1 See, for instance, Marcia A. Weigle and Jim Butterfield, Civil society in reforming Communist regimes: the logic of emergence, Comparative Politics, October 1992, pp. 1–23; Andrew Arato, Civil society against the state: Poland 1980–81, Telos, No. 47 (1981), and his Revolution, civil society, and democracy,: in Zbigniew, Rau (ed.), The Reemergence of Civil Society in Eastern Europe and the Soviet Union (Boulder: Westview Press, 1990)Google Scholar. Also written in this framework are Gail W., Lapidus, State and society: toward the emergence of civil society in the Soviet Union, inSeweryn, Bialer (ed.), Politics, Society, and Nationality inside Gorbachev's Russia (Boulder: Westview Press, 1989), pp. 121–147Google Scholar; Ivan, Szelenyi, Socialist Entrepreneurs: Embourgeosiement in Rural Hungary (Madison: University of Wisconsin Press, 1988); and Frederick Starr, Soviet Union: a civil society, Foreign Policy, No. 70 (Spring 1988), pp. 26–43.Google Scholar

2 For views of the mono-organization of Communist systems see Harry T., Rigby, Stalinism and the mono-organizational society, in Robert C., Tucker (ed.), Stalinism: Essays in Historical Interpretation (New York:W.W. Norton, 1977)Google Scholar, and Carl J., Friedrichand,Zbigniew, Brezinski, Totalitarian Dictatorship and Autocracy (Cambridge, MA: Harvard University Press, 1956)Google Scholar

3 the construction of corporateness in a Chinese socialist factory

3 Mayfair Yang, Between state and society: the construction of corporateness in a Chinese socialist factory, The Australian Journal of Chinese Affairs, No. 22 (July 1989), pp. 31–60, at p. 59.

4 For this view as applied to China see Martin, Whyte, Urban China: a civil society in the making? in Arthur, Rosenblum (ed.), State and Society in China: The Consequences of Reform (Boulder: Westview Press, 1992)Google Scholar; Thomas B., Gold, The resurgence of civil society in China, Journal of Democracy (Winter 1990).Google Scholar

5 Zbigniew Rau, Introduction, in Rau, The Reemergence of Civil Society, p. 4

6 Interview in Les Temps Modernes, quoted in Arato, Civil society against the state, p. 23.

7 In this article the term traditional Communist system will refer, in the case of the PRC, to the basic features of the Maoist system as it had developed prior to post-Mao reforms.

8 For general background on the Chinese work unit see Martin, Whyte andWilliam, Parish, Urban Life in Contemporary China (Chicago: University of Chicago, 1984)Google Scholar. For a treatment of the power dynamics within state-owned industrial enterprises see Andrew, Walder, Communist Neo-Traditionalism: Work and Authority in Chinese Industry (Berkeley: University of California, 1986). For a discussion of the nature of and particularly the socialist origins of the work unit see Lu Feng, Danwei, in Social Sciences in China, Fall 1989.Google Scholar

9 For an interpretation of the work unit which stresses its community aspect see Gail, Henderson andMyron, Cohen, The Chinese Hospital: A Socialist Work Unit (New Haven: Yale University Press, 1984), and Brantly Womack, An exchange of views about basic Chinese social organization: transfigured community: neo-traditionalism and work unit socialism in China, The China Quarterly, No. 126 (June 1992), pp. 313–332. For an interpretation of socialist systems as rooted in a corporatist structure see Daniel Chirot, The corporatist model and socialism, Theory and Society, No. 9 (1980), pp. 363–381.Google Scholar

10 The most systematic argument to this effect is Walder, Communist Neo-Traditionalism.

11 The term private sector will be used here as a broad reference to the non-state sector, including collective, privately-owned, foreign-owned and joint venture firms. The term privately-owned enterprise (siyou or siying qiye) will refer to privately-owned enterprises more specifically.

12 For a discussion of the Chinese work unit in historical perspective see David Strand, Political protest in Beijing: civil society and public sphere in China, Problems of Communism, May-June 1990, pp. 1–19.

13 For discussion of the changes in industrial work units in the era of reform see Andrew Walder, Factory and manager in the era of reform, The China Quarterly, No. 118 (June 1989), pp. 242-264, and Walder, Workers, managers and the state: the reform era and the political crisis of 1989, The China Quarterly, No. 127 (September 1991), pp. 467–492.

14 Parish and Whyte, Urban Life in Contemporary China

15 Lucien Pye, The state and the individual: an overview interpretation, The China Quarterly, No. 127 (September 1991), pp. 443–466. Andrew Walder, Communist Neo-Traditionalism.

16 Wolfgang, Streeck andPhilippe C., Schmitter, Community, market, and state – and associations? The prospective contribution of interest governance to social order, in Streeck and Schmitter (eds.), Private Interest Government: Beyond Market and State (London: Sage Publications, 1985), p. 21. For a discussion of collective responsibility in the context of rural China historically see Kung-chuan Hsiao, Rural China (Seattle: University of Washington Press, 1960)Google Scholar. For examples of the persistence of this pattern in the Republican era see David, Strand, Rickshaw Beijing (Berkeley: University of California Press, 1989)Google Scholar. For the pattern of local self-government in the Republican era, see Philip, Kuhn, Local self-government under the Republic: problems of control, autonomy, and mobilization, inFrederic, Wakeman andCarolyn, Grant (eds.), Conflict and Control in Late Imperial China (Berkeley: University of California Press, 1975)Google Scholar. For a more general discussion of historical change and continuities see Frederic Wakeman, Models of historical change, in Perspectives on Modern China, p. 73.

17 For a discussion of Chinas cellular social organization which focuses primarily on rural society see Vivienne, Shue, The Reach of the State (Stanford: Stanford University Press, 1988), and Shue, State power and social organization in China.Google Scholar

18 This approach shares basic assumptions with the school of historical institutionalists. For a discussion of this analytic approach see Kathleen Thelen and Sven Steinmo, Historical institutionalism in comparative perspective, in Sven, Steinmo, Kathleen, Thelen andFrank, Longstreth (eds.), Structuring Politics: Historical Institutionalism in Comparative Analysis (Cambridge: Cambridge University Press, 1992), pp. 1–32.Google Scholar

19 For a discussion of the relationship between culture and structure see Andrew Nathan and Kellee S. Tsai, Factionalism: a new institutionalist restatement, The China Journal, No. 34 (July 1995), pp. 157–192; Bruce Dickson, What explains Chinese political behavior? The debate over structure and culture (review article), Comparative Politics, Vol. 25, No. 1 (October 1992), pp. 103–118.

20 This approach is also informed by Michael Taylor's discussion of the relationship between structure, culture, and human action in evaluating social change. See Structure, culture, and action in the explanation of social change, in William Jones, Booth, Patrick, James andHudson, Meadwell (eds.), Politics and Rationality (Cambridge: Cambridge University Press, 1993).Google Scholar

21 The Stone Group, the largest company in the sample, had ,100 employees nation-wide and 1,000 in its Beijing branch. The second largest was Lian Xiang, with 800 employees in Beijing, 300 in the rest of China and 300 in Hong Kong. Four other companies had 60–400 employees. Three had less than 50. Stone and Lian Xiang were the oldest companies, having been founded in 1984. Three others were founded in 1989 and the rest in 1992 or later. Formal interviews were conducted with 25 informants. These included company owners, managers, deputy managers, middle level managers, software specialists, salespersons, and rank and file technicians. In addition to the formal interviews I gained important insights from informal discussions with friends, former employees of high-tech firms, and participants in the International Computer and High-Tech Exhibit which took place on 7–9 September 1993 at the Beijing International Exhibition Hall. Of the 11 companies five were minban firms and six were quanmin firms. Minban firms are collective enterprises established through private initiative and investment. They generally have no leading organ and are in principle autonomous in all areas of management including finance, personnel and investment. The quanmin firms in this study were generally established through the initiative of a university, research institute or other institution. Both types of firms operate for profit, are largely responsible for their profits and losses, and are autonomous in their management decision-making.

22 A manager who had joined the Stone Group shortly after its establishment explained how within the first year of operation the company was able to make enormous profits from a very small initial capital (10,000 yuan) by importing printers from Japan and selling them at a higher price (No. 4).

23 For a discussion of China's high technology zones see Erik Baark, China's new high technology development zones: the politics of commercializing technology, 1982–1992, Washington Journal of Modern China, Vol. 2, No. 1, pp. 83–102

24 Ibid. Beijing operates a development zone (kaifa qu) in Xicheng district which now competes with Haidian in attracting high technology firms.

25 For background on China's high technology sector see Richard Suttmeier, China's high technology: programs, problems and prospects, and Erik Baark, Fragmented innovation: China's science and technology policy reforms in retrospect, in Joint Economic Committee, Congress of the United State (ed.), China's Economic Dilemmas in the 1990s: The Problems of Reforms, Modernization, and Interdependence (Armonk, NY: M.E. Sharpe, 1993), pp. 546–564 and pp. 531–545 respectively.

26 The cost to the company of constructing this apartment building was reported to be around 10 million RMB (No. 15).

27 Some apartments were used as dormitories for unmarried employees. There were furthermore a certain portion of employees who had more than one family member working for the company. 500 apartments were thus able to accommodate a majority of Stone–s employees.

28 Companies sometimes rented their office space from a state work unit. The Kaida Computer Technology Institute rented office space from the Chinese Academy of Sciences at an annual rate of 50,000 RMB. The Xin Tiandi Company rented its offices from the sanitation department of Haidian district government at an annual rate of 150,000 RMB, although they were also expected to provide goods in kind and services in exchange, including computers and software products. One of Stone's retail stores rented its office space from a former state-owned vegetable store that decided it could earn more money by renting its land out than selling vegetables (Nos. 6 and 14). However, companies only occasionally bought or rented residential apartments, usually only for their top executives.

29 In 1993, it cost an average of 1,000–2,000 RMB per square metre for a company to build an apartment complex itself, through contractors, with an average one-bedroom apartment costing approximately 100,000 to 150,000 RMB. The cost of buying the same apartment was said to be approximately 250,000 to 300,000 RMB. While some companies did rent or buy a small number of apartments, they tended to do so only in the first years after their establishment. Stone had bought one floor of an apartment building in Haidian district in its early years in order to provide housing for its top-level personnel. However, the company subsequently constructed all its own housing.

30 Only fixed-asset or working capital loans were reported to be available, which do not cover housing construction, a regulation that applies to state-owned enterprises as well as foreign-owned, joint venture and private enterprises. Author's interviews, PRC, 1993. Housing now a key perk for MNCs to attract skilled workers in China, Business International, Business Asia, 20 January 1993.

31 The length of time varied, but it was often 5–6 years before a company could assume this financial burden. Companies were under particular pressure to provide housing for recent college graduates who, unlike employees who transfer from a state work unit and are often able to retain their danwei housing, can only rely on friends or relatives for housing. Some companies set up dormitory type arrangements for their single employees. Author's interviews, PRC, 1993.

32 Enterprises of all ownership types face the challenge of removing employees from company housing when the employee stops working for the company. Foreign companies which have had this problem have made employees sign contracts agreeing to relocate should they leave the company. See Housing for Chinese employees: accommodating answers, Business International, Business China, 6 September 1993.

33 According to one manager, if an employee broke his or her leg, the company would be very likely to pay for the hospital bills, and was even more likely to do so for a core (gugan) employee (No. 9).

34 The localized approach to medical coverage was further evident in the fact that the different branches of a company tended not to pool their resources but had to each cover its own employees (No. 19). Each of the 22 companies under the Zhongke Group had to pay directly for the medical expenses of its employees (No. 2). Stone also did not have a comprehensive medical insurance policy for all the employees in its various branches, but had each local branch cover its own employees (Nos. 15 and 17).

35 The employees of this company were on average very young. As a result it had lower medical costs. And the managers said they had not yet had any major hospitalizations (No.16).

36 Stone's vice-president in charge of personnel stated proudly that Stone employees could go to any hospital they wanted, see any doctor, and buy their medicine from any pharmacy and be reimbursed with, he implied, very few questions asked. He saw this freedom of choice that Stone offered its employees as a point of superiority over the medical coverage offered by state-owned enterprises (No. 15).

37 The one-time payment option saved the company money in the long run but required greater liquidity. Large companies were thus in a better position to adopt this payment method. At the time of the interview Stone was shifting from the monthly option to the one-time payment option (No. 15).

38 The cost of providing lunch was not insignificant. The price of a lunch was 5'8 yuan, a total of 100-160 RMB per employee/month, a sum which represented roughly one-fifth to one-quarter of an average salary. It is also interesting that employees generally ate lunch together in the office in a manner reminiscent of the traditional danwei. Author's interviews, PRC, 1993.

39 One Stone vice-manager explained employees, particularly ‘core cadre,2019; lose the habit of using public transportation, so that the company tries to take care of their needs by providing them with the use of a car (No.15).

40 While the purpose of the trip was said to be management training, closer inquiry suggested that it was more of a leisure trip than a professional one. Participants attended a one to two-day training seminar, but then spent much of the rest of the month going to Disney World and a host of other such tourist sites. The cost of these trips was estimated to be thousands of dollars for each participant (No. 15).

41 The ability of large companies to offer housing to employees gave them a definite advantage in the labour market. One might wonder how small companies survive. One answer is that they found alternative ways to attract and retain their skilled employees, such as to give employees a share in the company. This is one factor which helps explain why many small companies have a fuzzy ownership structure. Many of them attract skilled employees by promising them an informal partnership. According to one informant, small companies have to do this to keep their qualified staff, who would otherwise go and set up their own company or work for a larger company (No. 6).

42 For a description of the welfare system as it operated in the Maoist era see Whyte and Parish, Urban Life in Contemporary China.

43 In 1993 an average one-bedroom apartment in the Haidian vicinity was said to cost 250–300,000 RMB. While the cost of buying a house in industrial nations tends on average to be three or four times the buyer–s annual salary, in China it is between 10 and 20 times. China's housing market continues to be dominated by the construction of luxury apartments. See Housing shortages and reform: China's first fully commercial housing scheme, International Executive Reports, Ltd., East Asian Executive Reports, China Section, Vol. 16, No. 10 (15 October 1994), p. 14.

44 In 1983, 25.5% of all housing was owned by government, 17.0% was owned by individuals, and 57.5% was owned by enterprises and other work units. See Yang Lu and Wang Yukun, Zhufang gaige: lilundefansi yu xianshide xuanze (Housing Reform: Reflections on Theory and Choices in Practice) (Tianjin: Tianjin renmin chubanshe, 1992), p. 89. See also Zhufang gaige yufanggaijinrong (Housing Reform and the Finance of Housing Reform) (Beijing: Zhishi chubanshe, 1992).

45 The percentage of total housing investment made by local government has continued to decline relative to spending by enterprises. For instance, from 1986 to 1988 the proportion of investment made by Shanghai municipality declined from 27.2% to 17.3% of the total, a pattern characterizing most large cities. See Zhufang gaige, p. 89. In 1990, 326,790,000 sq. metres of housing was owned by public housing departments, in contrast to 1,185,120,000 sq. metres owned by work units. See Zhufang gaige, p. 89.

46 For documentation of managerial abuse which has developed in the reform era see Anita, Chan, PRC workers under capitalism with Chinese characteristics, China Information, Vol. 5, No. 3 (Spring 1991), pp. 75–82Google Scholar. See also Keith, Forster, The Wenzhou Model for economic development, China Information, Vol. 5, No. 3 (Winter 1990–91),pp. 53–64.Google Scholar

47 During interviews at company headquarters I observed on a number of occasions employees calling or coming to a manager to inquire and lobby for their housing and other benefits. Author's interviews, PRC, 1993.

48 For a description of the fenpei allocation system as it applied to housing see Yang Lu, Zhufang gaige.

49 Because the fenpei system does not conform to current government housing policies, managers were generally hesitant to use the term fenpei, which has strong connotations of the old system used by the danwei.

50 Ironically, these schemes were often presented as fulfilling the goals of housing reforms. For a general discussion of the aims and goals of housing reforms see Yang Lu, Zhufang gaige, pp. 173–203. On problems in the implementation of housing reforms see Cities plagued by shortage of housing, International Herald Tribune, 30 May 1994, and China phases out welfare-oriented housing system, Agence France Presses, 19 June 1995.

51 Housing shortages and reform: China's first fully commercial housing scheme, International Executive Reports. Individuals were reported to be able to get a mortgage loan from a bank, but only if their company acted as a financial guarantor. And one manager said his company was under no obligation to do this for an employee, and would be even less likely to do so if they thought the employee was going to leave the company (No. 3).

52 This was the case at the Lian Xiang company, which docked 30% of employee wages (Nos. 1 and 3).

53 Some companies stipulated that any employee who sold an apartment they had bought from the company would have to turn over any profits from the sale to the company. This has also been reported in the press. See Housing shortages and reform: China's first fully commercial housing, International Executive Reports.

54 A manager at the Beida Fangzheng Company said that most employees were not really sure how much they had accumulated in their individual accounts (No. 16).

55 The fact that health care is relatively inexpensive in China made this possible, and was further facilitated by the young average age of the workforce in high-tech firms. The manager in charge of personnel at Stone said that it only cost the company 360 yuan per employee per year to cover 100% of their medical expenses (No. 15).

56 The owners/managers of the Kaida Company, a small software company, were unusual in their commitment to a more equitable, non-discretionary, approach to the provision of medical coverage. This company set up a medical fund into which employees and the company contributed. In an effort to have medical coverage disbursed in an impartial manner the owner decided that the funds should be managed by the company's employees' union and the employee and staff representative committee. The approach taken by this company may have been related to the fact that its owner had spent many years working in Japan, spoke Japanese fluently, and had consciously adopted what he saw as a more modern style of management (No. 14).

57 It was difficult to confirm how the retirement system worked in practice since most high-tech firms had not been around very long, had a very young workforce and had not yet had anyone retire.

58 For a description of the distribution of housing in state work units see Corinna-Barbara Francis, The Paradoxes of Power in the Chinese Workplace (Ph.D. dissertation, Columbia University, 1993), especially ch. 6.

59 Employee dossiers were often placed with local personnel exchange centres (rencai jiaoliu zhongxin), allowing local authorities to monitor the hiring and firing of all the employees by local companies. Author's interviews, PRC, 1993.

60 This delegation of responsibility to the enterprise helps explains the continuation of paternalistic approaches to personnel management in non-state firms. Because of their responsibilities for the behaviour of their employees, enterprises have an incentive to regulate employee behaviour through disciplinary and normative means. The Stone company, for instance, made job candidates go through a rigorous personality test as part of their job application. And it required employees to go through a training programme in ethical behaviour, in which they learned the company code of ethics and Stone company culture, including how to be a good employee and an ethical individual (Nos. 15 and 17).

61 Andrew Arato and Jean Cohen have made a similar critique of the East European civil society theories. See their Civil Society and Political Theory (Cambridge, MA: MIT Press, 1992).

62 A number of scholars have stressed this point. Dorothy Solinger argues that the shared interests between individual entrepreneurs and bureaucrats is such as to make it difficult to differentiate clearly between them. See Dorothy, Solinger, Urban entrepreneurs and the state: the merger of state and society, in Arthur Rosenblum, (ed.), State and Society in China: The Consequences of Reform (Boulder: Westview Press, 1992). For similar views see also David Wank, Private business, bureaucracy, and political alliance in a Chinese city, The Australian Journal of Chinese Affairs, No. 33 (January 1995), pp. 55–71, and Yialing Liu, Reform from below: the private economy and local politics in the rural industralization of Wenzhou, The China Quarterly, No. 130 (June 1992).Google Scholar

63 For a similar view see Strand, Political protest in Beijing.