1 Sachs, J., “Comment on Gelb, Jefferson and Singh” in Blanchard, O. J. and S., Fischer (eds.), NBER Macroeconomics Annual 1993 (Cambridge, MA: MIT Press, 1993), pp. 139–140.Google Scholar

2 Sachs, J. and Woo, W. T., “Structural factors in the economic reforms of China, Eastern Europe, and the Former Soviet Union,” Economic Policy, No. 18 (April 1994), pp. 114–15.Google Scholar

3 The Sichuan case is especially interesting because of its size, its contribution to national agricultural production – about 9% of agricultural value-added in 1978 – and its role as a pioneer of economic reform. For an account of Sichuan's development in the Maoist era, see Bramall, C., In Praise of Maoist Economic Planning (Oxford: Clarendon Press, 1993).Google Scholar

4 Net agricultural output value measured at 1990 constant prices increased at an annual rate of 0.7% between 1965 and 1976. Calculated by ordinary least squares (OLS) from data in Zhongguo guojia tongjiju (Chinese State Statistical Bureau – hereafter SSB), Quanguo gesheng zizhiqu zhixiashi lishi tongji ziliao huibian (A Collection of Historical Statistical Materials on China's Provinces, Autonomous Regions and Centrally-Administered Cities) (Beijing: Zhongguo tongji chubanshe, 1990). In Sichuan's case, the most visible indicator of failure was the displacement of rice by sweet potatoes in the rural diet. According to Donnithorne, ”Rice is very much the preferred staple in Sichuan… to have to substitute wheat, maize and even the despised sweet potato for rice, constituted a serious decline in perceived standards of living” (Donnithorne, A., “Sichuan's agriculture: depression and revival,” Australian Journal of Chinese Affairs, Vol. 12 (July 1984, p. 63).CrossRefGoogle Scholar

5 The term “reform” is used throughout this article to encompass institutional change, deregulation and readjustment. “Readjustment” is defined as a reallocation of investment between and within sectors of the economy; this follows the usage of the term adopted by most Chinese economists in the late 1970s.

6 Sachs, in commenting on an article by Gelb, Jefferson and Singh, notes with approval their view that “… the rural reform was quite radical, and had ‘somewhat of a big bang character’ ” (Sachs, “Comment,” p. 140).

7 Yuming, Sheng, Intersectoral Resource Flows and China's Economic Development (London: Macmillan, 1993).Google Scholar

8 Sachs and Woo (“Structural factors,” p. 115) argue that China's economic reforms as a whole “were gradual in several respects” and that “the impetus for reform came from above and below” in the countryside during the late 1970s. Nevertheless, rural reform is portrayed as “breathtakingly rapid,” and the dismantling of collective property in 1981 is seen as “almost spontaneous.”

9 For an introduction to the differences between these systems of agricultural production, see Riskin, Carl, China's Political Economy (Oxford: Oxford University Press, 1987). pp. 286–290.Google Scholar

10 The trend here is calculated by OLS between these two peak years from data in Sichuan tongjiju (Sichuan Statistical Bureau), Sichuan tongji nianjian 1990 (Sichuan Statistical Yearbook) (Beijing: Zhongguo tongji chubanshe, 1990), p. 139, hereafter abbreviated as SCTJNJ 1990. In Chongqing prefecture, grain output in 1976 was 8% below its 1974 level and 2% below that recorded in 1966. Chongqing shi jihua weiyuanhui (Planning Committee for Chongqing Municipality), Chongqing shijingji zonghe zhi (Summary Gazetteer on the Economy of Chongqing) (Chongqing: Chongqing chubanshe, 1991), p. 219. In Peng xian on the Chengdu plain, the rice yield was 16% below its 1975 level, which was itself a rather poor year. Peng xian zhi (Peng County Gazetteer) (Chengdu: Sichuan renmin chubanshe, 1989), p. 236.

11 Bramall, In Praise, p. 325.

12 Zhonggong Sichuan sheng wei yanjiushi (Research Unit of the Sichuan Committee of the Chinese Communist Party), Sichuan sheng qing (Conditions in Sichuan) (Chengdu: Sichuan renmin chubanshe, 1984), p. 571.

13 Bramall, In Praise, ch. 2.

14 Gan Shulong, Sichuan sheng nongye ziyuan yu quhua (Agricultural Regions and Natural Resources of Sichuan) (Chengdu: Sichuan sheng shehui kexueyuan chubanshe, 1986), p. 66.Google Scholar

15 Stone, Bruce, “Developments in agricultural technology,” The China Quarterly, No. 116 (December 1988), pp. 767–822.CrossRefGoogle Scholar

16 The provincial government seems to have been galvanized into action by a severe drought on the Chengdu plain during the spring of 1969. Dangdai Sichuan shuili (Irrigation in Modern Sichuan) (Chengdu: Sichuan renmin chubanshe, 1991), p. 67.

17 JPRS, Agricultural Geography, p. 43.

18 China Today: Sichuan, Vol. 1, pp. 173–74.

19 SSB, Historical Statistics, pp. 14 and 702.

20 In Shehong xian, the first part of the year was cold and wet, culminating in torrential downpours during the summer which left 33 people dead. Jingyan xian suffered a violent hailstorm in July, which wrecked 2.4 million jin of grain {Shehong xian zhi (Shehong County Gazetteer) (Chengdu: Sichuan daxue chubanshe, 1990), p. 20;Jingyan xian zhi (Jingyan County Gazetteer) (Chengdu: Sichuan renmin chubanshe, 1990) p. 39). More generally, the harvest was particularly poor in northern Sichuan.

21 For a critique of double-cropping, see Xiaotong, Fei, “Sichuan: calamity and revival,” China Reconstructs, Vol. 28, No. 1 (1979), pp. 59–63.Google Scholar

22 China Today: Sichuan, Vol. 1, pp. 191–92. Even though the plastics needed to reduce frost damage became increasingly available in the late 1970s, double-cropping was still not a viable option because the introduction of hybrid rice extended the period needed between planting and harvesting even further.

23 Calculated from SSB, Historical Statistics, p. 694.

24 China Today: Sichuan, Vol. 2, pp. 228–29.

25 David, Shambaugh (ed.), “Zhao Ziyang’ s ‘Sichuan experience’: blueprint for a nation,” Chinese Law and Government, Vol. 15, No. 1 (Spring 1982), pp. 3–126.Google Scholar

26 Wendou, Xie, “Dui Sichuan nongye fazhan zhanlue yi xie kanfa” (“Some thoughts on Sichuan's agricultural development strategy”), in Sichuan, Jihua Jingji Xuehui (Sichuan Institute for Economic Planning) (ed.), Jihua jingji de lilun yu shijian (The Theory and Practice of Economic Planning) (Chongqing: Chongqing chubanshe, 1984), p. 209.Google Scholar The process of deregulation was nevertheless extremely gradual. For example, the restrictions on the number of fowl that each household could raise were not removed until January 1979 (Domes, J., “New policies in the communes,” Journal of Asian Studies, Vol. XLI, No. 2 (February 1982), p. 260).Google Scholar

27 China Today: Sichuan, Vol. 1, p. 192.

28 In Guanghan, xian, double rice-cropping accounted for 78% of rice area in 1976. By 1983, the figure was down to barely 2%. Guanghan xian zhi (Guanghan County Gazetteer) (Chengdu: Sichuan renmin chubanshe, 1992), p. 112.Google Scholar

29 SCTJNJ 1990, p. 130.

30 Chinese sources offer conflicting accounts of the area allocated to private plots in the province. See Bramall, In Praise, p. 269.

31 China Today: Sichuan, Vol. 1, p. 446.

32 Subsistence was defined here as an annual distributed collective income of 50 yuan and a grain ration of 400 jin; see Xie Wendou, “Agricultural strategy,” p. 205.

33 China Today: Sichuan, Vol. 1, pp. 193 and 445.

34 Wushan xian zhi (Wushan County Gazetteer) (Chengdu: Sichuan renmin chubanshe, 1991), p. 251.

35 Liangshan Yizu zizhizhou gaikuang (A Survey of Liangshan Yi Minority Autonomous Prefecture) (Chengdu: Sichuan minzu chubanshe, 1985), pp. 268–69.

36 SCTJNJ 1990, p. 361.

37 Shehong Gazetteer, p. 389;Xinlong xian zhi (Xinlong County Gazetteer) (Chengdu: Sichuan renmin chubanshe, 1992), p. 173. This policy was a continuation of those that had been pursued ever since the early 1960s. In 1962, net grain procurements in Wushan accounted for 10.6% of grain production (as procurements were measured in units of processed grain – unlike production – this percentage understates the true burden). By 1974. the figure was down to 3.3% and net procurements were negative in 1971 and 1975 (Wushan Gazetteer, p. 252).

38 In Anhui, some 1,200 teams had already adopted the baochan dao hu system by the end of 1978; see Xuanchuanju, Zhonggong Zhongyang Xuanchuanbu (Propaganda Department of the Central Committee of the CCP), Jidang Zhongguo nongcun de biange (The Transformation Surging Across Rural China) (Beijing: Guangming ribao chubanshe, 1988), p. 224.Google Scholar On the question of whether the impetus for baogan came from the farm population or central government, see Watson, A., “Agriculture looks for ‘shoes that fit’: the production responsibility system and its implications,” World Development, Vol. 11, No. 8 (August 1983)CrossRefGoogle Scholar, Unger, J., “The decollectivization of the Chinese countryside: a survey of twenty-eight villages,” Pacific Affairs, No. 58 (Winter 1985–86),CrossRefGoogle ScholarKelliher, D., Peasant Power in China (New Haven: Yale University Press, 1992) andGoogle ScholarZweig, D., Agrarian Radicalism in China, 1968–1981 (Cambridge, MA: Harvard University Press, 1989).CrossRefGoogle Scholar This evidence suggests that it was led from below in some counties but imposed from above in others. Much the same seems true of Sichuan (see below).

39 The case for baochan as an alternative to both collective farms and baogan is made in Minquan, Liu, “Commune, responsibility system and China's agriculture,” in Fan, Qimiao and Peter, Nolan (eds.), China's Economic Reforms (London: Macmillan. 1994).Google Scholar

40 China Today: Sichuan, Vol. 2, p. 238. The precise dating of the beginning of baochan dao zu in Sichuan remains unclear. According to China Today: Sichuan, Jinyu was the first commune to adopt the system in January 1978. However, an alternative source (Zhonggong Sichuan sheng wei yanjiushi, Conditions in Sichuan, p. 210) gives the date as the autumn of 1978. A mild source (Xinhan, Cui, Guozhong, Fan, Zhankui, Shi and Zuyuan, Xiong, Sichuan chengshi jingji (The Urban Economy of Sichuan) (Chengdu: Sichuan kexue jishu chubanshe, 1985), p. 412Google Scholar, reports that the small group was used by some teams in Guanghan's Xigao commune in the summer of 1977, and that this approach was adopted experimentally by Jinyu commune in the autumn.

41 For the debate in Cangxi county, see China Today: Sichuan, Vol. 2, pp. 239–240.

42 Wushan Gazetteer, p. 111.

43 Liangshan gaikuang, p. 166. Note that Sichuan still lagged far behind Anhui, where 71.2% of teams had adopted baogan by the end of 1981. See Jiarui, Yang Xun And Liu, Zhongguo nongcun gaige de daolu (Rural China's Reform Path) (Beijing: Beijing daxue chubanshe, 1986), p. 121.Google Scholar

44 Tongjiju, Zhongguo Guojia (SSB), Zhongguo fenxian nongcun jingji tongji gaiyao, 1980–1987 (Outline Rural Economic Statistics for China's Counties) (Beijing: Zhongguo tongji chubanshe, 1989).Google Scholar

45 Oi, Jean C., State and Peasant in Contemporary China (Berkeley: University of California Press, 1989).Google Scholar

46 China Today: Sichuan, Vol. 1, p. 199.

47 Shehong Gazetteer, p. 22.

48 Xinjin xian zhi (Xinjin County Gazetteer) (Chengdu: Sichuan renmin chubanshe, 1989), p. 110.

49 Endicott, Stephen, Red Earth (London: I. B. Tauris, 1988), pp. 136–37Google Scholar; Zhongxin, Chengdu Shi Renmin Zhengfu Jingji Yanjiu (Economic Research Centre for the People's Government of Chengdu City), Chengdu jingji (The Economy of Chengdu) (Chengdu: Xinan zhengjing daxue chubanshe, 1987), p. 75.Google Scholar

50 Jingyan Gazetteer, p. 157.

51 Qianwei xian zhi (Qianwei County Gazetteer) (Chengdu: Sichuan renmin chubanshe, 1991), p. 143.

52 Putterman, Louis, Continuity and Change in China's Rural Development (Oxford: Oxford University Press, 1993)CrossRefGoogle Scholar, ch. 7.

53 For these difficulties in the context of a Guangdong village, see Potter, S. H. and Potter, J. M., China's Peasants (Cambridge: Cambridge University Press, 1990) pp. 166–172.CrossRefGoogle Scholar

54 Agricultural and industrial output data have been separately adjusted to 1980 prices, and then re-aggregated.

55 It can be argued that, by promoting (near) grain self-sufficiency in the late 1980s and early 1990s, the leadership of the CCP is ignoring China's comparative advantage. There is undoubtedly something in this in the sense that per capita arable land is in short supply, and could perhaps be better used to produce high value-added cash crops. But the proposition is by no means self-evidently true. The usual “small country” assumption – that a country can import unlimited quantities of a commodity without affecting its terms of trade – is manifestly not applicable in China's case. If China were to meet the bulk of its grain requirements by imports, the world grain price would probably rise dramatically and this in turn would require an enormous expansion in Chinese exports. The dangers in such a strategy are considerable. On the one hand, Chinese living standards would be sensitive to crop failure in grain-exporting countries such as Canada and the United States. China would also be extremely vulnerable to political pressure. On the other hand, it is very far from obvious that OECD countries would countenance manufactured imports on the scale required to pay for China's grain imports. The East Asian countries have already run into these sorts of problems, and they are all smaller than China. Ultimately, much depends on the ability of OECD countries to convince China that it has little to fear. As Anderson points out, “The strategy of greater feedgrain and fibre import-dependence will be feasible in terms of its foreign exchange cost so long as the rest of the world is prepared to do business with China. The more signals other countries can send China to assure it that relying on grain and fibre imports would not leave the country vulnerable to political pressures from outside, the less wary China will be…” (Anderson, K., “China's industrialisation and fibre self-sufficiency,” in C., Findlay (ed.), Challenges of Economic Reform and Industrial Growth: China's Wool War (Sydney: Allen and Unwin, 1991) pp. 95–96).Google Scholar

56 I am especially grateful to an anonymous referee of The China Quarterly for reminding me of the need to avoid inferring too much from trends in grain yields.

57 Xinlong county provides a classic example of the dangers inherent in concentrating on grain yields to the exclusion of all else. Conditions in that county – located high on the Himalayan plateau in western Sichuan – are far better suited to animal husbandry and forestry than to grain production. It is therefore not surprising to find no increase in the yields of highland barley and wheat (the two main grain crops) in the county in the late 1970s and early 1980s. By contrast, however, there was a massive increase in GVAO as animal husbandry increased dramatically. One cannot put this down to baogan; collectivization was only completed in this area (the same is true of most of western Sichuan) in the mid-1970s and the process of decollectivization did not get under way until 1982–83. But it does illustrate the way in which grain yields are not always a good measure of agricultural performance.

58 The importance of irrigation is underlined by the experience of the arid province of Gansu, where grain yields remained below their 1975 level until 1986 despite the introduction of baogan. Bu, Nongye (Ministry of Agriculture), Zhongguo nongcun jingji tongji daquan (Complete Statistics on the Economy of Rural Economy) (Beijing: Nongye chubanshe, 1989), pp. 184–86.Google Scholar

59 SCTJNJ 1990, p. 147 and SCTJNJ 1993, p. 243.

60 BBC, Summary of World Broadcasts, FE/W1092/A/5, 23 July 1980.

61 Neijiang shi zhi (Neijiang City Gazetteer) (Chengdu: Bashu shushe, 1987), p. 69; Neijiang shi guotu ziyuan (Natural Resources of Neijiang Municipality) (Chengdu: Sichuan kexue jishu chubanshe, 1987), pp. 297–98. Even the Chengdu plain, protected as it was by the Dujiangyan, was not immune; there were about 200 deaths there as a result of flooding (BBC, Summary of World Broadcasts, FE/W1157/A/3,28 October 1981 and FE/W1148/A/5, 26 August 1981;China Today: Sichuan, Vol. 1, pp. 208–209).

62 Yanjiuzu, Guojia Kewei Quanguo Zhongda Ziran Zaihai Zonghe(Research Group for the National Scientific Committee on Major Natural Disasters), Zhongguo zhongda ziran zaihai jijianzai duice (Major Chinese Natural Disasters and Strategies to Reduce Them) (Beijing: Kexue chubanshe, 1993), p. 263.Google Scholar

63 There was, for example, a severe drought in the central Sichuan basin in 1977, and Santai was one of the xian affected (Irrigation in Modern Sichuan, p. 13). The 1977 grain yield was therefore depressed, with the result that post-1977 yield growth in Santai may be exaggerated compared with counties where 1977 was a more normal year.

64 Though having said that, there is no evidence to suggest that Sichuan farmers declined the opportunity to leave their collectives when given the opportunity. Whatever may be said of their economic merits, collectives were widely seen as fundamentally coercive institutions.