The Green Revolution, Employment, and Economic Change in Rural Java 9789814376792

Citation preview

Occasional Paper No. 84

The Green Revolution, Employment , and Economic Change in Rural dava A Reassessment of Trends under the New Order

The Institute of Southeast Asian Studies was established as an autonomous organization in May 1968. It is a regional research centre for scholars and other specialists concerned with modern Southeast Asia, particularly the multi-faceted problems of stability and security, economic development, and political and social change. The Institute is governed by a twenty-two-member Board of Trustees comprising nominees from the Singapore Government, the National University of Singapore, the various Chambers of Commerce, and professional and civic organizations. A terrman Executive Committee oversees day-to-day operations; it is chaired by the Director, the Institute's chief academic and administrative officer. The ASEAN Economic Research Unit is an integral part of the Institute, coming under the overall supervision of the Director who is also the Chairman of its Management Committee. The Unit was formed in 1979 in response to the need to deepen understanding of economic change and political developments in ASEAN. The day-to-day operations of the Unit are the responsibility of the Co-ordinator. A Regional Advisory Committee, consisting of a senior economist from each of the ASEAN countries, guides the work of the Unit.

The Green Revolution, Employment, and Economic Change in Rural .Jav A Reassessment of Trends under the New Order

Chris Manning Fll,.., Uni'IW'fity of Sourh Au«nl~

ASEAN ECONOMIC RESEARCH UNIT

INSTITUTE OF SOUTHEAST

IAN

Cataloguing in Publication Data Manning, Chris The green revolution, employment, and economic change in rural Java: a reassessment of trends under the New Order. (Occasional paper/ Institute of Southeast Asian Studies; no. 84) I. Green revolution- Indonesia- Jawa. 2. Agriculture - Economic aspects - Indonesia- Jawa. 3. Agricultural labourers- Indonesia - Jawa. 4. Agricultural wages- Indonesia - Jawa. I. Institute of Southeast Asian Studies, Singapore. II. Title. III. Series. 1988 DS50 I 159 no. 84 ISBN 981-3035-02-1 ISSN 0073-9731

Published by Institute of Southeast Asian Studies Heng Mui Keng Terrace Pasir Panjang Singapore 0511 · All rights reserved. No part of this publication may be reproduced, stored .in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior consent of the Institute of Southeast Asian Studies. ©

1988 Institute of Southeast Asian Studies

The responsibility for facts and opinions expressed in this publication rests exclusively with the author and his interpretations do not necessarily reflect the views or the policy of the Institute or its supporters.

Typeset by International Typesetters Printed and bound by Kim Hup Lee Printing Pte. Ltd.

Contents List of Tables Introduction I Consequences of Rural Economic Change: II The Conventional Wisdom III The Green Revolution in Java: Small and Large Farmer Participation and Land Concentration Land Concentration and Landlessness Trends in Land Owned and Controlled Tenancy IV Labour Use in Wet Rice Farming Overall Trends in Employment and Labour Use Factors Influencing Labour Use Labour Contracts and Wages in Rice Farming V Tebasan and Kedokan The Bawon Share and Agricultural Wage Rates Exclusionary Labour Contracts VI Rural Labour Markets, Income Distribution, and Poverty Rural Labour Markets Income Distribution and Poverty VII Conclusions

vn 1 4 10 13 18 24 29 29 38 50 52 55

58 61 62 68 72

~~

w

References The Author

82 96

List of Tables I. Trends in Participation in BIMAS, Adoption of Modem Varieties, Fertilizer Use per Hectare, and Yields per Hectare by Farm Size Operated, Java-Bali, 1971 - 78 2. Distribution of Farms and Land Owned on All Farms a nd Sawah Farms by All Land Owned, Java, 1973 3. Distribution of Households and Land Area by Area of Land Operated, Gini Coefficients, and Landlessness in Java ( 1973. 1976, and 1983) 4. Landlessness in Java 1979- 83: Results of Village Studies 5. Changes in the Distribution of Farms / Agricultural Households by Area of Land Controlled, Java. 1963- 83 6. Distribution of House holds, Sawah Land, and Land Transactions by Area of Sa wah Owned in 1978. Gemarang and Sukokembangs ri Villages ( 1978) 7. Sa wah Land Sales by All Households and Currently Landless, and Accumulation of Sawah by Large Landowners, Gemarang Village, 1950- 78 8. Distribution of Employment and Distribution of Increment in Employment, Rural Java, 1971 - 82 9. Labour Use in Rice Cultivation, Selected Villages in Rural Java, Various Years 1875- 1980 10. Average Labour Use in Dry Seaso n Rice Culti vati on in Gemarang and Sukosari, 1969 and 1978 II. Average Labour Use per Year in Rice Cultivation in Gemarang Village, 1968169 and 1977 j78 12. Number of Hand Tractors a nd Estimat ed Area of Sawah Cultivated by Tractor in Selected Kabuparen in West Java

(1982) 13. Number of Households and Area of Sawah Cultivated by 14. 15. 16. 17.

Different Techniques, Mariuk Village, West Java (Wet Season 1979 I 80) Changes in Bawon Shares and Size of Ba won Payments, Selected Villages in Java, 1968j69- 1979j 81 Index of Real Wages in Hoeing and Weed ing in Six West Java Villages, 1967- 77 Earnings per Hour in Various Activities in Java, 1975- 78, Very Poor Households Only Real Wage Increases in Seven West Java Villages, 1977- 83

II

14

16 18 19

22

23 30

32 34 35

40

41 56 57 63 67

I

Introduction

Recent studies of economic c ha nge in wet n ee growmg commumttes m rural Java in the early 1980s suggested a rem a rkable turnabout in the fortunes of poorer. landless ho useho lds compared with the rather dismal picture given by village studies conducted in the la te 1960s a nd 1970s. 1 After decades of stagnation. it has been suggested that rural wage rates have increased significantly. widespread la bo ur sho rt ages have been reported at peak periods of labo ur demand in agriculture. and many poorer rural households appear to be benefiting es pecially from employment in urban locations (Collier et a l. 1982h: Wiradi. Manning, and Sri Ha rtoyo 1984: Kas ryno 1983: Wo rld Ba nk 1985). The contrast with reported devel o pments in the first fifteen years of the New Order is difficult to believe for ma ny observers of economic change in rural Java. Fo r many. the densely populated island has for ow r a century represented a classic case of rural la bo ur surplus and since 1965 a prime exa mple of la bour displacement. inequa lities. and rural polarization occurring as a consequence of the green re\ olution. It suggests a reassessment may be required of the nature o f changes which were generally believed to have occurred in rura l a rea.•.; in the early years o f the Soeharto government. and of the implied disctmtinuit y in rural economic change in this early period compared with more recent developments in the early 1980s. This pa per attempts such a reassess ment. concentrating on the consequences of economic developments in rural Java in the period of military ruk to the early 1980s. I will argue that identilication of commercialization of rice farming as a central fa~:tor contributing to rural inequalities and labour displacement in agriculture has been to some extent misplaced and provides an overly simplistic pi~:ture of the processes of rural economic change in rural Java, especially in the early and mid- 1970s. This seems. in part. to be related to a tendency to overstate the polarizing tendencies created by recent agricultural change as a reaction to Geertz's ( 1963) pi~:ture tlf slowly ~i n vo luting". undifferentiated rural communities in the face of growing population pressure on land. It springs also from a tendency to generalize from the green revolution experience of other Asian co untries with different agrarian structures and economic environments. ln addition. in the wake of the army's political takeover and the massacres which occurred in 1965- 66, ~ ideological opposition to the Soeharto regime. its avowedly open door to foreign capitaL and attempts to resurrect market

2

THE GREEN REVOLUTION, EMPLOYMENT, AND ECONOMIC CHANGE

forces in rural areas, probably led some observers to conclude that increasing rural conflict and polarization of wealth was inevitable. Patterns revealed in some micro studies (especially Franke 1972; Budhisantoso 1975) were presented to suggest that the benefits from the green revolution had gone mainly, if not exclusively, to larger farmers and the new economic policies did little to encourage marginal pea.Sants to take advantage of the new rice technology. This, in combination with backwash effects on employment, was seen as unmistakable evidence that such polarization was already in process by the mid-1970s (see especially Hinkson 1975; Gordon 1978; Palmer 1977b). Although the Bimbingan Massal (mass extension) or BIMAS intensification programme favoured large landowners who took advantage of the new technology much earlier than small farmers, I will suggest that the large bulk of empirical evidence do not confrrm a trend towards polarization as the dominant pattern of agrarian change in rice growing communities in Java in the 1970s. In particular, it will be argued that generalizations drawn largely from selected village studies concerning trends in rural employment, incomes, and wealth between and within economic classes have frequently paid too little attention to the complex effects of agricultural change and to the influence of economic developments outside agriculture and in urban areas. They have also tended to neglect the critical importance of varying responses to economic change associated with diverse local labour market conditions. While a number of developments have had deleterious effects on employment and incomes especially of landless labourers, and the government fixation with expanding rice production has almost certainly increased regional inequalities, some of the negative effects believed to be linked to the green revolution and the processes underlying them are open to question. Others appear never to have been as widespread as initially implied or predicted. Closer attention to special conditions of agrarian structure, agronomic conditions, and broader macro-economic environment affecting economic and social change in rural Java suggests that landless and marginal peasants have probably fared much better than has often been believed to be the case, and probably better than in some neighbouring countries. Certainly no "take-off' occurred in the 1970s in terms of dramatically increased rural wage rates or manufacturing-led labour absorption outside agriculture (Lluch and Mazumdar 1983). But the relatively favourable macro-economic environment linked closely to Indonesia's position as an oil exporter has enabled the diffusion of economic benefits, even though the government has for the most par:t paid scant attention to policies which directly influence the welfare of the rural poor. This occurred furthermore in spite of centralized government formulation and implementation of rural policies.

INTRODUCTION

3

The paper opens with an overview of what might be considered the ~'conventional wisdom" regarding the effects of the green revolution on rural employment, incomes, and wealth in rural Java to the late 1970s, and the prevailing view regarding the structure of rural employment and labour markets. This is followed by a mote detailed account of the distribution of benefits among small and large farmers, including the issue of land concentration, and the impact of technological and institutional changes on employment and incomes of labouring families. We then tum up to more general developments in rural employment and briefly examine trends in incomes, income distribution, and poverty.

Notes This paper was written while the author was a research fellow in the School of Social Sciences, Flinders University. Thanks are extended to colleagues in Asian Studies, especially Lance Brennan, for encouragement and to Janet Pascoe and Vickie Mansell for typing the manuscript. I would also like to thank Graeme Hugo. Hal Hill, Sisera Jayasuria, Paul Alexander, and Gavin Jones for helpful comments on an earlier draft of this paper. Any errors or omissions remaining are, of course, entirely the author's responsibility. I. Most empirical studies have dealt primarily with wet rice (sawah) communities in Java and these are the major focus of this survey. See especially White ( 1978). Collier ( 1981), and Collier et al. ( 1982a) for a summary of the major findings of the village studies.

II

Consequences of Rural Economic Change: The Conventional Wisdom

One important coincidence for rural social and economic change under Soeharto was the discovery of new high yielding rice varieties (HYVs) at the International Rice Research Institute (IRRI) in the Philippines in the rnid-1960s on the eve of the army's accession to power, and the major political and social upheavals which led to the elimination of the Indonesian Communist Party and the influential peasants' front, Barisan Tani Indonesia (BTl). The means to dramatically raise rice output with increased inputs of fertilizer in irrigated wet rice or sawah areas carne at a time when landless and poor peasants were deprived of powerful patrons and policies orientated to land redistribution. The significance of the coincidence was not lost on scholars of recent Javanese political and economic history. Mortimer (1984, p. 102) writing in 1975 noted, for example, that "the terms of exchange, then, are turning sharply against the economically weak just as the economically strong are finding strong incentives and opportunities to exploit their advantages more freely". Mortimer viewed the coincidence of two sets of factors- prospects of large income gains to landowners from applications of inorganic fertilizer to new rice strains and the elimination of political and social forces representing the interests of poor peasants and landless labourers - as a potentially powerful combination leading to polarization of income and wealth and social conflict. t Despite initial euphoria regarding the potential which the green revolution offered for overcoming food shortages and poverty, by the early 1970s a number of studies pointed to similar propects of increasing inequalities associated with the spread of new seed varieties in many other Third World countries. The major elements in this process were: early adoption by large farmers who had better access to key inputs of irrigation water and credit; displacement of labour and discrimination in employment against certain groups of labourers (especially women) as mechanization accelerated in many countries; and accumulation of land and other assets by larger farmers. 2 These processes were well documented especially in the literature on agrarian change in India and Pakistan (see for example Frankel 1972i Johnston and Kilby 1975). Although several studies suggested that after an initial lagged response small farmers

4

CONSEQUENCES OF RURAL ECONOMIC CHANGE

5

adopted quickly and net employment effects were positive (Berry and Cline 1979; Barker and Cordova 1976; Ruttan and Binswanger 1978), empirical research in a range of countries continued to emphasize the deleterious effects of the green revolution on employment and the distribution of income and wealth. 3 In Asia, major exceptions to these unfavourable trends were noted in South Korea and Taiwan where land reform had produced a relatively equitable distribution of holdings (Griffin 1974; Booth and Sundrum 1985). The implications were clear for Java, where, despite the small average size of farms, land distribution was relatively equal - much closer to the celebrated "bimodal" rather than "unimodal" land distributions -and landlessness widespread: the green revolution might be expected to offer little prospect of positive gains to the majority of marginal and landless peasants. Micro studies in particular have indeed identified a long list of alarming developments in Java especially in the decade of the 1970s. White (1978, p. 8) provides a summary of what he considers to be the major findings of empirical research during this period: A perusal of available case studies of agrarian and other social-economic developments in rural Java would point out the following changes in recent years: unequal distribution of the direct and indirect benefits of new biological and chemical technologies in rice production. new technologies in cultivation, weeding. harvesting and processing which cut the costs for the larger farms but reduce the employment and income opportunities of labourers; more frequent harvest failures resulting from the new varieties' vulnerability to drought, flood and particularly to pests, which have affected the income of small farmers more seriously than those of large farmers; declining real agricultural wages; unequal access to agricultural and other forms of subsidized government credit, . . . . increasing landlessness and an acceleration in the purchase of agricultural land by wealthy villagers and urban elites: ....

In three respects, in particular, the case of Java has been of interest to observers of agrarian change associated with agricultural modernization in Asia: (a) the contrast between recent developments and the unusual historical experience of agrarian change in Java popularized by Geertz (1963) in his well-known thesis of agricultural involution; (b) the negative employment effects of "institutional" change induced by the green revolution; and (c) implications of a marked pattern of rural labour market "segmentation" for trends in rural poverty and income distribution. I will briefly summarize the major elements of each before turning to a more detailed examination of empirical research. The belief that Java had experienced a rather unique process of agricultural involution from colonial times through to the early y·ears of independence was a widely accepted view among social scientists till the early 1970s when various studies began to point to the inequalities which

6

1 HI· ( oR F I N RFVO I Li ll O '. FMI'I OYM EN T . t\N D FCONOMIC C ll 1\

tiF

ex isted in rural Java, a nd to tendenc ies contri buting to increas ing polarilation of income and wea lth 5 Geert7 referred to the co nce pt of "shared poverty" in particu lar as th e specific cu ltural norm by which Javanese acco mmodated th e growi ng rura l population upon a n increas ingly overcrowded la nd base with few opportunities for empl oy ment outside agriculture, a mechanism (i n contrast to th e ex perience of many other T hird World countries) which minimi?ed social a nd eco nomic different iation (see also Geert7 1984) . Although historia ns a nd a nthropologists in particu lar have subseq uently questioned key aspec ts o f the agricu ltural involution th esis and the ex istence of shared poverty as a peculiar "social" institution (especiall y criti ci7ing the sca nt attenti on which GeeT17 paid to class stru ct ure a nd landl ess ness), developments since 1965 have neverth eless been frequently viewed as a marked departure from past patterns of rural change 6 Co llier ( 1981), for exa mple, des pite a certain ambiva lence as to whether developments und er Soeharto have been qualitatively different from the pas t. ap pea rs to co nclud e in the affirmative in summing up so me of the evidence on rural economic cha nge: 7 T he above menti oned evidence ind icates that techno logy cha nge a nd substitution of capita l for la bor is beginning to permeate ma ny as pects of J ava's rice economy ... . the notion that the eco no my of .I ava nese society can rema in immune to the penetrati o n of new techn o logy must be put a~ id e in order to full y und erstand the current level of receptivity to new labo r displacing tec hnologies.... (p. 166) A review of emerging tre nds in the orga nin t ion of we t ri ce cult i\'a t io n o n Java suggests that the imperati ves of ef'li ciency a nd pro flt a hilit ~' are bt:ginn ing to exact their toll in the erosion of traditi ons.... ( p. 17 1)

"Inst itutional" cha nge - cha nges in syste ms of empl oyme nt a nd remuneration of wage la bour - in agriculture induced by-t he adve nt of new tec hn ology is a seco nd impo rta nt theme in th e lit erature on J ava a nd has received so me attenti on in more general wo rks o n co nsequences of th e green revolution (Ruttan a nd Binswanger 1978, pp. 40 1 :.' ; Scott 1976, p. 2 11 ; Booth a nd Sundrum 1985, pp. 219- 20). According to severa l so urces, changes noted es pecia lly in harvesting operati ons have led to substantial loss of jobs as a co nsequence of weakening patron-client or co mmun al ti es.~ To quote one stud y frel.juently referred to in the literature dealing with worse ning employme nt op portun1t1es ow in g to the introduction of the tebasan co ntract harvesting system (farme rs selling their crop prior to harvest to middlemen or penehas): With these population increases a nd the rural communit v\ tradition a l efforts to provide for all its membe r~ , the village's econo my .ca n no longer provide adeq uate subsistence. The socia l res ponsibilit y of the fa rmers increases beca use of the greatl y increased num ber or la ndless people. However, the introduction of the new ric~: techno logy may induce the

CONSEQUENCES OF RURAL ECONOMIC CHANGE

7

farmers to think and act more commercially. Consequently, the reciprocal principles of the patron-client relationship begin to be felt as a heavy burden and the farmers want to avoid it. This desire to improve one's position in relation to the community welfare is in direct conflict with the traditional values of the village community. (Collier et al. 1974. p. 175)

It is noteworthy that one major factor cited as contributing to change mentioned in both this and the previous quotation is a more "commercial" attitude on the part of rice farmers. The argument advanced by Collier al}d others is that given a greater dependence on purchased inputs and marketed output. and increased demand for a wide range of consumer goods, farmers begin to give more emphasis to cost minimization (and profit maximization) and thereby are more prepared to search for ways to cut labour costs with little regard to the effects on employment. 9 Turning to the third subject, two issues have been highlighted in the literature on "segmented" labour markets in rural Java in the 1970s. First, the structure of employment for individual households and household members has been noted for its diversity described by White ( 1976a, pp. 280- 81) as "occupational multiplicity" and featuring a wide range in earnings between various activities. with the unusual pattern of wage rates in agriculture and earnings in home industry placed at the top and bottom respectively of the earnings "ladder" (see also Lluch and Mazumdar 1983). 10 Various combinations of farm work. agricultural wage labour, and non-agricultural jobs are chosen by individual households to meet minimum income targets with non-agricultural jobs. as in other countries. playing a major role in off-peak periods in agriculture. Second, the employment and earnings structure has been described as segmented between various classes: poorer households having limited access to high wage activities in agriculture and are forced into extremely low productivity occupations especially in the slack season (Hart 1986a. Chapters 5 and 6) . An important emphasis in the village studies has been employment in low income non-agricultural work within the village or in neighbouring rural locations rather than in urban areas, the latter receiving far more attention in studies of circular migration and commuting (Hugo 1978, 1982; Mantra 1981). It will be argued that there are important grounds for questioning the general argument on all the three issues discussed above. Despite rapid growth in rice production and major changes in cultivation techniques, various elements of the response to these developments have close links with changes that had already been observed in the Soekarno era and in the colonial period. This applies particularly to the case of "institutional.. change. much of which occurred in various forms prior to the green revolution and appears to have re-emerged in selected areas more in response to the changing balance of labour supply and demand than as a

L m ~uence ol the ad,en t ol ne" tec hnu..J UC'> of rice famlmg. It will abo be ~u~e.ted that too much em pha'i ' h a~ heen placed on th e new bra nd of

u mmc-rcw l thm lung amon~ I arme r~ . dow n-play ing th e import ance of t('(·hmCI I rca"' m lm~ed to adopw no t new rice vari e ti ~ and whi ch help to e pl:11n tntn du tinn of huth ne" tec hn ology and empl oyment a nd wage ) tern Cit\en the political en\' ironmer.t a nd int ern ational experi ence. the rde\ant 4UC'\ tton would a ppea r to he wh y have institutions cha nged so ' lowly and the di,tribut io n of inco me a nd wealth not deteri orated much m r . ra th er th an to ~ tre.~ the ~ peed a nd seve rit y of change. Fma ll ~ . whil the co nce pt of Ia hour mark et segmental ion i, extremel y va lu a ble lor und er~ Lan din g th e process of inco me det ermination in rural area' . the pt~ rt icubr 1nrm f ~eg-m e ntat ion identifi ed in the literal ure fail s to ta l-e account of th e mu ch grea ter di versit y of participation. es pecially a mong p or r rur:.s l h o u ~e h o ld~. in non-agricultural activities in Ja va. a tr nd wh i h accek ra ted during th e 1970s. In particular. more stress might havt het>n give n to parti ci pati on of runll hou seholds in higher earning a t i , · i u e~ in urhan areas. requiring a reassessment of this particular version ,f the IHhour -.egment ati on hypoth es is. This has important implications for an overa ll eva luati on of cha ngi ng pauerns of incomes and wealth in many Ientatt\e of all\ illages in Ja, a. the micro level data highlight the \ ery ~ma ll sl.le of mllSt o perational sa\\'ah holdings but still marked tneljualitv 10 the distributi o n Ll f land.u One important point o f relc\'ance to the later discussion of technolugll'al and ilbtitullorwl l' ha nge co ncerns the difference in land o perating patterns bct\\t."en lowland amJ upla nd illages shown in Table 3. The fl1rn1cr are ma rk ed b~ a relatively high proportion of larger operators (22 per ce m) Clmtrulhng over half of all :iGII'tJh in contrast to a very small pro po rtion of large o perators in the medium and upland villages; where the large majority l1f house ho lds worked los than 0.5 of a hectare of sawah

TABLE 3. DISTRIBUTION OF HOUSEHOLDS AND LAND AREA BY AREA OF LAND OPERATED, GINI COEFFICIENTS, AND LANDLESSNESS IN JAVA (1973, 1976, AND 1983)

..._ 0.

Land Area Operated (ha) Village Type/Land Area

o.s VILLAGE SAMPLE (1983) (Sawah land only) Lowland (3 villages) % Households %Land Medium (2 villages) %Households %Land Upland (3 villages) % Households %Land Villages All % Households %Land ALL JAVA Sawah Land, 1976b %Households All Land, 1973d %Farms %Land

• Unweighted village mean. From SAKERNAS 1976.

b

SoURCE:

O.S-1.0

1.0+

All Operators/ Fanns N %

Mean per Area Operator/ Farm (ha.)

Gini Coefficient"

%Landless (Non-operators)

-1

:c (TJ

,.,0 ,.,z (TJ (TJ

(TJ

54 20

24 24

22 56

100 100

340

0.(;7

0.48

42