1. See Oi, Jean C., Rural China Takes Off: Institutional Foundations of Economic Reform (Berkeley: University of California Press, 1999).Google Scholar

2. There were elections during the Maoist period but they were not competitive and therefore failed to have legitimacy. See Burns, John, “The election of production team cadres in rural China, 1958–74,” The China Quarterly, No. 74 (06 1978), pp. 273–296.Google Scholar

3. Party secretaries may be popularly elected by Party members within the village.

4. There are also other bases of power, for example, those derived from family and kinship hierarchies such as clan ties.

5. Rozelle, Scott, “Stagnation without equity: patterns of income and inequality in China's rural economy,” The China Journal, No. 35 (01 1996), pp. 63–92.Google Scholar

6. These two organizations are the successor to structures found in earlier periods of communist rule. The villagers' committee is the successor to the management committee of the agricultural producers' co-operative and to the management committee of the production brigade; the villagers' assembly is the successor to the general assembly of the agricultural producers' co-operative and the representative assembly of the commune members.

7. Exceptions include Thurston, Anne F., Muddling toward Democracy: Political Change in Grassroots China, Peaceworks No. 23 (Washington, D.C.: United States Institute of Peace, 1998)Google Scholar; and Lawrence, Susan, “Democracy, Chinese style,” Australian Journal of Chinese Affairs, No. 32 (07 1994), pp. 61–68.Google Scholar

8. The Report on the Villagers Representative Assemblies in China (Beijing Research Group on the System of Village Self-Government, 12 1994), p. 1.Google Scholar

9. “There could not have been a villagers' committee, and hence true village self-government, without the villagers' assembly.” Ibid. p. 5.

10. See Peng Zhen's speech at the 20th session of the Standing Committee of the Sixth NPC in March 1987, quoted in ibid. p. 4.

11. Ibid. p. 4.

12. Ibid. p. 116.

13. According to MoCA, the average villagers' assembly would be attended by 600 to 1,800 people, all members of the village 18 years and older. Ibid. p. 2.

14. Ibid. p. 114.

15. Regulations stipulate that representative assemblies are the only body with the power to elect and remove members of the villagers' committee and work out of the village conventions and village pledge. Ibid. p. 115.

16. Zhenyao, Wang, “Village committees: the basis for China's democratization,” in Vermeer, Eduard B., Pieke, Frank N. and Chong, Woei Lien (eds.), Cooperative and Collective in China's Rural Development: Between State and Private Interests (Armonk, NY: M. E. Sharpe, 1998), p. 253.Google Scholar

17. For example, Shandong requires no less than 30, as does Liaoning. Fujian states that it should be no less than 35 in villages with more than 1,000 households and no less than 25 in villages with less than 1,000. The Report on the Villagers Representative Assemblies, p. 100.Google Scholar

18. For example, some places in Hebei state that the number of members should be generally 2–3% of the villagers aged 18 and above, in others there should be 30–60 members. While in Shanxi, there should be 20–60 members, but in villages with less than 400 no representative assembly need be established, instead there should be a full villagers' assembly.

19. The Report on the Villagers Representative Assemblies, p. 100.Google Scholar

20. Oi observed villagers' representative meetings in Hunan and Henan in the summer of 1994. The ministry requires that villages keep detailed notes of the issues discussed and the decisions made during these meetings.

21. China Interview (hereafter CI) 82494.

22. Some assemblies also allocate seats for retired village heads.

23. According to some accounts, these do not count in the total number of village representatives – but it is unclear what this actually means.

24. In Oi's fieldwork, some villages said that the village Party secretary should not be there unless he is an elected villagers' representative. CI 81694.

25. The deputies to the People's Congress and members of the CPPCC within a village usually only number 2–3. The Report on the Villagers Representative Assemblies, p. 103.Google Scholar

26. Ibid. p. 79.

27. Ibid. pp. 80–85.

28. CI 82494.

29. The Report on the Villagers Representative Assemblies, p. 49.Google Scholar

30. In October 1983, the Central Committee of the CCP issued the “Circular on Separating Government from People's Communes and the Establishment of Township Governments,” which stipulated detailed requirements for the establishment of these committees.

31. Study on the Election of Villagers Committees in Rural China (Beijing: Ministry of Civil Affairs, 1993), p. 1.Google Scholar

32. Zhonghua renmin gongheguo cunmin weiyuanhui youguan fagui, wenjian ji guizhang zhidu huibian (Laws, Documents and Regulations of the PRC Concerning Villagers' Committees) (Beijing: Minzhengbu, 1995), p. 40.Google Scholar

33. See Oi, Jean, State and Peasant in Contemporary China: The Political Economy of Village Government (Berkeley: University of California Press, 1989).Google Scholar

34. There usually were insufficient numbers of Party members in a team to form a Party branch.

35. For an elaboration and documentation of this system see Oi, State and Peasant.

36. For a discussion of team leaders as gatekeepers of their team members economic as well as political well-being see Oi, State and Peasant.

37. Rozelle, Scott and Boisvert, R. N., “Quantifying Chinese village leaders' multiple objectives,” Journal of Comparative Economics, Vol. 18, No. 1 (02 1994), pp. 25–45CrossRefGoogle Scholar; Rozelle, Scott and Guo, Li, “Village leaders and land-rights formation in China,” American Economic Review (05 1998), pp. 433–38.Google Scholar

38. Rozelle, Scott, “The economics of village leaders in Reform China,” Ph.D. dissertation, Cornell University, 1991.Google Scholar

39. Rozelle, and Li, Guo, “Village leaders and land-rights formation.”Google Scholar

40. The disbursement of such fees is often already earmarked, providing leaders with little, if any, discretionary control.

41. While all would have liked to shift into non-agricultural pursuits to generate revenue, physical, locational and human capital constraints have precluded all villages from benefiting equally from the reform opportunities. See Oi, , Rural China Takes OffGoogle Scholar for details on how village enterprises developed.

42. In some villages, firms contribute fees and informal taxes to the village fiscal accounts. Most villages keep them separate; these revenues are used for many of the same pro-community purposes, such as investment in the community infrastructure and social services. Xiao-yuan, Dong, “Public investment, social services, and productivity of Chinese household farms: a stochastic frontier analysis,” paper presented at the Allied Social Science Associations Meeting, Chicago, 3–5 01 1998.Google Scholar

43. Although we treat villages as though they fit neatly into these four cells, in reality most fall somewhere in between.

44. There is another important linkage between the structure of income in a village and the dynamics of local politics, namely the extent to which local political institutions are able to influence the distribution and use of the resources that generate different types of income (henceforth the “degree of political insulation”). This is discussed later in this section. The additional explanatory factor is assigned a secondary role in this article, not because it is any less important but mainly because our data set precludes us from offering any concrete empirical evidence on its importance. By omitting the degree of political insulation from our empirical work we are taking the risk of confounding our results, and as a result caution should be used in the interpretations that we derive from the data.

45. Oi, , Rural China Takes Off.Google Scholar

46. Rozelle, , “Decision-making in China's rural economy: defining a framework for understanding the bahaviour of village leaders and farm households,” The China Quarterly, No. 137 (03 1994), pp. 99–124.Google Scholar

47. Brandt, Loren, Li, Guo and Rozelle, Scott, “Land in China: a discussion of fact, fiction, and the issues,” Working Paper, Department of Agricultural and Resource Economics, University of California, Davis, 1999.Google Scholar

48. The reasons why villagers allow this institutional arrangement to emerge, of course, also need to be considered, but further in-depth examination of this institution is beyond the scope of this article.

49. CI 82394.

50. Rozelle interview 079702.

51. See Oi, , Rural China Takes Off.Google Scholar

52. The Report on the Villagers Representative Assemblies and Study on the Election of Villagers Committees.

53. See Oi, Jean C. and Rozelle, Scott, “Democracy and markets: the link between participatory decision-making and development in China's rural reforms,” paper presented at the annual meeting of the Association for Asian Studies, Chicago, 13–16 03 1997.Google Scholar

54. Article 4, Organic Law.

55. Leaders in highly industrialized villages had two responses when asked in open-ended questioning about their views on elections. Some said they had insulated themselves from elections. Others said that the enterprises had already provided so much for villagers that they not only did not fear elections, they welcomed them as a sort of additional legitimization of their power.

56. We cannot rule out, however, the possibility that these villages may be the ones that the upper levels are targeting to implement elections in an effort to rebuild village government and resources. Such villages may have individual rich households, but the villages as a collective are often poor.

57. Endogeneity in this case means the concern that 1995 income and 1995 village elections may be simultaneously determined. That is, elections may be leading to greater incomes, as well as the reverse. If so, we may not be measuring the true causal relationship between income and elections. To avoid this problem we need to adopt an “instrumental variable” effect. This statistical method relies on finding a measure that is related to income in 1995, but is not “caused by” elections in 1995. One such measure that is both available and has this property is village income in 1988. To the extent that income in a village in 1995 and 1988 are correlated and to the extent that there is no way that village elections in 1995 could have affected income in 1988, this measure is an acceptable instrument that we can use in our analysis.

58. CI 82494.

59. See Oi, Jean C., “Economic development, stability and democratic village self-governance,” in Brosseau, Maurice, Pepper, Suzanne and Tsang, Shu-ki (eds.), China Review 1996 (Hong Kong: Chinese University Press, 1996), pp. 125–144.Google Scholar