Abstract
As the Chinese Communist Party has loosened its grip in a more market-oriented economy, why have membership and the economic benefits of joining risen? We use three national household surveys over 11 years to answer this question for wages in urban China. Individual demand for Party membership is treated as an investment in ‘political capital’ that brings monetary rewards in terms of a wage premium that has risen in recent years. However, this does not explain why the wage premium is higher for the personal characteristics that reduce the probability of membership. Rationing with a scarcity value for members with those characteristics provides an explanation.
Acknowledgements
The authors are grateful for research support from the CCK Foundation and DfID (under ESCOR grant R7526) and for supporting the research and for comments received from referees, and from seminars and conferences at CERDI-CNRS, Université d'Auvergne, and the Universities of Gothenburg, Nottingham and Oxford.
Notes
1. For example, from their sample of 2,096 people in Shanghai and Tianjin in 1993, 18 per cent of whom had joined the Party, Bian et al. (Citation2001) found no-one who had withdrawn from membership.
2. This is pithily expressed in two quotes from CCP members cited by Rosenthal (Citation2002): ‘What does the Communist Party stand for now? Nothing. Stability maybe. But really no ideals at all’ and ‘There is almost no-one now who believes in the Party for its ideals.’
3. However, information on income before and after joining the Party may not be informative: people do not leave the CCP (so reducing ‘events’) and the economic benefits may flow only some time after joining.
4. The evidence is tentative: Wald tests show that the change is significant only at the 11 per cent level.
5. This result implies that there was a positive correlation between the unobservables determining wages for non-members and the unobservables determining CCP membership.
6. Strictly speaking, P is an approximation of the percentage wage differential; uses the transformation exp (P)−1 (see Halvorsen and Palmquist, Citation1981).
7. Pair-wise Wald tests indicate that almost all of these differences are significant at the 5 per cent level.
8. The results of this analysis are reported in Appleton et al. (Citation2004).