1. See Whyte, Martin King and Parish, William L., Urban Life in China (Chicago: University of Chicago Press, 1984)Google Scholar; Gold, Thomas, “After comradeship: personal relations in China since the Cultural Revolution,” The China Quarterly, No. 104 (December 1985), pp. 657–675CrossRefGoogle Scholar; Yang, Mayfair Mei-Hui, “The art of social relationships and exchange in China,” unpublished doctoral dissertation (University of California at Berkeley, 1986)Google Scholar; Walder, Andrew G., Communist Neo-Traditionalism: Work and Authority in Chinese Industry (Berkeley: University of California Press, 1986).Google Scholar

2. Children of affluent families tended to be assigned jobs in the privileged state sectors because their high-status fathers had better connections with the authorities who assigned jobs. See Lin, Nan and Bian, Yanjie, “Social connections (guanxi1) and social resources in the process of status attainment in urban China,” paper presented at the Sunbelt International Conference of Social Networks, San Diego, 1989.Google Scholar

3. See descriptions and analyses of the use of guanxi in developing getihu and other types of private businesses in Gold, Thomas, “Urban private business and social change,” in Deborah, Davis and Vogel, Ezra F. (eds.), Chinese Society on the Eve of Tiananmen (Cambridge, MA: Harvard University Press, 1990), pp. 157–178.CrossRefGoogle Scholar

4. Nee, Victor, “Social inequalities in reforming state socialism: between redistribution and markets in China,” American Sociological Review, No. 56 (1991), pp. 267–282CrossRefGoogle Scholar; Nee, Victor and Su, Sijin, “Institutional change and economic growth in China: the view from the villages,” Journal of Asian Studies, No. 49 (1990), pp. 3–25.CrossRefGoogle Scholar

5. He, Guang and associates (eds.), Xiandai Zhongguo laodongli guanli (The Labour Force Management in Contemporary China) (Beijing: Chinese Social Sciences Press, 1990)Google Scholar, ch. 1; Tang, Yunqi (ed.), Zhongguo laodong guanli gailan (A General Inquiry into China's Labour Management) (Beijing: Chinese Cities Press, 1990)Google Scholar, ch. 1.

6. In Zhongguo laodong guanli gailan, pp. 67–69, Tang Yunqi described job assignments in terms of a set of formal procedures and rules such as labour quotas, recruitment criteria, regulations to prevent discrimination against women or the disabled, and labour contract systems. These descriptions can also be seen in, among others, Rui, Guangzhao (ed.), Zhongguo de laodong zhengce he zhidu (Chinese Labour Policies and System) (Beijing: Economic Management Press, 1988)Google Scholar; Han, Fanglun, Ren, Zhitao and Tian, Maosheng (eds.), Xiandai shiyong renshi guanlixue (Modern Practical Personnel Management) (Beijing: Chinese Personnel Affairs Press, 1991).Google Scholar

7. See Whyte and Parish, Urban Life in China; Walder, Communist Neo-Traditionalism; Nan Lin and Yanjie Bian, “Social connections and social resources”; Deborah Davis, “Urban job mobility,” in Davis and Vogel, Chinese Society on the Eve of Tiananmen, pp. 85–108.

8. Although it is possible that respondents might more easily recall the use of guanxi in finding first jobs in recent years than earlier periods, those who entered the labour force before 1949 reported the highest percentage of using guanxi to find first jobs. So, the dramatic increase in reported use of guanxi for finding first jobs from the Cultural Revolution decade to the 1977–88 period should be interpreted as the result of Deng Xiaoping's decentralization policies.

9. Because of persistent regional differences in China, data from other Chinese cities, particularly those in Guangdong and Fujian provinces where economic reforms have been more far-reaching, may generate somewhat different results. Nevertheless, Tianjin, with its leading role in state industries in the north, its proximity to the national capital, and its administrative status as one of the three autonomous municipalities (with Beijing and Shanghai), represents a core locality in a planned economy during a reform period.

10. These include the labour contract system implemented in 1986 and the growth of private-sector jobs since early 1992. While the latter awaits fruitful research, the former has been documented in several studies including White, Gordon, “The politics of economic reform in Chinese industry: the introduction of the labour contract system,” The China Quarterly, No. 111 (September 1987), pp. 365–389CrossRefGoogle Scholar; Davis, , “Urban job mobility”; “‘Skidding’: downward mobility among children of the Maoist middle class,” Modern China, No. 18 (1992), pp. 410–437CrossRefGoogle Scholar; Davis, Deborah, “Job mobility in post Mao cities: increases on the margins,” The China Quarterly, No. 132 (December 1992), pp. 1062–85.CrossRefGoogle Scholar

11. Personal interview No. 33, 37 years old, university economics lecturer.

12. Personal nterview No. 5, 31 years old, typist working in the general office of a state-owned chemical material factory.

13. According to He Guang (Xiandai Zhongguo laodongli guanli, p. 152), from 1978 to 1983 12 million state and collective workers retired, and 80% of the vacant positions were filled by the children of retirees through the dingti practice.

14. Personal interview No. 3, 28 years old, purchasing agent in a state-owned soy sauce plant, who took his father's job through dingti.

15. Not every type of work unit was authorized to use neizhao to recruit new workers. The method was largely used to recruit unskilled workers, and then only when there were large numbers of employees’ children waiting for jobs. Usually an employee could submit only one child's name for neizhao during his or her entire lifetime. National statistics show that from 1978 to 1987, nearly 50% of new workers were recruited through the dingti and neizhao practices. See He Guang, Xiandai Zhongguo laodongli guanli, p. 156.

16. Personal interview No. 25, 27 years old, train conductor, who was recruited through neizhao after graduation from high school.

17. Personal interview No. 23, 31 years old, cargo van loader, who left his state farm job and was hired by his father's former work unit in a transport company.

18. He Guang, Xiandai Zhongguo laodongli guanli, p. 34.

19. Personal interview No. 8, 30 years old, lawyer whose work unit is the first law firm affiliated with the municipal government.

20. Personal interview No. 10, 24 years old, building construction worker affiliated with an employment service centre.

21. Bian, Yanjie, Work and Inequality in Urban China (Albany, NY: State University of New York Press, 1994).Google Scholar

22. Sixty-nine respondents were excluded from this study sample. Among them were full-time students who had no work experience, home makers, military personnel, and those who did not provide information on their first occupations or work units which are the necessary variables for the present study.

23. Lin, Nan and Xie, Wen, “Occupational prestige in urban China,” American Journal of Sociology, No. 93 (1988), p. 830.Google Scholar The Lin-Xie formula is: − 5.188+ (13.873*education) + (.262*monthly salary). The procedure of calculating the Lin-Xie scores for this study was that, first, average education level and monthly salary of the respondents were obtained for each of 16 occupational categories used in the survey (see Table 2), and, second, based on these aggregate averages and the formula, the scores were calculated for these occupations. Finally, each respondent was assigned an “occupational status score” according to his or her occupation on the 16-category scale.

24. See Whyte and Parish, Urban Life in China; Walder, Communist Neo-Traditionalism; Davis, Deborah, “Unequal chances, unequal outcomes,” The China Quarterly, No. 114 (June 1988), pp. 221–242.Google Scholar

25. Walder, Andrew, “Property rights and stratification in socialist redistributive economies,” American Sociological Review, No. 57 (1992), pp. 524–539CrossRefGoogle Scholar; Yanjie Bian, Work and Inequality in Urban China, ch. 8.

26. See Honig, Emily and Hershatter, Gail, Personal Voices: Chinese Women in the 1980s (Stanford, CA: Stanford University Press, 1988)Google Scholar; Johnson, Kay Ann, “Women in the People's Republic of China,” in Chipp, Sylvia A. and Green, Justin J. (eds.), Asian Women in Transition (Philadelphia: Pennsylvania State University Press, 1976), pp. 62–103Google Scholar; Stacey, Judith, Patriarchy and Socialist Revolution in China (Berkeley: University of California Press, 1983)Google Scholar; Whyte, Martin King, “Sexual inequality under socialism: the Chinese case in perspective,” in Watson, James L. (ed.), Class and Social Stratification in Post-Revolution China (Cambridge: Cambridge University Press, 1984) pp. 198–238Google Scholar; Wolf, Margery, Revolution Postponed: Women in Contemporary China (Stanford: Stanford University Press, 1985)Google Scholar; Zhou, Zheng, “Why fewer women at leading posts?” Beijing Review, 7–13 March 1988, pp. 27–31Google Scholar; Lin, Nan and Bian, Yanjie, “Getting ahead in urban China,” American Journal of Sociology, No. 97 (1991), pp. 657–688.CrossRefGoogle Scholar

27. The question used in the survey reads: “We would like to learn about someone who provided the greatest help or influence for you to get a job when you first entered the work force.” Although the pronoun “someone” means guanxi in the Chinese context, I chose not to use guanxi because it was a sensitive word. Party policies opposed the use of guanxi or “back door” practices in job assignments, as in other government-imposed programmes. Although it would be inappropriate to use the word guanxi in a questionnaire, it is fine in personal interviews, depending on the subject matter and the context in which the interviews are conducted. Throughout the data analysis which follows, the term “personal contacts” or “contact” is used. Also, since the question was about the respondent's actual first urban job, efforts to contact guanxi who did not lead to the final outcome were not likely to be counted. In other words, personal contacts might actually be more frequently used than reported here.

28. Nan Lin and Yanjie Bian, “Social connections and social resources.”

29. Lin, Nan, “Chinese family structure and Chinese society,” Bulletin of the Institute of Ethnology at Academia Sinica, No. 65 (Spring 1989), pp. 59–129.Google Scholar

30. During my trip to Beijing, Shanghai and Tianjin from March to May of 1992, 1 was most impressed by the increased use of guanxi for exchanging resources among friends and acquaintances. Material rewards were necessary even for relatives who were asked for help. My two hosts, both of whom were well-known sociologists in China, told me specifically what types of gifts I should give to my interviewees.

31. In two previous drafts of the article, figures for contacts’ work unit rank were mistakenly calculated by using a scale from 0 (not ranked) to 5 (ranked higher than bureau). The current calculation is based on a scale of 1 to 6.

32. For sociological research on strength of tie measures, see Granovetter, Mark, “The strength of weak ties,” American Journal of Sociology, No. 78 (1973), pp. 1360–80CrossRefGoogle Scholar; Marsden, Peter V. and Campbell, Karen E., “Measuring tie strength,” Social Forces, No. 63 (1984), pp. 482–501.CrossRefGoogle Scholar

33. Ensel, Walter M., “Sex, social ties, and status attainment,” doctoral dissertation, Department of Sociology, State University of New York at Albany (1991)Google Scholar; Lin, Nan, Ensel, Walter M. and Vaughn, John C., “Social resources and strength of ties: structural factors in occupational status attainment,” American Sociological Review, No. 46 (1981), pp. 393—405CrossRefGoogle Scholar; Lin, Nan, Vaughn, John C. and Ensel, Walter M., “Social resources and occupational status attainment,” Social Forces, No. 59 (1981), pp. 1163–81CrossRefGoogle Scholar; Nan Lin and Yanjie Bian, “Social Connections and Social Resources.”

34. A small number of the respondents failed to answer this question, and most of them did not know who their contacts were. All of these respondents used indirect ties.

35. The use of personal contacts in the employment process favouring individuals with certain characteristics has been known as selectivity bias. Studies conducted in the United States (see, e.g., Marsden, Peter V. and Hurlbert, Jeanne S., “Social resources and mobility outcomes: a replication and extension,” Social Forces, No. 66 (1988), pp. 1038–59CrossRefGoogle Scholar) have shown that the use of personal contacts in a job search is fairly random, because neither the status attainment model variables nor any network characteristics predict the use or non-use of personal ties. Lin and Bian found a similar pattern in their Tianjin study, in which the same set of predictors as in cited American studies were used. Following Berk, Richard A. (“An introduction to sample selection bias in sociological data,” American Sociological Review, No. 48 (1983), pp. 386–398)CrossRefGoogle Scholar, I included these characteristics in the models used to estimate the strength of tie to and status characteristics of the contacts.

36. The theory was first advanced in Lin's chapter, “Social resources and instrumental action,” in Nan, Lin and Peter, Marsden (eds.), Social Structure and Network Analysis (London: Sage, 1982).Google Scholar Empirical studies conducted to test its hypotheses in North America and Western Europe include Lin, Ensel and Vaughn, “Social resources and strength of ties”; Lin, Vaughn and Ensel, “Social resources and occupational status attainment”; Marsden and Hurlbert, “Social resources and mobility outcomes”; Degraaf, Nan Dirk and Flap, Hendrik Derk, “With a little help from my friends,” Social Forces, No. 67 (1988), pp. 452–472CrossRefGoogle Scholar; Wegener, Bern, “Job mobility and social ties: social resources, prior job, and status attainment,” American Sociological Review, No. 56 (1991), pp. 60–71.CrossRefGoogle Scholar The same hypotheses were tested with survey data from China and Taiwan and consistent statistical results were produced. For China, see Nan Lin and Yanjie Bian, “Social connections and social resources.” For Taiwan, see Ruimei, Xiong, Qingshan, Sun and Zhisong, Xu, “Strength of ties and job change behaviors of employees in manufacturing industries,” Sociological Journal of the National University of Taiwan, Vol. 18 (November 1986), pp. 1–24.Google Scholar

37. Wong, Siu-Lun, Emigrant Entrepreneurs: Shanghai Industrialists in Hong Kong (New York: Oxford University Press, 1988)Google Scholar, ch. 2.

38. Blau, Peter M., Ruan, Danqing, and Ardelt, Monika, “Interpersonal choice and networks,” Social Forces, No. 69 (1991), pp. 1037–62.CrossRefGoogle Scholar

39. It is methodologically correct to predict these social resource variables from respondents’ education and Party membership, even for those respondents whose fathers are the contacts, because education and Party membership determine the likelihood of contacting certain individuals for help.

40. Different results were reported by Lin and Bian in “Social connections and social resources,” using their 1985 Tianjin study. In this study, education was not a significant predictor of the contact's sector, nor of the contact's occupation. The father's education and occupation had only marginal correlations with the two social resource variables considered, which did not produce significant regression coefficients. Only the strength of tie proved to be negatively associated with the social resource variables.

41. Deborah Davis, among others, has provided consistent analyses of job mobility in urban China in a series of articles: “Urban job mobility,” “‘Skidding’: downward mobility among children of the Maoist middle class,” “Job mobility in post-Mao cities.” For earlier work, see Walder, Communist Neo-Traditionalism, pp. 69–70.

42. At the national level, there were more than 200 such systems before 1985. They were greatly reduced to about 20 major systems after a wage reform in 1985.

43. This policy was first implemented in the early 1970s, following a Maoist ideology to select for college enrolment workers and peasants (who were thought to be politically pure or “red”) rather than high school students (who were trained by intellectuals thought of as academically oriented or “white”). The policy continued through the 1980s, but with a different intent. From 1977 to 1980, it was designed to select academically capable youth who had been denied the opportunity of a college education during the Cultural Revolution (1966–1976). At the same time, the selection of high school graduates for college was resumed. From 1980 on, a large number of young workers received further training from television education centres (known as TV universities), workers’ universities, branch schools of regular universities, and various vocational schools.

44. Personal interview 37, 43 years old, a computer system analyst when interviewed in 1992, who had been sent to the countryside during the Cultural Revolution and then returned to Tianjin for a state factory job (as a manual worker). She became a maths teacher in her factory's education programme before going to Tianjin Workers University for further education on computer science.

45. See Walder, Communist Neo-Traditionalism, ch. 2; Davis, “Urban job mobility”; Nan Lin and Yanjie Bian, “Getting ahead in urban China;” Yanjie Bian, Work and Inequality in Urban China, chs. 3 and 5.

46. Davis, “Urban job mobility.”

47. Ibid., for similar but more detailed descriptions of job transfers.

48. Personal interview no. 25.

49. This question involves two separate issues about changes in occupation and workplace. Because of this measurement error, I cannot determine specifically how many respondents moved between particular work units and how many moved between occupations. I can only describe movement between and within work-unit sectors (state versus non-state), work-unit rank (6 ranks), and occupation (16 categories), based on the information of first and current jobs, and whether the respondents used personal contacts in changing their jobs.

50. My guess is that personal contacts are more heavily used in job changes through individual applications rather than through organized job transfers. Evidence for this is in the greater use of a personal contact in job changes in the 1977–88 period (82.3%), in which job changes through individual applications prevailed. Also, most job changes did occur in this period (70%), stimulated by a more open labour policy.

51. To test whether success in job changes and upward mobility for cadres, Party members, and those with high-status first occupations was due to a more effective use of guanxi by these individuals, interaction terms of these variables with the dummy variable for the use of contacts were constructed. However, simple correlations of these interaction terms with both dependent variables, job change and upward mobility were small in magnitude (all below 0.05) and statistically insignificant. These results indicate that guanxi did not work more favourably for individuals with greater political power.