; the house has a right, the country has a right… The first step towards a cure is a knowledge of the disorder. We have asserted these truths in our factory legislation; and I have on my side the authority of all civilised nations of modern times; the practice of this House, the common sense of the thing, and the justice of the principle… It is right, Sir, that the country should know at what cost its pre-eminence is purchased (Lord Ashley, 1840).

Child labour used to loom large as one of the more reprehensible aspects of British industrialisation. Today however the children who toiled in early mills, mines and manufactories are scarcely visible in mainstream accounts of the industrial revolution. The standard economic history textbook for this period lists only six references to what are squeamishly called ‘juvenile workers’.1 As a topic of research, children’s role in industrialisation has become passé.2 It is easy to see why modern economic historians are disinterested in child labour. They are guided by neoclassical economics, which interprets outcomes in terms of decision making, by free agents. But children did

252

Child labour

not choose work, they were directed by parents and commandeered by poor law officials. They do not fit easily into the mainstream conceptual framework. Moreover, modern economic history is quantitative and the record of children’s work is fragmentary and problematic. Before the mid-nineteenth century censuses there are no national counts of child workers and even the censuses systematically under recorded them. Child labour was hidden within the family, lost in undocumented casual employment, and deliberately falsified in regulated workplaces. Researchers are left with the patchy and sometimes suspect evidence presented to Royal Commissions and Parliamentary Inquiries. But an economic history, which emphasises what is readily theorised and easily measured, may be unbalanced. Perhaps the neglect of child labour has produced too benign an interpretation of the industrial revolution. Putting the historical record straight will require much new work on the extent and importance of children’s employment in the early industrial economy. Currently basic questions remain unanswered. How important was child labour in the industrial revolution? In particular, what proportion of children in each age group worked? At what age did children usually go to work and did this vary according to their geographical location and father’s employment? Did the development of factory industry give a boost to child employment or was domestic industry more child intensive? Did apprenticeship remain important in channelling and training young workers or had the institution faded by the beginning of the nineteenth century? Most difficult of all, how were young workers remunerated? Did they themselves receive their wages or were they subsumed in family wages? Were child apprentices paid at all or was their labour simply set against the training received?

Empirical evidence on child labour and problems with the data sources Recently, partly prompted by the strong concern with child labour in contemporary developing countries, and partly reflecting the resurgence of the pessimist interpretation of the effects of the industrial revolution on the standard of living of working people, there has been new interest in these questions. For the period classically identified with the industrial revolution, historians have to rely on patchy and localised sources. Some of these are very familiar to historians. For example, Carolyn Tuttle in her recent book referred to the evidence from well-known government enquiries into textiles and mining supplemented by surveys by contemporary authorities but subject these to a particular systematic reappraisal.3 The evidence suggested astonishingly high child employment shares in some industries for various years from 1800–1850. Children (aged under thirteen) and young people (aged thirteen to eighteen) made up between one-third and two-thirds of all workers in many textile mills in 1833 and regularly over one-quarter in many mines in 1842.

Humphries

253

Researchers have also turned to new and sometimes indirect evidence. My own work (with Sara Horrell) used accounts of working class family incomes and their sources to investigate child labour and children’s contributions to their families’ budgets.4 We found that children’s participation rates varied with their father’s occupation, but that throughout industrialisation one-quarter of all children in families headed by men worked. Over the period from 1787 to 1872, the decline in participation rates for children of fathers in most occupational groups was offset by the shift of adults towards jobs (and locations) associated with higher child activity rates. The absolute number of children working, of course, increased with the population growth of the period. We also uncovered evidence of younger children working as a general phenomenon in early industrialisation exaggerated by the shift of the adult population towards occupations in which children’s age at first participation was lower than average. The relative youth of children working in factories was strongly indicated. But these trends did not show up in higher overall child participation rates for children in families in the early nineteenth century because simultaneously children left home at earlier ages. Note that our estimates refer only to children living in families headed by men. Children living outside such a family structure, in institutions or in families headed by women, had higher participation rates and started work at a younger age.5 Thus our estimate that twenty-five percent of children in families were at work from the late-eighteenth until the mid-nineteenth centuries is very much a lower limit for all children whatever their family circumstances. The evidence explored here comes from a well-known source, workingclass autobiographies. Hitherto historians have used working-class autobiographies in two ways. Some autobiographies, usually the wellknown and accessible examples of the genre, have been used for illustration and substantiation, to inject some colour and immediacy into the historical narrative. In contrast, the distinguished historians of working-class autobiography have focussed intensively on the texts themselves to discover ‘how these working men and women (…) understood their lives, and thereby to reach a better understanding of the working class as a whole during the industrial revolution’.6 As both Burnett and Vincent note the majority of working-class autobiographers had something to say about childhood employment and children’s introduction to the labour market. These topics did not occasion one of those enigmatic silences, which characterise the genre. Instead there is a gold mine of information on pressures to work, links between the family and the labour market, the nature of first jobs, remuneration and apprenticeship. The strategy pursued here has been to extract this material for a sample of children whose experiences are recaptured through their own later writings and to use it to develop a partially quantitative study of child labour. The sample of 112 boys was taken from Burnett, Vincent and Mayall’s The Autobiography of the Working Class: An Annotated Critical Bibliography. For each child, a record of date of birth, location, father’s job, family structure, age at starting work, first job, subsequent childhood employment,

254

Child labour

apprenticeship, and adult employment was established. In some cases recorded in the Annotated Bibliography evidence on these variables was incomplete. In about half the cases the actual biographies were consulted which sometimes remedied the gaps in the records and always provided greater interpretative depth. The sample size is small and subgroups selected for examination often included only a handful of boys. Moreover even within the selected sample of autobiographers the sample was not random. Priority was given to readily available material, which could then be consulted directly. However a number of unpublished manuscript autobiographies were also read. An important selection criterion was the child’s date of birth: he had to have lived through the industrial revolution. The records were then coded, processed and analysed. The present study does not pretend to offer hard and fast findings. It is intended as a pilot for what could be achieved from this kind of approach to the material. By moving back and forth between the quantitative evidence extracted from the autobiographies and the interpretive depth provided by the writings themselves the objective is to uncover the childhood experiences of work in both their family and industrial context, as a guide to further research.

Age at starting work Starting work was clearly perceived as a significant milestone in autobiographical experience. Most of the autobiographers noted their age at starting work, the circumstances surrounding this event, and their first job.7 These stark facts provide the background for Vincent’s sensitive analysis of the autobiographers’ recognition that they were members of the working class and therefore denied the carefree innocence and childish amusements enjoyed by the more privileged. Bitterness dawned shaded only by the children’s pleasure in contributing to family survival in circumstances where they felt loved and valued and by a developing sense of solidarity with the class whose hard destiny they shared. The sample shows the importance and ubiquity of children’s work. Figure 11.1 presents the participation rates associated with each age group. Few children started work as young as six but by the age of eight more than a third of the children in the sample were working and by thirteen almost ninety percent. By the age of sixteen all the children in the sample had started work. Table 11.1 provides evidence on age at starting work for the sample as a whole and by father’s occupation. Average age at starting work was 10.21 (n = 96). But differences according to father’s broad occupational group are apparent.

0

10

20

30

40

50

60

70

80

90

5

6

7

8

9

10 11 12 Age at starting work

Figure 11.1. Cumulative percentage of working children by age, all children

Cumulative percentage

100

13

14

15

16

Humphries 255

256

Child labour

Table 11.1. Age at starting work by father’s occupation Father’s occupation

Mean

Sample size

Agriculture Mining Factory Domestic industry Trades Casual White collar Soldier Sailor Services Total known

10.53 10.33 10.75 9.19 12.18 8.67 10.50 10.50 12.50 11.57 10.45

19 9 4 21 11 6 2 2 4 7 85

Standard deviation 2.59 2.87 2.06 3.25 2.09 2.58 2.12 2.12 2.65 1.99 2.78

Source: see text

The sons of domestic manufacturers, such as handloom weavers and framework knitters, and casual workers, such as railway navvies and general labourers, started work at younger ages that the sons of workers in other occupational groups. In contrast the sons of artisans, sailors, and men with jobs in the services sector were older when they entered the labour market. The sons of agricultural workers, white-collar workers such as schoolteachers, and men in mining or metallurgy began work at intermediate ages.8 These ages of first participation by fathers’ occupation are very similar both in levels and rankings to estimates of ages at starting work made by Horrell and Humphries using very different data.9 The main difference in the findings is the later age at starting work for the sons of factory workers reported here in comparison with the extreme precocity of employment for this group reported by Horrell and Humphries on the basis of their family budgets. The estimate here is based on only three cases, which may not be typical. Moreover it relates to the sons of factory workers whereas Horrell and Humphries’ data was for children of both sexes. But the evidence from the autobiographies should not be read as undermining the Horrell and Humphries finding that children in factory districts started work at very young ages even by contemporary standards. Many if not most of the children of domestic manufacturers in the autobiographies, who started work a year younger than was typical, were employed in the early textile mills and manufactories. Figure 11.2 translates this youthful working back into the participation rates by age for the sons of domestic manufacturers taken from the sample for comparison with Figure 11.1. For these unfortunate children the big jump to over one-third working occurred at the age of six, two years before the same participation rate was reached for children in the sample as a whole. Given the variation in the age at which children started work with their fathers’ occupation, the sample average will reflect the sample composition in terms of paternal occupations, which may not be representative of the adult male labour force. A weighted average of the ages of starting

0

10

20

30

40

50

60

70

80

90

5

6

7

8

9

10 11 12 Age at starting work

Figure 11.2. Cumulative percentage of working children by age, according to father’s occupation

Cumulative percentage

100

13

14

15

16

Humphries 257

258

Child labour

work where the weights represent the actual proportions of the broad occupational groups circa 1800 would provide a better guide to average experience. Weighting the averages by broad occupational group gives an age of starting work of 10.57 not far away from the simple average of 10.21.10 What determined the age at which children started work? Did it vary by time period and by employment opportunities within the local labour market? Were children pushed into the labour market by pressure on family income? Horrell and Humphries found some evidence to suggest that children did go to work at younger ages in the first phase of the industrial revolution, a trend promoted partly by younger children working in factories. But age at starting work began to rise subsequently. Carolyn Tuttle implies something like the same pattern in her recent book, in which she ascribes the importance and persistence of child labour in the British economy to developments in technology, particularly textile technology.11 Tuttle argues that the technological innovations of the industrial revolution, particularly in the textile industries, were designed to increase the employment of children not only as helpers but also as substitutes for adult works in the principal processes. In addition the close workplaces of the industrial revolution and the nature of the machinery (wooden and close to the ground) put a premium on the small size of children and their ability to work in confined quarters. The nature of the workplaces in mining, especially in thin seam pits, also gave children situational advantages. Tuttle supports her argument by linking key innovations, especially in textiles to specific jobs and to the ages and gender of workers in those jobs. Christine MacLeod has also shown that the substitution of child for adult labour was the explicit intention of some inventors of textile machinery.12 Thus authors using different sources have hypothesised that industrialisation and especially the development of factories gave a boost to child labour which persisted long after the initial phase of factory production when rural water-driven mills had relied on child paupers. A similar trend is observable in the data from the autobiographies, but the sample sizes are too small for statistical significance and the time trend is affected by the sample composition in terms of fathers’ occupational group. Table 11.2 Trends in age at starting work Date of birth < 1820 1820-1844 >1844

Sample size

Mean age

Sd

Sem

46 31 19

10.37 9.71 10.63

2.69 2.84 2.87

.40 .51 .66

Source: see text

An expansion of the autobiographical data set will allow further testing of the hypothesis that industrialisation intensified the economic use of

Humphries

259

children. A larger sample will also cast light on whether this was a general trend associated with younger children working in most occupational groups or whether it was a result of the employment structure swinging towards sectors where children were employed at particularly young ages.

Family circumstances The clearest evidence of the effect of family poverty on age at starting work is revealed by the significant effect of the father’s absence. The sample was subdivided into cases where the father was present during the son’s childhood and cases of absence whether this was caused by death, desertion, or tramping in search of work. Boys whose fathers were missing began working on average aged 9.46 years (n = 22) in comparison with boys whose fathers were present who began work aged 10.65 (n = 69), a difference that is statistically significant. The premature ejection into the labour market following a father’s death or desertion was clearly recognised as a misfortune by the children involved. This gave the father’s loss a bitter edge that compounded any emotional loss the child might have felt. George Elson expressed this most clearly when commenting on the death of his father soon after the coronation of Queen Victoria. ‘My thoughts as I stooped to kiss the marble forehead of the lifeless being who was once my beloved father, formed a futuitive prescience of evil that was to be amply borne out in fact. These thoughts arose and remained in my own breast as I stood in the parlour witnessing the coffin brought downstairs. I said, “there goes the support and comfort of my young life; with him gone all is lost”. These were young but vital recollections’.13 Edward Rymer put it more bluntly about his father’s desertion ‘Nothing remained to us but to face the world and fight for existence’.14 The implications for his brothers and himself were immediate. ‘(…) my mother, though toiling hard to keep the home together, had to avail herself of any small labours we elder boys could render’.15 Sometimes fathers withdrew both emotional and economic support even while remaining on the periphery of sons’ lives. Thus John Buckley’s father, a yeoman farmer, turned to drink on the death of his wife when John was only two, leaving the little boy to be raised by his well-meaning but frail grandmother and then his ‘God-fearing’ but harsh uncle. Before too long, when John was ten, his father’s headlong downward social mobility produced its inevitable result. ‘Father had lost his little property and I must come home to work’.16 Sometimes it was the death of a sibling that necessitated the change. When John Bennett was aged nine his older brother Stephen who was then thirteen to fourteen and working with his father died. As a consequence John was taken from school ‘to supply his place’.17 Parish relief on which many female-headed households relied was probably conditional on children starting work at as young an age as possible especially when spiralling poor rates prompted overseers to pursue harsh economies. John Castle, for example, whose widowed mother

260

Child labour

received poor relief to raise her two sons in the 1820s started work in a silk factory at the relatively young age of nine. The adverse effects on a child’s prospects of a large number of siblings, especially younger ones, was acknowledged in the autobiographies and shows up in the quantitative analysis. The numbers of brothers and sisters are known for sixty-eight boys in the sample. The age at starting work for only children was 11.93 compared with 9.63 for boys with one or more brothers or sisters, a difference that is statistically significant. With a cut point at higher parities the difference remained but was no longer statistically significant. Thomas Wood clearly related his experience of early working to his position as the eldest in a family of ten brothers and sisters, which meant that there was usually four or five at home too young to work.18 Dependent siblings compounded the pressures on older children deprived of a father’s support. When Edward Rymer’s brother John became crippled and unable to work, probably as a result of his early employment drawing coals underground, the loss of his earnings was a harsh blow: ‘the struggle for dear life for our family had become terrible’.19 The autobiographies also point to factors, which operated to protect children from a premature introduction to the labour market even when they had lost their fathers, though the size of the fatherless sample is too small to explore this quantitatively. It follows from what has been said above that only sons or boys with few siblings fared better under these cruel circumstances. Thomas Cooper and William Lovett were both without fathers: from an early age in Cooper’s case and from before birth in Lovett’s case. But they were both only children, which might help to explain their relatively late entry into the labour force and their associated prolonged education. Moreover both appear to have benefited from industrious and enterprising mothers. Mrs Cooper, in particular was assiduous in exploiting opportunities to earn. Cooper’s autobiography provides a moving illustration of how a mother’s spirit and determination could be all that stood between these boys and exploitation and overwork. On one occasion when Mrs Cooper, accompanied by Thomas, was travelling around Gainsborough hawking some little work boxes that she had made, they were accosted by one Cammidge, a master chimney sweep and his two apprentices bowed down under huge soot bags. Cooper tells the tale. ‘He began to entice my mother into an agreement for me to be his apprentice, and took out two golden guineas from his purse and offered them to her. She looked anxiously at them, but shook her head and looked at me with tears in her eyes: and I clung tremblingly to her apron, and cried, ‘Oh, mammy, mammy! Do not let the grimy man take me away!’ ‘No, my dear bairn, he shall not’, she answered; and away we went – leaving the chimney-sweep in a rage, swearing and shouting after my mother that she was a fool, and he was sure to have me, sooner or later…’.20 But the difficulties facing women who were left to raise several dependent children must have seemed insurmountable. In these circumstances the defeated attitude manifest by David Love’s mother, left with

Humphries

261

seven children was probably more typical. ‘Being thus distressed, forlorn and destitute, with one babe at the breast, and myself about three years old, she was forced to beg her bread and no doubt told her mournful tale to everyone that it might excite their pity and relieve us in our distressed condition.’21

Apprenticeship The institution of apprenticeship has been caught up in the same ebb and flow of interest as child labour generally. It was regarded as crucial to an understanding of the industrial revolution by an earlier generation of scholars, but has generally been ignored by modern economic and social historians. Keith Snell, one of the few modern historians to integrate an appraisal of apprenticeship into his study of early-industrial labour markets, raised the important question of the persistence of apprenticeship, albeit in a less regulated form, long after the repeal of the Statute of Artificers.22 It is possible that apprenticeship as an institution was an important factor in the precocious structural change away from agricultural employment characteristic of the British industrial revolution.23 The autobiographies testify to the resilience of apprenticeship. Out of the sample of 112 boys forty-five were known to have begun what they termed an apprenticeship though in some cases this may not have been formal and in many other cases was not completed. In fifty-two cases there is enough information to rule out the possibility of an apprenticeship and in fifteen cases there is not enough information to make a judgement. Even if all the cases in the latter category are counted as never apprenticed, it remains that a substantial minority of boys was formally or informally apprenticed. Nor does it look like the proportion of boys apprenticed was in dramatic decline in the nineteenth century, though the serious investigation of trends would require a much bigger sample. The majority of those apprenticed were formally indentured. Concentrating attention on boys whose fathers’ broad occupational group was known, it emerges that boys whose fathers were themselves in trades were most likely to be apprenticed. Boys whose fathers were miners or metal workers were unlikely to be apprenticed, as were boys whose fathers were domestic manufacturers. While some boys were apprenticed to family members (father, brothers, uncles and grandfathers all feature) the majority of boys were apprenticed to nonrelatives. Throughout the period of the industrial revolution, working people persisted in seeing an apprenticeship in a trade as the key to economic security and social advancement for their sons. Both parents and sons hankered after apprenticeships, particularly in what they perceived as good trades, as a crucial investment. Such judgement characterised the community. When on his way to ‘Oxenbridge’ to be apprenticed to a carpenter and joiner, lonely and motherless John Buckley could not hold back the tears, the waggoner encouraged him with ‘Cheer up, my cockolorum, a

262

Child labour

joiner is better than a farm labourer’.24 William Farish’s mother and father were very disappointed not to be able to apprentice William to the trade of his maternal grandfather (carpentry), instead having to exploit his labour at an early age in the family’s handloom weaving activities.25 But apprenticeships, particularly ones in flourishing and prosperous trades required a premium. The problems that such a financial investment presented in poor families are discussed in several autobiographies. Buckley’s apprenticeship was delayed by his father’s inability to salvage enough money for the premium from the ruin of his financial affairs.26 James Burn saw an apprenticeship not only in terms of the acquisition of a skill but also as a way to provide some structure to his hitherto chaotic life and to put the wanderings and destitution of his childhood behind him in a long struggle back to respectability. But he was frustrated in his search for a berth by his initial inability to find either a premium or a person to stand security for him. Even Thomas Cooper’s resourceful mother was stymied. ‘My mother had tried, at my entreaty, to get me apprenticed to a painter, and had endeavoured to get me entered as a clerk at one or the other of the merchants’ establishments; but in every case a premium was demanded and my poor mother had none to give’.27 Thomas went to work as an unindentured shoemaker. One important point to emerge clearly from the autobiographies is that by the time boys were apprenticed they had often, indeed almost invariably, had significant labour market experience. Consequently to equate their average age at apprenticeship with their age at starting work would be to badly overestimate the age at which they entered the labour market. Buckley’s experience here was typical. He worked as a farm boy first within a ‘roundsman system’ and then in a permanent berth for two to three years before starting his apprenticeship aged thirteen.

Earnings The current sample is too small to provide estimates of wages. But enough autobiographers mention actual earnings and their growth over time to suggest that a larger sample could be used to document age-earnings profiles over time and by sector. As it stands, the fragmentary data provide some insight into children’s relative earnings. Joseph Arch started work as a crow-scarer, earning 4d a day for a twelve-hour shift. After two or three years, he graduated to ploughboy at 6d a day, and at fourteen was selected for the job of stable boy at 8s. a week, as much as his father then earned.28 After his experience ‘on the rounds’ in the 1830s, John Buckley aged eleven managed to secure a permanent place on a local farm paying 2s. a week. At the same time and place adult male agricultural labourers could earn 9s. or 10s. a week. Tom Mullins, working as a farm-boy twenty years later than Buckley, earned £3 a year plus his keep graduating to £5 in a year or so.29 George Barber, though only nine years old in 1869 was sent to work on a lonely farm for £2 a year and board. The anonymous navvy’s

Humphries

263

first job was minding two little boys for a farmer for which he got breakfast and a penny a day.30 On the death of his father, Robert Lowery, aged nine was taken out of school and sent to pick the brasses out of the coals at the mouth of the local mine. From working thirteen hours a day at this activity he earned 5s. a week.31 Some thirty years later, Edward Rymer and his brother, not yet ten years old, both earned 10d a day working underground.32 Thomas Wood started work in a woollen mill aged eight in 1830 earning 1s. 6d. per week, the same starting salary as seven year old Tom Mullins earned at a rope works at Leek in 1870. When Wood left in 1835 he was earning 4s. 9d.33 William Dodd reports the wages earned by children of various ages in cotton factories in great detail.34 John James Bezer’s first job was as an errand boy in London in the 1820s at which he earned 3s. a week for very long hours. This and similar evidence suggests that controlling for sector and time, boys’ age-earnings profiles were steeply sloped in all but the craft occupations for which an apprenticeship was needed. In many jobs, but especially those for which physical strength was at a premium, boys’ earnings could approach those of adult males by the time they reached their late teens. The sacrifices families made in withdrawing boys from the free labour market and placing them in apprenticeships in terms of the earnings foregone were frequently acknowledged. At the same time apprenticeship was seen as the key to higher lifetime earnings. Robert Lowery described his widowed mother’s decision to move back to North Shields. ‘Had she been merely anxious to live on our earnings without reference to the future, in two or three years the joint earnings of myself and brothers would have brought in a very good income’. But Mrs Lowery urged the change saying, ‘you and your brothers can have no opportunity of being anything here, we must make a struggle to get you to sea; it will be a hard one for us, but I trust in God’.35 In their first jobs, boys’ earnings were small in absolute terms. But even these pittances could be of considerable importance, especially if the family budget lacked a contribution from an adult male. As Edward Rymer put it, ‘money earned by myself and brother had a wonderful effect on the scanty home exchequor’.36 The autobiographies also cast light on the debated issue of whether apprentices were paid wages or simply received their room and board. It seems that by the nineteenth century wages were paid, certainly to apprentices in their teens. These wages were lower than those obtained by journeymen in the trade and lower than the wages boys could command in some unskilled labouring jobs. The level of remuneration also depended on whether the apprentice boarded in or out. An older model of apprenticeship in which the apprentice’s wage was assumed to be equivalent to his room and board survived even in cases where the apprentice lived out. John Buckley, for example, an apprenticed carpenter in the 1830s lodged with his uncle while serving his term and his wages were paid directly to this uncle. In this case, as in many others, masters accommodated apprentices’ need for pocket money by tolerating

264

Child labour

or even encouraging additional income earning activity. In Buckley’s workshop the wood shavings were considered the property of the apprentices who recycled this waste, selling it on for 2d. or 3d. a sack and so earning 5d a week pocket money.37 James Burn, when an apprentice hatter, during periods of slack trade at harvest time got work as a reaper. He used the money to purchase much-needed clothing as his apprentice’s salary after boarding and lodging was only a shilling a week for the first year rising in increments of 6d each year thereafter.38 Several apprentices bargained with their masters for a mutually agreeable combination of wages and board. For example, John Bennett an apprentice carpenter in the early 1800s faced a master who wanted him to ‘bed and board’ and be given 6s. a week for three years. But Bennett was unhappy with the food provided. ‘The living did not suit me’. He went back to an earlier arrangement whereby he received 16s. a week wage. Bennett was a good worker who had already received considerable training from his father. Perhaps this explains his master’s apparent ready compliance. The autobiographies suggest that the institution of apprenticeship continued to provide a framework within which training was obtained long after the repeal of the Statute of Artificers in 1814 which made it no longer possible to prosecute anyone practising a trade without completing a legal apprenticeship. There are cases of boys falling foul of their legal requirements as apprentices and on occasion dealing with their dissatisfactions by breaking agreements. John Buckley, for example, was arrested by the constable and taken before the magistrates for absconding from his apprenticeship. Other autobiographers were in conflict with their trade associations over failure to complete apprenticeships. William Lovett, who had served his time in the dying trade of rope making, faced difficulties in transferring to the prestigious and exclusive trade of cabinet making. Lowery too faced opposition from fellow workmen having abbreviated the normal training for a tailor. But there is also evidence as suggested above that apprenticeship could be a flexible institution providing training in exchange for labour in a variety of customised packages to suit the needs of both masters and men.39 However apprenticeship should not be confused with the child labour market. For most children work began several years before the normal age of indenture. Moreover the direct and opportunity costs of apprenticeship made the latter an impossible dream for many children.

Conclusions Despite economic historians’ attempts to shave growth rates and emphasise continuity, representations of the first industrial revolution remain heroic and empowered: ‘Prometheus unbound’. This research returns to some older disturbing images of what British industrialisation involved. Autobiographies of the period highlight the way in which work overshadowed childhood. I have not said much about the qualitative nature

Humphries

265

of this work, for example, its hours or intensity, but concentrated on its measurable dimensions. These suggest that child and youth participation rates were massively in excess of today’s Third World and probably well above child participation rates in other nineteenth-century industrialising economies. Table 11.3 Participation rates for children ten to fourteen 1950 World Africa Latin America Asia Europe Ethiopia Brazil China India Italy Britain (from bios) Britain 1851 census

1960

27.57 24.81 38.42 35.88 19.36 16.53 36.06 32.26 6.49 3.52 52.95 50.75 23.53 22.19 47.85 43.17 35.43 30.07 29.11 10.91 60-70 approximately

1970

1980

1990

2000

22.30 33.05 14.60 28.35 1.62 48.51 20.33 39.03 25.46 4.12

19.91 30.97 12.64 23.42 0.42 46.32 19.02 30.48 21.44 1.55

14.65 27.87 11.23 15.19 0.10 43.47 17.78 15.24 16.68 0.43

11.32 24.92 8.21 10.18 0.04 41.10 14.39 7.86 12.07 0.33

36.6

Source: ILO (1996), Census returns of England and Wales, and Autobiographies. (boys only)

Recently economic historians have argued that England’s ability to industrialise not only first but at a low per capita income, comparatively speaking, owed much to an intensification of labour input: to an ‘industrious revolution’ which preceded and overlapped with the industrial revolution. Crafts has suggested that the macroeconomics of the late eighteenth century suggest an increase in industrial labour input rather than a rise in labour productivity in industry, at least in the initial phase.40 De Vries has linked evidence of increased labour input to the availability of new products and new possibilities for consumption.41 Interestingly, he also emphasised the multiplication of jobs for women and children as an important component of increased industriousness. Voth has provided some concrete empirical evidence on time use, ingeniously derived from court records, to show a decline in St. Monday and in the number of annual holidays in the second half of the eighteenth century.42 The exceptionally intensive use of child labour documented in this paper accords with this new interpretation. The first industrial revolution owed much to a legal and cultural context, which permitted child labour, and to a technological environment, which fostered it. Economic historians have linked the earliest development of cotton factories to child labour (and whoever says the industrial revolution says cotton!). But the importance of child workers extended beyond manufacturers’ initial reliance on child paupers to work rural water-powered mills in the eighteenth century. The intensive use of child labour not only played a strategic role in the transition

266

Child labour

to the factory but also sustained industrial expansion well into the nineteenth century.43 Putting the child worker back at the centre of industrialisation draws attention to other factors whose importance may have been misread or neglected. For example, coincident with the first industrial revolution Britain fought a series of wars, which demanded increasing numbers of prime age males to serve in the army and navy. These circumstances left numbers of families de facto without male heads and exposed the children in them to earlier working. The demographic events of the era, along perhaps with the retreat of the Old Poor Law, also combined to boost the supply of cheap child labour. This supply was greedily seized in industries and mines. Inventors, those Promethian heroes of the industrial revolution, thought up new ways to substitute children for more expensive adults in leading sectors of industry. Elsewhere children laboured alongside adults rendering them and their machines more productive. It was a unique set of conditions and it produced an exceptional outcome. But this story is not just one about the underpinnings of the macroeconomics of the industrial revolution. The autobiographies at the heart of the study favour a history from below. The autobiographers were marked (often literally) by the events they experienced, but they also shaped the world that they encountered. This is obvious in the case of the writers who became famous as labour leaders, advocates of adult education, preachers, trade unionists, cabinet ministers and poets. Autobiographers are a selected sample. The upwardly mobile are overrepresented even if that mobility was often channelled by class background into class-specific activities. But even the humble voices from obscurity often tell us about struggles for a better life, if not for themselves then for their children. Given the costs both directly and in terms of earnings forgone, the pursuit of apprenticeships demonstrates significant self-sacrifice. Similarly although regular and extended school attendance was rare, and education was all too often crowded out by paid work, few children went completely without schooling. Finally although the occasional brutal parent (usually father or stepfather) is remembered, these figures are outnumbered ten to one by memories of caring parents and especially loving mothers. Such parents were often persuaded that early working provided useful training for children but they could distinguish between work that helped and work that harmed their offspring. While they tried to secure the former, sadly they were often forced by poverty and need to accept the latter. If industrialisation did involve an intensified use of child workers then the implications demand to be investigated. Historians have already suggested that deterioration in literacy levels in early industrialisation undermined the ability of the British economy to maintain growth rates in the later nineteenth century.44 Perhaps this argument is more general, with the intensification of child labour undermining the health and training of the workforce with adverse future implications. The cost of the

Humphries

267

country’s pre-eminence may have been much greater than even Ashley thought.

Notes 1. Roderick Floud and Donald McCloskey, The Economic History of Britain since 1700, 2 vols (Cambridge, 1994), 1. 2. Per Bolin-Hort, Family and the State: Child Labor and the Organization of Production in the British Cotton Industry (Lund, 1989). 3. Carolyn Tuttle, Hard at Work in Factories and Mines. The Economics of Child Labor During the British Industrial Revolution (Boulder, Colorado, 1999). 4. Sara Horrell and Jane Humphries, ‘ “The Exploitation of Little Children”: Child Labour and the Family Economy in the Industrial Revolution’, Explorations in Economic History, 32, no. 4 (1995), pp. 485–516. 5. Jane Humphries, ‘Female-headed Households in Early Industrial Britain: The Vanguard of the Proletariat’. Labour History Review, 63:1 (1998), pp.31–65; Horrell and Humphries, ‘Exploitation’. 6. David Vincent, Bread, Knowledge and Freedom: A Study of Nineteenth-Century Working Class Autobiography (London, 1981); John Burnett (ed.), Useful Toil: Autobiographies of Working People from the 1820s to the 1920s (London, 1974); John Burnett, David Vincent and David Mayall, The Autobiography of the Working Class: An Annotated Critical Bibliography (Brighton, 1984). 7. Out of an initial sample of 116 autobiographers, 96 (83 percent) recorded their age at starting work. 8. Despite the small sample sizes, several differences in age at starting work are statistically significant. For example, the hypothesis that the sons of tradesmen started work at the same age as the sons of domestic manufacturers can be rejected at conventional levels of significance. 9. Horrell and Humphries, ‘Exploitation’. 10. For the employment weights used to construct the aggregate figures see Sara Horrell and Jane Humphries, ‘Old questions, new data, and alternative perspectives: families’ living standards in the industrial revolution’, Journal of Economic History (1992), pp. 849–880, n. 40. The main sources used to calculate the national proportions of males in each occupation were Phyllis Deane and W. A. Cole, British Economic Growth 1688–1959 (Cambridge, 1962), p. 60 and Peter H. Lindert, ‘English Occupations 1670–1811’, Journal of Economic History, 40 (1980), pp. 702–704. 11. Tuttle, Hard at Work. 12. Christine MacLeod, Inventing the Industrial Revolution: The English Patent System (Cambridge, 1988). 13. George Elson, The Last of the Climbing Boys: An Autobiography (London, 1900). 14. Edward Allen Rymer, The Martyrdom of the Mine, or 60 Years’ Struggle for Life (Middlesbrough, 1898). 15. Rymer, Martyrdom, p.3 16. John Buckley (pseudonym of John Charles Buckmaster), A Village Politician: The life story of John Buckley (London, 1897). 17. John Bennett, Untitled Manuscript and Typescript, 36097, Bristol Record Office. 18. Thomas Wood, ‘The Autobiography of Thomas Wood 1822–1880’, in Burnett, Useful Toil. 19. Rymer, Martyrdom. p.4 20. Thomas Cooper, The Life of Thomas Cooper Written by Himself (London, 1872), pp. 9–10. 21. David Love, The Life, Adventures and Experiences of David Love (Nottingham, 1823–24), p. 3; see also Vincent, Bread, p. 65.

268

Child labour

22. Keith Snell, Annals of the Labouring Poor. Social Change and Agrarian England 1660–1900 (Cambridge, 1985); see also Joan Lane, Apprenticeship in England, 1600–1914 (London, 1996). 23. Jane Humphries, ‘English Apprenticeship: A Neglected Factor in the First Industrial Revolution’, in Paul David and Mark Thomas (eds), The Economic Future in Historical Perspective (Oxford – London, 2002). 24. Buckley, Village Politician, p. 74. 25. William Farish, The Autobiography of William Farish – The Struggles of a Handloom Weaver with some of his Writings (London, 1996 – org. 1889), p. 15. 26. Buckley, Village Politician, p.78. 27. Cooper, Life of Thomas Cooper, p. 39. 28. Joseph Arch, Joseph Arch, the Story of his Life told by Himself. Edited and with an Introduction by Frances the Countess of Warwick (London, 1898). 29. Tom Mullins, ‘Unpublished Autobiography’, in Burnett, Useful Toil. 30. Anonymous Navvy, Untitled, in Burnett, Useful Toil, p. 40. 31. Robert Lowery, ‘Passages in the life of a temperance lecturer connected with the public movements of the working classes for the last twenty years. By one of their order.’ Reprinted in Robert Lowery, Radical and Chartist (edited and with an introduction by Brian Harrison and Patricia Hollis), (London, 1979), p. 45. 32. Rymer, Martyrdom, p. 3. 33. Wood, Autobiography, p. 313. 34. William Dodd, A Narrative of the Experiences and Suffering of W[illiam] D[odd]. A Factory Cripple. Written by Himself. (London, 1968 -org.1841). 35. Lowery, ‘Passages’. p. 46. 36. Rymer, Martyrdom, p. 4. 37. Buckley, Village Politician. 38. James D. Burn, The Autobiography of a Beggar Boy. Introduced by David Vincent (London, 1978). 39. Humphries, ‘English Apprenticeship’. 40. Nick F. R. Crafts, British Economic Growth during the Industrial Revolution (Oxford, 1985). 41. Jan De Vries, ‘The Industrial Revolution and the Industrious Revolution’, Journal of Economic History, 54 (1994), pp. 249–270. 42. Hans-Joachim Voth, ‘Time and Work in Eighteenth-Century London’, Journal of Economic History, 58 (1998), pp. 29–58; Hans-Joachim Voth, Time and Work in England 1750–1830 (Oxford, 2000). 43. See also Tuttle, Hard at Work. 44. Michael Sanderson, Education, Economic Change and Society in England 1780–1870 (Cambridge, 1991).

NOTES ON CONTRIBUTORS

Harald Deceulaer works currently at the State Archives of Belgium (Brussels). He published several articles on the social history of the Low Countries, e.g., about guilds, proto-industrialisation, neighbourhoods, town-countryside relations and consumption, amongst others in the International Review of Social History (1998), Textile History (2000), History of Technology (2001) and the Revue d’Histoire Moderne et Contemporaine (2002). He published on the clothing trades in the Southern Netherlands in the seventeenth and eithteenth centuries (Pluriforme patronen en een verschillende snit. Sociaal-economische, institutionele en culturele transformaties in de kledingsector in Antwerpen, Brussel en Gent, ca 1585 – ca 1800, Amsterdam, 2001). Lina Gálvez Muñoz is Lecturer in Economic History at the University Carlos III in Madrid and Visiting Fellow at the Centre for International Business History at the University of Reading. Previously she worked in Reading (1998 to 2001) and Seville (2001). She was a member of the Unilever History Project from 1998 to 2001. She has published articles on gender and business, tobacco industry, state-owned enterprises, labour markets and multinationals. With Geoffrey Jones she published Foreign Multinationals in the United States, Management and Performance (Londen, 2002), and La Compañía Arrendataria de Tabacos: Cambio tecnológico y empleo femenino (Lid, 2000). She is currently editing a book on gender and labour market with Carmen Sarasúa (Privilegios o eficiencia, mujeres y hombres en los mercados de trabajo en España siglos XVIII-XX). Henny Gooren is affiliated with the University of Nijmegen where she participates in the same project as Hans Heger (see below). As a social and economic historian she focusses on the development of employment in the Netherlands during the first half of the nineteenth century. She published (with H. Heger) Per mud of bij de week gewonnen (Groningen, 1993) on the development of wage systems in agriculture in the Dutch province of Groningen and with H. Heger and P. Klep, ‘Labour relations, culture and fertility in Dutch agricultural communities, 1888’ in: Economic and Social History in the Netherlands (Amsterdam, 1994).

270

Notes on Contributors

Hans Heger is currently employed at the University of Nijmegen and the Dutch Institute of Scientific Information (NIWI) where he is involved in building a nation-wide database of social, economic and demographic information at the municipal level. In addition his research concerns the development of wage systems in the Netherlands during the nineteenth century. He published with H. Gooren (see above). Sakari Heikkinen is a Senior Lecturer in the Department of Social Science History at the University of Helsinki, Finland. His current research deals with Finnish economic crises in the twentieth century, seen from an institutionalist point of view. He is also interested in business history. His publications include Labour and the Market. Workers, Wages and Living Standards in Finland, 1850–1913 (Helsinki, 1997), and Paper for the World. The Finnish Paper Mills’ Association – Finnpap, 1918–1996 (Helsinki, 2000) and with Visa Heinonen, Antti Kuusterä and Jukka Pekkarinen, The History of Finnish Economic Thought, 1897–1917 (Tammisaari, 2000). Michael Huberman is Professor of History at the Université de Montréal and a research associate at the Centre interuniversitaire de recherche en anaylse des organisations (CIRANO), and the Centre interuniversitaire de recherche en économie quantitative (CIREQ). He has been a visiting fellow at the University of California, Davis; Clare Hall, Cambridge; and the Laboratoire d’Économie Appliquée (INRA), École Normale Supérieure, Paris. He is the author of Escape from the Market: Negotiating Work in Lancashire (Cambridge, 1996), and with Robert Lacroix, Le partage de l’emploi : Solution au chômage ou frein à l’emploi (Sainte-Foy, Québec, 1996). He has also written on unions and strike behaviour, and occupational segregation and the gender wage gap in Canada. He is currently researching globalisation and the rise of the welfare state. Jane Humphries has a BA in Economics from Cambridge University, and she received her Ph.D. from Cornell. She has taught economics at the University of Massachusetts and at Cambridge where she was University Lecturer from 1980 to 1993 and Reader from 1993. While working in Cambridge she was a Fellow of Newnham College, the only remaining all-female college in Cambridge, where she was also an undergraduate. In 1998 she moved to be Reader in Economic History at Oxford University and Fellow of All Souls College. Her research has focussed on women’s work and family lives both historically and in the present day. She has published many articles, which bring a feminist perspective to economics and economic history. Her most recent work is on child labour in industrialisation with particular reference to the British industrial revolution. She published the chapter ‘Household Economy’, in the forthcoming Cambridge Economic History of Britain edited by R. Floud and P. Johnson.

Notes on Contributors

271

Paul Johnson is Professor of Economic History at the London School of Economics, where he has taught since 1984. His research interests include the history of savings and welfare, the economics of old age and pensions, and the relationship between laws and markets in Victorian England. He has written or edited seven books, including Saving and Spending (Oxford, 1985) and Old Age from Antiquity to Post-modernity (London, 1998). Steven King is Chair of the History Department, Research Coordinator and Director of the Humanities Research Centre at Oxford Brookes University, England. His research interests span the period 1650 to 1920 and he writes in four major areas – the history of industrialisation, population family and kinship, poverty and welfare and medical history. He is currently working on two major projects, ‘The medical history of the north of England, 1650–1850’ and ‘The experience of being poor, 1700–1850’. Recent publications include: S. A. King and C. J. E. R. Payne (eds.), The Dress of the Poor, 1700–1850 (Pasold, 2002) and S. A. King, Poverty and Welfare in England 1700–1850: A Regional Perspective (Manchester, 2000). Craig Muldrew is a University lecturer in the faculty of History, Cambridge University and a fellow of Queens’ College, Cambridge. His research has focussed primarily on investigating the economic and social role of trust in the development of the market economy in England between 1500 and 1700. He has published The Economy of Obligation: The Culture of Credit and Social Relations in Early Modern England (London, 1998). He has also written articles in the field of legal history concerning debt litigation and its relationship to the nature of community and social relations. His most recent article is ‘Hard food for Midas’, Cash and its Social Value in Early Modern England’, Past and Present, 170 (2001), pp. 78–120. His current project involves examining the development of the concept of self-control and its effect on the structure of community as well as how banks came to be trusted in eighteenth century England. Reinhold Reith is Professor of Economic and Social History at the Department of History, and Dean of Studies at the University of Salzburg. He studied History and Political Science at the University of Constance and taught at the Universities of Berlin and Vienna. His major interest and empirical research concerns artisans, wages, labour, environment and technology. Publications include Lohn und Leistung. Lohnformen im Gewerbe 1450–1900 (Stuttgart, 1999); he edited Praxis der Arbeit. Probleme und Perspektiven der handwerksgeschichtlichen Forschung (Frankfurt, 1998). Peter Scholliers teaches contemporary social and economic history at the Vakgroep Geschiedenis, Vrije Universiteit Brussel, Belgium. His research deals with the history of the living standard in Europe in the nineteenth and twentieth centuries (wages, cost of living, food, labour, material

272

Notes on Contributors

culture). He edited Food, Drink and Identity. Cooking, eating and drinking in Europe since the Middle Ages (Oxford, 2001), and wrote Wages, Manufacturers,and Workers in the 19th-century factory. The Voortman cotton mill in Ghent (Oxford, 1996). Leonard Schwarz is a senior lecturer in the Department of History at the University of Birmingham. He is the author of London in the Age of Industrialisation. Entrepreneurs, labour force and living conditions, 1700–1850, of the chapter on eighteenth- and earlier nineteenth-century London in The Cambridge Urban History of Britain, vol. 2 and of various articles on wages, living conditions, servants and urban history in eighteenth and nineteenth-century England. Patricia Van den Eeckhout is Professor at the Vrije Universiteit Brussel where she teaches social and economic history and discourse analysis. She has published historiographical surveys of social history and contributions on family income and women’s work, housing policy, leisure, the clothing industry and social and economic policy. She edited a 1,400–page guide to Belgian archives and sources, Patricia Van den Eeckhout and Guy Vanthemsche (eds) Bronnen voor de studie van het hedendaagse België 19de–20ste eeuw (Brussels, 1999).

INDEX

absenteeism, 218–20 Addens, N.G., 140 agriculture Finland, 184 Groningen, 146–7 Alchian, Armen A., 54, 73 alehouse, 165, see also drinks Anderson, Michael, 167 apprenticeship, see workers artisans, 28, 33, 37, 46, 84, 129, 131, 155, 174, 256 Ashley, Lord, 251 Ashworth, Henry, 69 Austria – Vienna, 125, 126 bakeries, 186 bakers, 36, 96, 160,171,186 Bank of Spain, 211, 213, 214 Becker, Gary, 202 beer, see drinks Belgium, 17, see also Netherlands (Southern) social historiography, 81 Belgium – Antwerp, 12, 27–52 guilds, 27 cultural practices, 37–8, 42 members, 40, 42 monopoly, 32 nations, 31 port conflicts, 31–2 labour market, 31, 37, 38 petition, 33 social relations, 32–3 wages, 34–6, 39, 42, 45 work organisation, 30, 33, 34, 38–43 work rules, 31, 33 work segmentation, 33 processions, 42 religion, 37, 42 tax roll (1747), 36 Belgium – Bruges, 33

Belgium – Deinze, 103 Belgium – Ernage, 144 Belgium – Geeraardsbergen, 33 Belgium – Ghent, 12, 14, 32, 33, 37, 81–109 conge, see work termination Conseil de Prud’hommes, 12, 82–3, 86, 87, 88, 90, 92, 94, 96, 97, 98, 99, 102, 104 France, 88 labour market, 86 livret de travail, 84, 86–9, 93 strikes, 97, 103 transport workers – pre-industrial, 32 wage, 86, 92, 100, 103 work conflicts, 86 work contract, 87–89, 99, 103 work – custom, 82, 86, 88, 93, 99, 104 work termination, 82, 86 conflicts, 83–86, 90, 91, 93–98 initiative, 102–103 period of notice, 84, 89, 92, 95 reasons – legitimate, 93–98, 100 workshop regulations, 88, 99–102, 104 Belgium – Lillo, 32 Belgium – Lokeren, 103 Belgium – Menen, 33 Belgium – Ostend, 46 Bell Telephone Company, 212 Benedict, Ruth, 77n.64 Berg, Maxine, 207 Bernhard, Ludwig, 114 Bernstein, Edouard, 118 Best, Henry, 157 Beveridge, William, 3, 232 Bleachers, 36, 101 board – related to work, 122–126, 131, 133–134, 144, 145, 159n19, 162, 263–264, see also meals, food board fee, 125 Booth, Charles, 235

274

Boulton, Jeremy, 22n.34 Boulton, Matthew, 11 Bourdieu, Pierre, 29 Bowley, Arthur L., 3, 232 Bowles S., 74 Brants, Victor, 89 Bräuer, Helmut, 122 breakfast, see meals, food Brentano, Lujo, 115 bricklayers, 36, 192 Britain, 3, 4, 18, 58, 60, 229–49, 265, 266, see also England, United Kingdom Britnell, Richard, 158 Brown, E. H. Phelps, 6, 8, 117 229, 230 Browne, Margaret, 8 Brunner, Otto, 117 Bücher, Karl, 114 budget – household, see family income building trade, 13, 14, 90, 117–9, 122, 186 wage development, 8, 28, 36, 37, 113, 120, 125, 155, 161, 164, 187, 229, 230, see also Brown, E.H. Phelps wage disputes Germany, 128–129 workers, 163 Burnett, John, 253 Burnette, Joyce, 16, 202, 231, 240 Butchers, 60, 160 butter carriers, 32, 35, 37–8 carpenters, 36, 128–30, 165, 192, 261, 264 cash, 15–16, 155–180, see also coinage casual labour, see workers CAT (Compañía Arrendataria de Tabacos), see tobacco manufacture – Spain charity, 11, 16, 168, 173, 178n.77, 218, see also poverty, Poor Law Charles V, Emperor, 32, 32, 49nn.24, 29 Charlesworth, Andrew, 10 child labour, see workers, wage China, 58 chips, see perquisites Churchill, Winston, 68 circulation, see liquidity Clark, Gregory, 207 clothing, 37, 38, 66, 170, 238, 264 as payment, 144, 156 clothing trade, 38, 120 coal carriers, 35, 37 coinage, 10–11, 127, 160, see also liquidity coins – amount of, 158–160 collective bargaining, see labour relations Colquhoun, Patrick, 161 Compañía Arrendataria de Tabacos, see Tobacco manufacture – Spain Conseil de Prud’hommes, see Ghent Conseil Supérieur du Travail, 89 consumption, 17, 29, 38, 123, 205–8, 265 co-operatives, 66 Conrad, Johannes, 114 coppersmiths, 93

Index

cost of living, 3, 42, 123, 129, 142, see also food prices Cottereau, Alain, 104 cotton, 230, 240, see also textile trade mule spinning, see Bolton, Oldham ring spinning, see Bolton, Oldham counter-reformation, 30, 37, 42, 47 Court of Requests (or Conscience), 164, 171–3 Crafts, Nick, 21n.17, 265 crane men, 38, 44 credit, 5, 8, 11, 15, 53, 156, 164–8, 171–4 Crowley, Ambrose, 174 cultural goods, 30 culture, 9, 28–9, 33, 42, 45, 54–5, 130, 165, 218 currency, see coinage, liquidity custom, 4–5, 15, 60, 81–4, 86, 88, 89–93, 98, 99–101, 103–4, 116, 118, 142, 155, 203, see also wage Czechoslovakia – Prague, 125, 126 Czechoslovakia – Brno, 126 Daelemans, Frank, 144 debts, 18, 156, 161, 163–4, 167, 172–4 Deceulaer, Harald, 12, 19, 105, 269 Demsetz, Harold, 54, 73 Denis, Hector, 89, 103 differentiation, see wage differentials Dirlmeier, Ulf, 122 dispute, 27, see also strikes, wage Ditton, Jason, 142 dockyards, 10, 11 Dow, Gregory K, 53, 59 drinks, 38, 42, see also guilds, alehouse, wagenon-monetary form as means of payment, 6, 8, 44, 123–5, 128, 146, 155, 162, 172 Duplessis, Robert, 13 Dyer, Christopher, 161 earnest money (“handpenning”), 144 earnings, see wage, wages economic actors, 115 economic history – culture in, 28 economics – institutional 29, 44 economy of makeshifts, 16, 155–156, 166–174 economy – world, 29 education, 235, 238 Eeckhout van den, Patricia, 10, 12, 19, 23n.49, 123, 272 Eisenberg, Christiane, 119 Ellison, Thomas, 61 Elsas, Moritz J, 3 Elson, George, 259 embezzlement, 167, see also perquisites, wages empathy (in Lancashire cotton mills), 63, 68 Engelhardt, Ulrich, 130 engineers, 14, 58, 65, 66, 93, 96, 97, 100,

Index

101, 212, 217, 238 England, 10, 15, 17, 18, 44, 86, 122, 155–180 England – Ashton, 62, 72 England – Bath, 172 England – Birkdale, 167 England – Birmingham, 164, 165, 173 England – Blackburn, 64 England – Bolton, 12, 13, 53–80 blacklisting, 67 economic performance, 55–58, 64 industrial relations, 54, 65 mentality, 63 paternalism, 66 risk aversion, 62–63 spinners’ Union, 61, 64 strikes, 64 technology – spinning, 71–72 wages, 56, 62 wage system, 54–55 work organisation, 62 workers – characteristics, 62 England – Bristol, 171 England – Calverley, 167 England – Cumberland, 169 England – Devon, 10, 234 England – Deptford, 155, 174 England – Derby, 234 England – Durham, 164 England – Essex, 170, 230 England – Exeter, 10 England – Garstang, 170 England – Gloucestershire, 142, 145, 234 England – Halifax, 168 England – King’s Lynn, 177n.65 England – Lancashire, 12, 13, 53–80, 160, 161, 163, 164, 170, 230, 238–244, 245 technology in textiles, 58 wage differentials, 230 systems, 55, 230 England – Leeds, 234 England – London, 18, 44, 46, 160, 164, 172, 235, 240–244 England – Longton, 170 England – Manchester, 164, 165 England – Myddle, 173 England – Newcastle, 177n. 65 England – Norfolk, 157, 234 England – North Shields, 236 England – Oldham, 12, 13, 53–80 Building & Manufacturing Co., 59, 66 Cotton Spinners’ Union, 59, 66 economic performance, 55–58, 64–65 industrial relations, 54, 65 loans (as deposit accounts), 60 Master Cotton Spinners’ Association, 59 mentality, 63 rates of return – limiteds, 61 risk aversion, 62–63 strikes, 65

275

Sun Mill Co., 59 technology – spinning, 70–72 wages, 56, 62 wage system, 54–55 work organisation, 62 workers – characteristics, 62 England – Paisley, 234 England – Rochdale, 72 England – Rossendale, 164, 165, 167, 169 England – Shropshire, 173 England – Somerset, 163, 177n.47, 234 England – Spitalfields, 246 England – Stockport, 62 England – Westmoreland, 168, 169, 170, 178n.76 England – Wiltshire, 238, 239, 240, 242–244 England – Yorkshire, 160, 162, 164,169, 176n.28 England – Yorkshire – West Riding, 161 Enlightenment, 30 Enquête du Travail (Belgium, 1886), 88–89 entitlements, 142, 144, 148, 155–6, 161, 173, see also perquisites entrepreneurs, 28–9, 32, 42, 44, 46, 115, 161, 164–7, 174 Everitt, Alan, 161 exchange, see monetisation factory and social legislation, 10, 66–7, 81, 86–9, 104, 169, 246 family economy, 18, 155, 167, 171, 202, 205, 215, 220, 244 family income, 5, 17, 18, 142, 167, 253, 258–259, 263 family life cycle, 18 farmhands, see servants Farnie, Douglas A., 60, 64 Feinstein, Charles, 21n.13, 230 financial revolution, 159 Finland, 17, 183–199 Finland – Helsinki, 187, 188 Finland – Ostrobothnia, 187 Finland – Oulu, 187, 188 Finland – Satakunta, 187 Finland – Vaasa, 187, 188 Finn, Margot, 172 fiscal tobacco monopoly (Spain), 205–209 food, 123–124, 145, 155, 163, 172, see also meals conflicts, 8–9, 28, 46, 123, 230, 264 manufacturing, 120, 133 prices, 9, 28, 42, 123, 127, 158 food – as wage, see wage – non-monetary Ford, Henry 219 Foster, John, 65, 78n.73 France, 3, 8, 11, 14, 122, 123, 127, 217, 225n.46 France – Marseille, 52n.116 France – Paris, 155, 175n.6

276

France – Toulon, 11 free riding, 18, 67 Galvez-Muñoz, Lina, 11, 12, 18, 19, 206, 269 gender, see workers, wage as analytical tool, 201, 204 Germany, 15, 113–138 Germany – Augsburg, 124, 127, 128, 130, 131 Germany – Baden, 133 Germany – Bavaria, 131 Germany – Berlin, 124, 126, 127, 131 Germany – Bremen, 122, 124, 127 Germany – Breslau, 132 Germany – Brunswick, 123, 125, 129 Germany – Cologne, 127 Germany – Constance, 118 Germany – Dresden, 125, 127, 132 Germany – Frankfurt, 125, 127, 129, 132 Germany – Hamburg, 119, 126, 127, 130, 131, 132 Germany – Hanover, 123, 125 Germany – Königsberg, 126 Germany – Leipzig, 124, 125, 126, 128, 132, 133 Germany – Mannheim, 133 Germany – Munich, 125, 126, 130 Germany – Nuremberg, 127, 128, 130, 131 Germany – Prussia, 126 Germany – Regensburg, 126 Germany – Saxony, 122 Germany – Speyer, 118 Gewinnprinzip, 116 Gintis, H., 74 giving notice, see work termination glass manufacture, 185, 186, 188, 189, 190, 192 gleaning England, 169 Groningen, 145 Goldin, Claudia, 186, 194 Gooren, Henny, 13, 15, 16, 19, 269 Gordillo, Rodríguez, 206 Gough, Richard, 173 Gowler, Dan, 140, 141, 148 grain porters, 35 Granovetter, M., 30 Griessinger, Andreas, 119, 126 guilds, 5, 19, 27–8, 87, 116, 118, 123, 128, 130, 132, see also Belgium – Antwerp drinking, 27, 33 feasting 27–8 Gustafsson, Bo, 29 Haines, Michael, 234, 246 Hatton, Tim, 142 Heger, Hans, 13, 15, 16, 19, 270 Height, 5, 17 Heikkinen, Sakari, 17, 270 Herzig, Arno, 119

Index

historical school of German economics, 113–114, 134 Hobsbawm Eric, 4, 8, 9, 12, 13, 118, 119, 203, 223n.19 Hopkins, Sheila, 6, 8, 117, 229, 230, 245 Horrell, Sara, 17, 169, 234, 253, 256, 258 household, see family Huberman, Michael, 12, 13, 23n.49, 222n.16, 270 Hudson, Pat, 21n.22 human capital, 202, 236, 244 Humphries, Jane, 17, 18, 19, 169, 256, 270 Hunt, Edward H., 229, 234 Hutton, William, 165, 172 identity, 27, 28, 29 ideology, 17, 29, 30, 47 incentives, see wage system India, 58 industrial disputes, 19, see also strikes industrial psychology, 142 industrial revolution, 4, 17, 18, 52, 167, 174, 196, 202, 246, 251, 258, 261, 264–6 industrialisation, 5, 7, 13, 15, 18, 183, 187, 195, 229, 230, 251, 256, 264 ‘industrious revolution’, 17, 18, 265 inflation, 8, 9, 119, 123, 127, 129, 160, 219, 229, 238, see also prices institutional change, 28 International Committee on Price History, 3 Ireland, 58 Italy, 122, 225n.46 Italy – Venice, 11 Japan, 71, 77n.64 Johnson, Paul, 18, 270 Joint Stock Acts, 1855 and 1862 (Britain), 59 Jones, S.R.H., 207, 224n.29 Joyce, Patrick, 65 Kaelble, Hartmund, 118 Kampffmeyer, Paul, 118 Kaschuba, Wolfang, 119 Kautsky, Karl, 118 Keynes, J. Maynard, 20n.3 King, Peter, 142 King, Gregory, 157, 161 King, Steven, 9, 11, 12, 15, 271 Knotter, A., 21n22 Kodak, 64 Kreps, David M., 68 Kruse, 61 Kuczynski, Jürgen, 3, 20n.3 Kussmaul, Ann, 157, 162 labour, see work Labour Court, 82, see also Ghent, Conseil de Prud’hommes labour market, 5, 8, 9, 10, 13, 14, 15, 16, 18, 19, 30, 31, 37, 38, 86, 116, 117,

Index

118, 122, 129, 130, 143, 147, 160, 166, 167, 170, 173, 195, 201–3, 205, 217, 221, 229, 231, 234–6, 240, 244, 253, 256, 258, 259, 260, 261, 264, see also workers, unemployment internal labour market, 248n.40 labour movement, 82, 118 labour relations, 10, 12, 13, 81, 87, 88, 90, 93, 101, 104, 139, 140, 144, 208, 214, 220, 237, see also strikes labourer, see worker origin of word, 157 Labrousse, Ernest, 119 Law, John, 157, 159, 178n.59 Lazear, E. P., 63 leisure, 27, 66, 115–6, 204, see also St. Monday Le Play, Frédéric, 89 Leunig, Tim, 72 Levasseur, Emile, 3 limited liability companies, 54–55, see also Oldham Lindert. Peter H., 79n.97, 161 Linebaugh, Peter, 155, 174 Lipartito, Kenneth, 212 liquidity, 156, 158–166, 168, see also coinage, coins literacy, 5, 236, 266 Lloyd George, 75n.24 Loader, Robert, 157, 162 Locke, John, 159, 164, 165 lockout, 215 machinery manufacture, 14, 65, 92, 93, 101, 186, 188, 189, 190, 192 McCloskey, D., 74n.7 MacLeod, Christine, 258 maids, see servants male breadwinner ideology, 17, 202, 205 managerial strategy, 16, 18 mantua makers, 172–173 Marcroft, William, 66 market economy, 10, 31, 161, see also monetisation Marxist economic model, 212, 226n.56 masons, 39, 128, 130 Massie, Joseph, 161 Master and Servant Act, England (1823), 86 Mathias, Peter, 160 Mayall, David, 253 meals, 6, 8, 15, 31, 42, 123–6, 146, see also food, board Meistergroschen, see Germany, coinage mentality, 29, 115–6 metal trades, 90, 120, 185, 238, 241, 256, 261 migration, see workers mobility Mitchell, Dr. James, 233 Mokyr, Joel, 17 monetisation, 158, 161 money, see liquidity, cash, coinage

277

moral economy, 8–10, 28, 230 mortality, 5, 42, 65 Muldrew, Craig, 9, 10, 11, 12, 15, 271 Nahrungsprinzip, 116 Napoleonic Code, civil, 87 navy – British, 166, 175 neo-classical economic theory, 19, 203, 204, 212, 231 Netherlands, 11, 15, 44, 46, 122 Netherlands – Amsterdam, 32, 44, 46 Netherlands – Groningen, 13, 15, 15, 139–153, 149 Netherlands – Groningen– Fivelingo, 145–147, 149 Netherlands – Groningen – Hunsingo, 145, 147, 149 Netherlands – Groningen – Oldambt, 145–147, 149 Netherlands – Groningen – Veenkoloniën, 146, 147, 149 Netherlands – Groningen – Westerkwartier, 147, 149 Netherlands – Groningen – Westerwolde, 145, 147, 149 Netherlands – Groningen –Woldstreek, 149 Netherlands – Middelburg, 32 Netherlands – Southern, 30, 31, 32, 33, 42, 44 Netherlands – Vlissingen, 32 Netherlands – Zealand, 32 networks, 11, 16, 45, 65, 156, 160, 165, 166, 167, 170, 174, 215 Neufeld, Michael J., 130 New Survey of London Life and Labour, 235 North, Douglas, 29 nuptiality, 58, 202, 229 nutrition, 5, see also Food Oxley, Deborah, 234 paid holiday, 15, see also social benefits paper manufacturing, 120, 186, 188, 189, 190, 192 Park, Rowland, 163 paternalism, 64 patriarchy, 230 paupers 12, 69, 156, 160, 170, 174, 258, 265, see also Poor Law, poverty, Speenhamland pay system, 54, 120–122, see also wage system in connection to food, 124 historical development, 13–14 payment – in kind, 5, 14, 15, 23, 24, 44, 122–123, 128, 129 131, 137 n.70, 144–146, 148, 158, 161–165, 168, 171, see also perquisites, wage – non-monetary wage, truck system payment – tariff lists, 13 pension, 15, 71, 169–70, see also Social benefits

278

Pepys, Samuel, 158, 166 perquisites, 4, 6, 10, 15, 142, 166–7 Petty, William, 158, 159, 177n.59, 245 Philippovitch von, Eugen, 114 piece-rates, see wage system piecework, see wage System Pittomvils, Kathlijn, 82, 86, 105 Pollard, Sidney, 227n.58 Poor Law, 11, 16, 64, 68–70, 160, 163–6, 169–71, 173, 245, 252, 266 poor relief, 6, 11, 12, 55, 66, 73, 160, 166, 168–70, 174, 178n76, 259, 260, port labour, see Antwerp Postel – Vinay, Gilles, 21n.22 Posthumus, Nicolaas W., 3 poverty, 11–12, 155, 168, 173, 235, 259, 266, see also paupers, Poor Law, Speenhamland premiums, see wage system Pribram, Alfred F., 3 prices, 3, 6, 9, 10, 13, 28, 42, 45, 61, 113, 119, 122–4, 125, 127–9, 158, 169, 171, 229, 238, see also cost of living, food prices Prins, Adolphe, 89 printing industry, 90, 96, 120, 186, 188, 189, 190, 192, 193, 194 productivity, 8, 14, 16, 17, 53, 54, 56, 115, 121, 174, 191, 195, 196, 201, 202, 214, 215, 217, 229, 235, 238, 244, 245, 265 profit sharing – textile industry, Lancashire, 54 protest, 7, 28, 46, 118, 211, 218, 220, see also strikes, conflicts, dispute Prothero, Iorwerth, 21n.22 pulp and paper industry, 185 Putterman, Louis, 53, 59 putting-out entrepreneurs, 164, 167, 224n.34 Randall, Adrian, 10 Reddy, William, 9 Reed, Mick, 163 Reininghaus, Wilfried, 118 Reith, Reinhold, 9, 13, 15, 19, 22n.34, 271 rents, payment of, 35 representation – cultural, 36–38 Ricardo, David, 22n.30 Royal Commission on Labour (1892), 59 Royal Commission on the Employment of Children in Trades and Manufactures (1843), 236 Rule, John, 8, 10, 203, 207 Ryder, Sir Dudley, 165 St. Monday, 124, 265, see also leisure Sarasùa, Carmen, 21n.22, 197 sawmill industry, 186, 188, 189, 190, 192 Schanz, Georg, 118 Scheldt (river), 32, 38

Index

Schmoller, Gustav, 14, 114, 115, 116,118 Schoenlank, Bruno, 118 Scholliers, Peter, 4, 197, 271 Schotter, Andrew, 73 Schwarz, Gerhard, 122 Schwarz, Klaus, 119 Schwarz, Leonard, 4, 15, 155, 197, 272 Scotland – Aberdeen, 234 Scotland – Glasgow, 234 Sears, 64 second industrial revolution, 202 self-exploitation, 17 servants in agriculture, 148–50, 156–7, 161–3, 172, 184, 186, 187, 191 domestic, 49n.29 living-in, 143–145 non-resident, 145 servants in Antwerp port, 36 Shakespeare, William, 157 Sharpe, Pamela, 10 shoemakers, 125 silver, see coinage Simiand, François, 3, 20n.3 Smith, Adam, 157 Snell, Keith, 169 soap boilers, 32 social benefits, 6 social capital – textiles, 58 Soly, Hugo, 19 Sombart, Werner, 114, 115, 116 Sonenscher, Michael, 8, 14, 155, 175n6 Spain, 11, see also tobacco manufacture Spain – Alicante, 215 Spain – Cádiz, 215 Spain – Seville 11, 201–227 Speenhamland system, see Poor Law, poverty, paupers spinners, 54, 100, 102, 163–4, 230, 233, see also Bolton, Oldham, textile trade standard of living, 4, 5, 17, 18, 113, 252 measurement, 4, 6, 17, 113 starvation, 42 State – role in economy, 10–11 Steffens, Sven, 105 Steinfeld, Robert, 86 Stinchcombe, Arthur L., 245 strikes, 5, 10, 66, 81, 97, 102–3, 117–9, 122, 124, 125, 126–7, 128–30, see also Bolton, Oldham, wage conflict Stürmer, Michael, 119 subsidy (England, 1525), 157 subcontracting, 6, 11, 31, 36, 38–43, 46, 192, 222n.13 tailors, 13, 97, 119, 120, 124, 127, 130, 132, 163, 172, 264 taste, 29, 216 Taylor, Frederick Winslow, 114, 140, 219 Taylorism, 219, 222n.13 technology, 7, 58, 70, 71, 121, 202, 208,

Index

212, 218, 221, 237, 238, 244–6, 258, see also Bolton, Oldham, tobacco manufacture textile trade, 120, 122, 125, 164, 171, 185, 186, 188, 189, 190, 192, 193, 207, 233, 238–240 Thompson, Edward P., 8, 9, 82, 223n19, 230 timber, 163, 175, 185, 192 time wage, see wage system tobacco consumption, 205–206, 208, 210, 215–217 tobacco manufacture Finland, 185, 186, 188, 189, 190, 192, 193 Spain, 11, 18, 206–207 ama de rancho, 209 Bank of Spain, 213, see also financing cigarreros, 206, 208, 210 cigarreras, 204, 206, 208, 209, 210, 214–216, 218 Compañía Arrendataria de Tabacos, 11, 204, 205, 207, 211–220 financing, 211–212 fraud, 210 gender, 208, 209 maestras, 209 management, 213 mode of production, 208, 214 organisation, 207–209, 211–212, 214 porteras, 209 tabaqueros, 209, 210 taller, 209 technology, 210, 213–218 Treasury, 210, 218, see also fiscal tobacco monopoly wage system, 209–211, 214 work relations, 208, 214–215 workers, 211, 215 tokens, see coinage Toms, S., 61, 77n.47 Torres, Garcìa de, 211 transport workers, 27–52 Treasury – Spanish, truck system, 16, 165, see also payment in kind Tuttle, Carolyn, 252, 258 Tyszka von, Carl, 3 unemployment, 7, 63, 71, 119, 203, 204, 224n.35, 229, 235 benefits, 63, 71 unions, see workers United Kingdom, 55, 60, 68 United States of America, 53, 64, 71, 186, 194, 222n.15, 225n.46, 226n.49, 236 United States of America – Michigan, 142 Verbauwen, Pol, 104, 105 Verein für Socialpolitik, 134

279

Victoria (Queen), 259 Vincent, David, 253, 254 Volkmann, Heinrich, 118 Vries de, Jan, 17, 22n.34, 117, 265 wage, see also Antwerp, Bolton, building trade, Ghent, Oldham, tobacco manufacture age, 7, 16, 18, 34–36, 238–246 conflicts, 15, 19, 33, 82, 116, 120, 122, 126–128, 130, see also strikes, disputes, work termination custom, 4, 7, 8–10, 12, 18, 117, 187, 195–196, 203, 229, 231, 244 data, 113, 187, 229, 232–3 determination, see wage formation development, 24n.68, 119, 139 differentials, 7, 8, 17, 18, 139, 183–4, 187–191, 196, 201–2, 229, 231, 235–46 discourse, 121 discrimination, 16–7, 191, 195, 201–2, 230, 231 dispute in court, 19, 172 efficiency wage, 54, 68, 114, 122 form, 3, 5, 7, 13–6, 113, 120–3, 131, 132, 139, 142, 143, 144, 148, 156–66, 203, 244, 246 evolution, 3, 8, 28, 113, 139, see also wage development formation, 8–12, 28, 34, 114–6, 156, 183, 236–7 gender, 7, 16, 144, 183–199, 201–221, 230–231, 238–244 historiography, 3–5, 35, 113–114, 117–118, 139 indirect, 15 invisible, 142 labour market, 5, 8–10, 18, 116–117, 122, 195–196, 203, 229, 330 labour (wage), 7, 15, 86, 115, 140, 142, 143, 147, 149, 157, 158, 162 litigation, 12, 32 minimum wage, 219, 237 non-monetary wage, 5, 6, 14–15, 122–124, 129, 137n.75, 140, 142, 145, 156, 162–163, 168, 171, see also payment in kind, board penalties, 14 poor relief and wages, 12 rate busting, 14 significance, 6, 156–7, 162, 169 skill, 10, 18, 196, 213, 229–31, 236 stable in the long run, 8, 18, 117 survey, 232–235 system, 3, 6, 13, 16, 53–5, 59, 73, 113, 139–41, 148, 209–11, 214 by results, 13, 115, 191, 203 employer’s choice, 201–203 gender, 183, 191–196, 209–211 incentives, 14, 15, 54, 67–8, 114, 140,

280

145–6, 215 methodology, 3, 4–5, 35–6, 45, 113, 142 piece rates, 113–114, 121, 126, 131–133, 192–193, 201–221, 207, 218–220, 244 premiums, 14, 116, 120, 121, 140, 203, 222n.10, 237, 245, 262 real, 4–5, 14, 17, 28, 46, 119, 123, 128, 129 terminology, 157 theory, 113–117, 140–141, 195, 234 time rates, 13, 113–114, 121, 132–134, 146, 191–193, 203, 244 wages – comparison – international, 5 Wall, Richard, 169 War – First World, 8, 55, 62, 218, 229 War – Napoleonic, 160, 208 War – Religious (1578–1585), 42 War – Second World, 4 War – Seven Years’, 121, 127, 132 War – Thirty Years, 128 weavers, 8, 10, 87, 93, 99, 101, 102, 103, 118, 127, 128, 133, 163, 164, 233, 235–6, 246, 256, see also Bolton, Oldham, textile trade Webb, Sidney. and Beatrice, 117, 118 Weber, Max, 29, 77n.64 Williamson, Jeffrey G., 142, 161, 230, 231 Williamson, O., 207 Wissell, Rudolf, 118 women, see workers, wage wood porters, 44 woodworkers, 92, 120 work, see also Antwerp, Bolton, Germany, Ghent, Oldham, strikes complaints, 44, 82–4, 87, 93, 97, 102, 103, 161 contract, 11, 38, 54, 62, 86–89, 93, 96, 97, 98, 99, 103, 116, 120, 125, 126, 129, 132, 157, 191, 192, control, 15, 19, 67, 130, 139–153, 207, 213 organisation, 15, 19, see also Antwerp port regulations, 19, 33, 87–90, 88–104, 104–5, 132, 133

Index

representations, 30 termination, 81, 89–93, see also Ghent social history, 6 workers, see also under specific occupations, Antwerp, Artisans, Bolton, Ghent, labour relations, Oldham age, 189–190, 236, 254 apprenticeship, 36, 94, 167, 216, 220, 223n.18, 245, 252, 253–4, 261–264, 266, autobiographies, 253–6, 258, 260, 261, 262–4, 265 children, 18, 236, 246, 251–268 earnings, 262–264 father’s occupation, 256, 259 participation rates, 252–253, 265 starting work, 254–260 work experience, 262 casual, 15, 31, 35, 36, 38, 39, 143, 146, 148, 187, 252, 256 congé, see quitting female and male employment, 184–7 firing, 19, 81, 98–101, 104, 218 harbour, see Belgium – Antwerp hiring, 5, 31, 67, 86, 87, 100, 162, 205, 217, 218, 254 immigrants, 58 mobility, 7, 117, 122, 130–1, 231 number of, 161, 183, 197n2 penny clubs, 165 quitting, 12, 81–109 skills, 203 social status, 6 unions, 5, 8, 14, 59, 60, 61, 64–70, 117, 118, 140, 192, 201, 205, 210, 214, 219, 220, 231, 232, 237, 266 violence, 33 women – historiography, 230 working time, 3, 16, 115, 128, 129, 191, 214, Wood, H., 232 Woodward, Donald, 6, 155, 162, 163 Wordie, J. Ross, 158 Woude van der, Ad, 22n.34 Zwiedineck-Südenhorst von, Otto, 114, 116