^1. Walder, Andrew G., “The informal dimension of enterprise financial reforms,” in U.S. Congress, Joint Economic Committee, China's Economy Looks to the Year 2000 (Washington: Government Printing Office, 1986), pp. 630–645Google Scholar; “Factory and manager in an era of reform,” The China Quarterly, No. 118 (06 1989), p. 254.Google Scholar

^2. For discussion of divided property rights and multilevel supervision, see Granick, David, Chinese State Enterprises (Chicago: University of Chicago Press; 1990), pp. 25–28, 243–44Google Scholar; Tidrick, Gene, “Planning and supply,” in Tidrick, Gene and Jiyuan, Chen (eds.), China's Industrial Reforms (New York: Oxford University Press, 1987), pp. 180–82.Google Scholar

^3. Yasheng, Huang, “Web of interests and patterns of behaviour of Chinese local economic bureaucracies and enterprises during reforms,” The China Quarterly, No. 123 (09 1990) pp. 431–458.Google Scholar

^4. The concept of the soft budget constraint has been used most notably to explain enterprise bargaining in socialist Hungary. See Kornai, Janos, “‘Hard’ and ‘soft’ budget constraint,” Acta Oeconomica, No. 25 (1980), pp. 231–246Google Scholar. It, along with the “web of factory interests,” has been discussed with reference to China by Walder, “The informal dimension of enterprise financial reforms,” pp. 630–645Google Scholar; Yasheng, Huang, “Web of interests,” pp. 431–458Google Scholar; Tidrick, , “Planning and supply,” pp. 198–99Google Scholar; Shirk, Susan, “The political economy of Chinese industrial reform,” in Stark, David and Nee, Victor (eds.), Remaking the Economic Institutions of Socialism (Stanford: Stanford University Press, 1989), pp. 336–38Google Scholar. A qualification is suggested by Granick, , Chinese State Enterprises, pp. 152–58Google Scholar, who argues that firms under provincial or municipal control experience reasonably hard financial constraints because lower levels of government cannot create money and are subject to the largesse of the central bank. This implies that budget constraints soften as money supply growth increases, and harden during austerity.

^5. The research design highlighted variables related to director background and factory features. Studies of single enterprises might consider micro-level variables that affect bargaining success, e.g. interpersonal skills of directors, factional networks, promotion histories, and director relations with specific leading agencies, Party organs, unions and workers.

^6. Kornai, , “‘Hard’ and ‘soft’ budget constraint,” p. 243Google Scholar, has suggested budget constraints may be softest for the biggest companies, but argues “further empirical investigations are needed in order that the hypotheses might be considered fully proven.” Passing references that associate firm size with bargaining position can be found in Fureng, Dong, “Increasing the vitality of enterprises,”Google Scholar in Tidrick, and Jiyuan, Chen, China's Industrial Reforms, pp. 54–55Google Scholar; Tidrick, , “Planningand supply,” p. 181Google Scholar; Prybyla, Jan S., “China's economic experiment: back from the market,” Problems of Communism, No. 38 (01–02 1989), p. 12Google Scholar; Shirk, , “The political economy of Chinese industrial reform,” p. 334.Google Scholar

^7. The questionnaire, in Chinese, was administered simultaneously to all delegation members and required about 45 minutes to complete. Questions for which responses required 0 to 5 ranking were explained beforehand and individual queries were addressed as they arose.

^8. On the rank equivalents of central, provincial, and county government organizations see Lieberthal, Kenneth and Oksenberg, Michel, Policy Making in China (Princeton: Princeton University Press, 1988), pp. 142–45.Google Scholar

^9. Currency conversions are calculated at the 1988 exchange rate of 3.72 yuan=US$1.

^10. A 1985 survey of 900 factories found the following age distribution of factory directors and deputy directors: 23% under 41; 74% between 41 and 55; 3% over 56. Guansan, Yang, Bin, Lin, Hansheng, Wang and Quhui, Wu, “Enterprise cadres and reform,” in Reynolds, Bruce L. (ed.), Reform in China: Challenges and Choices (Armonk, NY: M.E. Sharpe, 1987), pp. 74–76Google Scholar. These slightly younger directors appeared in a sample that included more deputy directors and more small and collective factories than mine.

^11. Chinese sources report that from 80 to 90% of factory directors have university-level education. See Yuanhu, Yang, “Dazhong xing qiye lingdaozhe jieshou xitong peixun” (“Large and medium-scale enterprise leaders receive systematic training”), Liaowang (Haiwaiban), 18 08 1988, p. 7Google Scholar; Chamberlain, Heath B., “Party-management relations in Chinese industries: some political dimensions of economic reform,” The China Quarterly, No. 112 (12 1987), p. 650.Google Scholar

^12. It must also be remembered that this study reports subjective assessments. Directors of small, highly profitable factories might under-report bargaining success to discourage the reallocation of funds to “backbone” enterprises. Or high-ranking directors of large factories might systematically exaggerate their influence. However, an objective assessment would involve defining occasions of bargaining success and determining how favourable outcomes arose from a mixture of bargaining success, autonomy and evasion. This would require a thorough understanding of relevant regulations and customary practice in each policy area and of the dynamic of innumerable interactions, together with a fine sense of Chinese modes of bureaucratic communication and expected outcomes. Relying on perceptions allows directors to weigh the extent to which bargaining affected outcomes.

^13. On imposed allocations and industrial companies acting as new bureaucracies, see Solinger, Dorothy J., “Industrial reform: decentralization, differentiation, and the difficulties,” Journal of International Affairs, No. 39 (Winter 1986), p. 115Google Scholar. On bribery, extortion and protection rackets, see Constance Squires Meaney, “Market reform in a Leninist system: some trends in the distribution of power, status, and money in urban China,” Studies in Comparative Communism, No. 22 (Summer–Autumn 1989), pp. 210–13.Google Scholar

^14. Factory output and administrative level were highly correlated (Pearson's correlation coefficient=0.73) in the factories studied. Samples with more large, section-level factories and more medium, bureau-level factories are needed to distinguish the influence of size from that of rank.

^15. If conducted again, the survey would distinguish securing from distributing compensation increases to discover if these varied independently.

^16. Many have argued that a primary objective of Chinese managers is to increase benefits and security of employees in the form of bonuses, housing, and employment for children. See Byrd, William and Tidrick, Gene, “Factor allocation and enterprise incentives,”Google Scholar in Tidrick, and Jiyuan, Chen, China's Industrial Reforms, pp. 61–62Google Scholar; Yasheng, Huang, “Web of interests,” pp. 444–46Google Scholar; Walder, , “Factory and manager in an era of reform,” pp. 249–253Google Scholar. On bargaining over bonus levels and evasion of bonus ceilings, see Walder, Andrew G., “Wage reform and the web of factory interests,” The China Quarterly, No. 109 (03 1987), pp. 36–38.Google Scholar

^17. On the early stages of implementing “tax for profits” (li gai shui), see Bachman, David, “Implementing Chinese tax policy,” in Lampton, David M. (ed.), Policy Implementation in Post-Mao China (Berkeley: University of California, 1987), pp. 119–156Google Scholar; Naughton, Barry, “False starts and second wind: financial reforms in China's industrial system,” in Perry, Elizabeth J. and Wong, Christine (eds.), The Political Economy of Reform in Post-Mao China (Cambridge: Harvard University Press, 1985), pp. 223–252.Google Scholar

^18. On negotiating tax levels and investment support, see Walder, , “Factory and manager in an era of reform,” pp. 254–260Google Scholar; Walder, , “The informal dimension of enterprise financial reforms,” pp. 633–35Google Scholar. On failure to collect central taxes in exchange for local fees, see Yasheng, Huang, “Web of interests,” pp. 452–53.Google Scholar

^19. The survey did not force directors to distinguish success reducing tax rates from success reducing the amount of income subject to the rate. Some may have focused on the fixed, nominal rate and neglected to report bargaining success shaving payments. Yet there is no convicing explanation why large, bureau-level directors would disproportionately focus on negotiating payments, while the others would focus on negotiating the rate.

^20. Arguments that link enterprise bargaining with inefficiency can be found in Walder, , “The informal dimension of enterprise financial reforms,” pp. 631, 634–36, 641Google Scholar; Shirk, , “The political economy of Chinese industrial reform,” p. 335Google Scholar; Tidrick, , “Planning and supply,” pp. 198–99Google Scholar; Prybyla, , “China's economic experiment,” p. 11Google Scholar; Swaine, Michael D., “China faces the 1990s: a system in crisis,” Problems of Communism, No. 39 (05–06 1990), p. 26.Google Scholar

^21. Ps are probabilities that Pearson's r is 0 under the null hypothesis.

^22. Directors of large-output (p<0.06) and bureau-level (p<0.05) plants were quickest to expand lateral relationships and to increase interaction with plants in other regions, most often suppliers or customers. Linked factories tendered regional and industry-wide complaints, and often coaxed local or central decision-makers to reconsider adverse decisions or excessive demands.

^23. Yasheng, Huang, “Web of interests,” p. 457Google Scholar, also argues that enterprise directors have doubts about decentralization and autonomy because they fear losing bureaucratic protection and support. On similar attitudes in socialist Hungary, see Laky, T., “Enterprises in bargaining position,” Acta Oeconomica, No. 22 (1979), p. 244.Google Scholar