1. They are known as tontines in Africa, chit funds in India, tanda in Mexico, kye in Korea, arisan in Indonesia, paluwagon in the Philippines, boxi money in Guyana, dhikuri in Nepal, bisi committees in Pakistan. Classic comparative reviews of rotating credit associations include Geertz, Clifford, “The rotating credit association: a ‘middle rung’ in development,” Economic and Cultural Change, Vol. 10, No. 3 (04 1962), pp. 241–263CrossRefGoogle Scholar; and Ardener, Shirley G., “The comparative study of rotating credit associations,” Journal of the Anthropological Institute, Vol. 94, No. 2 (1964), pp. 201–229.Google Scholar Additional sources are compiled in Low, A., A Bibliographical Survey of Rotating Savings and Credit Associations (Oxford: Oxfam and Centre for Cross-Cultural Research on Women at Queen Elizabeth House, Oxford, 1995).Google Scholar

2. In 1998, only 0.4% of all loans extended by state banks went to the private sector. If loans extended by officially sanctioned non-banking financial institutions are included, the percentage of formal sector lending to “private, independent businesses” in China increases to 0.8%. “Fourth quarter financial statistics,” Jinrong shibao (Financial Times), 25 01 1999, p. 1.Google Scholar

3. The pawnshops relied on donated funds and extended credit to both wealthy and impoverished peasants. Although pawnbrokering by monasteries faded with the decline of Buddhism in the Ming (1368–1644) and Qing (1644–1911) periods, civilians adopted it as a regular form of business serving all tiers of society. However, with the rise of traditional and modern banks towards the end of the 19th century, the role of pawnbrokering as the primary source of credit diminished accordingly. See Yang, Lien-sheng, Money and Credit in China: A Short History (Cambridge, MA: Harvard University Press, 1952), p. 71.CrossRefGoogle Scholar

4. Ibid. p. 75.

5. Ibid. pp. 75–76.

6. An excellent analysis of the different types of hui is Pairault, Thierry, “Approches tontinieres (deuxième partie): formes et mécanismes tontiniers,” Études chinoises, Vol. 9, No. 2 (automne 1990).Google Scholar

7. Of course, depending on the length of the hui and local inflation, later members might actually receive a lesser amount of money than they originally contributed in real terms.

8. In the variant described by Kulp, however, the organizer always pays for the feast. Kulp, D. H., Country Life in South China: Phoenix Village, Kwantung, China, Vol. 1 (New York: Teachers College, Columbia University, 1925), pp. 190–96.Google Scholar

9. Fei, Hsiao-Tung, Peasant Life in China: A Field Study of Country Life in the Yangtze Valley (New York: Oxford University Press, 1946), pp. 270–73Google Scholar; and Gamble, Sidney D., Ting Hsien: A North China Rural Community (New York: International Secretariat, Institute of Pacific Relations, 1954), pp. 267–270.Google Scholar

10. This is the dominant form in present-day Fujian.

11. Interview No. 30, Fuzhou, Fujian, 7 June 1994. (A comprehensive list of the surveys and interviews referenced in this article is available from the author upon request.) For example, if the leader of a group borrows 100 units of currency in the first month of the arrangement, that person might pay 20% interest to the pool amortized over the term of the hui, but receive a net amount of 100–100* i (n-1) in the first month, where i is the interest rate and n represents the total number of members. Assuming constant interest rates and straight-line amortization of interest, the net monthly amount received for the remaining members would be calculated as follows: P (n-1) – [i* P (n-1)]/(t-m), where, P = payment of each member, n = number of members, i = annual interest rate, t = total number of months of the hui cycle, m = the month of the hui. The last person who borrows would not have to pay interest and might actually earn or lose interest in real terms, depending on inflation.

12. Gates, Hill, China's Motor: A Thousand Years of Petty Capitalism (Ithaca: Cornell University Press, 1996), p. 32.Google Scholar Tawney points out that most of the rural moneylenders were “landowners, merchants, and dealers,” and farmers often obtained credit by pledging the sale of a portion of their prospective harvest to creditors at a discount. Tawney, R. H., Land and Labour in China (London: George Allen and Unwin, Limited, 1937), pp. 60–62.Google Scholar

13. Ibid. p. 62, cited in Fei, Hsiao-Tung, Peasant Life in China, p. 264.Google Scholar Tawney's point was that such indiscriminatory lending lead to usurious interest rates, however. Another study observes that better-off farmers in Hunan were able to avoid high-interest grain loans by forming “co-operative credit societies” in the immediate aftermath of the Taiping Rebellion. Perdue, Peter C., Exhausting the Earth: State and Peasant in Hunan, 1500–1850 (Cambridge, MA: Harvard University Press, 1987), p. 243CrossRefGoogle Scholar, citing Liling xianzhi (Liling Gazetteer), 05 1948.Google Scholar

14. Fei, Hsiao-Tung, Peasant Life in China, p. 267.Google Scholar The hui should be distinguished from the credit societies established as part of the broader agricultural movement in the 1920s. With the assistance of the China International Famine Relief Commission and the Agricultural College of the University of Nanjing, the first one, the Feng Ren Co-operative Credit Society of Vegetable Growers, was established in September 1923. See Tawney, , Land and Labour, pp. 92–96.Google Scholar

15. Fei, Hsiao-Tung, Peasant Life in China, pp. 268–69.Google Scholar The popularity of hui among men is echoed in the other works cited above. The only reference that the author could find to female involvement was a parenthetical reference in a discussion of “money loan societies” (yihui) in Hong Kong during the late 1800s. Ball, J. Dyer, Things Chinese: Being Notes on Various Subjects Connected with China, 3rd ed. (London: Sampson Low, Marson and Company, Limited, 1900 [Ist ed., 1891]), p. 536.Google Scholar Ball proceeds to use examples of male heads of associations through the rest of the discussion, however; for example, “should any member die before drawing his loan, his wife or children, or in default of them, nearest of kin, may continue on in the Association in place of the deceased…” p. 543.Google Scholar

16. I spent a total of 18 months in the field over the three-year period, including pre-dissertation field research during the summers of 1994 and 1995, and formal dissertation research from March 1996 to June 1997.

17. I also conducted research in the central-northern provinces of Henan and Hebei, but they are beyond the scope of this article.

18. Addendum to Survey No. 121.

19. This is calculated based on the 273 valid survey responses in the south.

20. The 1990 provincial-wide survey of women's social status focused on urban-rural differences rather than north-south, regional ones. See Fujian funü shehui diwei diaocha (Survey of Women's Social Status in Fujian) (Fujian: Zhongguo funü chubanshe, 1995).Google Scholar National surveys administered in preparation for the Fourth World Conference on Women in Beijing, 1995, similarly focused on urban-rural cleavages. See for example, Chunfang, Tao and Yongping, Jiang (eds.), Zhongguo funü shehui diwei gaikuan (An Overview of the Social Status of Chinese Women) (Beijing: Zhongguo funü chubanshe, 1993)Google Scholar; Zhongguo shehui kexueyuan renkou yanjiusuo (Institute of Population Studies, Chinese Academy of Social Sciences), Dangdai Zhongguo funü diwei chouyang diaocha ziliao (Sampling Survey of Women's Status in Contemporary China) (Beijing: International Academic Publishers, 1994).Google Scholar An overview of the results from the Women's Federation and Institute of Population Studies surveys is Hung, Jean K.M., “The family status of Chinese women in the 1990s,” in Kin, Lo Chi, Pepper, Suzanne and Yuen, Tsai Kai (eds.), China Review 1995 (Shatin, Hong Kong: The Chinese University Press, 1995), ch. 12.Google Scholar

21. Note, however, that even within the wealthier southern coastal province like Fujian, substantial variation exists in standard of living. See Hui, Zhao and Yu, Wei, “Zhanwang molai, renzhong daoyuan: Fujian juxing jiti gongsi” (“Making greatest possible efforts to eliminate poverty: Fujian Juxing Corporate Group”), Fazhan yanjiu (Development Research), No. 8 (1996), pp. 6–11Google Scholar; and Ji, Zhou and Bingwen, Liu, “Shilun Fujian pinkun diqu de kedai fazhan” (“On the continuity of poverty-stricken areas' development in Fujian province”), Fazhan yanjiu. No. 2 (1996), pp. 32–33.Google Scholar A case study of Anxi as a successful example of poverty alleviation in the 1980s is Lyons, Thomas P., Poverty and Growth in a South China County: Anxi, Fujian, 1949–1992 (Ithaca: Cornell East Asian Program, 1994).Google Scholar

22. For more detail, see Crane, George T., The Political Economy of China's Special Economic Zones (Armonk, NY: M.E. Sharpe, 1998)Google Scholar; Pak, Chong-dong, China's Special Economic Zones and Their Impact on its Economic Development (Westport, CT: Praeger, 1997)Google Scholar; and Sun, Xiuping, New Progress in China's Special Economic Zones (Beijing: Foreign Languages Press, 1997).Google Scholar

23. Studies focused on the floating population have found that informal borrowing and lending occurs within migrant communities, but not necessarily in the form of rotating credit associations. See Ma, Laurence J.C. and Xiang, Biao, “Native place, migration and the emergence of peasant enclaves in Beijing,” The China Quarterly, No. 155 (09 1998), pp. 546–581Google Scholar; Solinger, Dorothy, Contesting Citizenship in Urban China: Peasant Migrants, the State, and the Logic of the Market (Berkeley: University of California Press, 1999)Google Scholar; and Zhang, Li, “Strangers in the city: space, power, and identity in China's ‘floating population’,” Ph.D., Department of Anthropology, Cornell University, 1998.Google Scholar

24. As of 1995, official statistics estimate that the floating population in Fuzhou exceeded 600,000 people at its peak, of which only 260,000 were officially registered. Fuzhou nianjian 1996 (Fuzhou Yearbook 1996) (Beijing: Zhongguo tongji chubanshe, 1997), p. 176.Google Scholar As of year-end 1995, Fuzhou municipality had a population of 5.6 million with nearly 1.4 million people living in the city proper.

25. Based on the 141 valid survey responses of microentrepreneurs in Fuzhou, the average store is open for 12 hours a day.

26. Some 17% of surveyed hui participants in Fuzhou indicated that they did not know all of the members of their hui.

27. Survey No. 120. To protect the identity of the interviewees, pseudonyms are used in all the cases discussed in this article.

28. The monthly 28,360 yuan in expenses includes rent (15,000), electricity (60), ICB fee (800), taxes (4,500) and storage facilities (8,000). Survey No. 120.

29. It is estimated that 95% of all illegal Chinese immigrants to the U.S. in the 1990s left from Changle. Chin, Ko-lin, Chinatown Gangs: Extortion, Enterprise, and Ethnicity (New York: Oxford University Press, 1996).Google Scholar

30. Interview No. 64, Changle, Fujian, 12 June 1996.

31. The Fujian Provincial CPC Committee, Fujian People's Government and Public Security Bureau have conducted numerous investigations to capture snake heads. Between 1993 and 1997, the Fuzhou Frontier Guard Unit caught nearly 1,000 migration organizers and transporters, including over 50 snake heads from abroad. Between 1993 and 1996, over 4,500 illegal migrants were caught and over 16,000 were repatriated from abroad. See Baozhang, Jiang, “Fujian zhandou toudu xianxiang” (“Fujian combats illegal migration”), Renmin ribao (People's Daily), 24 05 1997, p. 5.Google Scholar

32. At least four teenage girls made the Golden Venture trip in 1993; U.S. immigration officials inadvertently released them to the gang that smuggled them in and gang raped them during the trip. “Chinese girls released to gang,” Newsday, 21 06 1993, p. 6.Google Scholar In May 1998, the Oops II powerboat carrying 23 Chinese men from Changle ended up in New Jersey. They had originally been part of a group of 50 men aboard the Oriental I, a coastal freighter, which had departed from Venezuela and picked up the men in Suriname. Half of the Oriental I group ended up in New Jersey, the other half in the Bahamas. See McCoy, Kevin, “First inside look at smuggling ring,” Daily News, 1 07 1998, p. 38Google Scholar; McFadden, Robert D., “22 illegal immigrants seized after a Jersey shore landing,” The New York Times, 1 06 1998, p. 1.Google Scholar

33. Survey No. 165.

34. Quanzhou also brags the oldest mosque in eastern China, built in 1009 for the local Muslim population.

35. Interviews No. 7, 18, 142, 143.

36. A similar phenomenon occurred in Taiwan. When there were large-scale hui failures during 1983–85, hui participation declined significantly; but participation levels had just about recovered to 1977 levels by 1991 (20–30% of all households). See Levenson, Alec R. and Besley, Timothy, “The anatomy of an informal financial market: Rosca participation in Taiwan,” Journal of Development Economics, Vol. 51, No. 1 (10 1996), pp. 45–68.CrossRefGoogle Scholar

37. Survey No. 321.

38. The people of Hui'an are classified as being Han Chinese though the women have a distinctive style of dressing. Women from the coastal villages of Chongwu township wear bright floral scarves on a wide headband that extends the height and width of the head, accompanied by a wide-rimmed yellow bamboo hat with a pointed top, a short jacket that barely reaches the navel, and loose ankle-length pants with a wide embroidered or silver belt. See Bingzhao, Jiang, “Hui'an diqu changzhu niangjia hunsu de lishi kaocha” (“An historical investigation of extended natolocal residence marriage customs in the Hui'an region”), Zhongguo shehui kexue (China Social Science) No. 3 (1989), pp. 193–203Google Scholar; and Huixiang, Lin, “Lun changzhu niangjia fengsu de qiyuan ji muxi zhi dao fuxi zhi de guodu” (“A discussion of the origins of extended natolocal residence customs and the transition from matrilineal to patrilineal society”), in Lin Huixiang renleixue lunzhu (Lin Huixiang's Anthropological, Works) (Fuzhou: Fujian renmin chubanshe, 1981), pp. 254–288Google Scholar, cited in Friedman, Sara, “Owing a debt: marriage, labor, and social change in southeastern China's Hui'an county,” Ph.D., Department of Anthropology, Cornell University, forthcoming.Google Scholar

39. Interviews No. 22, 45, 142, 143.

40. One anthropologist also found that some women pray to sets of tiny dolls every night before going to bed and put the dolls under their pillows with the hope that their grievances may be alleviated while they are sleeping. Interview No. 45.

41. Interview with Carolyn Ng Fung Yee, Department of Anthropology, Chinese University of Hong Kong, 28 March 1996. Although Sara Friedman notes that collective suicides among women existed in the past, she did not come across specific incidences of them during 1995–97 fieldwork in Hui'an. Friedman, “Owing a debt,” ch. 3. Apparently, the incidence of sister suicides has declined since 1949. For more detail, see Huixiang, Lin, “A discussion of the origins,”Google Scholar and Jian, Qiao, Guoqiang, Chen and Lifang, Zhou, (eds.), Huidongren yanjiu (Research on Eastern Hui'an People) (Fuzhou: Fujian jiaoyu chubanshe, 1992).Google Scholar I thank Sara Friedman for pointing me to these sources.

42. This trend is by no means exclusive to Fujian. See, for example, Xianfan, Meng, “‘Nangong nügeng’ yu Zhongguo nongcun nüxing de fazhan” (“‘Men at work, women on the farm’ and Chinese rural women's development”), Funü yanjiu (Research on Women), No. 4 (1995), pp. 48–51.Google Scholar

43. Interviews No. 22, 24.

44. Survey No. 340.

45. Liu, Ya-ling, “Reform from below: the private economy and local politics in the rural industrialization of Wenzhou,” The China Quarterly, No. 130 (06 1992), pp. 293–316Google Scholar; and Parris, Kristen, “Local initiative and national reform: the Wenzhou model of development,” The China Quarterly, No. 134 (06 1993), pp. 242–263.Google Scholar

46. “Red hat” enterprises are defacto privately-owned operations that register themselves as “collectives” for favourable access to key inputs, including credit; the practice is called “wearing a red hat” (dai hongmaozi) because such businesses only have the appearance of being collectively held by the people. “Hang-on” household enterprises (guahu qiye) are variants of this practice, whereby private concerns pay off state-owned enterprises to use their office space, letterhead and account numbers.

47. For more detail on Wenzhou's informal financial sector, see Tsai, Kellee S., “Curbed markets? Financial innovation and policy involution in China's coastal south,” Weatherhead Center for International Affairs Working Paper Series, No. 98–96, Harvard University, 05 1998.Google Scholar

48. Even monthly payments of only 1,000 yuan were considered astronomical in 1985 given that the average urban salary in Zhejang province at the time was less than 200 yuan. Zhejiang nianjian 1985 (Zhejiang Statistical Yearbook).

49. According to transcripts from the 1984 “Huanghua” court case, a person named Nan started organizing juhui in 1977 to alleviate poverty in Huanghua township, Yueqing county. By 1982, Nan was involved in several hui and lost the ability to keep track of them, and at year end became completely illiquid. Nevertheless, Nan continued to participate in hui in order to finance the other hui. During the spring of 1984, they completely collapsed, bringing 367 households into 1.1 million yuan of collective debt. As of 1991, 60% of Nan's debt remained outstanding. Zhenning, Zhang and Chunhua, Mao, Wenzhou jinrong xianxiang toushi (Perspectives on Wenzhou's Financial Phenomena) (Zhejiang: Zhejiang daxue chubanshe, 1992), p. 21.Google Scholar

50. Unlike conventional hui, which involve a group of people who pool their resources, taihui and paihui entail bilateral relationships between a leader and an investor who is promised high returns on his/her monthly contributions. For example, an investor might “invest” 3,000 yuan for three months and then receive 18,000 yuan from the leader in the fourth month. Since such high rates of return were not premised on productive investments in Wenzhou, the taihui/paihui leaders found themselves having to “invest” in taihui/paihui to sustain the payments. Complex networks of taihui/paihui emerged based on unrealistically high returns.

51. Yu, Li, “Jinrong wanhuatong” (“Financial kaleidoscope”), in Shizhang, Yu (ed.), Wenzliou gaige moshi yanxin yinxiang (New Reflections on Wenzhou's Reform Model). (Wenzhou: Zhonggong Wenzhou shi wixuanchuangu, 1989), pp. 49–62.Google Scholar The two small counties that were spared from the fervour of hui were geographically remote from the others.

52. For example, if the organizer of a 10,000 hui defaulted, at least eleven households would be affected. If the organizer of a one million yuan hui defaulted, then at least 50 10,000 hui would be dragged down (comprising at least 550 households). The largest reported hui involved 100,000 participants and 100 million yuan (over US$12 million). Jinrong, Ma, “Wenzhou Jinrong shichang” (“Wenzhou's financial market”), in Wenzhou shichang (Wenzhou's Market) (Beijing: Zhonggong dangshi chubanshe, 1995), p. 414.Google Scholar

53. The aggregate scale of hui reached one billion yuan (over US$123 million).

54. The hui collapses were not mentioned in the local press because the matter was “dealt with internally” (neibu chuli).

55. Interview No. 142, Quanzhou, 16 November 1998.

56. For an excellent collection of writings by western and Chinese scholars on the construction of gender (from the 16th century to the present), see Gilmartin, Christina K., Hershatter, Gail, Rofel, Lisa and White, Tyrene (eds.), Engendering China: Women, Culture, and the State (Cambridge, MA: Harvard University Press, Harvard Contemporary China Series, No. 10, 1994).Google Scholar

57. Croll, Elisabeth, Chinese Women Since Mao (London: Zed Books, 1983), pp. 31–32Google Scholar; Entwisle, Barbara, Henderson, Gail E., Short, Susan E., Bouma, Jill and Fengying, Zhai, “Gender and family businesses in Rural China,” American Sociological Review, Vol. 60 (02 1995), pp. 36–57CrossRefGoogle Scholar; and Jacka, Tamara, Women's Work in Rural China: Change and Continuity in an Era of Reform (New York: Cambridge University Press, 1997).CrossRefGoogle Scholar

58. For example, Granovettor proposes that the degree of trust among individuals is related to the “social embeddedness” of their interaction. Granovettor, Mark, “Economic action and social structure: the problem of embeddedness,” American Journal of Sociology, Vol. 91 (11 1985), pp. 481–510.CrossRefGoogle Scholar For an earlier articulation of embeddedness theory, see Polanyi, Karl, The Livelihood of Man (New York: Academic Press, 1977), ch. 4.Google Scholar Different motivations underlying trust are discussed in Gambetta, Diego (ed.), Trust: Making and Breaking Cooperative Relations (Oxford: Basil Blackwell, 1988).Google Scholar

59. In another context, Robert Putnam argues that civic associations and other community organizations (including rotating credit associations) are more popular in northern Italy due to the greater stock of social capital in the north. Putnam, Robert D., Making Democracy Work: Civic Traditions in Modern Italy (Princeton: Princeton University Press, 1994).Google Scholar

60. Judd, Ellen R., Gender and Power in Rural North China (Stanford: Stanford University Press, 1994), pp. 208–209.Google Scholar

61. Jicai, Sha, Dangdai Zhongguo funü jiating diwei yanjiu (Women's Domestic Status in Contemporary China) (Tianjin: Tianjin renmin chubanshe, 1995).Google Scholar Cf. Aslanbeigui, Nahid and Summerfield, Gale, “Impact of the responsibility system on women in rural China: an application of Sen's theory of entitlements,” World Development, Vol. 17, No. 3 (03 1989), pp. 343–350.CrossRefGoogle Scholar They employ Amartya Sen's theory of entitlements to illustrate that under the HRS, a woman's controllable income may differ from her contribution to production inside and outside the household.

62. Yan, Yunxiang, The Flow of Gifts: Reciprocity and Social Networks in a Chinese Village (Stanford: Stanford University Press, 1996).Google Scholar

63. Judd, , Gender and Power, p. 205.Google Scholar

64. For example, Hechter, Michael, Principles of Group Solidarity (Berkeley: University of California Press, 1987), pp. 107–111.Google Scholar

65. This phenomenon is also documented in Croll, , Chinese Women Since Mao, pp. 31–32Google Scholar; Wolf, Margery, Revolution Postponed: Women in Contemporary China (Stanford: Stanford University Press, 1985), pp. 103–111.Google Scholar In the case of business owners from Wenzhou, however, the entire family tends to relocate at the same time since the nature of their business relies on the labour of women and children. Zhang, Li, “The interplay of gender, space, and work in China's floating population,” in Henderson, Gail and Entwisle, Barbara (eds.), Redrawing Boundaries: Gender, Space, and Work in Households; and Work in China (Berkeley: University of California Press, forthcoming).Google Scholar

66. Women account for 90% of the Chinese Grameen Bank replications' members since they are deemed to be better credit risks than men. Interviews No. 42, 158–166. In other contexts, a number of poverty alleviation and women-in-development programmes have also found that women have higher repayment and lower default rates than do men. For example, Bakhoum, I. et al. , Banking the Unbankable: Bringing Credit to the Poor (London: The Panos Institute, 1989)Google Scholar; Berger, Marguerite and Buvinic, Mayra (eds.), Women's Ventures: Assistance to the Informal Sector in Latin America (West Hartford: Kumarian Press, 1989)Google Scholar; Buechler, Simone, Credit Approaches and Women Entrepreneurs (New York: UNIFEM, 1993)Google Scholar; Sutoro, Ann Dunham, KUPEDES Development Impact Survey (Indonesia: BRI, 1990)Google Scholar; and Vyas, Jashree, “Banking with poor self-employed women: Mahila SEWA Bank's Experience,” paper presented in a seminar on “Savings and Credit” organized by Action Aid Bangalore, India, 16–17 01 1992.Google Scholar

67. The primary reasons for exclusion are poverty and olher indicators of social marginalization such as gender and ethnicity. For example, Ardener, Shirley and Burman, Sandra (eds.), Money-Go-Rounds: The Importance of Rotating Savings and Credit Associations for Women (Oxford: Berg Publishers Limited, 1995)Google Scholar; Bouman, F. J. A., “The ROSCA: financial technology of an informal savings and credit institution in developing counties,” Savings and Development, Vol. 4, No. 1 (1979), pp. 253–276Google Scholar; Bouman, F. J. A., “Rotating and accumulating savings and credit associations: a development perspective,” World Development, Vol. 23, No. 3 (03 1995), pp. 371–384CrossRefGoogle Scholar; Janelli, Roger L. and Yim, Dawnhee, “Interest rates and rationality: rotating credit associations among Seoul women,” Journal of Korean Studies, Vol. 6 (1988)CrossRefGoogle Scholar; Kurtz, Donald V., “The rotating credit association: an adaption to poverty,” Human Organization, Vol. 32, No. 1 (Spring 1973), pp. 49–58CrossRefGoogle Scholar; Light, Ivan, Kwuon, Im Jung and Zhong, Deng, “Korean rotating credit associations in Los Angeles,” Amerasia Journal, Vol. 16, No. 2 (1990), pp. 35–54CrossRefGoogle Scholar; and Wu, David Y. H., “To kill three birds with one stone: the rotating credit associations of the Papua New Guinea China,” American Ethnologist, Vol. 1 (1974), pp. 565–584.CrossRefGoogle Scholar

68. The RCFs were established by the Ministry of Agriculture in 1978 to provide a grassroots source of credit to rural households; the People's Bank of China does not recognize them as legitimate financial institutions, however.

69. This point is made in Ardener, Shirley, “The comparative study of rotating credit associations,” Journal of Royal Anthropology, No. 94 (1964), pp. 201–229Google Scholar; and Hechter, Michael, Principles of Group Solidarity (Berkeley: University of California Press, 1987), pp. 108–111.Google Scholar

70. Similar criticisms have been levelled against the frustration-aggression theories of revolution. In other words, relative deprivation in and of itself does not explain what translates frustration into action. Representative works include the following: Davies, James C., “Toward a theory of revolution,” American Sociological Review, Vol. 27 (02 1962), pp. 5–19CrossRefGoogle Scholar; Feierabend, Ivo K., Feierabend, Rosalind L. and Nesvold, Betty A., “The comparative study of revolution and violence,” Comparative Politics, Vol. 5, No. 3 (04 1973), pp. 393–24CrossRefGoogle Scholar; and Gurr, Ted Robert, Why Men Rebel (Princeton: Princeton University Press, 1970).Google Scholar

71. The failure of China's party-state to liberate women is articulated most powerfully in Andors, Phyllis, The Unfinished Liberation of Chinese Women, 1949–1980 (Bloomington: Indiana University Press, 1983)Google Scholar; and Wolf, , Revolution Postponed.Google Scholar Journalistic accounts of women's oppression in China are too vast to list here. Most have focused on the coercive nature of the state's population control efforts. More general discussions of the plight of women in rural China include Rosenthal, Elisabeth, “Women's suicides reveal rural China's bitter roots: nation starts to confront world's highest rate,” The New York Times, 24 01 1999, pp. A1, A12Google Scholar; and Tyson, Ann Scott and Tyson, James L., “Long days, hard labor for women left on the farm,” Christian Science Monitor, 5 08 1992, p. 10.Google Scholar The now-classic theoretical antidote to such “top-down” analyses of structural oppression is Scott, James C., Weapons of the Weak: Everyday Forms of Peasant Resistance (New Haven, CT: Yale University Press, 1985).Google Scholar

72. Judd, , Gender and Power, pp. 212–239Google Scholar; and Judd, Ellen R., “‘Men are more able’: rural Chinese women's conceptions of gender and agency,” Pacific Affairs, Vol. 63, No. 1 (Spring 1990), pp. 40–61.CrossRefGoogle Scholar A revealing volume on the experiences and representation of women in urban areas is Honig, Emily and Hershatter, Gail, Personal Voices: Chinese Women in the 1980's (Stanford: Stanford University Press, 1988)Google Scholar; and The Ford Foundation, Reflections and Resonance: Stories of Chinese Women Involved in the International and Preparatory Activities for the 1995 NGO Forum on Women (Beijing: Ford Foundation, 1995).Google Scholar

73. Judd, , “Men are more able,” p. 47.Google Scholar

74. In Korea, Laurel Kendall finds that the perception of women's role as matchmakers is highly polarized, i.e. the Evil versus the Good matchmaker. Yet they perform an integral function because of the constructed centrality of family involvement in social weddings. Laurel Kendall, Curator at the Museum of Natural History, Adjunct Professor at Columbia University, “Ambiguous heroines or what Korean matchmakers taught me about the crisis in anthropology,” presentation at Columbia University, 12 December 1994.

75. Cf. Zhou, Kate Xiao, How the Fanners Changed China (Boulder: Westview, Press, 1996).Google Scholar For a case in another context of how the confinement of peasant women to small-scale agriculture may paradoxically increase their confidence and foster organization into labour gangs, see Hart, Gillian, “Engendering everyday resistance: gender, patronage and production politics in rural Malaysia,” Journal of Peasant Studies, Vol. 19, No. 1 (10 1991), pp. 93–121.CrossRefGoogle Scholar Hart finds that men are more subservient to their employers than women because of their exposure to political patronage relations and inability to fulfil societal standards of male household responsibilities (p. 114).

76. A number of women-run NGOs have emerged in the past few years. Most involve successful women entrepreneurs or leaders who seek to help other women through sharing information and ensuring the protection of women's rights and interests. For a working list of women's NGOs in China, see Ford Foundation, Interim Directory of Chinese Women's Organizations (Beijing: Ford Foundation, 08 1995)Google Scholar; and United Nations Development Programme (UNDP), Gender and Development in China: A Compendium of Gender and Development Projects Supported by International Donors (Beijing: UNDP, 1998).Google Scholar

77. I do not assume that agency necessarily leads to normatively positive outcomes for the actors. For example, women in Hui'an also exercise agency in organizing sister societies, which represent a source of comfort and camaraderie for women, but have also ended in collective suicides (at least in the past).

78. It is worth noting, however, that men are indeed more likely to participate in hui involving larger sums of money than in the smaller ones. Furthermore, the hui with larger contributions are also more likely to collapse partially because the size of the collective pot may be large enough to finance a one-way ticket out of town and the costs of setting up a new residence.

79. For a discussion of how emotional deviations from calculating self-interest have been explained by Smith, Hume, and others, see Holmes, Stephen, “The secret history of self-interest,” in Mansbridge, Jane J. (ed.), Beyond Self-Interest (Chicago: University of Chicago Press), pp. 267–286, especially pp. 275–280.Google Scholar

80. A 1992–93 survey finds that this more likely to be the case in urban than rural areas. Yang, Xu, “Chinese women enjoy equal say,” China Daily, 15 01 1994, p. 3.Google Scholar

81. Laurel Kendall raised the issue of social cost in the context of why women are more likely than men to perform the role of matchmaking in Korean society. Kendall, , “Ambiguous heroines.”Google Scholar

82. Elster, Jon, The Cement of Society (New York: Cambridge University Press, 1989).CrossRefGoogle Scholar

83. Evans, Harriet, Women and Sexuality in China: Dominanl Discourses of Female Sexuality and Gender Since 1949 (Cambridge: Polity Press, 1997).Google Scholar Cf. Ruan, Fang Fu, Sex in China: Studies in Sexology in Chinese Culture (New York: Plenum Press, 1991).CrossRefGoogle Scholar

84. See Gilmartin, Christina, “Violence against women in contemporary China,” in Lipman, Jonathan N. and Harrel, Steven (eds.), Violence in China: Essays in Culture and Counterculture (Albany: State University of New York Press, 1990), pp. 203–226Google Scholar; and Honig, and Hershatter, , Personal Voices, pp. 277–286.Google Scholar

85. See citations in Tsai, Kellee S., “Women and the state in post-1949 rural China,” Journal of International Affairs, Vol. 49, No. 2 (Winter 1996), pp. 493–524.Google Scholar

86. For example, Adams, Dale W. and Canavesi, Marie L., “Rotating savings and credit associations in Bolivia,” in Adams, Dale W. and Fitchett, Delbert A. (eds.), Informal Finance in Low-Income Countries (Boulder, CO: Westview Press, 1992), pp. 313–324Google Scholar; and Baydas, Mayada M., Bahloul, Zakaria and Adams, Dale W., “Informal finance in Egypt: ‘banks’ within banks,” World Development, Vol. 23, No. 4 (04 1995), pp. 651–661.CrossRefGoogle Scholar Adams has also documented the use of ROSCAs among employees of the Development Finance Corporation in Belize and St Kitts/Nevis; commercial and central banks in Belize, Papua New Guinea, Bolivia, the Dominican Republic and the Philippines; and among economists in the International Monetary Fund in Washington D.C. Adams, Dale W., “Taking a fresh look at informal finance,”Google Scholar in Adams, and Fitchett, , Informal Finance, p. 14.Google Scholar

87. Vélez-Ibañez, Carlos G., Bonds of Mutual Trust: The Cultural Systems of Rotating Credit Associations among Urban Mexicans and Chicanos (New Brunswick, NJ: Rutgers University Press, 1983), p. 22.Google Scholar

88. Coleman, James C., Foundations of Social Theory (Cambridge, MA: Harvard University Press, 1990), pp. 306–307Google Scholar; Putnam, , pp. 161–178.Google Scholar

89. Geertz, , “The rotating credit association,” p. 263.Google Scholar

90. Ardener, , “The comparative study,” p. 221.Google Scholar

91. Additional informal financial institutions are discussed in Tsai, Kellee S., “Banking behind the state: private entrepreneurs and the political economy of informal finance in China, 1978–1998,” Ph.D. dissertation, Columbia University, New York, 1999.Google Scholar

92. For example, Light, Ivan and Bonacich, Edna, Immigrant Entrepreneurs: Koreans in Los Angeles 1965–1982 (Berkeley: University of California, 1988), pp. 244–259.Google Scholar

93. See Ardener, and Burman, , Money-Go-RoundsGoogle Scholar and Leshkowich, Ann Marie, “Tightly woven threads: gender, capital, and kinship among cloth and clothing traders in Ho Chi Minh City's Ben Thanh market,” Ph.D., Department of Anthropology, Harvard University, forthcoming.Google Scholar