On the Verge of Convergence: Social Stratification in Eastern Europe 9789633864906

Citation preview

ON THE VERGE OF CONVERGENCE

ON THE VERGE OF CONVERGENCE Social Stratification in Eastern Europe HENRYK DOMAŃSKI

.. � ' � ►

CEU PRESS

◄ ' �

Central European University Press

First published in Polish as "Na progu konwergencji" by

Wydawnictwo IfiS PAN, Warszawa, 1996 English edition published in 2000 by

Central European University Press Október 6. utca 12 H-1051 Budapest Hungary 400 West 59th Street New York, NY 10019 USA

© 1996 by Henryk Domański and Wydawnictwo IfiS PAN Distributed in the United Kingdom and Western Europe by

Plymbridge Distributors Ltd., Estover Road, Plymouth, PL6 7PZ United Kingdom

All rights reserved. No part ofthis publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the permission ofthe Publisher. ISBN 963-9116-81-5 Cloth ISBN 963 9116-82-3 Paperback ISBN 978-963-386-490-6 ebook Library of Congress Cataloging in Publication Data A CIP catalog record for this book is available upon request

Printed in Hungary by Akaprint, Budapest

CONTENTS

1 2 3 4 5 6 7 8

List of Tables

vii

List of Figures

x

lntroduction: The Constrained Plurality of Stratification Systems Peasant Societies-Market Societies: The Touch of Modemization Two Transformations and Social Mobility Social Mobility Pattems: A Basic Continuity Economic Stratification: Similarities, Differences, and Emerging Change The 'Owners' Debate: Nomenklatura or Self-Recruitment? Income Distribution Culture and Lifestyle Religion-A Stage on the Road to Modernization?

69 91 107 129 145

Conclusions: Modem and Traditional Social Structures

159

Appendix

167

References

173

Index of Authors

183

1 9 25 45

LIST OF TABLES

Table 1. Distribution ofsocio-occupational categories for father's father, father, respondent's first job, and respondent's occupation in 1988 and in 1993. Men. Bułgaria Table 2. Distribution ofsocio-occupational categories for father's father, father, respondent's first job, and respondent's occupation in 1988 and in 1993. Men. Czech Republic Table 3. Distribution of socio-occupational categories for father's father, father, respondent's first job, and respondent's occupation in 1988 and in 1993. Men. Hungary Table 4. Distribution of socio-occupational categories for father's father, father, respondent's first job, and respondent's occupation in 1988 and in 1994. Men. Poland Table 5. Distribution of socio-occupational categories for father's father, father, respondent's first job, and respondent's occupation in 1988 and in 1993. Men. Russia Table 6. Distribution ofsocio-occupational categories for father's father, father, respondent's first job, and respondent's occupation in 1988 and in 1993. Men. Slovakia Table 7. Total mobility rates. Percentages ofmobile men in 19481952 and 1952-1963, and in 1983-1988 and 1988-1993 Table 8. Results of fitting models to three-way men's tables of origin category by destination category by time (1948-1952 and 1952-1963) Table 9. Results offitting models to three-way men's tables of origin category by destination category by time (1983-88 and 1988-93) Table 10. Men's inflow rates to the intelligentsia, lower nonmanuals, owners, skilled workers, unskilled workers, and farming Table 11. Coefficients ofcanonical correlations between occupational categories offathers and sons/daughters in 1988 and 1993

17 17 18 18 19 19 33 36 37 41 54

viii Table 12. Discriminant weights on first and second discriminant functions for tables of mobility from fathers' to sons' categories in 1988 Table 13. Discriminant weights on first and second discriminant functions for tables ofmobility from fathers' to sons' categories in 1993 Table 14. Mean monthly family incomes per capita (in US dollars) Table 15. Mean monthly individual incomes (in US dollars) Table 16. Ownership ofmotor vehicle, separate deep-freeze, microwave oven, personal computer, satellite receiver, telephone. Synthetic index ofmaterial position (100- owning all 6 items, O - none) Table 17. Ownership ofsavings Table 18. Ownership of shares Table 19. Logistic regression analysis ofbeing an owner in 1993. Effects ofmembership of nomenklatura in 1988, ownership of firm in 1988, years of schooling, and sex Table 20. Logistic regression analysis ofbeing an owner in 1993. Effects offather's father's and father's ownership offirm Table 21. Multiple regression analysis ofincome from main job. Metric coefficients divided by grand mean ofincomes x 100 Table 22. Monthly incomes of men and women, and gender gap Table 23. Monetary returns for graduates ofuniversity, some high schools, and secondary, basie vocational, and elementary schools. Metric coefficients in multiple regression analysis divided by grand mean x 100 Table 24. Monetary returns for graduates ofuniversity, same high schools, and secondary, basie vocational, and elementary schools in the private and public sector. Metric coefficients in multiple regression analysis divided by grand mean x 100 Table 25. Multiple regression analysis ofincome from main job. Coefficients of semipartial correlation Table 26. Going to libraries, museums, opera, theater, listening to classical music, reading serious books. Means ofsynthetic index of cultural participation Table 27. Multiple regression analysis ofrespondent's occupational status on cultural participation and selected variables. Partial correlation coefficients and zero-order correlations Table 28. Number of books

56 57 70 78 82

86 88 97 103 11O 113 124

125

126 133 134 136

lX

Table 29. Multiple regression analysis ofcultural participation. Partial correlation coefficients Table 30. Means ofincome, education, occupational prestige, cultural participation, and political involvement among atheists and religious denominations. Bułgaria Table 31. Means ofincome, education, occupational prestige, participation in culture, and political involvement among atheists and religious denominations. Czech Republic Table 32. Means of income, education, occupational prestige, cultural participation, and political involvement among atheists and religious denominations. Poland Table 33. Means of income, education, occupational prestige, cultural participation, and political involvement among atheists and religious denominations. Slovakia Table 34. Changes in religiosity. Bułgaria, Czech Republic, Poland, Slovakia

139

Table Al. Percentage of car owners Table A2. Standardized effects ofnomenklatura membership in 1988, ownership of firm in 1988, education, sex, age, and size of place ofresidence in logistic regression ofbeing an owner in 1993-94 Table A3. Standardized effects in logistic regression ofbeing an owner in 1993-94 Table A4. Standardized effects in logistic regression ofbeing an owner in 1988 Table AS. Religion in childhood by religion in 1994. Poland Table A6. Religion in childhood by religion in 1993. Bułgaria Table A7. Religion in childhood by religion in 1993. Czech Republic Table A8. Religion in childhood by religion in 1993. Slovakia

170 170

147 148 149 149 154

171 171 171 172 172 172

LIST OF FIGURES

Figure 1. Bułgaria, 1988. Discriminant weights for sons' categories in 1988 on first and second discriminant variate Figure 2. Bułgaria, 1993. Discriminant weights for sons' categories in 1993 on first and second discriminant variate Figure 3. Czech Republic, 1988. Discriminant weights for sons' categories in 1988 on first and second discriminant variate Figure 4. Czech Republic, 1993. Discriminant weights for sons' categories in 1993 on first and second discriminant variate Figure 5. Hungary, 1988. Discriminant weights for sons' categories in 1988 on first and second discriminant variate Figure 6. Hungary, 1993. Discriminant weights for sons' categories in 1993 on first and second discriminant variate Figure 7. Poland, 1988. Discriminant weights for sons' categories in 1988 on first and second discriminant variate Figure 8. Poland, 1994. Discriminant weights for sons' categories in 1994 on first and second discriminant variate Figure 9. Russia, 1988. Discriminant weights for sons' categories in 1988 on first and second discriminant variate Figure 10. Russia, 1993. Discriminant weights for sons' categories in 1993 on first and second discriminant variate Figure 11. Slovakia, 1988. Discriminant weights for sons' categories in 1988 on first and second discriminant variate Figure 12. Slovakia, 1993. Discriminant weights for sons' categories in 1993 on first and second discriminant variate

58 58 58 59 59 59 60 60 60 61 61 61

IN TRODUC TION

THE CONSTRAINED PLURALITY OF STRATIFICATION SYSTEMS

THE CONCEPT of the 'postcommunist society' concisely depicts a com­ mon heritage. This book has been written with the intention of consider­ ing how long this term will continue to encapsulate really common fea­ tures. I ask whether, in the first half of the 1990s, there were still simi­ larities between Bułgaria, the Czech Republic, Hungary, Poland, Russia, and Slovakia. I analyze the social structure of these six countries using survey data from research carried out in 1993 within the ambit of an intemational project which was designed to grasp the social transforma­ tions taking place in Eastern Europe after 1989. In the Appendix, I pro­ vide detailed information on the project, and discuss the strict compa­ rability of the data collected. Since, in all countries, the surveys covered random national samples of adults, there are solid grounds for believing that the diagnosis and conclusions presented in this book are reasonably comprehensive and valid. What I shall discuss concems cross-national similarities and differ­ ences along the pivotal axes of social stratification and macrostructure. I will focus on pattems of occupational movements, social fluidity, and rigidities and distances between basie socio-occupational strata. Taking these components into account, I will attempt to outline a stratification ladder in East European societies. The question must be faced: To what extent do the intelligentsia, lower nonmanuals, private entrepreneurs, skilled and unskilled workers, and farmers tend to hold similar positions in the hierarchies of income, materiał standards of living, cultural par­ ticipation, and so on? How much do they differ across our six nations? Although I refer to data snapshots at a given point in time, my analy­ ses address not only the statics of social stratification but also tendencies to transformation over time. Insofar as social dynamics are concemed,

2 I attempt to go beyond a narrow East European context. Changes are looked at from two standpoints: (i) the convergence of societies and (ii) their emerging differentiation. From its very inception, the notion was passionately contested from various theoretical positions. If famous names count in terms of signifi­ cance and popularity, one need only read the roll-call of researchers who claimed a growing similarity between Eastern Europe and the West: Aron, Bendix, Lipset, Zetterberg, Dahrendorf, Bell, and Galbraith. In the decades which followed, the theory of convergence gained further support (see Treiman, 1977). The adversarial camp included students of stratification in communist societies. What, then, is the specific contribution which I hope to make in the following chapters? Certainly, the watchword of 'convergence' refers to the postcom­ munist societies which are approaching the standards of the Westem world. I look at the question of convergence, examining how well it fares in the light of the evidence, but also in relation to the revelations of previous studies. The thesis to the effect that convergence is going on between different political, economic, and cultural systems was formu­ lated in the 1950s. Convergence theories of the 1960s and 1970s pre­ dicted that the two rival political and economic systems would inevita­ bly move towards and assimilate one another. The communist East was to be enriched with market elements, while the economic order of West­ em capitalism had already adopted elements of state intervention in pro­ duction and distribution. In fact, both Eastern and Westem societies be­ carne more 'mixed' , although the process was more advanced in the lat­ ter. Concessions were made to political liberalization, national inde­ pendence, decentralized forms of ownership, a competitive economy, and the foundation of self-governing bodies in enterprises, fostering 'economic democracy'; nevertheless, Western 'admixtures' in terms of economic and political organizations were not allowed or were with­ drawn. Hungary and Poland were the only countries which managed to win moderate reform measures. The rapid flow of events which accompanied the upheavals of 198991 made the transition of East Central European countries to Western patterns a real issue. Economically and politically, these countries have no choice but to copy the institutional framework of the capitalist world. They are dependent on it mostly in the economic sphere-progress de­ pends on assistance and support from the West-and all postcommunist regimes have become the object of paternalistic Westem strategies in the

3

political, economic, and military domains. The question arises whether the convergence going on in terms of political democracy and the capi­ talist market economy is as overwhelming as it seems-and is it also paradigmatic in terms of social stratification? This brings us to the influence of social stratification on other aspects of life in the industrial world. In the case of systemie transformation in Eastern Europe, the problems of social mobility, distribution of materiai goods, social justice, recruitment to new elites, and the looming 'new middle class' which are now on the agenda, have raised a number of theoretical and empirical questions (Stark, 1992; Kovacs, 1994; Lane, 1996; Linz and Stepan, 1996; Offe, 1996; Szelenyi et al., 1996; Mar­ shall, 1997). The variety of findings paints a fragmented and unfocused picture. It is difficult to formulate a precise image of the reshaping so­ cial stratification, either because of the lack of detail in particular studies or because of the impossibility of achieving comparability across the different analyses. Nonetheless, an impressive start has already been made. Since the present volume makes no pretence of being a textbook on class analysis in postcommunist countries I will not attempt a sys­ tematic review of the !atest analyses. However, two aspects of the dis­ cussion are worth mentioning in outline. The first deals with the implications of these data for understanding the different paths of convergence. Research in this branch extends be­ yond convergence between the postcommunist and capitalist worlds. Under the label 'transition to democracy' three groups of countries have been considered in the comparative perspective of political modemiza­ tion processes since the Second World War: (i) the 'postwar democra­ cies' (such as Italy, Austria, Japan, and West Germany); (ii) Greece, Portugal, and Spain, which underwent democratization in the 1970s; and (iii) the authoritarian regimes of South America which collapsed in the 1980s (O'Donnel et al., 1989). East Central European countries consti­ tute a fourth group, broadening the scope of comparative analyses of transition. What differentiates the three above-mentioned cases from transitions to democracy following the breakdown of communist re­ gimes is that the modemizing processes of the former are of a political kind, whereas the emerging democracies of the East face the additional task of reforming the economy. The implications ofthis for class forma­ tion reside in the need to transfer state-owned assets to private hands, the creation of an entirely new class of owners, and the 'installation' of mechanisms of income distribution organized around the principles of a competitive market.

4

The second point concems the interplay between the convergence of, and the differences between, countries. Despite important features which all East Central European countries have in common, they are rooted in different historical soil. The planned economies of the Czech Republic and Hungary, among the most successful in 1989, might ex­ pect steady economic growth thanks to market reforms. Poland, despite deep and urgent economic problems in 1989, has managed, through radical reform, to join the Czech Republic and Hungary in the economic vanguard of Eastern Europe. Bułgaria, by contrast, began with one of the most impoverished economies of the region, and its halting reforms have brought great hardship without accomplishing a systemie transition to the free market. The main problem in Russia is a strong pro-statist bias and its associated excessive regulation, particularly in respect of foreign trade, which-as in many developing countries-breeds ineffi­ ciency and rent-seeking (see Balcerowicz, 1995). Throughout the period under consideration, 1992-93, real economic activity in Russia con­ tracted at unprecedented rates: GDP fell 29 per cent from 1991 levels and industrial production fell 31.3 per cent over two years (Ericson, 1995). Slovakia lies somewhere between these two groups: its economy showed promise, but the commitment to reform of the goveming elites wavered. Despite the troubling political climate, the fledgling Slovak economy has fared well, and its 1 O per cent inflation and 6.6 per cent growth in 1995 were impressive. Still, Slovakia stands alongside Buł­ garia-and Romania-as the states which have yet to implement large­ scale privatization. Bułgaria and Romania, the region's poorest coun­ tries in 1989, saw their GNPs drop by about 10 per cent each year from 1989 to 1992. The Bulgarian and Romanian economies did grow in 1995 (by 2.5 per cent and 4.5 per cent respectively), but suffered the region's highest rates of inflation (60 per cent and 32 per cent respec­ tively) (see Economic Intelligence Unit [EIU] ViewsWire, 1996). Fi­ nally, Slovakia, Bułgaria, and Romania have received comparatively little foreign investment (according to the Financial Times of 15 April 1996). A preliminary examination of the states of the region reveals Poland, the Czech Republic, and Hungary as economic success stories, while Bułgaria and Romania are struggling to reverse steep declines in production. As already mentioned, Slovakia lies somewhere in between. lt is elear that intercountry disparities and differences represent the re­ verse side of convergence. Notwithstanding its pivotal role in my discussion, I refer to the term 'convergence' in only two places: in the lntroduction and in the Con-

5 clusion. Everything which comes in between does concem the 'con­ vergence' question, but boiled down into digressions and hypotheses. I cannot go beyond merely hypothetical interpretations due to the lack of comparable data from Westem countries which might be subjected to serious examination. However, data with a limited degree of compara­ bility do exist and I draw on them in seeking generał regularities which seem common to Eastern Europe and a wide range of postindustrial so­ cieties. Thus, my treatment of the convergence issue takes the form of references to cross-country distinctions and similarities, and so is re­ stricted to commentaries on the central problem, the comparison of six East European countries. The latter, though, revolves around social uni­ versals and differentiation within the East European region. As far as the universals of social stratification are concemed, the problem resides in the degree to which those nations drew close to each other during their common past under communist rule: that which brought them together must have been embedded, above all, in a uni­ form political and economic system which left its stigmata on the shape of social barriers and distances. This thesis may, of course, be chal­ lenged. Nevertheless, we shall see that postcommunist societies preserve certain unique features in comparison with developed capitalist coun­ tries, a fact which-as I shall argue-may be attributed to the effect of a common institutional context inherited from the recent past and greatly influencing social hierarchies and divisions. The results which I report in the following chapters allow one to formulate such generalizations. They show that postcommunist societies maintain disparities in relation to the West in both the shape and the mechanisms underlying income distribution. In Chapters 4 and 6, I elaborate on the effects of the com­ mand economy in this area. One question which is peculiar to Eastern Europe concems the con­ version of members of the farmer communist nomenklatura into owners of firms. The nomenklatura problem has become a controversial issue in politics and has been vigorously debated in the mass media. In Chapter 5, I deal with this question on an empirical basis. It is elear that, while the movement of farmer nomenklatura members into business was by no means negligible, in all six countries it was substantially exceeded by the self-recruitment of owners. However, East European uniqueness must be set against the back­ ground of processes which are shared by East and West. This highlights areas of social stratification which remain impervious to the effects of the political and economic context: in other words, they are cross-

6

systemie. In this study, one will find that such features have been dis­ closed in many prior studies, expressed in-for example-the relentless effects of family background on socioeconomic attainments across countries which, for several decades, differed in type of economy and political system. In light of the evidence presented in Chapter 3, we can claim that intergenerational mobility patterns in Eastern Europe did not diverge in a substantial way from those described in capitalist countries. In the area of cultural participation too, there are similarities. I deal with these in Chapter 7. In Bułgaria, Poland, Hungary, and the farmer Czechoslovakia, it appears to be the intelligentsia which is primarily involved in cultural activity, in contrast to other basie class segments, as is the case with professionals in the United States and Great Britain, the counterparts of our intelligentsia. According to Bourdieu, intergenera­ tional transmission of cultural capital tends to underlie the cultural dominance of the upper classes. I provide some insights into this ques­ tion for Eastern Europe by testing the strength of cultural inheritance in our six nations. Additionally, I highlight 'cultural distances' between the intelligentsia and other basie strata in this perspective. It will not come as a surprise that things which look inherently uni­ form in a Western perspective may be overlooked or rejected by insiders who emphasize elear distinctions between, say, Russians, Hungarians, or Czechs. Uneven standards of living are indisputable, as were the na­ tional divisions which led to the separation of Czechs and Slovaks into different states. Strong differences in religiosity are also evident. Some kinds of religiosity entail traditional orientations which, under particular circumstances, may constitute obstacles to the transition of East Euro­ pean societies to market-oriented systems. In Chapter 8, I examine the pace of secularization in postcommunist societies: how they differ and where secularization is the most advanced. In fact, the unequal progress of secularization may be subject to different interpretations. According to one ofthem-which I follow-widespread religiosity, as displayed in the attitudes of East Europeans, may slow down adaptation to modern capitalist structures. Polish society is peculiarly subject to delay on the road to a market society because of its retentive religiosity-in marked contrast to the Czech Republic, where, in 1993, more than 50 per cent of the adult population declared their atheism. As may be elear from this overview, I shall attempt to portray the social macrostructure in East European countries within the framework of the transformations in the political and economic systems of the re­ gion. Given the strictly comparable data-sets of the six countries, gen-

7

eralizations are easier. One must acknowledge that the question of sys­ temie transformation in postcommunist societies makes for a natural, promising, and fruitful theoretical framework for the analysis of social hierarchies. The challenge of attempting to account for the relationship between changes going on in stratification systems and economic and political institutions cannot be avoided. My treatment of this issue is inherently limited in that I rely, in the main, on hypothetical interpreta­ tions of cross-country coincidences between, on the one hand, the extent of mobility, economic inequalities, and openness of social structures, and, on the other, macroscopic characteristics of economic development, cultural traditions, and so on. I cannot apply any rigorous tests in this area. What warrants this type of speculative explanation is that relation­ ships between macrosystemic parameters can never be ultimately veri­ fied. We can only compare various phenomena registered in time, and in different historical and systemie contexts, in order to ensure that rela­ tionships between them really did exist. My impression, gained after reading the first draft of this book, is that many of the findings presented identify the disintegration of an East European uniformity enforced over recent decades. The postcommunist constellation of social barriers, class distances, and channels of recruit­ ment to newly emerging positions highlights unprecedented, unique features of stratification with the emphasis on similar problems in all countries in the region. Over and above this, one must deal with na­ tionally specific transition paths, the course of which will be determined not only by the communist history of the individual countries over the last fifty years, but also by their histories over centuries. Almost ten years of research into the sociology of social stratification have shown convincingly that the transition in East Central Europe is not following any single logic of 'market modernization', but is progressing along di­ verse routes determined by country-specific route dependencies and institutional contexts (Parrot, 1997; Mach and Dievald, 1998; Nee and Matthews, 1996; Szelenyi et al., 1997). Looked at in this way, asking about the future prospects of the postcommunist societies is tantamount to asking not only whether it is possible to emulate capitalist democra­ cies in respect of culture and history, but also whether postcommunist social development will set out along a range of trajectories of social stratification. This leads us to expect constrained plurality. Although it may well be that, in the course of the transformation of these societies, similarities will emerge once again, they will be shaped by a logic hith­ erto absent-that of the capitalist market.

CHAPTE R 1

PEASANT SOCIETIES-MARKET SOCIETIES: THE TOUCH OF MODERNIZATION I INTEND in this chapter to establish the direction of change. To this end, we shall follow long-term trends in occupational differentiation. Occu­ pational structures are a valid-perhaps the best-reflection of social distances and class divisions. In recent centuries, the occupational system has undergone a con­ version from a pre-industrial to a modern, postindustrial stage. This transformation was, in fact, a long-lasting movement away from tradi­ tional structures. One hundred and seventy years ago, only England had moved beyond the stage of a predominantly agricultural society with (by 1831) no more than 25 per cent of the economically active population working in agriculture (Thompson, 1992, 159). In other countries, this percentage exceeded one-half of working adults, reaching in some so­ cieties a level of 70-80 per cent. By the time ofthe industrial revolution, agricultural workers were still the overwhelming element in the social structure. Development of a factory system brought about the replace­ ment of these rustics with manuał industrial workers, and the growing capitalist market engendered the numerical expansion of small owners outside agriculture. Now, in the late twentieth century, it is the profes­ sions, managerial cadres, and skilled sales and service categories which are growing at the fastest pace. This has become the main developmen­ tal trajectory of socio-occupational structures in the Western world. I will discuss these long-term paths of development in an attempt to determine the main direction of societal transformation in East Central Europe. Which countries have already abandoned peasant-like occupa­ tional structures or have been moving away from them? How far are they advanced in the development of structures more typical of contem­ porary market-oriented systems? Which mechanisms could underpin

10

these transformations? Of special interest is the question of how they might have been affected by the consecutive political changes in this region in the late 1 940s and at the end of the 1980s. We will look for these transformations by examining the changing proportions of basie socio-occupational strata. I will compare percent­ ages of the intelligentsia, top-level managerial staff, clerical workers, pri­ vate entrepreneurs, working-class categories, and peasants in five suc­ cessive elements of the biographies of persons surveyed in 1 993: start­ ing from occupational distributions in the generation of their grandfa­ thers, followed by those of their fathers, then distributions established at the time oftheir first entrance into the labor force, in 1 988 and in 1 993. The time-span is great and the social space will be mapped in terms of basie class and strata divisions. The relevant metaphorical name of these processes is that of 'social metabolism'. The 'founding fathers' of the great sociological systems of the nineteenth century drew analogies with mechanics and nature in their descriptive analyses of social dynam­ ics on a macroscale. I will refer to similar processes in this chapter, though drawing on only one aspect of social metabolism, that restricted to the growing size of particular segments of the social strata and the decrease of others. This will be a first step in an attempt to establish common pattems and differences in social stratification in East Central Europe.

THE UNIVERSAL LO G IC OF INDUSTRIALISM MODIFIED BY POLI TICAL CON TEXT AND TRADITION Under this heading, I outline the main conclusion of my analysis of the changing shape of socio-occupational structures in our six countries. The data which I will now present suggest that processes endogenous to industrial development trajectories, as initiated in the nineteenth cen­ tury, are also exhibited in East Central Europe. At the same time, it ap­ pears that, despite these common roots, it was the different historical circumstances and political fluctuations in this region over the last fifty years, along with the installation of the command economy, which carne to shape the direction of changes. The logic of industrialism collided, first, with the logic of central planning and state monopoly, and re­ cently-since 1989-the inherent laws of industrialism have been modified by the logic of the natura! retrieval of social hierarchies sup­ pressed under communist rule.

11

The inherent laws, referred to above, resided in the universal, evolu­ tionary changes of socio-occupational structures characteristic of indus­ trial societies. According to some influential theories, industrial socie­ ties in recent decades have moved on to a higher stage, that of postin­ dustrial systems. Both terms accentuate technological features, neverthe­ less, the changes were much more comprehensive and encompassed the evolution of the institutional framework of the economy, political sys­ tems, values, orientations, and lifestyle. All of these changes may be properly analyzed in many other, slightly different theoretical perspec­ tives. We could replace, for example, 'postindustrialism' , as coined by Bell ( 1973), with such concepts as 'service society' , 'post-Fordism' , 'consumption society' , 'civil society' , or 'postmodemism' , depending on which aspect of the long-term tendencies is to be placed in the fore­ ground (see Esser, 1990; Hardiman, 1990; Featherstone, 199 1; Gilbert and Burrows, 1993 ; Crompton, 1993 ). Even the label 'disorganized capitalism' has gained popularity: the exponents of the vision of the dis­ organization of the contemporary capitalist system are pessimistic about the progressive nature of the changes. In their analysis, the center of gravity is transferred from the changes themselves to the dramaturgy of the developments, where a tone critical of the status quo and gloomy predictions prevail (Offe, 1985; Lash and Urry, 1987; Beck, 1992). Turner's claim ( 1988) expresses yet another tone in its concise overview of the mechanisms involved in social status, namely, that contemporary societies have become a battlefield between opposing factions. Each of them fights for status and-consequently-a 'politics of status' becomes constitutive of the main axes of social conflict. After demolition of the old status system by capitalism, differentia! statuses have been re­ emerging and once again create the terrain of social divisions, as in feu­ dał society. In fact, any interpretation of the trajectories of the occupational sys­ tems in East Central Europe in terms of universal tendencies should re­ fer to more generał concepts, moving beyond questions revolving around 'postindustrialism' . In the 193 0s Fisher ( 1935), and later Clark ( 1957), put forward one of the earliest conceptions of sequential devel­ opment. They pointed out the close associations between structural transformations and socioeconomic developments. In brief, in the first stage of development of the capitalist market, the traditional peasant economy was replaced by a more efficient form of farming; subse­ quently, its contribution to the economy faded, to be replaced by manu­ facturing, which became the dominant factor. Next, the role of modem

12

technologies in manufacturing grew, paralleled by a decline in the posi­ tion of extractive and, later, processing industries. Finally, manufactur­ ing gave way to multifarious social services: initially, materiał services associated with transport were in high demand, but a growing demand for an effectively performing logistics and the rising needs of consump­ tion enforced development of social services in administration, banking, health, education, recreation, and leisure. The theory of sequential stages aimed, chiefly, to explain the me­ chanics of a developing economy. In fact, these analyses were over­ whelmingly economic, as far as the macroscopic perspective was con­ cemed (Singelman, 1978; Gustaffson, 1979; Gershuny and Miles, 1983; Zagórski, 1986). This did not rule out interesting supplementary contri­ butions by authors who described sequential transformations of occupa­ tional structures (Wright and Martin, 1987; Szafran, 1992; Esping­ Andersen et al., 1993). I will dwell on this expanded schema of evolu­ tionary change in pursuing the trajectories of occupational systems in East Central Europe. But let us recall, first, what constituent processes made up the 'inherent' developmental trends in occupational structure which I referred to above. First, insofar as the traditional macrostructure was abandoned, it re­ leased great waves of manpower from agriculture to manufacturing. In turn, manufacturing provided the service sector with more and more la­ bor in further phases of the transformation. In effect, in developed economies, both a total and a relative decrease of employment in agri­ culture took place. It declined in some countries from 70-80 per cent of all those economically active in the incipient stage of industrialization to 2-1 O at the present day. This percentage looks like a limit for the poten­ tia! efficiency ofagriculture in the modern economy. The natura! consequence of an inflow to manufacturing was, ini­ tially, a growth in the number of manuał workers employed in produc­ tion-this appears as a second universal trend. In developed capitalist societies, the percentage of manuał workers climbed to 50 per cent and more. Afterwards, this ascending trend gradually slowed down. Relative to the type and pace of changes in the economic structures of various countries, it peaked at 70-80 per cent. In the 1950s and 1960s, officia! statistics started to show a relative decrease in the number of manuał workers and, in the next decade, in the United States and a number of other countries, manuał workers lost their leading position, in terms of number, to nonmanual workers (Singelman and Browning, 1980; Singelman, 1978).

13

This moderate decline, preceded by a continuing increase in the rela­ tive share of nonmanual workers, marks a third tuming point in the se­ quence of structural transformations. Both in total numbers and in per­ centages the growth of nonmanual workers resulted from the transfer of 'free' manpower from manufacturing to administration and services. Greater productivity in manufacturing left ample scope for outflow. 'Freed' reservoirs of labor reinforced municipal infrastructures, the de­ mands of the public utilities, as well as the industrial and state bureauc­ racy-with the latter expanding due to the increase in the regulatory functions of government bodies. Another impetus carne from the post­ war economic boom (end of the 1940s and the early 1950s) which gave rise to an increase in materiał welfare on a mass scale. Westem societies witnessed an explosion of chain-stores, restaurants, show-business and tourist services in response to mass demand. In relation to occupational distribution, it resembled the incessant mechanism of a suction-pump. At the same time, services penetrated the productive sector. An effective administration became a functional requirement for the operation of manufacturing plants, transport, and building firms. This numerical ex­ pansion was accompanied by interna! restructuring. Within the differen­ tiated category of nonmanual workers, routine clerks were leaving the stage. They had epitomized the traditional model of the 'white-collar worker' performing simple, repetitive tasks. lnstead, managerial cadres, professionals, semi-professionals, and technicians underwent a dynamie growth. The fourth of the global tendencies in occupational structures, which paralleled the consolidation of the postindustrial system, is that of con­ traction-in terms of relative size-in the category of small and me­ dium-sized owners. Their share of the labor force stabilized at a level of 10-15 per cent in the majority of capitalist economies. As industrial so­ cieties matured, the small-business sector gradually withered before the advance of super-efficient large firms enjoying ever-increasing econo­ mies of scale. The numerical decline was seemingly inexorable. For ex­ ample, small owners accounted for 9.3 per cent in 1984, whereas in 1940 they had represented 20.9 per cent, and in 1880, sixty years earlier, this percentage had amounted to 41.8 per cent ! In 1985, in the nine countries of the European Union, they accounted for 12.6 per cent of the population actively employed (Steinmentz and Wright, 1989, 984-85). It is worth stressing a remarkable exception in the generally bleak prospects for small enterprises, namely, the renaissance of the small­ business sector in British society in the 1980s, a widely documented

14

process. They grew from 7.7 per cent in 1 980 to 1 2 per cent in 1 985. This reversal of a seemingly inescapable historical downward spiral has provoked controversy. Among the explanations offered is that much of this activity is only temporary, consisting of the establishment of new enterprises as an altemative to unemployment or the threat of unem­ ployment. Second, there have been changes in managerial strategies which, it has been argued, have had the effect of enhancing opportuni­ ties for small-scale economic activities. A fashionable managerial strat­ egy in the 1 980s was for companies to subcontract such activities as ca­ tering, cleaning, and security. Thirdly, it has been argued that the renais­ sance of small business was in a sense a spurious growth, since the sta­ tistics of owners included free professional subcontractors and home­ based categories who, in fact, have been satellites of big companies: they are built into their structures and are not autonomous (see Curran and Blackburn, 1 99 1 ). Above all, the numerical growth of the category registered as owners has not been impressive and, apart from in the United Kingdom, has been hardly discemible. Let us establish whether socio-occupational structures in East Central European countries have followed the same pattems. We shall then in­ quire to what extent our six societies differed over the decades as regards relative proportions of farmers (smallholders), owners, and categories of manuał and nonmanual workers, and the pace at which they changed. We shall dwell on the distributions presented in Tables 1 through 6, which express the direction of shifts in occupational structures. I em­ ployed a set of categories which is regarded in cross-national studies as the most valid description of basie class-or, interchangeably, occupa­ tional strata-divisions in modem industrial societies. This is an EGP categorization, used by students of social stratification in recent years (see Erikson and Goldthorpe, 1 992). (In fact, Erikson, Goldthorpe, and Portocarero constructed it for cross-national studies on social stratifica­ tion and mobility-see Erikson, Goldthorpe, and Portocarero, 1 979.) I collapsed the full EGP version into ten categories. As regards the criteria used by Erikson et al., they tend to differentiate positions within labor markets and production units in terms of employment relation­ ships. The EGP version adapted in my study aims to reflect the differ­ entiation of economic rewards associated with jobs, the unequal bargain­ ing power of occupational categories, their materiał standard of living, social status, and other attributes which divide people into 'better placed' and 'worse placed' in the social structure. Neo-Weberians con­ sider these determinants as critical for class position in modern market-

15

oriented societies, and Erikson et al. explicitly adhere to Weber's class theory. EGP has also proven its validity outside capitalist Westem so­ cieties, nam.ely in Poland. In comparison with standard Polish categori­ zations, EGP performed well as regards its explanatory power-in a statistical sense-in relation to schooling, incomes, cultural consump­ tion, and differentiation ofother aspects ofsocioeconomic status. A cautionary note is needed, however. Looking through the distribu­ tions-stretching from the grandfather's generation to 1993-we can determine no more than the direction of changes. We ought to bear in mind that the distributions established for grandfathers, fathers, and those starting their first job do not identify a given month or year, but refer to long periods oftime. In particular, the occupational distributions for grandfathers bring together the grandfathers of respondents who, in 1993 , were 20 years old and those who were 69. In the former case, their occupational activity peaked in 1930-1950, and in the latter group it covered the second half of the nineteenth century. Thus, it may be seen that the distribution of grandfathers collects cases far apart in time. Neither distributions for grandfathers nor those for fathers or for the first job represent meaningful historical points of time. However, this does not rule out the possibility of outlining the dynamics of change, and its direction and pace. The occupational distribution for grandfathers prop­ erly precedes in time 'average fathers'. The latter, in tum, reflects socio­ occupational distributions earlier than those established for those start­ ing their first job. Certainly, this analysis covered only males: it makes sense to compare occupational distributions for grandfathers and fathers with those for grandsons and sons. One should in this case exclude women due to gender differences in occupational structure. Following these five distributions, we can see that at least one of the global paths of occupational transformation is also featured in East Central European countries. I am referring to the developmental stages characterizing manuał workers. Everywhere, in time, they accounted for more and more of the economically active population, after which they stabilized and their relative numbers fell. If we restrict this analysis to men, we can barely trace the continuous development of the nonmanual categories which make up the core of the transformation of occupational structures in the industrial world. Certainly, women, who account for a significant proportion of professionals, clerical workers, and adminis­ trative staff, have contributed to this developmental trend. In East Cen­ tral Europe the tum of the 1980s saw some stabilization of the małe proportion of nonmanual categories. Conversely, farming categories and

16

nonagricultural business owners found themselves, demographically, in decline, fading away as in all industrial societies. Nonmanual workers, broadly defined, include the first three catego­ ries distinguished in Tables 1 to 6, from the intelligentsia through cleri­ cal workers. In the generation of grandfathers only in Russia and Hun­ gary did they account for more than 1 O per cent. In the five other coun­ tries, this proportion fell to 8.5 per cent-in the case of the former Czechoslovakia-at best, and approached 4-5 per cent in Bułgaria, Po­ land, and Slovakia. Over time, the percentage of nonmanual workers, taken overall, increased, and in 1 993 reached 20-32 per cent across the six countries. It stood łowest in Bułgaria, at 20.4 per cent. By contrast, in Russia it amounted to ałmost 32 per cent of małes in 1 993. It shoułd be noted that I refer to 1 993 for the sake of convenience-the Połish data come from 1 994 (I will follow this referentiał convention in the remainder of the present vołume). As regards manuał workers outside agriculture, manuals in the strict sense include categories of skilled and unskilled workers since 'technicians and foremen' cover, apart from manuał supervisors, also lower-grade technicians who overlap with nonmanual workers. In the generation of grandfathers-our starting point-manuał workers prevailed over nonmanuals to a substantial de­ gree, although they were outnumbered by farmers. In rełative terms, they were most highly represented in the territory of the contemporary Czech Republic (totaling 44. 8 per cent). This is in accord with the his­ torical evidence that, with respect to industrialization and urbanization, this region of East Central Europe was the most highly advanced. By the 1 870s, the urban population in the Czech lands exceeded 50 per cent of the whole (Wereszycki, 1 975, 1 45). This fact necessarily accelerated the numerical growth of manuał workers. Relatively, the lowest rate of grandfathers belonging to the working class was in Bułgaria ( 1 5.8 per cent). In Poland, they accounted for 2 1 .6 per cent, a figure derived from the retrospective reports of adult males in 1 994. At consecutive points of biographical time, the percentages of manuał categories increased. In the majority of countries, this upward trend stopped in the period preceding 1 988-it fell between that year and some earłier point of time which averaged the dates on which persons surveyed in 1 993 started their first job. We may assume-łooking through the prism of Clark's and Fisher's classical sequential model-that the occupational structures of postcommunist societies, which emerged on the basis of ex­ tensive industrialization, have already reached their peak in this regard. Henceforth, the service sector shoułd increase in importance.

17 Table 1 . Distribution ofsocio-occupational categories for father 's father, father, respondent 'sfirstjob, and respondent 's occupation in 1988 and in 1993. Men. Bułgaria (%) Socio-occupational categories Higher professionals and managers Semi-professionals Routine nonmanual, sales and service workers Owners with employees Owners without employees Lower-grade technicians Skilled workers Unskilled workers Agricultural laborers Farmers Total

Father's . F ather F"Irst Job 1 988 f:ather 1 .0 2.6 1 .7 6.5 O.O 0. 1 6.2 9.6 2.7 69.6 1 00.0

5.3 5.7 3 .2 1 .7 5.1 2.8 1 2.7 23.3 27.9 1 2.4 1 00.0

4.5 7.5 2.7 0.8 2.2 2.0 26.9 29.5 1 9.2 4.7 1 00.0

8.0 10.1 2.9 1 .4 2.3 4. 1 25.1 33.5 1 1 .6 1 .0 1 00.0

1 993 7.7 9.5 3 .2 3.8 5.5 3.6 23 .6 3 1 .8 9.6 1 .6 1 00.0

Table 2 . Distribution ofsocio-occupational categories for father 's father, jather,

respondent 'sfirstjob, and respondent 's occupation in 1988 and in 1993. Men. Czech Republic (%) Socio-occupational categories Higher professionals and managers Semi-professionals Routine nonmanual, sales and service workers Owners with employees Owners without employees Lower-grade technicians Skilled workers Unskilled workers Agricultural laborers Farmers Total

Father' s . b 1988 F ather p 1rst · Jo f:ather

1 993

1 4.2 1 3 .7 3 .4 0.2 0.9 7.2 26.8 28. l 5.1 0.3 1 00.0

1 1 .4 1 3 .0 3.8 4.8 9.5 5.6 22.9 24.4 3 .9 0.7 1 00.0

1 .7 2.6 4.2 1 1 .6 O.O 1.1 20.8 24.0 7.2 26.6 1 00.0

9.5 1 1.1 5.0 0.9 1 .2 6.6 20.0 30. 1 9.3 6.4 1 00.0

6.3 9.3 1 1 .4 0.1 1 .5 2.6 39.9 27.6 7.2 I .O 1 00.0

In the case of manuał workers, their increase, stabilization, and, fi­ nally, the decline of their relative proportion were constituents of a pro­ gression which conformed to the classical model of macrostructural transformations. This does not seem to be the case with farmers and owners (the term 'owners' as used in this book should be understood in the sense of 'owners of nonagricultural businesses'). In East Central Europe, they were transformed according to a logic which departed, to some extent, from the typical trajectory of the postindustrial stage.

18 Table 3 . Distribution ofsocio-occupational categories for father 's father, father, respondent 'sfirstjob, and respondent 's occupation in 1988 and in 1993. Men. Hungary (%) Socio-occupational categories Higher professionals and managers Semi-professionals Routine nonmanual, sales and service workers Owners with employees Owners without employees Lower-grade technicians Skilled workers Unskilled workers Agricultural laborers Farmers Total

Father's . Father F"1rst J0b 1988 fiather 2.7 3.9 4.4 5.2

o.o

0.4 1 2.2 1 1 .3 1 2 .4 47.4 1 00.0

6.7 5.7 4.3 2.7 5 .4 4.2 1 8 .0 22.3 1 7. I 1 3 .7 1 00.0

3 .2 9.8 6. 1 8.8 4. 1 3 .4 0.2 1 .4 1 .2 3.9 6.3 1 .3 40.8 3 1 .5 2 1 .5 26.4 1 8.0 6.7 1 .7 3 .6 1 00.0 1 00.0

1 993 8.8 1 0 .3 3.6 3.8 8.8 5 .4 30. 1 22.2 5.2 2.3 1 00.0

Table 4. Distribution ofsocio-occupational categories for father 's father, father, respondent 's first job, and respondent 's occupation in 1988 and in 1994. Men. Poland (%) Socio-occupational categories Higher professionals and managers Semi-professionals Routine nonmanual, sales and service workers Owners with employees Owners without employees Lower-grade technicians Skilled workers Unskilled workers Agricultural laborers Farmers Total

Father's . . Father Frrst Job 1988 father 0.8 1 .8 2.2 4.3

o.o o.o

1 0 .4 1 1 .2 3.0 66.3 1 00.0

6.2 5. 3 3.1 2.4 4.8 2.9 3 5 .8 20.4 5 .3 3 1 .0 1 00.0

4.3 6. 1 3.8 0.4 1 .6 2.9 3 5 .8 22.9 1 6.8 5.5 1 00.0

10.1 7.5 4.4 2.7 3 .4 5 .4 27.7 24. 1 4.3 1 0. 5 1 00.0

1 994 8.5 7.0 4.8 6.0 7.9 4.3 2 5 .2 2 1 .2 3.3 1 1 .8 1 00.0

It is true that, in both cases, we find analogies with developmental trends involving farmers and owners in Westem societies. Analogously with the West, in five of our postcommunist countries the percentage of farmers declined, except for the nontypical case of Polish agriculture. However, this decline was underpinned by mechanisms quite different to those in the capitalist West. The numerical decline of farmers in East Central Europe departed from the classical sequence in that it resulted from the mass collectivization of private farms which took place in the

19 first half of the 1950s (with the exception of Poland). Collectivization accelerated the pace of decline in private forming. Notably, there was a big slump in the percentage of farmers in the workforce in one particular period: between the generations of grandfathers and fathers. We may suppose that, even without the political intervention of the communist authorities, the inherent rules of the industrial economy would sooner or later have resulted in 'depeasantization', if not so rapidly. Table 5 . Distribution ofsocio-occupational categories for father 's father, father, respondent 'sfirstjob, and respondent 's occupation in 1988 and in 1993. Men. Russia (%) Socio-occupational categories Higher professionals and managers Semi-professionals Routine nonmanual, sales and service workers Owners with employees Owners without employees Lower-grade technicians Skilled workers Unskilled workers Agricultural laborers Farmers Total

Father's . F ather Fust JOb 1988 father 6.5 3 .2 3.5 5.8

O.O

0.7 1 7.2 1 0.2 4.5 48.5 1 00.0

16.1 5.8 2.6

I .O

2.0 6.5 24.5 1 9.7 1 9.3 2.5 1 00.0

I l .O

9.4 2.5 0.4 0.8 2.2 34.3 22.3 1 5 .8 0.4 1 00.0

20.9 1 0.5 2. 1 0.9 1.1 5 .4 29.0 2 1 .3 8.4 0.4 1 00.0

1 993 1 9.9 9.7 2.4 3.3 2.4 5.2 28.0 2 1 .6 7.3 0.3 1 00.0

Table 6. Distribution ofsocio-occupational categories for father 's father, father, respondent 's firstjob, and respondent 's occupation in 1988 and in 1993. Men. Slovakia (%) Socio-occupational categories Higher professionals and managers Semi-professionals Routine nonmanual, sales and service workers Owners with employees Owners without employees Lower-grade technicians Skilled workers Unskilled workers Agricultural laborers Farmers Total

Father's Father First job 1 988 father 1 .2 1 .5 1 .8 6.5

o.o

0.6 1 2 .2 1 6.5 8.4 5 1 .4 1 00.0

7.4 7.6 4.4 0.4 1 .2 5.7 0.2 27.7

IS.I

1 .4 1 00.0

6.7 8.3 4.5 0.3

I .O

2.5 42.7 23.6 8.6 1 .8 1 00.0

1 3 .0 1 0. 8 3 .9 0.6 0.6 8. 1 33.1 23.8 5.5 0.6 1 00.0

1 993 1 0. 5 1 0.7 4. 1 4. 1 5.5 7.3 30.9 2 1 .8 4.5 0.6 1 00.0

20

As regards owners, we can still trace analogies with the West; these, however, plainly conceal different political and economic causes of change. The relative proportion of owners decreased in the generation of fathers as a result of nationalization at the end of the 1940s. In the next period with which we are concemed, the time when respondents started their first job, private ownership tended to recover, except in the Czech Republic and Slovakia. After 1988 this upturn in the trend became a veritable explosion. As in the case of farmers, the restrictive administra­ tive measures imposed on owners may be regarded as a counterpart of the market forces which brought about the decline of the 'old middle classes' in capitalist economies. However, the seeming similarities be­ tween the recessions witnessed by both 'capitalist' and 'communist' owners broke down completely after 1988. The dominant tendency in the West was to remain at a standstill, with the proportion of owners not exceeding 10-15 per cent. Instead, postcommunist societies saw a marked revival of private business. After the collapse of communism, the institutional impediments to development were to an extent removed and the number of small businesses enjoyed a spectacular upsurge. This growth trend clearly-but perhaps temporarily-departed from the Westem pattem. These macrostructural tendencies overlapped with a number of typi­ cal trajectories in occupational careers. One of them is evident in the changing occupational distributions. I am referring to the relative de­ crease in the proportion of farmers in the distribution established for first jobs. This phenomenon appeared only in Poland because only there did private farming exist under communist rule (collectivization was abandoned after 1956). However, this decrease did not signify a tuming point in the long-range transformation of the occupational system. In particular, one should not conclude that the percentage of farmers among the actively employed population in Poland suddenly declined, only to increase afterwards-in 1988, as may be seen in Table 4-from 5.5 per cent to 10.5 per cent. The latter increase reflected the termination of temporary outflows from the agricultural sector of sons and daughters of farmers, a regular occurrence often identified in studies of the life­ careers of farmers (see Andorka and Zagórski, 1980). Farmers took temporary, nonagricultural-mainly manual-jobs, only to return to the farm later on, a return which was generally permanent. It seems likely that identical processes of return mobility took place within the intelligentsia and managerial cadres. In almost all countries­ the exception was Bułgaria-the proportion of this category fell at the

21 time of the first job and increased thereafter up to 1988. It is a 'norm' that if representatives of the top social strata leave their original catego­ ries, they do so only temporarily. Usually, they experience downward mobility at the beginning of their life-careers. Later on, they move into top positions, which-in a sense-are assigned to them by virtue of so­ cial origin and the 'iron rules' ofreproduction of privilege which main­ tain class barriers and stratificational hierarchies basically intact. Our data lend support to this regularity. The return mobility of farmers also gives rise to the consolidation of social distances between them and nonagricultural occupational strata. Returns to the farm result, in the main, from the difficulties faced by peasants in their acculturation to urban life. Equally, one cannot discount the factor of emotional ties to the land inbred in peasant families: they often find it difficult to aban­ don the country for the alien urban environment. Oddly enough, the relative size of the intelligentsia and managerial cadres diminished between 1988 and 1993. This occurred in all five countries (that is, with the exception of Bułgaria). Undoubtedly, this has nothing to do with the decline in the proportion of the intelligentsia and managers observed in postindustrial countries, which may be described as a long-term secular trend in occupational transformation. The sources of the temporary-as I assume it to be-decline in the 'service class' of postcommunist societies lay in the peculiar develop­ ments which occurred between 1988 and 1993. For example, the process may be attributed to the so-called circulation of elites, both political and occupational, the latter being composed of the highest managerial cad­ res. Representatives of both elites were included-at least theoreti­ cally-in our samples. The outflow from the former elite, which must have occurred after 1988, would have brought about a reduction in the relative size of the intelligentsia and managerial cadres. The 'circulation of elites' notwithstanding, the intelligentsia may be trying to find a new place in a changing labor market: for example, some are likely to have moved into the business world. This is another explanation of its de­ creasing size at the onset of systemie change.

SPURIOUS ANALOGIES BETWEEN EAST AND WEST Transformations of occupational systems are of particular interest within the framework of our study in that they may help to indicate the extent to which postcommunist societies are integrated in modem capitalism.

22 We have already established that: (i) the developmental trajectories of our six East Central European countries have been subsumed under gen­ erał evolutionary trends leading to the establishment of postindustrial systems; (ii) within these generał trends, unique features emerged which may be attributed to · the policies pursued by individual communist states; (iii) changes in political and economic systems at the end of the 1980s brought about a further modification in the transformations of the occupational macrostructure-there was an increase in the relative numbers of owners and an unexpected slump in the relative size of the intelligentsia and managerial cadres. Although the relative growth in the number of owners seems to por­ tend a convergence between postcommunist and market societies, it has not led to their homogenization. In particular, the Polish case­ characterized by a significant percentage of farmers among the actively employed, both before and after the collapse of communism-points to distinctive routes which transformations may traverse. East Central European countries have also faced a temporary decline in the relative size of higher managerial cadres, administrative officials, and profes­ sionals-in sum, the intelligentsia. This has not propelled East Central Europe onto a divergent path of structural transformation, but it never­ theless makes it elear that occupational systems, while subject to univer­ sal rules of division of labor, remain under the influence of develop­ ments connected to their historical context. It would be extremely difficult to predict coming trends, partly be­ cause we did not find many clear-cut changes in occupational distribu­ tions for the six countries. But even when such tendencies are detect­ able, it is difficult to distinguish long-term trends-those indicative of the modemization of social structures-from short-term processes which obscure the inherent logic of development. The latter may consist of deliberately implemented measures-dictated, for example, by the urgent need to eradicate all vestiges of the communist economy. Predictions conceming the development of professionals and mana­ gerial staff are necessarily the most vague. In 1988-93, this category declined in relative size, contrary to expectations. This may serve as an exemplary case of a conjunctural effect, resulting perhaps from the cir­ culation of an occupational elite and its transition to private business activities. Obviously, the long-drawn-out collapse of a universal trend may also be involved. The future of owners as a group is also obscure. At the end of the 1980s, their proportion increased in all East Central European countries.

23

This clearly reflected the effects of systemie transformation ensuing from the removal of formal impediments to privatization and the opera­ tion of market rules. Does it also portend the entry of postcommunist societies anto a trajectory leading to a market system? It may, at least in the early part of the 1 990s. Nevertheless, we know that in developed capitalist countries, the situation is the opposite : small business carne to be a rather stagnant category, in the demographic sense, and its contri­ bution to the national product relatively small. Bearing this in mind, one could scarcely regard the numerical expansion of owners in East Central Europe as a forerunner of modernization. Their numerical increase there proves something rather different: profound upheavals in their political, economic, and social systems.

CHAPTE R 2

TWO TRAN SFORMATI O N S AND S O C IAL M OBILITY

WE SHALL try to assess the dynamics of mobility rates in our six East Central European nations in two historical periods: first, at the tum of the 1940s; second, forty years later, at the tum of the 1980s. The ulti­ mate concem of this chapter is a comparison of the effects of the two systemie ruptures on the social metabolism. We shall establish whether the transitions from the quasi-capitalist, postwar phase in the 1940s and from communism in the l990s transformed the processes through which particular individuals were allocated different positions in the division of labor. Pattems of social mobility follow their own logic, which does not necessarily respond to institutional transformations, even if they are systemie. As far as the latter really affected mobility, we might rea­ sonably expect to find evidence of this shift in the first half of the 1950s and ofthe 1990s. In the first instance, a shift in pattems of social mobility might result from the transformation of socio-occupational structures. In the years following the Second World War the communist leaders of Poland, Hungary, and other countries in the region initiated a major reconstruc­ tion of the social order. Particularly crucial, in their view, was rapid in­ dustrialization driven centrally by the communist state, the nationaliza­ tion of manufacturing, transport, and the majority of private firms in other industries, and, starting from the late 1940s, the collectivization of private farms. These measures compelled a transfer of manpower from agriculture to heavy industry, together with the introduction of large numbers of workers and peasants into govemment and industrial bu­ reaucracies. We may recall the implications of these structural changes for the composition of specific classes in Poland. As of the early l970s, men of

26

peasant and ofworking-class origin made up 30 per cent and 26 per cent of nonmanual workers respectively (Zagórski, 1978, 132). As far as the consequences of these mass transitions for class mobility are concemed, Erikson and Goldthorpe (1992, 1O 1) showed that the two most rigid barriers-those separating the intelligentsia from manuał workers and from peasants-were almost nonexistent in Poland or at least clearly weaker than in any of the other eight nations included in their study. Notably, both barriers were apparent in Hungary, which experienced a similar trajectory of socialist transformation in the early postwar era. It would appear that policies directed towards shaping new hierarchies and pattems of mobility in Poland did the trick, resulting in a temporary weakening of social rigidities. At the same time, strict limitations were placed on private entrepreneurship: while private firms continued to ex­ ist, they did so only in vestigia! form. We must determine the extent to which the reconstruction concomi­ tant with the emergence of a market society in the 1990s compared, in scope, with that of the 1940s and 1950s. Ofparticular importance here is the rapid expansion of the private sector after 1989, engendering a mas­ sive inflow to the category of owners, prompting an overall increase in mobility. For example, in 1994 owners accounted for 10.4 per cent of all those actively employed in Poland. This proportion stood at only 4.3 per cent in 1988, before the systemie upheaval. In our treatment of the dy­ namics of mobility in postcommunist societies, we shall focus on the influx to the social category of 'owners', an important element in the formation of new social strata in this region. One widely held view of comparative macrosociology is that its prime objective must be to demonstrate differences between aspects of social structure and then to account for these differences through analyses in which nations serve as the basie units of observation. As regards studies of social mobility, one can scarcely find systematic cross-national variation in respect of both total mobility and its rela­ tive rates. In other words, any variation is scarcely attributable to dif­ ferences in, for example, the level of economic development or of de­ mocracy in contrast to totalitarianism in the political system (see Erik­ son and Goldthorpe, 1992). The existence of such relationships sketched in a number of studies (for example, Tyree at al., 1979) has been contested by recent national mobility inquiries which introduced higher standards of data comparability (Erikson and Goldthorpe, 1992). In a cross-time perspective too, students of mobility have em­ phasized the absence of directional trends (Featherman and Hauser,

27 1978; Ganzeboom and de Graff, 1984; Payne, 1993; Yamaguchi, 1987; Jones et al., 1994). The question inevitably arises as to whether these results are appli­ cable to postcommunist societies, given that all the findings referred to above reported on social mobility in stable societies which had not faced systemie changes: the East Central European countries have a distinctly different set of experiences. The most pertinent question is whether mobility rates altered significantly with the rise of the communist sys­ tem, after its collapse, and in the early stages of the creation of a new social order. Furthermore, if they did, was the trend towards greater openness, particularly in respect of the increased flow into the 'owners' category in recent years, the rising 'old middle class'? Indeed, one cannot overlook the difference between the 'pheno­ typical' level of actually observed mobility rates and the 'genotypical' level of the pattem of relative mobility opportunities which underlies these rates (see Featherman et al., 1975). Looked at in terms of the 'phenotypical' level, changes can easily be anticipated, precisely be­ cause observed rates are greatly influenced by the structure of the divi­ sion of labor and, in turn, by effects deriving from a range of economic, technological, and demographic circumstances, all of which are known to vary over time. Insofar as mobility is considered net of all such changes, the thesis of the basie invariance of fluidity pattems in time has received empirical support also in respect of communist societies. Studies on long-term trends carried out in the 1980s in Poland, the for­ mer Czechoslovakia, East Germany, Hungary, and Russia proved that mobility regimes had not altered in a substantive way over recent dec­ ades and that they conformed to the pattems described in the West (see Haller and Mach, 1984; Andorka, 1990; Boguszak, 1990; Marshall et al., 1995; Marshall, 1996). Let us look now at the consequences of the two transformations. Did they give rise to any change in the dynamics of social mobility? Did mobility barriers open up? Is it true that the inflow into the 'owners' category significantly affected the transitions taking place after 1989? Finally, which ofthese two waves was greater in terms of the volume of transitions, on the proviso that the substantial shifts in mobility attrib­ uted to the transformation of the political system really occurred?

28

H YPOTH ESES: TRANSITION TO AND EXIT FROM COM M UNISM Social mobility tables reflect both relative opportunities for movement and the constraints of occupational origins and opportunities. Sociolo­ gists have for some time recognized this duality and have attempted to distinguish total movements between socio-occupational categories­ which include both structural constraints and opportunities-from rela­ tive rates which may be attributed to 'circulation', 'exchange', or 'pure' mobility. The latter encapsulate mobility rates net of changing distribu­ tions of origin and destination categories. Analyses of these aspects ad­ dress different theoretical and substantive issues. While total­ absolute-rates can be used to map configurations of basie social dis­ tances as established by determinants of movement and the 'inheritance' of position, relative rates refer to the openness of specific social strata and global social structure. Previous cross-time mobility studies which covered long periods of time reported changes in total movements over decades (Glass, 1954; Svalastoga, 1958; Featherman and Hauser, 1978). It was convincingly proved that such changes, insofar as they took place, were 'pheno­ typical' in nature, that is, they were mediated by a wide variety of eco­ nomic, technological, demographic, and political influences largely ex­ ogenous to the dynamics of social stratification. Above all, they derived chiefly from transformations of origin and destination categories. The driving force of occupational transformations in Westem countries was the economic boom after the Second World War. Its counterpart in East Central Europe was the mass mobility associated with extensive indus­ trialization, followed by a decline in total flows (Andorka and Zagórski, 1980). Although the decline was detected as early as the 1960s, we might expect the transformation of economic and political systems to have oc­ casioned a new growth of mobility. Since we are seeking the effects of these institutional changes, a good referential base for the 1990s is the 1980s, the immediately preceding period of communism. The present study will compare and contrast mobility rates in 1983-88 and 1988-93. As far as the first transformation is concemed, we will compare mobility rates in 1948-52 and 1952-63. One might expect that in the 1990s the total volume of mobility would increase as the outcome essentially of structural changes con­ comitant with systemie transformation, including the emergence of new

29

jobs and skills. Development of capitalist markets in postcommunist societies carried with it the expansion of the financial sector, including banking, and of marketing and a wide range of other services. Ensuing demand gave rise to the creation of occupational roles which had had no counterparts in the communist economy. One example of a new busi­ ness sector is private security, which employed 200,000 persons in Po­ land in 1996: in terms of numerical size, it was the third largest occupa­ tional category, after teaching and mining. There was also a rapid ex­ pansion in the category of owners of private businesses. In studies of social mobility, changes in occupational distribution are considered as belonging to the 'demand' side of the rules governing the flow of persons through the life cycle. Newly created positions tend to 'attract' mobile persons. The ' supply' side consists of relative advan­ tages afforded to individuals by different class origins, which may be thought of in terms of economic, cultural, and social resources. In the interplay of supply and demand characteristic of the 1990s, the expan­ sion of business may be attributed a decisive role. Representatives of the intelligentsia, working-class categories, and farmers (smallholders) wit­ nessed the tangible effects of growing opportunities to succeed in busi­ ness. After 1989, the 'entrepreneurial spirit' and 'possessive individual­ ism'-which had been blocked under communism by administrative obstacles-found an outlet. New pattems of mobility might also result from changing educational channels and new forms of job training. The rules of the capitalist market tend to convert the generał knowledge received in schools into practical skills. Since the beginning of the 1 990s, in Hungary, Poland, and Czech society, new vocational courses have been developed which are based on Westem models. These have provided individuals with new opportunities and encouraged them to set out on new occupational career paths. But as the newly emerging education system has been the subject of a certain amount of experimentation and modification, it has been difficult to es­ tablish institutions and mechanisms which might release even greater flows. Nevertheless, educational restructuring is a fairly new element in a changing context which should reshape the structure of occupational op­ portunities. At the same time, there has been an increase in various parts of the welfare state, such as new pension schemes or insurance systems which differentiate life-careers in the West. The same is true of wage bar­ gaining strategies adopted by trade unions against employers and the manner in which they attempt to restrict access to particular jobs or firms (see Esping-Andersen et al., 1993). In East Central Europe, new labor-

30

market mechanisms have begun to consolidate and their mobility effects have gradually come into play. As regards mobility rates, we will test four hypotheses. First, one may predict that the total number of movements, as attributable to the systemie changes, increased-these movements were higher in 1948-52 than in the maturing evolutionary stages of the communist system. With respect to the second transformation, it is likely that 1988-93 saw mare intense social mobility than in the period immediately preceding the collapse of communism. In brief, social mobility was greater on the verge of each transformation of the political system. Our second hypothesis concerns the relative weight of each trans­ formation, assessed in terms of the size of the mobility flows they pro­ duced. There are reasons to expect that the imposition of the communist system brought about mare mobility than its collapse in the 1990s. First, the reshuffles of the social structure at the tum of the 1940s were un­ derpinned by deep changes in the economy and in occupational distri­ butions. It seems that three parallel processes-extensive industrializa­ tion, collectivization of agriculture, and nationalization of major indus­ tries-released larger flows than those prompted by economic privati­ zation in the 1990s. In fact, privatization was the only macrostructural feature which might be set against the three mentioned above which characterized the situation four decades earlier. Furthermore, the com­ munist system was very much imposed. Harsh administrative measures were taken to create a 'New Man' and a new social order. The influence of political and ideological criteria over the channeling of persons into and within the educational and occupational system was overwhelming: the advancement of a substantial portion of the working class and the peasantry was actively promoted, while serious career disruptions were inflicted upon the intelligentsia and the farmer bourgeoisie. The new macrostructural arrangements were implemented exogenously with re­ spect to the logic of social stratification. Contrariwise, the social changes which have got under way since 1989 can be said to have pro­ ceeded relatively 'smoothly', much mare at an evolutionary than at a revolutionary pace. Finally, the mass shift in manpower which resulted from industrialization involved the largest segments of society, the peasantry and the working class. Its scope was much greater than the social transformations brought about by privatization and the new op­ portunities emerging with the rise of the market economy. In fact, new opportunities could be exploited only by those already possessing supe­ rior cultural and social capital.

31

Third, relative mobility during both political ruptures remained basi­ cally unchanged. Growth in absolute rates therefore resulted from changing occupational distributions as implied by the increasing size of some socio-occupational segments and the diminution of others. Fourth, as regards the transformation from communism to a market­ like society, inflows to private business might be expected to exceed inflows to other occupational strata. This possibility will be considered in the context of the theoretical debates in the sociological literature re­ garding class and strata formation.

MOBILITY RATES In pursuit of the effects of systemie transformation on rates of mobility, we used the following strategies. In order to assess whether the transi­ tion to communism gave rise to massive flows we utilized the data on father's occupation, comparing occupational mobility between 1948 and 1952 with occupational mobility between 1952 and 1963. As far as the effects of the fall of communism are concemed, we compared the career mobility of respondents between 1983 and 1988 with their mobility between 1988 and 1993. If our first hypothesis is correct, there should be more mobility between 1948 and 1952 than between 1952 and 1963, and more mobility between 1988 and 1993 than between 1983 and 1988. If the second hypothesis is correct, there should be more mobility between 1948 and 1952 than between 1952 and 1963, but little or no difference between 1983-88 and 1988-93. lt is a matter for further de­ bate to what extent these potentia! changes may be affected by transfor­ mations in the political and economic systems, or derive simply from the 'endogenous' logic of social structuration. Certainly, if any disrup­ tion in occupational careers occurred, there should be no difference between either pair of mobility matrices. In our treatment of mobility in the earlier period of 1948-63, we had to restrict our analysis to five countries, excluding Russia. This is due to the lack of data from Russia for 1948 and 1952 conceming father's oc­ cupation. Two-way (6 x 6) EGP categorizations for 1948-52 yielded only 83 men in Russia, while for the 1952-63 matrix we obtained only 163 cases, making reliable estimates impossible. There are two technical problems with the earlier comparison. First, a sample of fathers is not a representative sample of the 1948-52 or 1952-63 populations. But while we must acknowledge this problem, it

32

should not prevent us from proceeding. The second difficulty is that 1948-52 covers only four years, while 1952-63 covers eleven: we would expect the normal processes of the social metabolism to bring about more mobility in the latter period. The obvious solution is to raise the 1948-52 transition matrix to the third power, which would give the expected 1948-60 table, on the assumption that the 1948-52 pattern continued without change (see Hodge, 1966). In this way we obtained a twelve-year period to compare with the eleven-year transition from 1952 to 1963, which is close enough to warrant comparison. We shall restrict our examination of mobility to men. This provides us with a commonly used reference point, bearing in mind that most substantive conclusions on this issue have so far been reached in the analysis of transitions affecting the małe population. Women' s mobility would be of interest insofar as meaningful modifications were necessary to our findings in respect of men. In fact, we replicated the same analy­ ses with the mobility tables of women. lt turned out that, in each coun­ try, the configuration of intergenerational mobility barriers in the fornale population closely fits the patterns for men; as a result, we will not dis­ cuss women separately, on the assumption that what we found for men more or less represents social mobility patterns overall. The version of EGP which I introduced in Chapter 1-collapsed into ten categories-I now collapsed into six strata: (i) higher-grade profes­ sionals, administrators, officials, managers of large industrial establish­ ments, and large proprietors (referred to, interchangeably, as the ' intelligentsia'); (ii) other nonmanuals, that is, lower-grade profession­ als, administrators, officials, higher-grade technicians, managers of small industrial establishments, and routine nonmanual employees in administration, commerce, sales, and services; (iii) small nonagricultural owners with and without employees; (iv) skilled workers; (v) unskilled workers; and (vi) farmers and agricultural workers.

OVERALL RATES: THE MODEST EFFEC TS OF TRANSITION In order to address directly whether mobility rates increased we com­ pare two sets of figures: the percentage of movers in 1983-88 and in 1988-93 and, separately, the percentage of mobile men in 1948-52 and 1952-63, as shown in Table 7. The percentages given there were calcu­ lated on the basis of 6 x 6 matrices of movements between occupational

33 categories for the periods 1 948-52-raised to the third power-and 1 952-63. We established the mobility rates for 1 9 83-88 and 1 988-93 ( 1 994 for Poland) in an analogous fashion. Hence, the mobility rates of persons are simply the percentages of men in our national samples found in cells off the main diagonal of the 6 x 6 mobility matrices-in other words, the percentage of men whose 'destination' category was different from their category of 'origin' . Table 7. Total mobility rates. Percentages of mobile men in 1948-1 952 and 1 952-1963, and in 1 983-1988 and 1 988-1993 1 948-1952 1952-1 963 1 983-1988 1 988-1993 Bułgaria Czech Republic Hungary Poland Russia Slovakia

21.1 3 8.4 38.1 27. 1

12.3 20.9 22.4 12.2

38.8

25. 5

1 2 .2 8.1 1 3 .4 9.7 1 1 .2

8.8

1 7.2 23 .6 1 9. 5 20.0 15.1 1 9.7

As regards the transition from communism, we can see that in all six countries overall mobility increased. In the małe population, mobility rates were higher in 1 988-93 than in 1 983-88. In tum, in the maturing stage of communism in the 1950s, rates of mobile men declined signifi­ cantly. lt was during the birth and ' early youth' of the communist sys­ tem, in 1 948-52, when the dynamics of mobility between basie socio­ economic strata intensified the most. The predictions of our first hy­ pothesis are therefore confirmed: the two fundamental breaks with the past in the political history of East Central Europe were accompanied by a discemible increase in the volume of flows in the social structure. Our second hypothesis concemed the relative weight of the systemie breaks for the transformation of mobility barriers. The dynamics of mobility proved most intense during the first transformation. In all five countries it peaked in 1 948-52, with as much as 3 8-39 per cent of men in Czechoslovakia and Hungary changing their occupational category. Poles and Bulgarians remained relatively more attached to their socio­ occupational positions. In the following years, mobility evidently slowed down. The overall rate of mobile men halved in Poland and Czechoslovakia in 1 952-63 as compared to 1 948-52. Three other so­ cieties witnessed a decrease of at least one-third.

34

If we tum to mobility in the period of transition from the communist system, one can compare its acceleration in 1988-93 as compared to 1983-88, with the pace of its decline in the 1950s, in the aftermath of the most concentrated flows. Differences between the two periods are discemible in the total volume of transitions. The collapse of commu­ nism was accompanied by much the same kind of increase in career mobility as that which occurred in the period of transition to this system, but on the verge of the emergence of capitalism in the postcommunist world, mobility rates decreased in comparison with the period of en­ forced industrialization, nationalization, and collectivization of agricul­ ture. Mobility dynamics were most intense in Czech society at the tum of the 1980s: 23.6 per cent of men changed occupational category between 1988 and 1993. The transitions intensified also in Slovakia and Poland. Mobility was least important in Russia. In the 1990s, the Czech Repub­ lic, Poland, and Slovakia had the highest overall mobility rates, Bułgaria and Russia the lowest. Our first conclusion is that, in parallel with the systemie changes in East Central Europe, mobility between basie segments of social stratifi­ cation increased. This might be the result ofthe mutual reinforcement of restructurations in social space and changing economic and political structures. Also, mobility barriers opened up more during the first trans­ formation than during the second. O P E N N E S S : T H E C O N STANT F L U X Strictly speaking, w e have not yet established whether there was a higher level of openness. What has been established is that there was a fairly systematic cross-time and cross-country pattem of change in abso­ lute mobility rates, viewed through the lens of our class schema. One may infer from this that the underlying cause in both 1948-63 and 1983-93 was shifts in the occupational structure. We are naturally led to the question of how mobility trends would appear if they could in some way be assessed 'independently' of this changing structural context. In studies of social mobility, these net rates 'allowing for' changing occu­ pational distributions are regarded as more direct measures of the open­ ness of the social structure. We shall therefore focus our attention on a detailed examination of the set of relative mobility opportunities-the 'mobility regime', as Hauser (1978) termed it.

35

In dealing with this question, sociologists have tak.en a number of different approaches. However, all of them rest on a distinction between 'structural'-or 'demand' or ' forced'-mobility, and 'exchange' (or 'circular' or 'pure' or 'relative') mobility. The former is defined as that part of total observed mobility which is directly attributable to changes in the structure of objective mobility opportunities, and the latter as that part which is not associated with such changes. The present account will also rely on this conceptual distinction. A number of writers have approached the problem of 'allowing for' structural changes by drawing on the application of log-linear models to the analysis of multivariate contingency tables (see Goodman, 1 972; Hauser, 1 978; Hout, 1980; Ishii-Kuntz, 1994). In comparing mobility between 1 948-52 and 1 952-63, and between 1 983-88 and 1988-93, we employ two models in particular which were applied in previous mobil­ ity studies. (i) The diagonals model. The simplest constrained diagonals model specifies a single parameter for all diagonal cells in the mobility table, testing the proposition that immobility exceeds what would be expected on the basis of perfect mobility by the same proportion in all occupational categories (see Goodman, 1 972, 661-71 ; Hout, 1980, 28). The diagonals model refers to a state in which change occurs only in occupational distri­ butions, but not in structural or exchange mobility: self-recruitment in six occupational strata accounts for all association in mobility tables. I made the comparison first for both 1 948-52 and 1952-63; and second for both 1983-88 and 1 988-93 in each country. Adequacy of fit for this model should cast much light on the question of openness in stratification sys­ tems since we are comparing transitions taking place over very short peri­ ods of time. It seems unlikely that radical changes in mobility opportuni­ ties emerged over fifteen (1948-63) and ten (1983-93) years. One may hypothesize that it was self-recruitment rather than circulation which shaped occupational careers during this time. (ii) The constant fluidi ty model (CFM). In this case, as the name implies, the effects of origin and destination vary in time, while the as­ sociation between them is constant. We thus have variations in absolute mobility between 1 948-52 and 1 952-63, but constant relative mobility. Altematively, changes in structural mobility account for all changes in overall observed mobility. The same is hypothesized for 1 983-88 and 1 988-93. Did postcommunist societies become more fluid at the beginning of the 1 990s, in parallel with the increase in overall mobility rates? Was

36

Czech society ahead in terms of relative flows, and were rigidities most marked in Russia in the 1 990s? More generally, did 'mobility regimes' change as a result of the systemie transitions, especially during the first political transformation, at the end of the 1940s? And if substantial changes really did take place, where did they take place? In fact, this would be an unprecedented finding, bearing in mind that relative mobil­ ity rates appeared to be more or less constant over time. This is what results of cross-time mobility studies carried out in various countries suggest. Table 8. Results offitting models to three-way men 's tab/es of origin category by destination category by time (1948-1 952 and 1952-1963)

Model3 ODT

OT DT

OT DT DIAG

02

2 754 0.00

1 80 93.0 0.00

66 97.4

- 1 20

p Czech Republic

2 54 1 1 .3 0.00

264 90.0 0.00

97 96.3

-96

1.1

95 96.6

- 1 08

Bułgaria rG 2

BIC for

OT DT OD

O.O l

02

2 629

p Hungary

0.00

2 539 3 .4 0.00

2 798

2 768

0.00

0.00

30 1 89.2 0.00

G2 r G2

2 447

2 445 0.9 0.00

1 77 92.8 0.00

83 96.6

-1 02

02 rG 2

940

922 2.3 0.00 50

1 24 97.0 0.00 49

36 99. 1 >0.05 25

-130

rG 2

02 rG2

p Poland -

OT DT OD

p Slovakia p df for all countries

0.00

0.00 60

O.Ol

O.O l

O.Ol

a O - category of origin i n 1 948-1952 and 1 952- 1 963 tables; D - category o f destination i n 1 9481 952 and 1 952-1 963 tables; T - time ( 1 = 1 948-1 952, 2= 1 952-1 963); DIAG (l=off-diagonal cells; 2=diagonal cells)

37 Table 9. Results offitting models to three-way men 's tab/es of origin category by destination category by time (1983-88 and 1988-93) Model3

Bułgaria Q2 rG2 p Czech Republic Q2 rG2 p Hungary Q2 rG2 p Poland Q2 rG2 p Russia Q2 rG2 p Slovakia Q2 rG2 p df for all countries

ODT

OT DT

OT DT DIAG

OT DT OD

5 288

5 219 1 .3 0.00

161 97.0 0.00

36 99.3 >O.OS

- 1 62

6 233 4. 1 0.00

1 97 96.7 0.00

45 99.3 O.o3

-1 59

4 478 1 .4 0.00

1 39 96.9 0.00

34 99.3 O. I O

- 1 62

4 1 76 1 .3 0.00

202 95.2 0.00

52 98.7 0.00

- 1 40

4 871 0.6 0.00

141 97 . 1 0.00

21 99.6 >O. I O

-1 76

5 920 2.3 0.00 50

181 97.0 0.00 49

52 99. 1 0.00 25

-1 5 1

0.00 6 498 0.00 4 543 0.00 4 23 1 0.00 4 903 0.00 6 060 0.00 60

BIC for OT DT OD

a O - category of origin in 1 983- 1 9 8 8 and 1 9 8 8-1 993 tables; D - category of destination in 1 9831988 and 1 9 8 8- 1 993 tables; T - time ( 1 = 1 983-1 988, 2= 1 9 8 8-1 993); DIAG ( ! =off-diagonal cells; 2=diagonal cells)

The results of applying these models are set out in Table 8 (the first transformation) and in Table 9 (the second). We considered our six countries separately, and in each case fitted both the diagonals and the CFM models to a three-way table which comprises six categories of origin, six of destination, and the two transitions (1948-52 and 195263, and 1983-88 and 1988-93). In Tables 8 and 9 I also present the re­ sults of applying two other models. The first tested the hypothesis of the

38

conditional independence of class origins and destinations. Usually, this model is employed to serve as a useful baseline, with reference to which we can assess how much of the total association between class of origin and class of destination the diagonals and CFM models are able to ac­ count for. The calculated G2 statistics are given in the first row of each country's cell and the diagnostic rG 2statistics (O < rG2 < 100) are given in the second row for each country. The rG2refers to 'coefficient of multiple determination' as implemented in log-linear modeling by Goodman (1972). The statistics conceming fit in the case of the condi­ tional independence model are shown in the first column of the tables. In the second column, I report statistics for the model, which assumes that occupational distributions for origin and destination changed be­ tween 1948-52 and 1952-63 (Table 8), and 1983-88 and 1988-93 (Table 9), while still maintaining that categories of origin and destina­ tion were independent. What, then, can we leam from Tables 8 and 9? The results do not de­ scribe a radical change, but something did change with respect to the dynamics of mobility characteristic of the decline of the communist system, followed by its collapse and the emergence of a new social or­ der. In accordance with the findings of previous studies, mobility barri­ ers in East Central European countries basically remained as open (or closed) in the 1990s as they had been in the preceding decade. As re­ gards mobility during the transition to communism, rather unexpectedly, the constant fluidity model does not reproduce core fluidity satisfacto­ rily in all five nations. The second model, which examines the hypothesis that occupa­ tional distributions changed in the two consecutive decades-the 1980s and the 1990s-produces values of G 2 , which are significant in all six countries. This explains no more than 1-4 per cent of the total association, as we can see in the second column of the tables. Even so, it improved the fit to a significant degree. The result is direct support for the thesis that, at the tum of the 1940s and the beginning of the 1990s, occupational distributions underwent significant transforma­ tion. Tuming to the diagonals model, we see that it performs fairly well. While it does not fit the observed data in each country in accordance with the conventional 0.05 criterion, statistical significance is not the only measure of substantive sociological significance. Allowing for constant immobility in 1948-63 and in 1983-93-exceeding what we would expect on the basis of perfect mobility-it accounts for no less

39

than 89-93 per cent of all associations. It must have been the case that self-recruitment prevailed over mobility in both 1948-63 and 1983-93. Nevertheless, the reduction in 0 2 proved unsatisfactory, which indicates that there was circulation between the six categories, in addition to the strong tendency to remain within them. It is appropriate, therefore, to consider whether the constant fluidity model, which assumes that circulation took place, will improve the fit. Prom the fourth column of Table 9, in which the results of fitting the CFM are presented, it is apparent that circulation between six occupa­ tional strata remained basically intact in 1983-93: the CFM reproduces the observed data almost exactly-for each country it accounts for 99100 per cent of the association between class of origin and class of des­ tination. Statistically significant deviations are present only for Czech society, Slovakia, and Poland. The caveat must be that in these countries some significant portion of the discrepancies between observed and ex­ pected values, which more detailed analysis could reveal, pertain to cir­ culation. Nonetheless, even in the Czech Republic, Poland, and Slovakia, cir­ culation rates appear to be largely captured by the core model of con­ stant fluidity over time at the tum of the 1980s. Stability predominated insofar as we are concerned with openness of social structure. It seems reasonable to suppose that the collapse of communism and the rebirth of capitalism in East Central Europe affected the occupational system by relaxing rigidities and closures in social space. Yet no support for this assumption can be found in the current data. Even the transformations of the political and economic system, with their concomitant institutional changes, did not suffice to make class barriers more fluid. In fact, there had been no such development even by the middle of the 1990s. Cer­ tainly, delayed potentia! for growth in circulatory rates of mobility does exist, and may have remained dormant only to emerge and to reshape life-careers in later years. Thus, the unchanged openness of social structures paralleled a slight increase in overall mobility after 1988. The sources of the ascendant trend consisted in changing occupational distributions-we have already shown that the class of proprietors demonstrated particularly rapid growth in the 1990s in all six societies. The shift towards greater overall mobility supports our third hypothesis, that increasing flows in the 1990s resulted from macrostructural changes, enforced chiefly by eco­ nomic transformations which created new positions in the division of labor. The mobility regimes remained stable.

40

This is even mare true of the transition from the communist system. Fluidity pattems did not come to be a sociological constant during the period when the system was being imposed. In all five countries, the G2 s returned by the constant fluidity model account for no less than 98 per cent of the G2 s returned by the independence model. Nevertheless, these results indicate significant deviations in the core, unchanging fluidity pattem, except for Slovakia. 1 It is true that the deviations are small and not sufficient to cause us to abandon the idea of basie continuity over time: stability in terms of pattems of opportunity, rather than dynamics, also predominated several decades ago. Nonetheless, despite the over­ whelming stability, the implementation of communism opened up the social structure to same extent.

INFLOW TO BUSINESS The uptum in overall mobility in the 1990s may be attributed exclu­ sively to the transformations of occupational structures. In this respect, the differentia specifica of the exit from communism became the ex­ pansion in the number of private businesses. But did the resulting rise in inflow to this category from the other strata account for the increase in overall mobility during this period? Table 1O shows the proportion of men across our six societies who in 1988 and 1993 found themselves in an occupational category different from the one they had been in five years previously. The table shows the inflow rates to the intelligentsia, lower nonmanuals, owners, skilled workers, unskilled workers, and ag­ ricultural categories. The generał impression gained from Table 1O is that in the 1990s owners clearly accounted for the highest proportion of newcomers rela­ tive to the other categories. Between 1983 and 1993, the class of lower nonmanuals-clerical workers, teachers in elementary schools, techni­ cians, shop assistants, receptionists, and so on-was also a significant extemal recruiter of labor. Indeed, in 1983-88, this category had the highest rate of inflow. For example, in Poland, the inflow rate to lower nonmanual occupations stood at 77.1 per cent in 1988 and in Russia it amounted to 86.4 per cent. Although in the 1980s the inflow to lower nonmanuals ranged from 77 per cent to 86 per cent in five of the six na­ tions, it had decreased greatly by 1993-except in Slovakia, where it had actually increased. Simultaneously, the rate of inflow into owner­ ship was generally either maintained or increased, Bułgaria being the

Table 1 O. Men 's injlow rates to the intelligentsia, !ower nonmanuals, owners, skilled workers, unskilled workers, andfarming Socio-occupational categories

Bułgaria

Czech Republic

Hungary

Poland

Russia

Slovakia

1 983-88 1 988-93 1 983-88 1 98 8-93 1 983-88 1 98 8-93 1 983-88 1 9 8 8-94 1 983-88 1 98 8-93 1 983--88 1 988-93

9. 1

1 7. 1

8.2

16.1

1 6. 7

1 6. 5

7.2

13.1

8. 1

1 5.7

8. 1

20.0

Lower nonmanuals

79.3

1 5 .3

80.6

1 9. 9

79.4

1 2.6

77. 1

1 .3

86.4

I I .O

1 7.0

2 1 .4

Owners

65.0

55.7

55.0

90.2

40.8

40.6

50.8

59 . 1

35.8

67.6

70.0

89.5

Skilled workers

1 0.8

l i.I

8.0

I I .O

7.9

I l .O

1 1 .5

8.6

I O.O

1 1 .3

6.4

8.0

8.7

1 4. 7

8.9

1 3 .6

1 2.2

1 2.3

1 4. 7

1 1 .4

1 1 .4

15.1

8.5

1 5.5

1 4. 1

1 5 .6

1 1 .7

19.1

1 6. 5

26.4

1 9.7

I O. O

8.5

1 2.3

4.3

20.5

Higher professionals-intelligentsia

Unskilled workers Farmers and farm workers

.i,.

42

only exception, experiencing a reduction of 1O per cent. In consequence, owners remained the most transient category at the beginning of the 1990s. They accounted for the highest influx in the Czech Republic and Slovakia, where only 1 0 per cent of owners found themselves in 1 993 in the same position as in 1988. The intelligentsia, lower nonmanuals, manuał workers, and farm categories were self-recruited to the extent of 78.6 per cent (in the Slovakian case) or more. If we look at the volume of mobility the issue of class formation in­ evitably arises. Discussion of the homogeneity of different strata in terms of their recruitment pattems is another way of looking at the de­ gree of their 'demographic identity' (Goldthorpe, 1987), that is, the de­ gree to which they have formed collectivities of individuals and families identifiable through the continuity of their association with sets of strata-positions over time. From this standpoint, a number of national differences can be identified as regards the self-recruitment of owners. In the Czech Republic and Slovakia, owners seem the least homogene­ ous category in terms of business experience. Nine out of ten owners in 1 993 had recruited themselves from another category of origin as of 1988. In Bułgaria, Poland and Russia, on the other hand, this proportion did not exceed 50 per cent. Given such heterogeneity of origin, one can scarcely predict the direction in which the formation of owners into a sociocultural entity might proceed. In Slovakia and Czech society, they were an amalgam of various social circles. In fact, this is a new social category, composed of social backgrounds, orientations, lifestyles, and value systems which originated in different strata. Mobility into busi­ ness, which derives from systemie transformations, has led to the disin­ tegration of this category as a class. As far as the consequences of social mobility for class formation in postcommunist countries are concemed, the considerable heterogeneity of owners seems the most striking by­ product of the systemie changes.

CON CLUSION : MORE OPENNESS DURIN G THE TRANSITION TO COMMUNISM THAN AFTER ITS COLLAPSE ? What do our findings on mobility tell us that is of relevance to our cen­ tral concem? The years immediately following the Second World War saw fundamental transformations in the economies and political systems of East Central European countries. These changes have their counter-

43 parts in the 1990s. The aim of the present study is to establish mobility rates in both critical periods. Overall rates appear higher in the 1990s than in the preceding decade: in terms of comparative macrosociology, this was to be expected. The six national economies with which we are concemed began to switch over to mechanisms govemed by the rules of the capitalist market, and the democratization of public life removed formal obstacles to prominent positions in politics and private business. Which of the two transformations did more to disrupt the hindrances to occupational mobility? According to our data, the transition to com­ munism in the late 1940s released more intensive flows between basie segments of the social structure than occurred during the exit from communism in the 1990s. Social mobility in East Central Europe re­ sponded to institutional transformation in a way predicted by Sorokin (1958). If we tum to mobility into specific social strata, we find, not surpris­ ingly, that the class of owners experienced the highest influx in the 1990s. Due to mass privatization, social space was 'freed up' for those who wished to go into business. As far as international comparisons are concerned, it is important to mention that in the 1990s mobility increased the most in Czech society and the least in Russia. Regardless of the generally low variation be­ tween our six countries in respect of increasing mobility, the same pat­ tern is repeated in the case of other changes in the occupational system. The relative size of the 'owners' category also increased the most in the Czech Republic (followed by Slovakia). lndeed, this category experi­ enced the highest influx in the 1990s in all six countries. Self­ recruitment into business was the highest in Russia-its social composi­ tion changed the least in comparison to the other five societies. As we are seeking consistency between the transformation of the economic system and the dynamics of mobility, the data give us reason to regard the lower level of mobility in Russia as indicative of a slower pace of social changes overall, in stark contrast to Czech society, whose social upheaval seems to have been the most striking.

44 Note 1 The statistics of fit for the models are shown, by country, in the third rows of Tables 8-9. As regards mobility in 1 948-63, the CFM model produces a satisfactory fit by conventional standards only for Slovakia, that is, as indicated by p values. This ex­ ceptional case can be explained in terms of a relatively small number of men-3 57 and 3 6 8 respectively-in mobility tables for 1 948-52 and 1 952-63. In the last col­ umn of both tables, BIC statistics are given-only for the CFM model-which are not sensitive to the number of cases employed in the analysis. The BIC (Bayesian Infor­ mation Criterion) is the statistic of a model fit which is independent of sample size and is especially recommended for selection between models if the number of cases involved in the analysis is very large, that is, when large N makes it almost impossi­ ble to get a satisfactory fit in accordance with standard criteria (statistical significance of G2 ). The rule in this case is that the best model is the one with the smallest BIC (see Raftery, 1 987). Applied to Slovakia, the fit of the CFM model falls short of sig­ nificance for 1 948-63 mobility tables.

CHAPTE R 3

SOCIAL MOBILITY PATTE RNS: A BASIC CONTINUITY

PITIRIM Sorokin was probably right in his contention that, on a long­ term view, mobility rates display no continuous direction of change but merely 'trendless fluctuation'. I would extend this statement by saying that not only mobility but processes of social stratification in generał tend to fluctuate. They are marked by continuity and not by abrupt changes, a fact which we clearly documented for our six East Central European societies in Chapter 2. lt seems that in areas strongly exposed to the impact of factors ex­ ogenous to the inherent logic of social stratification, even a couple of years may witness rapid 'ups and downs', a fact convincingly proved by the data, particularly on income distribution. Government welfare pol­ icy, resulting in redistribution of economic benefits (Esping-Andersen, 1990), or centrally imposed allocation of financial resources by the so­ cialist state (Domański, 1990), exemplify what we refer to as the opera­ tion of exogenous forces. While social dynamics constitutes the obverse side of continuity, rapid breakup and upheavals rarely occur in stratifi­ cation systems. Mechanisms exist which serve to anchor social stratification. By pur­ suing their dynamics over time we can obtain some insight into the fi.m­ damental question: Do really substantial restructurations occur? In what follows, we shall focus on the dynamics of mobility pattems among the basie segments of stratification systems. Mobility pattems, together with their underpinnings, are traditionally regarded as forces which stabilize social structures. Analyses of this kind aim primarily to determine the map of transitions between socio-occupational strata considered as constitutive of the stratification system. Ultimately, however, they ad­ dress the more generał issue of which social categories are more, and

46

which less, distant from one another in a social space defined in terms of patterns of inter- and intragenerational movements. Looked at in this perspective, the configuration of barriers and distances shows the gen­ erał shape ofsocial stratification. Weber (1968) and Sorokin (1958) laid out the theoretical rationale for taking mobility patterns as a synthetic outline of the contours of so­ cial stratification. In attempts to update these classical formulations, leading neo-Weberians, such as Giddens (1973), Parkin (1979), Ste­ phens ( 1 979), and Goldthorpe (1987), have specified the composite factors of social mobility which affect social structuration. Among the mechanisms of allocation to consecutive ' echelons' of the stratification ladder, the significant roles of cultural and economic capital, 'good' or ' bad' family background, and finally market capacities are identified as factors which either facilitate 'getting ahead' or put obstacles in its way. There are various strategies of obstruction and forms of corporatism which-as applied by trade-union associations or professional bodies­ tend to solidify social distances between classes, strata, and occupa­ tional categories. In a concise recapitulation advanced by two adherents of the theory of industrial society, mobility is regarded as a crucial me­ diating process between social structure and the creation of identities, interests, and actions. It determines where, and with what degree of sharpness, lines of cultural, political, and social division are drawn (Erikson and Goldthorpe, 1 992, 2). Solid grounds exist for believing that an analysis ofmobility patterns will illuminate the problem of whether continuity or change prevails in synthetically portrayed social divisions in East Central European coun­ tries. It seems reasonable to assume that, if any substantial restructura­ tions in these societies have occurred, they carne into play at the core which keeps social stratification in dynamie equilibrium. All in all, analysis of mobility patterns should make our conclusions more robust in relation to the dilemma: stability or change? I shall focus on the cross-time consistencies and cross-national simi­ larities found in previous comparative analyses performed across a number of countries, differing in terms of their level of economic devel­ opment, political systems, and cultural traditions. What these studies revealed were elear barriers between: (i) the agricultural and nonagricul­ tural populations; (ii) blue- and white-collar workers; and (iii) the occu­ pational elite (managers, high government officials, professionals) in contrast with the remaining nonmanual categories. In some countries, a distinct position in the structure of intergenerational movements was

47

held by entrepreneurs. Lastly, the most generał mechanism goveming intergenerational transitions proved to be the location of individuals in the hierarchy of socioeconomic status (Domański and Sawiński, 1991; Sawiński and Domański, 1989). Similar results were obtained in studies which used different theoretical approaches and analytical techniques (see Goldthorpe and Payne, 1985; Breen and Whelan, 1985; Hout, 1988; Luijx and Ganzeboom, 1987; Erikson and Goldthorpe, 1992). Such empirical evidence confirmed the relevance of utilizing mobil­ ity pattems to map out the shape of social stratification at the macro level. The findings referred to above are not unique to social mobility, but generally reflect pivotal features of stratification systems: pattems of intergenerational movements closely resemble distances generated by distributions of rewards, resources, and assets strategie for life-careers, and conform to pattems of marital choice, selection of friends, and many other forms of social structuration (Laumann, 1973; Treiman, 1977; Mitchell and Critchley, 1984; Hollinger and Haller, 1990).

H YPOT HESES: W HAT MI G H T C HAN GE ? In which direction might the shape of social divisions be transformed, and which aspects of these changes might be discemed in pattems of mobility? Any attempt to address these questions must be considered against the background of the generał context of systemie changes in East Cen­ tral Europe. The East Central European countries have slowly, but firmly, abandoned the command economy, approaching a system domi­ nated by the capitalist market. As a result, the natura! referent · in our search for structural changes will be contemporary market democracies. Certainly, if stratification systems in East Central Europe have un­ dergone substantial transformation, this will be detectable in areas which were mostly subordinated to the logic of 'socialist' stratification and regarded as peculiar to this system. Particularly imprinted on the shape of social stratification in East Central Europe was a state-con­ trolled system of economic benefit distribution. In the years prior to the transformation, this underwent change in the direction of an enhanced role for the rationality of market-like rules of distribution. The data seem to indicate two new elements which might reshape mobility chan­ nels in East Central Europe. The first is growing economic inequality, the second, new rules governing income distribution. Cross-time analy-

48

ses for Poland revealed that, in the 1990s, inequality of earnings was much higher than in the preceding decade. This tendency is reflected in the growing span of differentiation. When measured in terms of a coef­ ficient of variation (V), it appears that in 1987 the span of differentiation was 0.44, increasing to 0.87 in 1992, and then falling back a little in 1994 to 0.77. Turning to the whole adult population, it appears that dif­ ferentiation of family incomes per capita rose from 0.42 in 1982 to 0.82 in 1994 (Domański, 1994). At the same time, in Hungary the disparity between family incomes of the highest and lowest deciles increased from 4.6 in 1987 to 6.3 in 1993 (Kolosi and Róna-Tas, 1993). In Czech society, the Gini coefficient for earnings rose from 0.19 in 1988 to 0.27 in 1996 (Vecemik, 1996). These developments resulted in growing distances between basie so­ cio-occupational strata. Of particular importance was the rise in earnings of the occupational elite in the 1990s: the average earnings of the intelli­ gentsia and cadres, notably underrewarded in the socialist economy, went up, and those of manuał workers and farmers (smallholders) fell markedly (Domański, 1994). These tendencies evidence a basie reorientation in the generał rules of reward-from socialist principles of distribution towards meritocracy. In the Czech Republic, Poland, and Slovakia-the relevant data come from these three countries-the demise of the command econ­ omy removed the central distribution of financial resources grounded chiefly in a non-economic approach. In the Polish case, analyses docu­ ment a slight increase in the net effect on earnings ofboth number ofyears of schooling and occupational status between the late 1980s and 1991-93. Furthermore, the net financial returns for supervisory positions improved from about 10 per cent above the grand mean in 1987, to 20 per cent in 1993 (Domański, 1994). Similarly, in the Czech Republic, the GDR, Hungary, and Slovakia, personal incomes became more consistent with level of education and occupational position (Kolósi and Róna-Tas, 1993; Machonin and Tucek, 1994; Mateju and Rehakova, 1993; Mateju and Lim, 1996; Headey et al., 1995). The consequences ofthe apparent redistribution of economic benefits might stimulate mobility. In particular, higher positions may come to be considered more attractive by members of the lower classes, who will as a result seek to improve their market position and secure higher eco­ nomic benefits on a competitive basis. At the same time, however, an increasing propensity to move upward might be counteracted by members of the established occupational elite. The East Central European intelligentsia, traditional in its mentality and

49

orientations is undergoing a transformation into professions. It is per­ fectly possible that the monopolization of skills and services will enable the former intelligentsia to control the supply side of labor, as in market societies. The growing reliance on credentials will become a precondi­ tion of entry to the professions, as well as a handy device for ensuring that those who possess cultural and intellectual capital, skills and exper­ tise, receive the best opportunities to transmit the benefits of profes­ sional status to their children. We can reasonably predict the emergence of strategies of closure and rules of meritocratic reward distribution. We would expect occupational strata to differ in their bargaining power, and that socioeconomic status has much to do with who is more and who is less successful in attaining better pay, insurance schemes, more secure jobs, and so on. Much more than in the past, the relentless forces of the capitalist labor market will shape succession to jobs and recruitment pattems in both an inter- and an intragenerational perspective. Our data make it possible to check whether, in the 1990s, a differentiation determined by socioeconomic status has grown in importance in structuring intra- and intergenera­ tional movements relative to preceding decades. Changes in distribution systems have been most tangible in recent years. Nevertheless, the effect on mobility, if it really occurred, was mediated by numerous factors. It may have been restricted to barriers separating off the top positions on the occupational ladder. We might ask which effects might have resulted from the reshuffling of political elites and which were more productive in terms of pattems of mobility, so constituting another signum temporum of systemie transformation in East Central Europe. The top political positions have undergone considerable changes in personnel, but it is difficult to say whether the barriers dividing the po­ litical elite from the other segments of the social structure are solidifying or becoming more open. Despite the dissolution of the socialist no­ menklatura, elite positions remain-as they do everywhere. Recruitment formerly based on allegiance to the Communist Party was replaced after 1989 by a system based on other political and ideological affiliations. What really took place was an exchange of incumbents. This affected individuals, who moved up or down, but not the positions themselves, which, as slots in a structure, persisted practically untouched. As far as possible shifts in mobility barriers are concemed, we still have to consider the rapid expansion of private entrepreneurial activity. lt should be remembered that, in view of the regular pattems character-

50

istic of many countries, both nonagricultural proprietors and, especially, farmers exhibit distinct pattems of mobility which determine their loca­ tion in social space (Sawiński and Domański, 1989). This clearly dem­ onstrates, by the way, that control over the means of production, so im­ portant to Marxist theory, was highly effective in shaping occupational mobility opportunities. The results of mobility studies have shown that the distance between agricultural and nonagricultural categories was the greatest in compari­ son to all other distances between basie occupational strata, proving that the factors underlying the pattems of transitions for the agricultural and the nonagricultural sectors were different. The distinctive position of farmers and farm laborers resided in their exceptionally high level of self-recruitment. A syndrome exists which restricts a massive outflow from agricultural categories (as well as preventing any inflow). This in­ cludes, first, the particular nature of the farm as a place whose mainte­ nance and operation is regarded by its members as a primary value, a view which leads to continuity and attachment. The second constituent of the stability syndrome in forming is the limited opportunities avail­ able for converting land, machines, livestock, and context-specific qualifications into marketable assets. Thirdly, socialization takes place largely in terms of attachment to 'the soil' and the farm. The role of education in socialization is very small, as a consequence of which aspi­ rations to upward mobility are relatively low and the universal norms of achievement preponderant in the nonagricultural strata rarely come into play. Fourthly, cultural distinctiveness is considered a positive value, an attitude which militates against changes in social position as incompati­ ble with the need for assimilation and integration-this is felt particu­ larly strongly by the members of traditional communities. Finally, geo­ graphical isolation means that a person's social passage to a nonagricul­ tural category requires a simultaneous change in place of residence, so putting up an additional barrier to mobility. These mechanisms are strongly embedded in the tradition-oriented structures and culture of East Central European countries; it seems unlikely that their weight had substantially reduced by the beginning of the 1990s. Obviously, one would expect them to act with unequal strength in, for example, the Czech Republic and Poland. In the latter, private forming existed through all periods of communist rule. In 1993, Polish peasants accounted for 12 per cent of the actively em­ ployed, still working almost entirely on smallholdings which will re­ quire considerable restructuring if they are to compete with Westem

51

producers. The peasant mentality is very traditional, fostering a ten­ dency to immobility. Apart from Poland, private farming in the other postcommunist so­ cieties has effectively been reinstated. The intriguing question arises: Which factors will shape recruitment into this category-in-the-mak:ing? Is 'peasantification' taking place or is a modem path of development being followed, inflow being driven by what we have referred to as the allocative power of socioeconomic status, which appears to affect the mobility of the nonagricultural strata dramatically? We should bear in mind the growing marketability of agriculture in East Central Europe, in parallel with the infrastructural modemization ofrural communities. Our analysis will shed light on how the first years of renewal of private farming in the postcommunist countries have influenced pattems of en­ try to the agricultural sector. Previous analyses have shown that, in some countries, nonagricul­ tural proprietors show up in a separate dimension of mobility space. Not surprisingly, their distinctiveness appeared more pronounced in capital­ ist countries-especially in France-in comparison to Hungary and Po­ land (Sawiński and Domański, 1989). Nonagricultural proprietors also stand out because of their considerable self-recruitment and inheritance. The ownership of a private firm or of economic assets generally consti­ tutes a specific type of resource, passed down in families of owners from parents to children. It seems that means of production are less con­ vertible than other forms of capital, so explaining the distinctive loca­ tion of proprietors in mobility space. Private entrepreneurs are the second newly emerging category in East Central Europe. But do entrepreneurs constitute a distinctive social segment in the structure of mobility pattems and is their social location similar to that of their counterparts in the advanced industrialized coun­ tries?

PATTERNS OF MOBILITY In our investigation of the effects of systemie transformation on the structure of mobility we compared configurations of intergenerational mobility pattems established for two points in time: (i) 1988, the year immediately preceding the collapse of communist rule, and (ii) 1993, when institutional changes in political systems and economies, once brought into effect, might have initiated some restructuration in stratifi-

52

cation systems. Any substantial differences in the configuration of the basie divisions between these two points in time, as reflected in the pat­ tems of intergenerational movement, will be regarded as indicative of changes in social stratification. It is a matter of further debate to what extent these potentia! changes may be affected by transformations in the political and economic systems, or derive simply from the natural logic of social structuration. We shall look in same detail at the mobility tables relating to fathers and sons, moving on from the short time-perspective of the decade 1 983-93 to pattems of intergenerational movement. The latter encom­ pass a mare comprehensive set of constraints and opportunities for the transmission of benefits and disadvantages, and so stand as a mare valid representation of basie divisions and cleavages in social stratification in comparison to work-life pattems. In order to figure out distances in mobility space, discriminant analysis will be used, which represents a family of methods particularly suitable for extracting the structure of flows in a two-way table (such as canonical correlation analysis, multi­ dimensional scaling, correspondence analysis, and so on). In order to grasp the configurations of mobility barriers in a compre­ hensive way, I expanded the sixfold EGP categorization, as applied in Chapter 2, into an eightfold version. A division was made between: (i) higher-grade professionals, administrators, officials, managers of large industrial establishments, and large proprietors, (ii) lower-grade professionals, administrators, officials, higher-grade technicians, managers of small industrial establishments, and supervisors of nonmanual employ­ ees, (iii) routine nonmanual empłoyees in administration, commerce, sales, and services, (iv) owners with or without empłoyees, (v) łower­ grade technicians and supervisors of manuał workers, (vi) skilled work­ ers, (vii) semi- and unskilled workers, and (viii) farmers and agricułturał workers. The discriminant anałysis utilized in this study aims to determine dif­ ferences between two or mare groups and a set of discriminating vari­ abłes (see Kłecka, 1 980). Since it considers the groups to be defined as a single nominał-łeveł variabłe-with each vałue denoting a different group--discriminant anałysis can be looked at as a technique which re­ lates one nominał variable to several interval-leveł variables. In this study, a discriminant anałysis of intergenerational mobility is performed by representing sons' categories in 1 988 and in 1 993 as nominał vari­ abłes, and fathers' categories when respondents were 1 4 years of age as a set of dummy variables (expłanatory ones).

53

Discriminant analysis fits discriminant functions which are a simple linear combination of the discriminating variables. For the first function, coefficients, referred to discriminant weights, are derived, so that the group means on the function are as different as possible. The weights for the second function are also derived to maximize the differences be­ tween the group means, but with the added condition that values on the second function are not correlated with the values on the first function. The consecutive functions are derived in the same fashion. Each discriminant function also gives score values for group cen­ troids of the nominał variable. These are points in a multidimensional space where all of the discriminating variables have their average values over all cases belonging to given groups. We will be interested only in the first two discriminant functions, which in fact account for the vast bulk of relations between 'origin' and 'destination' categories in mobil­ ity tables. The canonical correlation coefficient will be the measure of association which summarizes the degree of relatedness between the groups and the discriminant function. In sum, discriminant analysis allows one to transform the observed mobility flows between class categories into distances in multidimen­ sional space. If mobility flows reveal the underlying location of the categories within social stratification-as has been suggested in previ­ ous theoretical and empirical studies-the discriminant analysis solution can be interpreted in terms of the structuration process. We shall present our findings in the same perspective.

MAP OF DISTAN CES It is a perennial question whether, and to what extent, parents transmit their social positions to their offspring. The strength of correlation be­ tween the occupational categories of fathers and their children is com­ monly taken to identify the degree of intergenerational inheritance of placement in the social hierarchy. In Table 11, coefficients of first canonical correlation are given. Scanning their values, one can see that in the 1990s both men and women tended to inherit the positions of their fathers no less strongly than in the late 1980s. Mobility routes have not been opened up-this is what a decline in correlations would indicate, although mobility barriers have not become more rigid either-since in that case the values of cor­ relation coefficients would have increased.

Ul .j:,.

Table 1 1 . Coejficients of canonical correlations between occupational categories offathers and sonslda ughters in 1 988 and 1993 Bułgaria 1 988

1 993

Czech Republic 1 988

1 993

Hungary 1 988

1 993

Poland 1 988

1 994

Russia 1 988

1 993

Slovakia 1 98 8

1 993

Canonical correlations between father' s occupational category and category of son/daughter in 1 988 or 1 993

Men

0.3 I

0.30

0.34

0.3 1

0.32

0.32

0.34

0.36

0.28

0.28

0.29

0.3 1

Women

0.32

0.32

0.35

0.34

0.30

0.32

0.34

0.33

0.25

0.24

0.30

0.29

55

It is difficult to trace substantial cross-national differences in the open­ ness of the social structure. However, in Poland the 'father-son link' is slightly closer and in the Czech Republic slightly looser. Cross-national similarities and stability over time prevail over differences and change. It appears that, at the beginning of the 1990s, people were following in their fathers' footsteps just as they had in the preceding decade. In the remainder of this chapter, we shall restrict our treatment of class structuration patterns to the inspection of małe social mobility. In fact, we replicated the same analyses with the mobility tables ofwomen. It turned out that, in each country, the configuration of intergenerational mobility barriers in the female population closely fits the patterns for men: as a result, we will not discuss women separately, on the assump­ tion that what we found for men more or less represents the structure of basie social divisions overall. Table 12 presents discriminant analysis solutions for our six coun­ tries. In each country' s two columns we show discriminant weights for class categories of sons in 1988 on the first (the dominant) and second discriminant function (that is, the next in terms of discriminatory power). Original, standardized values of weights were transformed into a O to 100 points scale. The discriminant weights for fathers' categories are not listed because of space limitations. The focus of our interest will be the recruitment patterns of the analyzed classes. In Table 13, we show analogous scores for categories of sons in 1993 (1994 in the case of Poland). By way of visual illustration, the score values are plotted in Figures 1 to 12 with vertical and horizontal axes corresponding to the first and second discriminant functions (scores for the second variate were scaled proportionally to the ratio between second and first canoni­ cal correlation). These twelve plots will serve as the basis for a discus­ sion on the dynamics and stability of social distances and divisions un­ der systemie change.

°'

Ul

Table 1 2 . Discriminant weights on first and second discriminant functions for tab/es of mobility from fathers ' to sons ' categories in 1988 Discriminant weights: Socio-occupational categories in 1 988

Bułgaria

Czech Republic

Hungary

Poland

Russia

Slovakia

1 st

2nd

1 st

2nd

1 st

2nd

1 st

2nd

1 st

2nd

1 st

2nd

1 00.0

13.1

91.1

5 1 .2

1 00.0

39.5

86.0

88.6

98.0

52.5

1 00.0

3 5 .4

Semi-professionals

94.5

50.7

8 5.4

42. 8

75.4

50.9

8 1 .7

700.0

77.3

63 . 1

86.9

29.3

Routine nonmanual, sales and service workers

67.3

93.2

8 1 .3

50.8

22.4

67.2

94.2

84.5

O.O

59.2

49.6

26.4

Owners

44.0

1 00.0

1 00.0

42.2

28.6

1 00.0

1 00.0

82.3

1 00.0

1 00.0

77.4

93.3

Lower-grade technicians

97.9

83.6

53.3

1 00.0

34.7

63.7

80.4

30.9

27.4

46.8

5 5 .6

1 00.0

Skilled workers

50.8

93.0

34. 1

75. 1

20.8

84.2

7 1 .2

O.O

O. I

79. 1

20.3

79.5

Unskil!ed workers

33.4

1 00.0

25.9

52.2

9.3

4 1 .6

52.7

1 2.2

0.9

54.0

1 5. 7

1 8.4

O.O

O.O

O.O

O.O

O.O

O.O

o.o

74. 8

0.3

O.O

O.O

O.O

Higher professionals and managers

Farmers and agricultural laborers

Table 1 3 . Discriminant weights on first and second discriminant fimctions for tab/es of mobility from fathers ' to sons ' categories in 1 99 3 Discńminant weights: Socio-occupational categońes in 1 993

Czech Republic

Bulgańa

Hungary

! st

2nd

! st

2nd

1 st

1 00.0

1 00 . 0

1 00.0

64. 9

1 00.0

Semi-professionals

95. 1

50.4

94.4

59. 5

Routine nonmanual, sales and service workers

93 .4

20.2

66.6

Owners

94.3

5 1. 4

Lower-grade technicians

95.5

Skilled workers Unskilled workers

Higher professionals and managers

Farmers and agricultural laborers

Poland

2nd

Russia

Slovakia

1 st

2nd

1 st

2nd

1 st

2nd

6.8

1 00 . 0

75.6

88.4

67.4

9 1 .7

48.2

74. 8

82.0

92.4

1 00 . 0

95.2

45.8

1 00.0

47.4

83.7

1 8. 1

89. 1

99. 1

50. 1

5 0. 8

6 1 .0

26.7

34. 1

75.7

71.1

56.4

1 00.0

85.0

54.9

1 00.0

1 00.0

74. 5

59.4

20.4

66.5

1 00.0

3 1 .0

64.3

77.0

49. 1

70.0

54.0

40.4

94. 1

63 .4

o.o

35.7

96.3

27.2

92. 1

69.9

o.o

1 6.9

74.4

1 7. 8

1 00.0

44.9

O. I

28. 1

70.8

1 5.5

47.4

8 5 .6

1 9. 1

1 8.4

67.8

1 5. 8

o.o

O.O

96.2

o.o

o.o

o.o

o.o

o.o

66.9

O.O

o.o

o.o

1 7.4

V, --.J

58 HP

•

LP LT •

•

RN

• •

SW PR

• •

UW

FA

•

Figure I . Bułgaria, 1 988. Discriminant weights for sons ' categories in 1988 on first and second discriminant variate Note: The exact values of coordinates are given in Tables 12 and 1 3 . The occupational categories are : HP - Higher professionals and managers; LP - Semi-professionals; RN - Routine nonmanual, sales and service workers; PR - Owners; L T - Lower-grade technicians; SW - Skilled workers; UW - Unskilled workers; FA - Farmers and agricultural workers.

HP •

LP LT • • PR RN •

•

SW

• •

UW

FA

•

Figure 2. Bułgaria, 1993. Discriminant weights for sons ' categories in 1993 on first and second discriminant variate PR • HP • LT SW • uw •

•

FA

•

Figure 3 . Czech Republic, 1988. Discriminant weights for sons ' categories in 1 988 on first and second discriminant variate

59

PR

RN • LT • SW

UW

•

FA

•

•

Figure 4. Czech Republic, 1993. Discriminant weights for sons ' categories in 1993 on first and second discriminant variate HP

•

LP

•

L':f PR RN •

•

SW

UW

•

FA

•

•

Figure 5 . Hungary, 1988. Discriminant weights for sons ' categories in 1988 on first and second discriminant variate HP

•

LP

•

• • RN •

PR

LT SW • UW FA

•

•

Figure 6. Hungary, 1993. Discriminant weightsfor sons ' categories in 1993 on first and second discriminant variate

60 PR RN• HP LP •

•

LT

•

•

SW

UW

•

•

FA

•

Figure 7. Poland, 1 988. Discriminant weights for sons ' categories in 1988 on first and second discriminant variate HP RN

• •

LP

•

SW UW

•

•

FA

•

Figure 8. Poland, 1 994. Discriminant weights for sons ' categories in 1 994 on first and second discriminant variate PR

HP

•

LP

•

•

LT

•

•

FA UW RN SW

•

•

•

Figure 9. Russia, 1988. Discriminant weightsfor sons ' categories in 1988 on first and second discriminant variate

61 PR

• • HP • LT • RN •

LP

UW • SW

•

FA

•

Figure 1 O. Russia, 1993. Discriminant weights for sons ' categories in 1993 on first and second discriminant variate

•

HP

LP

•

•

PR LT

RN

•

SW

•

UW

•

FA

•

•

Figure 1 1 . Slovakia, 1988. Discriminant weights for sons ' categories in 1988 on first and second discriminant variate

• •

LP HP

PR

•

LT UW

•

FA

RN

•

• •

SW

•

Figure 1 2. Slovakia, 1 993. Discriminant weightsfor sons ' categories in 1993 onfir�·t and second discriminant variate

62

C L A S S I C A L H I E RA R C H Y Inspection of the plotted configurations of discriminant weights gives the generał impression that, in each country, the configuration of dis­ tances approximates the classical hierarchy of socioeconomic status: that is, an ordering of the class categories on the status scale provides a strong explanation of the position of most categories within the structure of intergenerational movements. There is an apparent tendency for higher professionals to occupy an extreme position at one end of the dominant axis. They are situated somewhere close to private entrepre­ neurs in the Czech Republic (1988), while in Poland and Slovakia they are closer to a number of other nonmanual strata, and in Russia to both nonmanuals and entrepreneurs. At the other end of the scale are agricul­ tural categories and manuał workers. Two main results are apparent. First, the structure of intergenera­ tional movements in East Central Europe strongly resembles the generał shape of basie distances and rigidities in the West which we know from previous studies. Secondly, it is not only the strong cross-national simi­ larity but also the lack of substantial changes in the critical period 198894 which dominate the scene. This conclusion carries a significant im­ plication, to which we shall return. For the moment, it is worth pursuing other regularities by considering the most salient distances between the class categories. A G R I C U L T U RA L/N O NA G R I C U L T U RAL D IVERGE N C E The distinctive placement in mobility space of farmers and farmworkers emerges sharply from earlier analyses. Our data also show that our six East Central European societies tend to conform to this regularity. Only in Slovakia are małe agricultural categories not clearly separated from manuał workers. In Hungary, the division between agricultural and non­ agricultural categories seems visible, but we can scarcely count it among the basie lines of social structuration. The four other countries exhibit a strong separation of agricultural categories from the remaining classes and this distance reproduces itself over time. In fact, one can recognize on the first and second discriminant func­ tions the same pattem, which displays a sharp contrast between farmers and all nonagricultural categories. They reinforce each other to make the agricultural/nonagricultural division still more evident. Plainly, the

63

considerable distance between these segments in all six countries results from the low level of intergenerational movements between agricultural and nonagricultural strata. The social mobility of farmers and members of non-farm categories is organized around distinct dimensions, a fact determined by the particular nature of social resources in farming, such as their lack of exchangeability on the nonagricultural labor market, as well as other factors already mentioned. Unquestionably, the mechanisms underpinning the unique recruitment pattems to farm categories are connected to their relatively highest rates of self-recruitment. For example, in 1993 the percentages of farmers and farm laborers who originated from farming in Bułgaria, Poland, and Rus­ sia stood at 76.1 per cent, 82.4 per cent, and 57. 1 per cent respectively. The self-recruitment rates for Czech, Hungarian, and Slovakian farmers and farm workers were markedly lower, amounting to 50 per cent, 47.6 per cent, and 41.7 per cent respectively. It may be that in communist Czechoslovakia and Hungary agricultural categories were mare closely integrated with the state sector, in keeping with the higher level of mod­ emization of occupational activities and roles. Farming in these countries seemed organized mare in accordance with the requirements of rules of work resulting from technological constraints than in Bułgaria, Poland, and Russia, where agriculture is still characterized by mare traditional practices. According to a claim made in early 1 995 by Prime Minister Klaus, Czech agriculture met Westem European standards and the Czechs would comply with European agricultural policy after the latter had been reformed (Zagrodzka, 1995). It may be that mobility routes between agri­ cultural and nonagricultural categories in the Czech Republic were, in fact, relatively open, in confirmation of our findings.

T HE BLURRED MANUAL/NON MANUAL DIC HOTOMY When looked at in a cross-national perspective, the mobility barrier between nonmanual and manuał categories appeared the second most conspicuous cleavage within the social structure (Sawiński and Do­ mański, 1 989). With respect to our six East Central European countries, the degree of this dichotomy was conspicuously unequal. A thorough analysis of Tables 1 2 and 1 3 shows that, in the early 1 990s, it consti­ tuted a visible demarcation line in Bułgaria, the Czech Republic, and Poland. The values of discriminant weights identifying manuał workers fall clearly below those for nonmanual categories.

64

It seems that this dichotomy has proved to be stable in both Czech and Polish societies in recent years. The social distance between these two broad segments was much the same in 1988 as in the 1990s. At the same time, in Bułgaria the contours of division between the working-class and nonmanual strata sharpened, in contrast to Russia and Slovakia, where, according to the configuration of weights in the two discriminant functions, this division remained blurred until 1993 : that is, categories of workers and routine nonmanuals overlapped. Fi­ nally, in Hungarian society, routine nonmanuals closely neighbor skilled and unskilled workers-which would imply that a substantial division is lacking. In hypothesizing about the plausible paths of restructuration between 1988 and 1993 , we pointed out some of their potentia! sources in deep changes in the distributive system. Could it be that the hlurring of mo­ bility barriers in Slovakia and Russia reflected a redistribution of in­ comes? That would be so if-for example-members of the working class found professional occupations more attractive because of rising financial incentives in the nonmanual sector-as already mentioned, in some other East Central European societies the earnings of the occupa­ tional elite increased in relative terms. This may have intensified inter­ generational flows from manuał to nonmanual categories at the begin­ ning of the 1990s. Since such direct linkages between the dynamics of distributive rules and social structuration occur rarely, we should perhaps not take this interpretation at face value. As in Slovakia and Russia, the 'manual/non­ manual' dichotomy dissipated, which may only reflect a 'trendless fluctuation' rather than a long-term trend with strong macrostructural underpinnings. The opposite was true of Bułgaria, where, in 1988-93 , the exchange between blue- and white-collar workers decreased. In this case, one may cite the effects of the purported professionalization of the intelligentsia and cadres, on the assumption that the strategies applied by Westem professionals are already present in East Central Europe. The claim could be made that the Bulgarian intelligentsia is seeking to se­ cure its prosperous position in the labor market by imposing barriers of entry to candidates from the lower strata. Notwithstanding variations in a cross-country perspective, the domi­ nant impression given by the data is one of stability in respect of the position of nonmanual categories relative to the working class. Our maps of intergenerational barriers to mobility also reveal other forms of reshaping, but all of them seem negligible. We are unable to interpret

65

them in any meaningful way, or to attribute them to transformations of the economic and political system. T H E O P E N O C C UPAT I O NAL E L I T E The same conclusion may be drawn with respect to the social location of higher professionals, managers, and top administrative officials, which we refer to as the occupational elite. As far as the main axis of differen­ tiation between socioeconomic categories is concemed, the salient posi­ tion of the occupational elite in capitalist societies seems sufficiently well documented. In East Central Europe, as our data show, this distinc­ tion is scarcely relevant to 1 988-94. Comparison of discriminant scores indicates that it is only in Hungary and-to a lesser extent-Bulgaria, that higher professionals stand at a distance from !ower white-collar categories. In the other four countries, the occupational elite tends to be close to the lower professionals, routine nonmanuals, and proprietors in the structure of intergenerational flows. Access to the occupational elite in the Czech Republic, Poland, Russia, and Slovakia seems relatively open. On this basis, we believe that the thesis concerning the invariably sa­ lient position of the occupational elite is in need of some qualification. Originally, it was based on findings from capitalist countries. What in fact occurs is that the position of the occupational elite is affected by a multiplicity of factors which do not operate with the same magnitude across countries. For example, in postcommunist societies a strictly formalized path of recruitment to the civil service has not yet been suc­ cessfully implemented. By the early 1990s, slight changes in the place­ ment of higher professionals, managers, and administrative staff in some East Central European countries were detectable, but these did not rear­ range the configuration of basic social distances in a meaningful way. OWN E R S In the two-dimensional space o f men' s intergenerational movements, it is not possible to identify a separate location for nonagricultural entre­ preneurs as against other class categories in recruitment patterns. Only in Russia-for both 1 988 and 1 993--did proprietors occupy a discerni­ ble position in the second dimension of mobility space.

66 In the mobility pattems of a number of capitalist countries-by now well studied-the specific character of recruitment to the entrepreneur category has emerged as a distinct dimension in the multidimensional space of intergenerational movements. The special position of proprie­ tors usually emerges on the third consecutive dimension-that is, it stands third in importance after socioeconomic status and the farm­ ing/nonfarming division. This was apparent in respect of capitalist so­ cieties and the same pattems appeared in our data, although-due to limitations of space-I shall not present these results here. Our findings do not show that pattems of entry to the proprietors category are less salient in East Central Europe than in Westem societies. This distinction exists also in postcommunist countries, reflecting the high degree of self-recruitment in the intergenerational transfer of property. Interest­ ingly, inflow frequencies for Russia-not presented here-show that not self-recruitment but a substantial flow from the occupational elite-that is, higher professionals and top administrative officials-appears to ac­ count for the unique position of owners in Russia. It bears repeating once again: the most important result of our find­ ings is that no substantial changes are detectable. The numerical expan­ sion of the category of small and middle-sized owners, pervasive throughout East Central Europe, did not bring any substantial restructu­ ration in recruitment pattems. In particular, the rules of entry to the group of owners did not change to such an extent that their position be­ carne decisively marked in mobility space. Looking at the configuration of discriminant weights, one can trace only a certain 'upgrading' of owners between 1 988 and 1 993 in Bułgaria and Hungary: in the 1 990s they tend to accompany nonmanual categories, while at the end of the preceding decade they were neighboring categories of the working class. What these tendencies reflect is, in fact, the changing composition of the social origin of proprietors-while in 1 988 they originated mostly from the lower classes, the early 1990s witnessed an increasing inflow of the sons of nonmanual strata.

C ON CLUSION : POLITICAL BREAKTH ROUGH AND MOBILITY The aim of our comparison was to establish mobility rates in a critical period of the 1 990s and to discover pattems of social structuration as exhibited in mobility space. We have noted where our findings appear to

67

support, qualify, or undermine the evidence assembled in previous analyses. As regards East Central European countries, knowledge of the generał contours of distances between basie class segments was, at that time, fragmentary, limited to individual societies, and weakly integrated with regularities established for capitalist countries. It represented a missing theoretical link. Of the two theoretical positions-the one claiming a basie similarity in pattems of structuration across contemporary industrial societies and the other pointing out differentials related to specific features of politi­ cal--economic systems-the results reported in the present study seem to favor the former. We found no evidence that our six East Central Euro­ pean nations were markedly different from capitalist countries. As far as intergenerational flows are concemed, it appears that in Bułgaria, the Czech Republic, and so on, they have come to be struc­ tured along the same core axes and divisions as in the UK, Germany, or the USA. Here we refer particularly to: position in the hierarchy of so­ cioeconomic status, division into agricultural and nonagricultural occu­ pations, the dichotomy 'manual/nonmanual' workers, the outstanding position of higher managerial and professional categories, and the sepa­ ration of nonagricultural proprietors. Our data indicate that these simi­ larities predominate over cross-national variations in respect of the rela­ tive magnitude of class distances and crystallization of the hierarchical divisions intimately linked to social mobility. These are the main implications of our findings in terms of compara­ tive macrosociology. However, the issue which directly informs this study is the effect of systemie transformations in East Central Europe on the dynamics of basie social cleavages. Overall, our comparison of mobility pattems between 1988 and the early 1990s did not reveal any consistent change: the temporal constancy in basie mobility barriers continues. The political break and upheavals in the economic system did not breach the broad similarity in the generał shape of social stratifica­ tion and in its underpinnings.

C H A PTE R 4

E C O N OMIC STRATIFICATIO N : SIMILARITIE S , DIFFERENCE S , AND EMERGING CHANGE

WHAT w e presented i n the preceding chapter may suggest far-reaching uniformity and continuity, but this would be an oversimplification. Let us consider instead an area of stratification which seems to bring out the more singular features of our six East Central European countries. Economic differentiation is less vulnerable to the homogenizing ef­ fects of what we have called the 'inherent logic of postindustrialism'. Indeed, both levels of incomes and the rules of their distribution reflect economic crises, the destructive impact of wars, long-standing welfare systems, state interventionism, wage bargaining between unions and employers, and so on: all in all, the historical, political, and economic context which is unique to each country (Esping-Andersen, 1990, 1993). The impact of politics on social stratification was overwhelming in communist societies, since they were subjected to an encompassing system of direct planning. Nevertheless, centralization of the rules of distribution in the hands of government was weaker in Hungary and Poland than in Russia. In the present chapter, while seeking cross­ national differences in economic stratification in East Central Europe, we will not lose sight of the traces of the different institutional heritages. Obviously, our six nations also differed in terms of level of economic development and the modemization of social structures. In fact, as we shall see, they differed in economic stratification to a slightly greater degree than in respect of their pattems of mobility-although the simi­ larities prevailed over the differences in this respect too.

70

FAMILY IN COMES Beginning one's account with the hierarchy of incomes gives one the advantage of precision. As compared to inequalities of power, social prestige, and other important dimensions of social stratification, in­ comes precisely locate socio-occupational strata on consecutive levels of social stratification, from the top to the bottom. Table 1 4 . Mean monthly fami/y incomes per capita (in US dol/ars) Socio-occupational categories

Higher professionals and managers Semi-professionals Routine nonmanual, sales and service workers Owners with employees Owners without employees Lower-grade technicians Skilled workers Unskilled workers Agńcultural laborers Farmers Mean Coefficient of variation (V) Correlation ratio (E)

Bułgaria

Czech Hungary Poland Republic

Russia

Slovakia

86

1 37

216

1 68

26

1 06

71 67

1 14 1 00

1 76 1 47

123 84

23 19

92 82

1 04 72 79 65 56 53 67 65 0.76 0.22

1 80 135 1 07 1 02 92 84 139 1 07 0.68 0.20

1 75 1 55 1 80 1 39 127 1 15 141 1 47 0.67 0.27

1 44 1 02 1 00 76 77 65 76 93 0.90 0.33

47 66 27 21 19 17 22 23 2.00 0.22

1 29 1 09 85 78 78 77 1 29 85 0.66 0.2 1

Table 14 gives the mean incomes for the basie socio-occupational categories. This ordering does not depart substantially from the hierar­ chy of socioeconomic status which we found in the configuration of mobility barriers. Mean incomes are shaped according to a classical stratification ladder which constitutes the core of contemporary market­ oriented systems. It is appropriate to call it 'classical' because status differentiation lies at the origin of the development of macrostructures typical of postindustrial societies; Turner (1988) gives a concise over­ view of the status-based roots of differentiation. Scanning the hierar­ chies of incomes we must take into account the fact that, although our samples were representative for adult populations, they covered no more than four or five thousand cases. As usual, representatives of the top level-the old aristocracy or the financial elite-were excluded; those at

71 the bottom o f the ladder in terms of poverty (a prominent part of the economic mosaic everywhere) were treated in the same way. The respondents were asked to report on their total household in­ comes-coming from all persons and from all sources-over the past three months. The incomes declared in this way were divided by the number of family members and-to make them comparable across countries-they were converted into US dollars. This allows us to underline the monetary divide between East and West. There was an evident gap between the incomes of the inhabitants of East Central Europe and of those in the advanced industrialized countries. Average incomes in Hungary were relatively the highest, yet they were still 17 times lower than that of the average American family. Mean annual family income per capita in Hungarian society stood at USD 1,764 in 1993, while in the USA it was USD 30,786 (according to the Statistical Abstract 1994 for 1992). If we consider the total of USD 276 which-if we can rely on the survey data-an average Russian family received, the East-West gap expands from 1: 17 to 1: 112. The contrast is comparable to that between the extremes of wealth and pov­ erty. The ratio of 1: 112 seems so great that it obliterates disparities within the postcommunist camp which until recently was unified by a common political and economic system and officia! ideology. However, it would be wildly inaccurate to regard all six countries as a single, undifferenti­ ated entity and to ignore the differences between them. Hungary and Russia are located at the two poles of the axis of mean incomes. Hungar­ ian family incomes were on average 6.3 times greater than in Russia, 2.3 times higher than in Bułgaria, and 1.7, 1.6, and 1.4 times higher than in Slovakia, Poland, and the Czech Republic, respectively. Average incomes, which approximate the standard of living in par­ ticular societies, are only one side of the coin. Also important is the dis­ tribution of incomes across individuals and socio-occupational strata. We shall now enquire about both the degree and the shape of economic inequalities. Let us start with the degree of inequalities, which can be expressed in terms of income dispersion. Strong concentration around the mean im­ plies low inequality and differentiation. Measures of 'dispersion' are given at the bottom of Table 14. We took the coefficient of variations (V) which divided standard deviations of incomes by means. It is worth comparing Vs with average incomes for specific countries. In doing so one can see that average incomes and

72

their differentiation by no means reflected each other. Paradoxically, income differentiation is, on the whole, strongest in Russia, despite low income levels: income differentiation in that country is very high, standing at 2.0. This translates to the standard deviation being twice as much as the mean. Evidently, distribution of incomes in Russia was highly skewed to the right. Everywhere, incomes are strongly concen­ trated at the lower end-below, rather than above the mean-and the Russian case exemplifies this regularity perfectly. Values of coefficient of skewness are suggestive. In Russia, this reached 20.1, 2.5 times as much as in the Czech Republic (skewness of 8.7), and 5 times more than in Hungary, where dispersion of incomes around the mean was rela­ tively the most equal (4.4). It is worth noting that in the society with, on average, the lowest in­ comes, the inequalities were the most pronounced. Note, also, that­ apart from the Czech Republic and Slovakia-the narrowest dispersion of incomes was displayed in Hungary. Thus, economic differentiation decreased across our six nations with an increase in the average level of incomes, and not the other way around. The striking Russian case strongly imprints on the specificity of the whole region. The exceptionally low average incomes were accompa­ nied by the highest differentiation. We have, therefore, good reason to think that-quite apart from the poor economic condition of the society as a whole-there still existed a vast area of poverty in contrast to the condition of the financial elite. The spectrum of income differentiation was extremely 'stretched', which diverged appreciably from what we found in Bułgaria, Poland, or Hungary. lt will suffice if we compare the mean incomes in the highest and lowest deciles in the six societies. This ratio stood at 171.9 in Russia, followed by Poland with only 65.1, and, successively, 53.1 in Bułgaria, 41.8 in Czech Republic, 38.1 in Slova­ kia, and 31.9 in Hungary. Now we can return to the shape of these inequalities. It appears to be similar across the six countries, despite unequal income differentiation. In the first half of the 1990s, agricultural workers received the lowest in­ comes and this is replicated, without exception, by all six societies. The economic situation of unskilled manuał workers was not much better. They occupied the penultimate position in the hierarchy, except in Poland and Russia. Peasants and skilled workers in Poland fared no better than unskilled workers and, in Russia, clerical staff did poorly as well. As regards nonmanual workers, they are differentiated into three strata. Everywhere, the intelligentsia and managers enjoyed the highest

73

incomes, exceeding categories of semi-professionals, middle-ranking administrative staff, and so on. Below them, and separated from the two higher categories, fall routine clerical workers (such as secretaries and filing clerks), and shop assistants. Tuming to the broad categorization into nonmanuals, manuals, and farmers (smallholders), we also find basie intercountry similarities. The families of the intelligentsia, semi-professionals, and clerical staff re­ ceived, on the whole, higher incomes than manuał workers. This applies also to the lowest-ranking clerical staff, although in many societies they happen to fare worse than the higher strata of the working class-that is, skilled workers-in other dimensions of economic stratification. Only in the Czech Republic and Russia did the incomes of clerical workers fall below those of skilled workers. Also, technicians and foremen did better than skilled workers. The strict location of the former within the 'manual-nonmanual' division cannot be specified in an unambiguous way since they include-apart from foremen-lower technicians, who, within the framework of a number of classifications, used to be counted as nonmanuals. As regards foremen, they perform supervisory tasks but conjoined with manuał work; usually, they do the same work as the workers whom they supervise-foremen in construction serve as a good example. lncomes of technicians and foremen were on a par with those obtained by the families of semi-professionals. In Bułgaria, Russia, and Hungary they were even slightly higher. Universal pattems prevail but economic stratification in particular countries is also marked by regional specificity. This is most visibly the case in respect of the position of private entrepreneurs-both owners of larger firms who employ workers and the self-employed who do not hire labor. Insofar as the class position of owners is disputed in sociological theories, it would be difficult to come up with an intelligent interpreta­ tion of their placement in the hierarchy of incomes in postcommunist societies. Analogies with the advanced industrialized countries might have been especially helpful. It may, for example, have been attractive to regard larger owners-those with employees-as a potentia! under­ pinning of the newly developing East Central European economic elite, by virtue (say) of their high incomes. Self-employed owners can be re­ ferred, in tum, to the traditional 'old middle classes'. The 'old middle classes' in modern capitalism are de facto in the 'middle', but in many cases the small owners of repair workshops, craftsmen, and grocers lose out to the working class in the competition for a share of the income pie (see Curran and Blackburn, 1991; Savage et al., 1992). What do the in-

74 come data imply as regards the class position of private entrepreneurs in postcommunist Europe? Cross-country similarities emerge, but only in a few cases. Incomes of larger businessmen and small owners are similarly contrasted to other so­ cio-occupational strata in Poland and Hungary. The mean incomes of pri­ vate entrepreneurs are relatively low. In Hungary and Poland, larger own­ ers received lower incomes than managerial cadres and the intelligentsia. And in Hungary they did almost the same as semi-professionals and me­ dium-level administrative staff. Small businesses proved less profitable as well: in Hungary, the self-employed without employees received signifi­ cantly lower incomes than manuał supervisors (technicians and foremen) and nonmanual workers (but only clerical staff). In Poland, the monetary distance between families in the small business and 'administration, and semi-professionals' stood at USD 21 monthly, which, as far as 1994 is concemed, was not bad. There is some analogy, in fact, with. the economic position of small and medium-sized owners in the UK, the USA, Ger­ many, and France-the majority of studies on the 'old middle classes' were carried out in these countries (see Scase and Goffe, 1982; Bechhofer and Elliot, 1978). Both in Polish and Hungarian society, private entrepre­ neurs entered the 1990s with quite decent monetary gains-which put larger owners among the higher reaches of income distribution-but as a whole they were no longer a financial elite. They had ceased to be the privileged category they had been in the 1980s. Over severa! decades of communist rule, they took a prominent economic position-within the possibilities provided by the system, of course-in the spectrum of basie socio-occupational strata. In the case of Poland, at least, these statements are empirically valid. Data based on results coming from nation-based representative samples has been available in Poland since the 1970s and 1980s. They document that owners really did form an economically privileged enclave (Domański and Sawiński, 1991; Domański, 1988 and 1990). Owners were still on top as communism began to decline. But at the beginning of the 1990s this pattem was modified. We can see that at that time the incomes of Polish and Hungarian owners stood below the mean incomes of managers and professionals. It was otherwise in Bułgaria, the Czech Republic, Russia, and Slovakia. This could be said to herald a slowly emerging decline from the favorable position of the past. Some modification might have taken place and the mechanisms of distribution typical of contemporary market systems may have begun to operate. Under the communist system, private entrepreneurs were privileged by

75 virtue of a well-established monopoly of selling scarce goods and serv­ ices to meet consumer demand simply because the bureaucratized and inefficient state sector could not satisfy mass consumption needs. This monopoly provided owners with unparalleled profits. Wages in the state sector were established on the basis of rigid rules and could not compete with incomes drawn from business. This monopoly underwent a natural erosion after the collapse of the socialist economy. A paradox emerges. First, history created a social system which, with a glut of contradictory rules, fostered a social enclave which, although operating in a politically hostile environment, eamed substantial profits. Then, after the fall of the system, fate decreed a new and contradictory scenario, which took those most interested in retaining their privileged conditions by surprise. Under the new circumstances, both the legal system and political infrastructures gave support to the development of private business. However, the rules of the capitalist market have estab­ lished a new barrier which is probably no less effective in limiting the economic role of the 'old middle classes' as the arrangements in the mature capitalism of the West. This is no longer an asset which tends to promote itself. Instead, it involves a multiplicity of threats and enforces a continuous struggle to keep one's privileged economic position with respect to other strata. This is what Polish and Hungarian owners have witnessed in the 1990s. Their incomes remained moderately high, but not excessive. It looks as if, in both countries, the dynamics of the income hierarchy fol­ lowed a pattern govemed by the logic of market systems and it appears that this pattern emerged earliest in Poland and Hungary. As a result, we must examine the opposite pole. In the former Czechoslovakia and in Russia, mean incomes of private entrepreneurs were far ahead of the other strata. In the Czech Republic and Slovakia, owners without employees did no worse than managerial cadres and professionals which, at the beginning of the 1990s, was astounding, even against the background of the region's current circumstances. Also, in the Czech Republic, incomes of farmers-small agricultural busi­ nesses-were unexpectedly high: they were on a par with the highest nonmanual strata. Obviously, owners of larger firms obtained the high­ est incomes-in the Czech Republic they outstripped the incomes of managers and the intelligentsia by 31 per cent, in Slovakia by 22 per cent, and in Bułgaria by 21 per cent. Nevertheless, these advantages pale when set against the incomes of small owners in Russia, who enjoyed the best and most privileged eco-

76

nomie position as compared to other strata. Their incomes exceeded the mean for the second category, larger entrepreneurs, by 40 per cent. Then carne technicians, foremen, managers, and the intelligentsia, whose in­ comes, overall, were much below the mean for larger owners. Let us note that the Russian intelligentsia neighbored categories traditionally placed in the lower half of the income hierarchy. As compared with other countries, its economic situation was exceptionally bad in Russia. It looks as if the pace of systemie transformation was too fast for engi­ neers, lawyers, academicians, and medical doctors, and that they failed in competition with the other strata. But why did smaller businesses profit more than larger owners? In pursuit of this question, we would have to know more about the broad economic and structural underpinnings of private-sector activity in Rus­ sia, much more than is contained in standardized information reported by respondents in surveys. The disturbing case of Russia may reside, to some degree, in errors cornmitted in the fieldwork. It might well be that owners with employees underreported their incomes, due-for exam­ ple-to the systematic omission of important items which they regarded as being irrelevant. Usually, respondents do not realize precisely how large their incomes are and give only estimates based on salaries, wages, and selected sources. Other errors may also be involved. Nonetheless, the threat of systematic error seems to be reduced in view of the fact that family income distribution in Russia is consistent with earned income distribution. The high family incomes of small owners conform to their high position in the hierarchy of individual incomes earned from work (we will deal with this topie shortly). So, although the data are suspect, the hierarchy of incomes in all likelihood does reflect existing pattems rather than artificial reporting. What remains to be explained is the relatively high position of own­ ers in Russia, Bułgaria, Slovakia, and the Czech Republic as compared to Hungary and Poland. Above all, different scenarios of economic de­ velopment seem to lie behind these two distinctive pattems, with une­ qual levels of economic prosperity in the foreground. Although the countries we are considering are all industrial and display what may be thought of as generic developmental characteristics, the structural con­ texts for economic disparities which they provide are quite strongly dif­ ferentiated on a case-to-case basis. In Russia, the gap in mass consurnption was greater and could not be filled so quickly. Private firms were more efficient than state ones, but privatization in Russia slowed down. Owners have not yet faced the ob-

77

stacle of competition, nor was the demand for consumption goods satis­ fied to the same extent as in Poland and Hungary. Looked at in this per­ spective, one may conclude that the Russian society of the early 1990s remained at an earlier stage of transformation to the market system. The marked prosperity of businesses also persisted in Bułgaria, Slovakia, and the Czech Republic. These favorable conditions faded away in Hungary and Poland, where small and medium-sized businesses main­ tained their relatively good economic situation only by virtue of inertia. The economic boom became a slump.

EARNED INCOMES In order to obtain a better understanding of economic stratification, we shall move on from total family incomes to eamings. These are the basie constituent of total incomes and their distribution underlies economic stratification-a fact well established by analyses of empirical data over many years in various countries (see Psacharopoulos, 1994; Featherman and Hauser, 1978; Kuznets, 1989; Atkinson and Micklewright, 1992). Postcommunist societies are no exception to the generał rule. In Ta­ ble 15 mean individual incomes from work are included for basie socio­ occupational strata. Before we scrutinize monetary distances between the categories, let us remark on the values of correlation ratios given in the last row. They indicate how intemally coherent are our EGP class categories with respect to income differentiation in each of the six countries. The higher correlations indicate that these categories are more distant from each other and more intemally homogeneous. The magni­ tude of class structuration in this sense is not low, taking into account that the correlation ratio reaches 1 .00 at its maximum value. What de­ serves more attention is that earnings differentiate the means for catego­ ries more strongly than family incomes. The correlations appear sys­ tematically higher than the respective values in Table 14 (except for Poland). It suggests what one might have expected-that differentiation in eamings tends to discriminate class segments more strongly (in a sta­ tistical sense) than the financial situation of families. This has important implications since earnings and family incomes are distributed according to different rules. Earnings depend more on individ­ ual pursuits and are embedded in the labor market, whereas family in­ comes come from all sources, not only from work. How much one earns results directly from one's occupational and supervisory position, skills,

78

firm, and sector of the economy. Other determinants include age, sex, number of years in the labor force, and level and kind of education, which affect incomes indirectly through occupational status. Then comes the oc­ cupational path that one follows after leaving school, which may either doom one to failure or facilitate success. Since these determinants are built into the labor market, this is a critical area, which contributes to the crys­ tallization of incomes within the framework of socio-occupational status. This is most evident in Hungary where, by the way, the grand mean of eamings is also highest (likewise with family incomes). Table 1 5 . Mean monthly individual incomes (in US dol/ars) Czech Hungary Poland Repu bi"1c

Russia

Slovakia

300

39

181

219 1 78

1 99 1 24

32 25

1 45 118

298 227 252 1 80 1 65 1 43 216 1 94 0.64 0.34

323 214 209 1 40 1 42 1 15 145 1 69 1.13 0.25

99 1 40 41 32 27 25 35 34 1 .40 0.25

25 1 214 1 60 137 1 25 114 1 96 141 0.65 0.3 1

Socio-occupational categories

Bułgaria

Higher professionals and managers Semi-professionals Routine nonmanual, sales and service workers Owners with employees Owners without employees Lower-grade technicians Skilled workers Unskilled workers Agricultural laborers Farmers Mean Coefficient of variation (V) Correlation ratio (E)

1 32

207

303

1 07 98

148 121

20 1 1 32 139 1 09 90 82 111 1 04 0.7 1 0.29

299 223 1 65 151 1 22 111 22 1 1 49 1 .00 0.27

We are still seeking what is universal against a background of the unique features of our six countries. The basie difference between hier­ archies of family incomes and earnings is that the rules for rewarding work tend to elevate owners to the detriment of the intelligentsia and managers. In Hungary, larger owners (with employees) gave way, only slightly, to managerial cadres and the intelligentsia, and in Poland they moved to the top. Apart from in Hungary and Poland, larger businesses stood far above the East Central European service class on the scale of earnings, but not in respect of family incomes. As regards the self-employed (without employees), they stood sec­ ond after owners with employees in the Czech Republic and Slovakia.

79

Future generations will see how prosperous it was to own even a small firm in East Central Europe at the beginning of the systemie transfor­ mations in the 1990s. Small owners did worse than the intelligentsia only in Poland and Hungary, but better-on average-than the represen­ tatives of all other strata. Only in Hungary were individual incomes of technicians and foremen still higher. Summing up, the intelligentsia and managerial cadres in Poland earned less than owners of large firms, in contrast to the situation in re­ spect of family incomes. The Hungarian service class did better than owners in terms of earnings, but not so well with respect to family in­ comes. In the six countries overall business paid more in terms of earn­ ings than in terms ofhousehold income per capita. We have suggested that income hierarchies in Hungary and Poland reflect generic characteristics of the 'old middle classes' in Westem so-:­ cieties. I would risk making one more generalization of this kind, in full awareness of the fact that I am plunging into the swampy ground of speculative predictions and hypotheses. lt appears that, in comparison with incomes, the distribution of earnings in East Central European countries diverges from the pattems typical of market-oriented societies. The pattems of differentiation of total (family) incomes across catego­ ries are more similar between countries than those of individual in­ comes. Obviously, these analogies are limited to relationships between in­ comes of representatives of the service class-which are moving up­ and private entrepreneurs. The economic position of the latter is deterio­ rating-at least in comparison to the intelligentsia, who did worse under the communist regime. We shall focus on one aspect of income redistri­ bution, but a pivotal one. If it reflects long-wave tendencies-and this is my assumption--one may expect more similarities between the situa­ tions of East Central European business and the Westem 'old middle classes' in coming years. An obvious conclusion follows: that the rules for distribution of earnings in postcommunist societies remain specific to this region be­ cause they--certainly more than family incomes-were shaped by the policy of the socialist state. Nowadays, we still encounter remnants of this heritage. Eamings, which used to be shaped by the rules ofthe labor market, are vulnerable to interference 'from outside'-for example, by govemment legal regulations or bargaining by trade unions. Family in­ comes, by contrast, are received from various sources by means of the collective efforts of all family members, which keeps them in relatively

80

persistent equilibrium. After all, it is easier for the govemment to raise the wages of miners, or, on the other hand, to put pressure on large firms because they provide them with substantial inputs from the state budget, than to change the predilections of married couples of the intelligentsia class for saving money rather than for engaging in lavish consumption like working-class couples. The intelligentsia and professionals find it easier to mobilize cash-this is an empirical fact, confirmed regularly by many studies. It is something that we would envisage as an insepa­ rable link of social stratification, resistant to interventions from 'outside', no matter if such attempts are made by the administration of a communist state or by a social-democratic cabinet proceeding according to the tenets of the welfare state. The Russian case clearly departs from this generał rule. In that coun­ try the intelligentsia and managerial cadres have an extremely low eco­ nomic status. Representatives of the latter may regard themselves as fi­ nancially deprived, but not only in comparison to owners. The incomes of the intelligentsia also fell-on average-below the mean incomes obtained by technicians and foremen. Moreover, they scarcely did better than other occupational strata. Can this be attributed to a complete demolition of the mechanisms goveming income distribution in post­ industrial societies? This is a strong claim, but possibly a true one. Long-standing political intervention may well have had a profound in­ fluence on this. The centralized policy of income distribution in the So­ viet Union lasted, after all, considerably longer than in its satellite countries and should be expected to be more persistent. Note that the relative position of managerial wages declined dramatically over the last fifty years: the difference between them and manuał workers fell from 1:2.6 in 1932 to 1: 1. 1 in 1986 (Lane, 1992, 168). There is one further vestige of the past in the hierarchy of earnings which lends support to the hypothesis concerning the significant impact of communist policy, namely, the monetary distance between the work­ ing class and nonmanual workers such as semi-professionals, adminis­ trative staff, sales clerks, and other white-collar workers. In the 1 990s, as in preceding decades, working-class categories did relatively better in terms of earnings than of family incomes. Conversely, with respect to mean household incomes skilled workers' families fared worse than­ for example-the families of office workers or sales clerks. At best-in the Czech Republic and Russia-the family incomes of skilled workers were on a par with those of clerical staff. As regards earnings, skilled workers did better than clerical staff in all six countries.

81

Notably, in Bułgaria, the Czech Republic, and Russia skilled workers were rewarded no less than semi-professionals, which in East Central Europe means middle-ranking administrative staff, nurses, and so on. More specifically, the main losers proved to be routine white-collar workers: in the Czech Republic, Poland, Slovakia, and Russia they earned less than unskilled workers, while in Bułgaria they made only slightly more than the lower manuał categories. Only agricultural work­ ers fell below routine white-collar workers-indeed, they were at the bortom of the pile in all countries. Let us also note that, in the past, skilled workers earned more than lower clerical and sales staff in East Central Europe. This situation was typical of the region over recent dec­ ades. However, the Hungarian case must also be borne in mind. lt was the only example of lower white-collar workers faring reasonably well in the earnings hierarchy. We have remarked on the economic stratification of Hungary severa! times because it frequently differs from the domi­ nant postcommunist partem. Thus, one more divergence has been added to the list, but a stipulation is also needed: the relatively berter position of routine nonmanuals does not push Hungary closer to the economic stratification generic to market-oriented societies. Even in the USA or the United Kingdom, clerical workers do not eam almost as much as skilled workers, as was the case in Hungary in the 1990s (USD 178 and USD 180 respectively). Routine clerical work pays less-this tends to be the norm. As regards earnings differentiation, the case of Hungary is somewhat unique.

MATERIAŁ RESOURCES Not only the possession of cash, but how we spend it determines our socioeconomic status. Differentiation according to materiał resources constitutes the third axis of economic stratification in East Central European countries. The typical list of materiał resources which deter­ mine standard of living consists of owning a home or apartment, along with household equipment of various kinds, including electrical appli­ ances and vehicles. Instead of analyzing each of these items separately, I synthesized them into one summary index composed of six items which were taken as valid measures of materiał standard of living (I de­ scribe the construction of this index in the Appendix).

82 Table 1 6. Ownership of motor vehicle, separate deep-freeze, microwave oven, personal computer, satellite receiver, telephone. Synthetic index ofmateria/ position (1 00 - owning all 6 items, O - none) Socio-occupational categońes

Higher professional and managers Semi-professionals Routine nonmanual, sales and service workers Owners with employees Owners without employees Lower-grade technicians Skilled workers Unskilled workers Agricultural laborers Fanners Mean Correlation ratio (E)

Bułgaria

Czech Hungary Republic

Poland

Russia

Slovakia

14.0

35.0

40.3

3 3 .8

1 7.5

37.3

1 2.3 1 2 .7

29.0 27.0

34.9 27.9

28.2 23 . 1

1 3 .6 1 1 .4

3 1 .0 27.4

2 1 .3 1 3 .2 14.0 1 2.6 1 1 .9 1 0.2 1 0.5 1 2.3 0. 1 5

43 . 3 3 5 .4 28.3 23 .0 20 .5 1 8.4 32.3 26. 1 0.34

45 .5 34.9 28.8 22. 1 1 6.6 1 2.4 23 .2 24.8 0.37

42.7 3 1 .9 24.7 18.1 1 6.4 12.0 1 7.7 22.4 0.37

23.6 1 5 .6 1 2.7 9.8 9.0 5 .4 14.3 12.2 0.30

43 .5 3 3 .2 30.8 22.9 1 9.9 1 6. 8 29.7 26.0 0.35

The data in Table 16 outline the contours of materiał differentiation. For each socio-occupational stratum I took the mean values of a six-item index, which included: car, microwave oven, separate deep-freezer, per­ sonal computer, telephone, and satellite TV receiver. I did not take into account ownership of home or apartment, since, at the beginning of the 1990s, these classical indices of welfare in the West did not reflect one's socioeconomic status in East Central Europe. The housing market did not exist under the communist regime. From the beginning ofthe 1990s, such a market began to develop, but was still far from being an area of safe investment. According to Saunders (1990), division into home owners and non-owners represents a core dichotomy in British society. Many other students of social stratification see home ownership as a key determinant of socioeconomic status, although they did not share Saun­ ders' extreme position (for example, Savage et al., 1992). As is evident from Table 16, the ordering of categories according to materiał possessions appears almost identical throughout the six East Central European countries. Running from top to bottom, we have: owners with employees, the intelligentsia and managers, self-employed owners, middle-ranking nonmanuals, technicians and foremen, routine clerks and sales staff, farmers, skilled workers, unskilled workers, and invariably at the bottom, agricultural workers.

83

These results, when set alongside those of Tables 14 and 15, strongly suggest that the hierarchy of materiał resources adds to the overwhelm­ ing gradation of class segments according to socioeconomic status. Owners with employees are consequently at the top. Even in Hungary they found themselves above the intelligentsia. This is not unusual. Af­ ter several decades of financial domination, business circles could afford to discount accumulated incomes to a larger extent than other strata. One can hardly add anything more to what we already knew about the shape of differentiation according to social status from our own findings and previous studies. However, particular intercountry differences in the ordering of cate­ gories with respect to materiał status are also apparent. Thus, we find a elear line of division between the poor southeast, which contains Buł­ garia and Russia, and the more wealthy northwest, comprising the four remaining countries. With regard to possession of materiał resources, the average citizen of Bułgaria and Russia had half as much as his counterpart in the Czech Republic, Poland, Slovakia, and Hungary-al­ though, to be sure, our index does not encompass all aspects of materiał standard ofliving. Prom Table 16 it can also be observed that Bulgarians had a low level of materiał status which corresponded with the low differentiation between upper and lower strata. The overall compression of materiał inequalities was strong. At the same time, in Russia the symbiotic rela­ tionship between low level of economic conditions and high variation is repeated once again. The index of materiał possessions which was on average low was paralleled by a bigger disparity between top and bot­ tom than in the other countries. Owners with employees were more than four times better off than agricultural workers in Russia, while in Buł­ garia they were only half as well off. The rates for the other countries were as follows: 3.7 in Hungary, 3.6 in Poland, 2.6 in Slovakia, and 2.4 in the Czech Republic. We argued earlier that the striking symbiosis of sharp inequalities of incomes with a low average level resulted in Russia from extreme con­ trasts between wealth and poverty. The mass pauperization in Russia heavily skewed the distribution of incomes to the bortom. This tendency is also discemible in the distribution of materiał resources. The peculi­ arity of the Russian case continues to be confirmed. In Table A l in the Appendix, I present percentages of car ownership across our class categories. This makes elear how the summary index of materiał standards of living translates into concrete items. The propor-

84

tion of car owners ranged from 40 to 50 per cent in most countries, but not in Russia, where in 1993 only one in every four families had a car. Apparently, East Central Europe still deviates significantly from West­ em standards of materiał possession. For example, in the USA, in the same year over 90 per cent of families owned at least one car. In the postcommunist world only owners approached the level of the USA, with 73-83 per cent of such families owning a car across the six nations. The intelligentsia and managers carne next with 61-68 per cent. Need­ less to say, the ordering of all strata was consonant with the hierarchy of socioeconomic status. The proportion of car owners was generally similar in all countries, perhaps with the exception of farmers, who had more cars in the Czech Republic (73.9 per cent) and in Slovakia (60 per cent). Incidentally, in both societies the percentage of car ownership was the highest in our sample. The Czech Republic was the only country in which more than half of all families had their own car. SAVI N G S A N D S HARE S Differentiation of savings and share ownership (including stocks, bonds, and investment trust certificates) completes our picture of economic stratification. Pattems of distribution of savings and valuable assets do not reflect only differentiation of materiał wealth. Incomes may be dif­ ferently consumed depending on one's willingness to secure and/or ac­ cumulate financial resources. Here, we move into the area of calculation of gains and losses: those who prosper need not invest on the stock mar­ ket-such decisions depend on life strategies, orientations, consumption habits, and personal predilections. Our task is to establish the extent to which differentiation in the ownership of savings and stocks across basie occupational strata will reveal what we expect to find-a social hierarchy in which owners and the intelligentsia remain at the top and unskilled manuał workers are at the bottom. Financial markets in East Central European countries offer a vast range of opportunities. Bold investments in valuable assets may be considered as indicative of adaptation to changes brought about by capitalist structures. Conversely, persons who restrict their financial strategies to keeping money in the bank may be condemned for tradi­ tionalism and backwardness. This is what economists refer to as the basie opposition between modem and traditional attitudes. According to this argument, investment

85 in shares tends to indicate individual initiative, an open attitude to the civilization of the service economy, and a readiness for risk taking. We have to follow the economists, yet without drawing finał conclusions about the nature of the motives behind investing or saving. The purchase of shares does not mark out only market-oriented attitudes, whether in­ novation or individualism. True, the challenge inherent in the Warsaw Stock Exchange must stimulate such attitudes, but it does not automati­ cally make them an unequivocal indicator of modernity. For example, in the first stage of capitalist revival, shares were distributed centrally, by virtue of the institutional rules prescri�ed by law and not on the initia­ tive of stockholders. In Poland, employees in firms undergoing privati­ zation automatically received 15 per cent of the shares by law: no inter­ est was required on their part and their share ownership tells us nothing about their individual initiative. It would be no less of an oversimplification to put the blame for backwardness on the owners of large bank accounts. Savings do not prevent depositors from investing their money in stocks, trusts, or bonds. We may usefully pause at this stage to say that, in light of our data, there was, in fact, almost no correlation between share ownership and bank savings, with the slight exception of Poland and Hungary. In Poland, the percentage of share owners among depositors exceeded the percentage of share owners who did not have savings-the former stood at 15 per cent, the latter at 4.5 per cent. Similarly, in Hungary depositors were more likely to own shares-the proportion of the latter increased from 3. 7 per cent to 13. 9 per cent. In the more concise terms of a coefficient of correlation, the relation­ ship between owning shares and saving money in banks amounted to 0.17 in Poland and Hungary. By comparison, in Russia, where share­ holders also tended to be depositors, the correlation was 0.11. In the Czech Republic and Slovakia, it fell to an almost insignificant value (O.OS), close to Bułgaria at O.Ol. The latter finding should not surprise us: one could hardly expect anything else in a country where the relative quantity of shareholders, broadly defined, did not exceed 1.1 per cent of all adults (see Table 18). In Bułgaria, there was practically no 'social space' for the differentiation of financial strategies ofthis kind. The percentages of depositors are given in Table 17. They are clearly differentiated across socio-occupational strata. lt seems that savings formed some sort of axis of social division. Also in Table 17 are re­ ported the percentages of persons who owned shares, which sheds some light on whether ownership of financial assets had stratified. We restrict

86

our analysis to the distribution of those with savings and shares, al­ though we would be able to draw more telling conclusions if we knew a lot more about the size of stockholdings and bank accounts. Needless to say, such data are not available: even if people were prepared to supply some information to a survey its credibility would be very doubtful. Table 1 7. Ownership ofsavings (%) Socio-occupational categories

Higher professionals and managers Semi-professionals Routine nonmanual, sales and service workers Owners with employees Owners without employees Lower-grade technicians Skilled workers Unskilled workers Agricultural laborers Farmers Total Correlation ratio (E)

Bułgaria

Czech Hungary Republic

Poland

Russia

Slovakia

74.3

96.3

7 1 .8

45.6

27.3

96.0

60.4 58.8

94.4 90.2

5 5 .2 42.4

30.6 22.2

, 20.0 1 7.7

93 .2 90.4

68. 1 42.4 65.7 48.6 45.6 45.5 54.8 52.5 0. 1 8

88.4 90. 1 94.0 90.0 85.5 87.4 1 00.0 90.3 0. 1 2

69. 1

39.8 26.3 1 8.6 1 4.0 1 2.8 7.4 9.2 20.2 0.26

43 .7 3 1 .9 23.2 1 7.7 14.8 23.2 46.2 20.7 0. 1 4

88. 1 9 1 .4 90.2 88.0 84.4 88.4 1 00.0 89.5 0. 1 2

SO.O

5 5 .2 40.2 3 1 .9 37.2 44 . 1 43.6 0.23

Bank accounts, the more traditional of the two forms of monetary allocation, differentiated socio-occupational strata slightly more in Po­ land and Hungary. Between-category distances were more sharply ex­ hibited in Poland, although, generally, intercountry differences were not large. This translated into a relatively higher value of the correlation ratio in Poland (0.26). The between-category variation could not be as high in the Czech Republic and Slovakia, simply because the percentage of depositors approached 90 per cent, with 100 per cent of farmers holding accounts. Russia once again appears to be a deviant case. As far as the rate of depositors is concemed, Russians are on a par with Poles (20.2 per cent), but the same percentage in both countries seems to reflect differ­ ent circumstances. The low ratio in Poland indicates rather a lower in­ terest among Poles in saving as a form of allocation of money than a lack of financial means. The latter fact may well explain the Russian case, however. We have already learned about the low materiał standard

87 of living in Russia, and we need look no further than the remarkably low incomes in this country to find the key to understanding why only one in every five citizens had a bank deposit. The remaining 80 per cent simply do not possess disposable financial resources which could be put in a bank account. We must acknowledge that the correlation between occupational category and savings is less striking than that between incomes and pos­ session of materiał goods. Savings 'discriminated' poorly between the intelligentsia, the working class, owners, and other socio-occupational strata. Classical indices of economic status did better, as evidenced in the lower correlation ratios in Table 1 7 as compared with those pre­ sented in Tables 1 4- 1 6. When we come to examine the variation of savings across categories, we are not, therefore, surprised that it weakly conforms to the hierarchy of socioeconomic status. The ordering of the socio-occupational hierar­ chy with respect to percentage of owners of bank accounts shows a dis­ parity with the classical stratification ladder. As far as the Czech Re­ public and Slovakia are concemed, it appears due to the minor differen­ tiation between categories. In the remaining four countries, similarities are fragmentary and, practically speaking, show up only in the position at the top of larger owners, managers, and the intelligentsia; the place­ ment at the bottom of unskilled manuał workers; and the marked dis­ tances between consecutive social levels. In fact, only in Poland and Hungary does the percentage of deposi­ tors stratify the hierarchy of basie socioeconomic segments in a similar way to incomes and possession of household goods. In Polish society, the percentage was highest among managers and the intelligentsia, with 46 per cent of them having bank accounts. They were followed by own­ ers with employees, of whom 40 per cent kept money in banks. Next carne nonmanual categories at the medium and lower levels, small busi­ nessmen, manuał workers, and, finally, farmers, of whom no more than 1O per cent put money in banks. The sequence is much the same in Hungary, although the high percentage of bank depositors among the intelligentsia, managerial cadres, and larger owners is more pronounced relative to the lower strata. It is apparent that in the four remaining countries percentages of de­ positors in particular categories exceed what one may predict from in­ comes and possession of materiał goods. Agricultural workers in Russia serve as a good example. Among them, saving money in banks was more popular than among routine clerks and manuał workers outside

88

agriculture, although both incomes and possession of household goods in agricultural households were relatively lower. In Russia, farmers also contributed to the decomposition of hierarchies of savings and incomes. We can see that depositors were overrepresented among Russian farm­ ers relative to mean incomes in this category. In Bułgaria, this was the case for skilled workers. These examples lend support to the thesis that a predilection for keeping money in a bank does not depend only on financial resources but also reflects individual choices, strategies, even tastes. Having a bank account does not necessarily imply a higher economic position. ·Furthermore, what is at stake is not only individual financial means and preferences to save or not to save, but also institutional constraints. The psychology of wealth aside, the higher propensity to save money in the Czech Republic and Slovakia as compared with Poland or Russia, must result from its profitability. lnsofar as government monetary policy is stable and inflation low, citizens can trust the banks, and bank accounts grow as a consequence. Table 1 8. Ownership ofshares (%) Socio-occupational categories

Higher professionals and managers Semi-professionals Routine nonmanual, sales and service workers Owners with employees Owners without employees Lower-grade technicians Skilled workers Unskilled workers Agricultural laborers Farmers Total Correlation ratio (E)

Bułgaria

Czech Hungary Republic

Poland

Russia

Slovakia

0.3

1 9.0

14.7

23.3

43 .5

37.7

1 .5 0.7

1 5 .4 1 2. 1

I I .O 9. 1

1 2 .6 5.3

4 1 .2 39.2

3 1 .9 28.8

7.3 2.3

2 1 .7 15.1 16.1 1 0.6 7.8 7.3 3 1 .8 12. 1 0. 1 3

7. 1 7.8 1 4.0 6.8 4.2 5.8 4.4 7.6 0. 1 2

18.1 8. 1 7.9 4.9 1 .9

72.9 37.7 3 7.2 42.7 32.2 23.9 37.5 39.2 0. 1 4

3 7.6 3 5 .2 28.8 24.4 2 1 .0 1 8.7 26. 1 26.9 0. 1 3

o.o

1 .2 0.6 1.1 1 .6 1.1

O. I I

o.o

1 .4 6.7 0.25

A glance at the proportions of shareholders among the various socio­ occupational strata leads to similar conclusions (see Table 18). This as­ pect of differentiation tends to converge with the classical pattem of socioeconomic stratification even less than that of savings (we are still

89 relying on the comparison of percentages in the respective categories). It is true that private entrepreneurs, the intelligentsia, and managers were represented among shareholders rather than semi-professionals, middle­ ranking administrative staff, clerks, and sales assistants, while manuał workers appeared to be least active on the stock market. However, the differences between the categories were not conspicuous overall. There were, instead, marked differences between countries in respect of the total percentage of persons who declared share ownership. How­ ever, one qualifying remark is in order. It may well be that the differ­ ences reflect a different understanding of 'stocks' , 'bonds', and ' shares' . I suspect that, in Russia, where the percentage of share owners was­ according to our survey data-surprisingly the highest, respondents re­ ported as valuable assets any 'papers' which certified some kind of ownership. Slovakia was the second country marked by a high percent­ age of share owners: this figure might include property restitution cou­ pons. In Slovakia, government action related to restitution intensified in the 1990s.

PRELIMINARY CON C LUSION Having outlined the basie barriers to social mobility and pattems of economic stratification, we may make some preliminary generalizations. First, social stratification in postcommunist countries still contains traces of the common economic and political past peculiar to East Cen­ tral Europe. Nevertheless, the peculiarities tend to be counterbalanced by certain generic developmental features characteristic of all postindus­ trial societies. Secorid, our findings also suggest that the differences in the configu­ ration of social distances between East Central European societies were overpowered by the similarities. What made East Central European patterns especially unique was the high economic position of owners of larger firms. They received the highest incomes compared to the other basie occupational strata and were better off in respect of the possession of resources which deter­ mined materiał standard of living. This differs from the advanced indus­ trialized countries where the 'old middle classes '-that is, the represen­ tatives of small business who correspond to 'our' owners-find them­ selves in the middle of the stratification ladder and not at the top. Only in Poland and Hungary did the family incomes of larger owners (that is,

90

those with employees ), fall below the incomes of the intelligentsia and managerial cadres. In Bułgaria, the Czech Republic, Slovakia, and­ particularly-Russia, the East Central European old middle classes were still the most economically privileged categories. Does this mean that Hungary and Poland were in the vanguard as far as the adaptation of their economic hierarchies to the stratification typi­ cal of developed capitalism is concerned? The cross-national variation in these patterns in East Central Europe appears to constitute evidence of the uneven pace of marketization. I shall now proceed to address the question of whether the conclusions I have reached as regards social mobility and economic stratification might require revision if a more comprehensive view of social stratification in postcommunist societies were taken.

C H AP TE R 5

THE ' OWNERS ' DEBATE : NOMENKLATURA OR SELF­ RE CRUITMENT ?

THE RECRUITMENT of owners requires careful investigation. We have already shown that, in the first half of the 1990s, businesspeople were economically the most privileged strata of all basie socio-occupational categories. In Chapters 2 and 3, we traced their inter- and intragenera­ tional trajectories and levels of self-recruitment at the end of the 1980s. We shall now examine the determinants of access to business at the be­ ginning of the 1990s, attempting to gain some insight into the mecha­ nisms of recruitment. We shall also take into account another path leading to engagement in business activities which was hotly debated in public discussions in the 1990s. The central concem of my study is straightforward: nomenklatura or inheritance of property? We shall attempt to determine which of these two channels of mo­ bility appeared the more probable entrance path to business in the first years of emerging capitalism in East Central Europe. For their part, the nomenklatura were endowed with specialized qualifications, experience of running state-owned socialist enterprises, and political capital. They made use of these assets by moving into business-this is the claim ad­ vanced by proponents of the transfer thesis. We also have a traditional entry-gate-continuity in these positions. In this case, assets consist in ownership of a firm at the beginning of an occupational career, inheri­ tance of the family business, or long traditions of ownership. Undoubt­ edly, these are considerable assets. But which of these attributes was more important, which less, and in what proportions?

92

W H Y NOMEN KLATURA? The debate is a strictly historical one, strongly embedded in the changes in the political system and informed by political struggles in the region. The political break was directly related to the circulation of elites. Above all, however, we are dealing with ownership of firms and so-in the common view-with lucrative positions in statu nascendi. The de­ bate revolved around such issues as whether it was fair that former communist cadres were able to reap the financial benefits of their posi­ tion. They ruled in the name of the Communist Party, and all of them contributed to the maintenance of communist rule. Such discussions created a climate of disenchantment, along with expectations which were expressed in terms of selective norms: who can and who should be inhibited from running a profitable activity which brings economic benefits and opens the door to wealth. Arguments of this kind appeared primarily in the political campaigns in the newly emerging democracies of East Central Europe. Fervent emotions were stirred up by the mass media and the public pronounce­ ments of political leaders. They alerted public opinion to the fact that the managers of state enterprises from the moribund communist period were taking advantage of their political assets and moving into the pri­ vate sector. It has to be admitted that such claims were, in large part, true. Some of them were based on real, although fragmentary, observa­ tions, not unusual in popular sociology. The representatives of the op­ posite side countered and appealed for moderation. All in all, the con­ troversy over the nomenklatura became a battle between contending political factions and also amongst those contesting the overall assess­ ment of the recent past. While the contested terrain of political options may be abandoned in sociological analysis, one cannot ignore the issue of recruitment to the newly emerging class. The sociological significance of this question lies also in the fact that it attracts widespread attention, which itself gives rise to a dynamie in social relationships: contested issues give impetus to potentia! mass action. The problem was not only political per se. It may also be refined in such a way as to disentangle what lies behind the alarming thesis about the inflow of former nomenklatura to the private sector. For our present purposes, we shall refer to former contributions on this issue made by authors of theories on social dynamics in postcom­ munist Europe-these theoretical debates were always set against the

93 backdrop of public discussions. Let us systematically consider the po­ tentia! sources of the transformation of the socialist nomenklatura into private entrepreneurs. Exponents of the first position emphasize what they call a conversion of political capital into economic capital. This theory holds that the pre­ rogatives of organizational power were exchanged for the property of firms. Members of the nomenklatura have metamorphosed into capital­ ists due to the fact that their positional power has allowed them to ap­ propriate productive assets and because the social networks in which they were involved have provided them with advantages in emerging markets. These arguments were illustrated by privatization in Hungary and Poland. According to this argumentation, practically speaking, if during the disintegration of communism somebody took advantage of his leading position in bureaucratic structures, no obstacles prevented him from establishing his own business (Hankiss, 1991; Staniszkis, 1991; Szalai, 1994; Szelenyi and Szelenyi, 1995). Political capital lost its importance irrevocably-as the authors cited above claim-but managers appointed by virtue of nomenklatura rules kept their influence over the operation of enterprises. For example, if their firms entered on the path of privatization, they were able to prede­ termine decisions conceming type of privatization, who was to buy the firm, and at what price. The loopholes in the privatization of state firms allowed cadres to purchase shares at incredibly low prices and to trans­ fer assets of state firms into privately owned companies. Following this scenario, the former manager became an owner. He could also make use of acquired economic capital in another way-for example, he could buy stock in other privatized enterprises. According to other adherents of the thesis about the conversion of political into economic capital, these mechanisms were established as early as the 1970s. Acute observers of political life in Poland assert that it was during this period that state officials and party functionaries started to set up shops and small manufactures not on their own behalf but on that of relatives and friends. The communist party elite resigned from political rule, but took care to substitute its political with economic power (Tarkowski, 1983). The second mechanism involved in the conversion of the nomenkla­ tura involved reliance on the transfer of the capital of knowledge and qualifications. In order to run a firm one needs skills of universal utility as applicable in various sectors of the economy. Technical competence derived from the management of state-owned enterprises might be easily

94 transferred to the operation of private businesses. Young communist technocrats attended educational institutions where they received a solid training in engineering, economics, or administration. Added to experi­ ence gained in managing enterprises, all these assets facilitated the de­ cision to start out on one's own. Certainly, the capital of knowledge strengthened the power of economic capital and the various strategies for entering business already mentioned. There might have been a third mechanism of exchange of 'capital' at stake, which I would describe as the circulation of ownership and cultural capital. Those who made up the managerial cadres in socialist enterprises belonged to the intelligentsia. They were endowed not only with manage­ ment and organizational abilities, but also with generał knowledge and social capital which helped them to orientate themselves well in the complicated world of social life. It is well known that the cultural capital of the intelligentsia provides them with a great allocative power as regards the highest and most profitable positions. Empirical data coming from many countries, including Poland, confirm this regularity. Both cultural and social capital derive from a 'good' family background, college educa­ tion, connections with influential people, and, in relation to one's occupa­ tion, from experience with a world of symbols which is conducive to openness, elasticity of thinking, and a broad cognitive perspective (Bourdieu, 1986; Di Maggio and Mohr, 1985; De Graaf, 1989). Słomczyński and Shabad ( 1997) showed that in Poland contacts with the West prior to 1988 gave people a significant advantage in gaining entrance to the private entrepreneurial class in 1993, after controlling for severa! important variables, such as education and gender. Not unexpectedly, the cultural capital of nomenklatura members might come to be an asset in their endeavors to acquire an optimal strat­ egy of conversion. For example, in going into partnership with a foreign company, the 'converted owner' was able to take advantage-not al­ ways but sometimes-of the opportunity to renew or transfer intema­ tional connections which he had established during his management of the socialist enterprise. By activating social networks formed in the course of years of administrative activity, farmer cadres were able to acquire valuable information on emergent markets as well as the credit necessary to run their own businesses. According to Stark, who had al­ ready, in the early 1980s, carried out research on informal patronage ties and protective networks in socialist enterprises in Hungary, these net­ works persisted in the 1990s. They were organized by the farmer mana­ gerial cadres, and nowadays the old, informal ties reside in privatized--

95 formally at least-structures. He claimed that a reintegration of the old elite of state managers had occurred, rendering it an elite of private business operators (Stark, 1992). I have referred only to some of the hypothetical explanations con­ ceming the mechanisms of transition from the nomenklatura to business. Proponents of this thesis-that the 'old middle class' in East Central Europe has its roots in former communist cadres-used to refer to 'the mass transitions which exceed the 'norm'. We shall determine whether this norm has been really exceeded, surveying the relationship between holding high managerial positions in 1988, just before the collapse of the communist regime, and membership in the category of owners in the 1990s. I will try to establish the relative proportion of former nomenkla­ tura members among owners of firms after some years of systemie trans­ formation-that is, in 1993, our 'destination' point. In order to assess the relative strength of the 'nomenklatura effect', one needs some reference points. How much more effective, then, was the 'nomenklatura route' as compared, for example, to the effect of edu­ cation in getting into the category of owners? In the case of education, the most valuable asset is a diploma from an institution of higher education. It would be interesting to know whether the higher educational qualifications proved more critical in access to the ownership of firms, relative to secondary schools, basie vocational schools, or graduation from the lowest, elementary level. Bearing in mind that educational channels allocate individuals on the labor market and determine occupational status, we must determine · whether educa­ tional level opened, or perhaps closed, the entry-gates to business. Self-recruitment will be our next reference point. We ask, first, whether people running a business in the 1990s had done so previ­ ously-in 1988? This allows us to see to what extent owning firms un­ der different sociopolitical and economic circumstances increased the likelihood of ownership after systemie upheaval. Certainly, this is only one aspect of the transfer of property over time. We expect that, apart from the transfers going on in the medium term, dated to 1988, continuity of ownership over a longer period of time also reshaped pattems of structuration in this area of social space. This involves the reproduction of ownership from a remote past and a return to property owned by parents and grandparents after a break of severa! decades Some telling examples are appropriate. There is a well-known con­ fectionery firm, Blikle, in Warsaw, which was founded in 1851. At the

96

beginning of the 1990s, Blikle was taken over from the state by a son of the last owner, after a break of several decades. Blikle had been na­ tionalized in the aftermath of the Second World War and reprivatized in 1 989. The successor was a physicist who gave up his scientific career for the sake of re-establishing the continuity of the famous family firm. Poland also witnessed the revival of the prestigious publishing firm of Gebethner and Wolff which, in the 1990s, made remarkably swift prog­ ress. Similarly, the Orgelbrand publishing house has been resuscitated as a private firm. Such comebacks have happened always and everywhere, but a return to business after the fa11 of the communist system in East Central Europe seems spectacular. An example of intergenerational succession in politics is József Antal1, the first prime minister of the Hungarian government after the anticommunist opposition carne to power. His fa­ ther had been a deputy minister in the prewar cabinet of Admirał Hor­ thy, and his chief adviser was a nephew of Istvan Bethlen, the prime minister of a conservative government in the 1920s, who died during the war in a Soviet prison. Szelenyi ca11s such retums to one's father's or grandfather's position the 'circulation of elites'. In this context, elites are composed of inheri­ tors of family assets which were lost after the coming to power of the communists. The inheritors became an economic 'elite' after having their property returned after the collapse of the communist regime. They are new among owners with respect to their occupational career but deeply rooted in private business by virtue of family background (Szelenyi and Szelenyi, 1995). Predictions related to the circulation of elites may be empirica11y tested since we have the reports of respondents in each country to tell us whether their grandparents owned firms. We rely on grandfather' s oc­ cupational status as particularly relevant in assessing the effect of circu­ lation because-according to Szelenyi-circulation covered newcomers to business in the 1990s who derived their roots from businesses founded in the precommunist period. The occupational activity of the grandparents of our respondents fell within this time frame (the young­ est of our respondents were in their twenties in the 1990s). Apart from coming back to business, ownership of firms in the 1990s might owe much to continuity in these positions. This concems both continuity of fathers' positions and an uninterrupted career of one's own. To obtain a more detailed understanding of transmission of prop­ erty throughout the life cycle, we examined whether operating a firm at

97 some point of the occupational career increased the likelihood of owner­ ship in the 1990s. It seems that long-lasting inclusion in this category tends to strengthen links with business. One may anticipate that every­ one who started off his occupational career in business owned one in 1988, and-additionally-had taken it over from his grandfather or fa­ ther, so giving him a better opportunity to stay on than to move on. We shall now address the problem of how the 'path through business' com­ pared with the transmission of former nomenklatura assets.

SELF-REC RU I TMEN T RATHER TH AN NOMEN KLATURA MEMBERSH I P OFFERED G REATER OPPORTUN ITIES The data presented in Table 19 directly address these questions. The table gives the metric coefficients in logit regression with the dependent variable-membership of the category of owners in 1993--coded 1 for owners and O otherwise. The positive values in the first row show the extent to which nomenklatura membership in 1988 increased the logit of being an owner in 1993 as compared to individuals outside nomenkla­ tura circles. In fact, metric coefficients indicate the relative odds of ac­ cess to business. The nomenklatura includes: the highest functionaries, state and communist party officials, leaders of trade unions and youth organizations, but predominantly the highest management of state­ owned enterprises and administrative offices, chiefly managers. Table 1 9. Logistic regression analysis of being an owner in 1993 . Effects of membership

o/nomenklatura in 1 988, ownership o/firm in 1988, years ofschooling, and sexa Effects of: Membership in nomenklatura in 1 988 ( 1 =member) Owner in 1 988 ( l =owner) Years of schooling Sex ( l =men)

. Bulgana

Czech Hungary Repubi'1c

Poland

Russia

Slovakia

0.90**

0.67**

0.62

0.65*

1 .20**

0.59

8.44** 0. 1 2* * 0.02

9.64 O.OS * 0.73 * *

4.56** 0.08* * 0.36

4.93 * * 0.06 0.77

7. 1 9* * 0 . 1 4* * 1 .3 5 * *

5.08** 0.08 0.98

** p