Rural factor markets in China after the household responsibility system reform

DISCUSSION PAPER Report No7: ARU 61 Rural Factor Markets in China After the Household Responsibility System Reform by Justin Yifu Lin Research Unit Agriculture and Rural Development Department Operational Policy Staff World Bank November 1986 The vLews presented here are those of the author(s), and they should not be interpreted as reflecting those of the World Bank,  The author is a consultant of the World Bank. However, the World Bank does not accept responsibility for the views expressed herein which are those of the authors and should not be attributed to the World Bank or to its -affiliated organizations. The findings, interpreta- tions, and conclusions are the results of research supported -in part by the Bank; they do not necessarily represent official policy of the Bank. The designations employed and the presentation of material in this, document are solely for the convenience of the reader and do not imply the expression of any opinion whatsoever on the part of the World Bank or its affiliates concerning the legal status of any country, territory, area or of its authorities, or concerning the delimitation of its boundaries, or national affiliation.  RURAL FACTOR MARKETS IN CHINA AFTER THE HOUSEHOLD RESPONSIBILITY SYSTEM REFORM Justin Yifu Lin Research Unit Agriculture and Rural Development Depart:- nt, the World Bank November, 1986  I. INTRODUCTION This paper is a preliminary study of rural factor markets in China. Transactions in factor markets were severely constrained by government policies in the past. Recent reforms in rural areas, however, have brought exchanges in factors to life again. Traditionally, Chinese peasants, like their contemporaries in other Asian countries, were not unfamiliar with factor markets. In fact, a whole spectrum of market exchanges in land, labor, and credit existed in complex forms in rural China before the socialist revolution. ' Nevertheless, the cooperative movement, starting in the 19509, collectivi4zed land, labor, and other resources and made market exchanges in land and labor between households impossible. Private credit exchanges were also severely limited. Rent and interest were taken as means of capitalist exploitation and the labor force was not to be treated as a commodity that could be bought or sold; therefore, market exchanges between two collective farming units were also prohibited. Under the collective system, a production team, usually consisting of about 30 neighboring households, was gonerally the basic unit of production .and income distribution. The team was entitled to all factors of production.' These factors were allocated under the unified management of a team leader with the exception of small private plots reserved for households use in their spare time. Peasants, working under the supervision o:i a 'See Wiens (1982) for detailed discussions.  2 team leader, were credited with work points for a day's work that they had done. At the end of a year, net team income was first distributed among team members according to basic needs, then the rest was distributed according to the work points that each one had accumulated during the year. This institution was found to be very inadequate in providing work incentives to peasants in a production team. a A new policy called the production responsibility system was introduced at the end of 1978 as one element of a package of reforms aiming at improving agricultural production in rural areas.' At first, this policy was designed to improve the management and incentive problems within, a team. However, it developed into a specific form now called. "the household responsibility system" that dissolved the production teams and restored individual households as units of agricultural production and income distribution. The household responsibility system evolved into the main feature of the recent reforms in the Chinese rural areas. Lin (1986) finds that the shL.ft from a production team system to the household responsibility system on the average increased the agricultural productivity 14%. This .jump in productivity explained about half of the growth in 2For a detailed discussion of the incentive problem in a production team, see Lin (1986). 3The other reforms included diversification of the rural economy, production specialization, crop selection in accordance with regional comparative advantages, expansion of local fairs, marked rises in state procurement prices, and rapid growth in the availability and better allocation of chemical fertilizer.  3 agricultural production between 1980 and 1983. The improvement in incentive, nevertheless, may have simultaneously created allocative inefficiencies., When the household .responsibility system was introduced, land and other resources in a team were in most cases allotted to each household in proportion to its size. Therefore, for the households in a team, their land-person ratio was equalized after the household responsibility reform. Households are at different stages in the life cycle. They thus have different endowments of family labor. In addition, households differ in abilities. An equal land-person ratio across households in a team thus does not fully equalize land-labor ratio across households. If each household faces the same production function, this egalitarian allocation of land will' result in disparities in the marginal products of land and labor across households.' These differences in marginal products represent an allocative inefficiency. Output can be increased if resources are reallocated. One possible way to take advantage of these opportunities is through direct government intervention, like land-reallocation among households. Nevertheless, government intervention can be 4 In addition to the disparity in marginal products created by the household responsibility system, the allocative inefficiency has another source in China. Like any other country, the endowment of land and other resources varies greatly from region to region. Since migration between rural areas failed to exist at any significant level in the past, the difference in land-person ratio as well as land-labor ratio has long been maintained. Therefore, marginal products in land and labor should also be different across regions. See the discussion in the next section.  4 ruled out as an alternative for the near future. When the household responsibility system was first introduced, the land contracts in general ranged from 1 to 3 yearrs. When an original contract expired, land was reassigned and adjusted according to changes.: in household size and labor endowment. This practice was soon found to be impractical. As land might be assigned away in next contract, each household thus lacked incentives to invest in land improvement and to maintain properly the soil fertility. To overcome this disincentive in land investment and land maintenance, the Chinese government has adopted a policy of lengthening the contract of land usage to each household for up to 15 years or longer. The other possibility for improving allocative efficiency is through market transactions. Market transactions can range from hired labor to land tenancy or may be packaged in complex contracts involving several transactions in different markets. Transactions in land and labor naturally will give rise to demand for credit. If factor transactions are costless, certain, unconstrained, and enforceable, then marginal products will be brought into equality by market transactions. However, as discussed by Binswanger and Rosenzweig (1986), factor transactions in rural areas are characterized by risk and beset with incentive problems. The existence of well developed rural factor markets cannot, therefore, be taken for granted. This paper is devoted to examining the extent and possible developments in rural factor markets in China.  5 Before going into any detailed discussions, three specific features that characterize China's rural factor markets need to be mentioned: a) The rural reform in China has gone through a first stage, which featured the individual household responsibility system. By the end of 1983, 94.5% of rural households in China had adopted this new system (Editorial Board of China Agriculture Yearbook 1984, pp. 66-67). The Chinese government launched a second-stage reform in 1984. The main theme of the second-stage reform is to transform a self-subsistence economy into a commodity production and exchange economy by way of readjusting the production structure in rural areas through market mechanisms. When the household responsibility system was first introduced, hiring labor,' subleasing land, and lending money at high interest rates were all explicitly prohibited.$ Since then there have been substantial changes. The first change came to the credit market. Private credit with a high interest rate is no longer categorically classified as usury in the 79th document issued by the State Council in 1981. Leasing out land to other farmers and. hiring workers within a limited number (less than eight) were also formally sanctioned in Document No. 1, issued by the Central Committee of the Communist Party of China in 1984. Transactions in -actor markets have been legalized. However, socialist sentiment is still deeply rooted in China. It appears unlikely, 5See "Several Issues Concerning the Further Strengthening and Perfecting of the Agriculture Production Responsibility System," in China Agriculture Yearbook. 1981 pp.409-411.  for example, that the government will force e person to be evicted from his house if he uses it aa collateral and fails to repay his loan. It is also unimaginable that public opinion will sympathize with the lender in the case of a default. b) The cultivated land per capita in 1949 was 2.7 mu and now it has shrunk to about 1. 6 mu. However, the percentage of labor force which remains in rural areas has been about the same during this period. Under the current price system and the average operational landholding, the value of marginal product of labor in agriculture, especially in cropping, is much lower than that in non-agricultural sectors; therefore., there is a general tendency. for the rural labor force to engage in nonf arm activities. A study shows that the average net income per worker in the suburbs of Shanghai in 1981 was Y 441 for agriculture, Y 1,003 for sideline production, and Y 1,625 for industry (Shi 1981). The differences in incomes across sectors should also be similar in other regions. This income differential will thus induce a tendency for the labor force to move out of the agricultural sector. c) There is no landless class in China's rural areas -today. The original production teams are still entitled to the ownership of land after implementing the household responsibility system. Each household is assigned the use right of a piece of land for a period of 15 years or more. This practice provides each household with a safe shelter. Unless a household is assured a nonfarm or noncrop job that produces an expected income that is greater than  7 working on its farm, a household will not leave its land. This practice also has impacts on the labor market. Farm households will only supply labor to other farms when the opportunity cost of not working on their own farms is low, i. e., the labor market will be constrained from the supply side, especially during peak periods. This paper is organized as follows: land markets, labor markets, and credit markets will be discussed in turn in the following sections. At the end of each section, some stylized facts of the concerned markets will be summarized and their related hypotheses will be presented as well. The main hypothesis of this paper is presented at the last section. II. LAND ALLOCATION AND LAND MARKET Marginal Products of Land. The differences in the marginal products of land and labor in China have two major sources. One is the differences in land endowment across regions. The other one is the egalitarian distribution of land after the household responsibility reform. Table IA shows that the eight provinces that have the lowest land-labor ratio possessed 39.6% of the total labor force in China in 1983; however, they only had 21.4% of the total cultivated land. On the contrary, the 9 land-rich provinces possessed only 10. 5% of the total labor force but were endowed 34. 1% of the total cultivated land. The peasants in land-rich Helongjiang Province on the average had about 17 times as much  TABLE lA LAND ENDOWM'ENT IN EACH PROVINCE (1) (2) (3) (4) (5)(6 Province Labor Cult. Land- % of Area Multiple Effective Land Labor Irrigated Cropping Land-Labor Ratio Index Ratio Guizhou 10,087 28,450 2.8 24 153 3.8 Sichuan 38,712 98,1093 2.5 47 181 3.9 Zhe.ijiang 14,030 27,249 1.9 84 252, 4.0 Yunnian 12,808 42,488 3.3 34 140 4.3 Guangdfong 19,061 47,130 2.5 65 200 4.4 G3uangxi 13,963 39,301 2.8 54 177 4.4 Fujia.ng 7,083 19,240 2.7 64 189 4.5 Tibet 825 3,437 4.2 53 93 4.6 Hunan 20,239 50,998 2.5 82 218 4.8 Henan 25,370 106,508 4.2 45 160 4.9 Shanghai 2,071 5,249 2.5 98 218 4.9 Anhui 17,478 66,518 3.8 50 177 5.9 Jiangsu 20,068 69,451 3.5 75 184 5.9 Shandong 24,988 107,728 4.3 63 146 6.1 Hubei 14,572 55,481 3.8 63 200 6.6 Tianjing 1,391 6,879 4.9 67 133 6.7 Beijing 1,372 6,343 4.6 81 151 6.9 Hebei 17,481 98,551 5.6 54 131 -7.3. Jiangxi 9,305 35,753 3.8 74 229 7.4 Shaanxi 9,124 56,377 6.2 33 127 7.6 Oinhai 1,104 8,640 7.8 27 87 7.7 Liaoning 6,4EJ8 54,814 8.4 19 102 8.9 Gansu 5,841 53,425 9.1 24 98 9.5 Shanxi 6,647 58,076 8.7 28 107 9.6 Ningxia 1,061 12,605 11.9 28 101 12.8- I.Mongolia 4,657 75,974 16.3 20 91 16.3 Jinin 3,845 60,895 15.6 18 100 16.3 Xingjiang 2,570 47,422 18.5 83 932 21.4 Hei-longjiang4,110 131,273 31.9 7 98 32.1 SOURCE: China AgTriculture Yearbook 1984. NOTE: (1) Agricultural labor force excluding workers in village- run industry, unit = 1,000 workers; (2) cultivated land unit 1,000 mu; (3) col.2/col.1; (4) X of cultivated land irrigated; (5) unit= %; (6) effective land-labor ratio is the land-labor ratio adjusted for irrigation and multiple cropping; its formula is: effective land-labor ratio= Land-labor ratio x (1 + % of area irrigated/4) x (1 s+( Multiple cropping index- 100)/2). See A. M. Tang, An Analytical and Empirical Inivestigation of Agriculture in Mainland China 1952-1980 (Seattle: UJniversity of Washington Press, 1984) for 'the rationale of for these adjustments.  TABLE 1B LAND ENDOWMENT IN EACH PREFECTURE, ANHUl PROVINCE (1) (2) (3) (4) (5) (6) Prefecture Labor Cult. Land- % of Area Multiple Effective Land Labor Irrigated Cropping Land-Labor Ratio Index Ratio Anqing 85 126 1.5 a8 215 2.9 Huizho 705 1,192 1.7 76 248 3.5 Anqing 2,094 4,859 2.3 66 211 4.2 Tongling 124 297 2.4 95 200 4.5 Wuhu 697 1,736 2.5 77 224 4.7 Maanshan 297 756 2.5 93 225 5.0 Chaohu 1,527 4,355 2.9 80 199 5.2 Liuan 2,004 6,706 3.3 68 171 5.2 Fuyang 4,263 16,915 3.9 24 166 5.5 Xuancheng 764 2,246 2.9 65 222 5.6 Hefei 1,172 4,183 3.6 72 172 5.7 Huainan 390 1,522 3.9 57 174 6.1 Suxian 1,669. 7,913 4.'7 29 158 6.5 Huaibei 416 2,208 5.3 11 160 7.1 Chuxian 1,297 6,250 4.8 68 165 7.4 Bangbu 844 4,542 5.4 47 162 7.9 SOURCE: Statistical Bureau, Anhui Province. NOTE: Definitions and units are the same as in Table 1A, except labor force here includes workers in village-run industry. Data are for 1983.  10 land as the peasants in land-poor Zheicjiang Province. Not only is the distribution of cultivated land unequal among provinces, but it is also unequal within a province. Table 1B shows that, in Anhui Province in 1983, ~31.8% of cultivated land located in nine prefectures that had 22. 9% of labor force. In contrast, the six prefectures that had the lowest land-labor ratio had 21.7% of the labor force but only 13.6% of the cultivated land. Although the differences within Anhui Province are not as large as the differences among provinces, the disparities are still quite substantial. The peasants in Huaibeishi have 3.6 time as much land as the peasants in Anqingshi. The differences in land-labor ratio reduce after adjusting for irrigation (proxy for land quality) and multiple cropping (proxy for climate and temperature). However, the differences are still very substantial as the last columns of tables IA and 1B suggest. Although, without empirical studies, it is difficult to say to what degree land endowments differ across neighboring production teams, the difference itself can be taken as a fact. The distribution of inherited intelligence of a large population approaches normal in any large sample. There is no a priori reason to believe that the average quality of the labor forces in two neighboring teams, which both have about 100 workers, would be significantly different.' It thus should not be too unrealistic to assume that 6Acquired ability through education will be different depending on the availability and quality of education.- In this respect, the labor force in urban areas will have a higher quality than in rural areas because, in general, the education in urban areas is better and. more accessible. However, within a  the quality of labor forces across teams and regions is the same. Consequently, much of the dilferences in the land-labor ratio represent an allocative inefficiency. Allocative inefficiency within a team, however, would arise from an opposite reason. Under the production team system, the team-owned land was divided into collectively farmed plots and private plots. Private plots were allotted to each household according to its size. The land that could be allotted for private plots varied from time to time. The average amount of land in private plots nationally was 5.7% in 1978. It rose to 7.1% in 1980 (Perkins and Yusuf 1984). After the introduction of the household responsibility system, the collectively farmed lan.d was contracted to individual households in two different categories. One was the "food ration plot. " The other one was the "responsibility plot." The difference between these two kinds of plots was that a household had to pay only state tax on the food ration plot, but it also had to pay the public accumulation fund, public welfare fund, and other duties to its team on the responsibility plots. As for the private plot, the state tax was also waived. Two different practices were used to contract the collectively owned land. The first practice contracted the land *strictly in proportion to the size of each household. The second one took into account both the size and the labor force of each household. However, the results of these two practices may not be rural area, the quality and availability of education faced by each team can be assumed to be roughly identical.  12 very different. A survey of a production team in Guangxi Province found that the household with the largest labor force only had 0.16 mu per capita more than the average of the team, and the household with the smallest labor force had only 0.078 mu per capita less than the average of the team, even though 70% of weight was given to the labor force in the contracts (Hu et al. 1984). Therefore, it can be assumed that the land-person ratio across households in a team is roughly equal no matter what practice has actually been adopted. Not only is the quantity of land per capita equal across households, but the quality of land owned by each person in a team is also the same. This is because land was first graded according to its quality, then each person received a piece of land from each grade. Therefore, each household in China after the individual household reform often owns more than 10 strips of land,7 Households in a team are at different stages of their life cycles and thus have different labor endowments. They also have different level of education, experience, and other abilities. As a consequence, the equal land-person ratio across households in a team generates a potential allocative inefficiency. A survey of 235 households in a village in Sichuan Province found that 25% of households with a rich labor endowment did not have enough land to farm; 6% of households did not have enough labor to work on their land; and 'The government has encouraged households in a team to exchange their land in order to reduce the fragmentation of land holding. However, a household still owns more than five pieces of land, in general (Editorial Boaxd of China Agriculture Yearbook 1985, p.285).  13 4.7% of households were good at other trades, so they did not want to work on their land (Research Unit of Nan Chong Prefectural Government, Sichuan Province, 1985). Land Transactions. Land transactions in China's rural areas after the household responsibility reform are restricted in their form of lease. The government has encouraged the households specialized in farming to consolidate their landholding. Table 2 summarizes several studies concerning the extent of land transactions in certain regions in China. Rows 1-3 are based on the surveys done at the end of 1983 and rows 4-5 are based on the data collected at the end of 1984. Column 2 is the percentage of households in an area that either leased out their land to other households or returned their land to their production teams. The land returned to a production team may be recontracted to other households. Column 3 shows the percentage of land in an area that was involved in land transactions. The percentage of land involved is le.>s thar. the percentage of households involved. This is due to the fact that most households only leased out or returned their responsibility plots and kept their food ration plots and private plots. All these studies found that land transactions were more active in areas closer to cities. Tianjin is the third largest city in China. But even by the end of 1984, only 8.3% of land