Content area
Abstract
Though more serious during the economic reform than in Mao-era, corruption has not totally got out of hand of the Chinese Communist Party (CCP). It is not too rampant to control and its destructive effects appear to have been limited from impeding the economic growth. To study the "resilience" that the CCP has shown in controlling corruption, this research starts from the point that most local administrators are agent of the central governments and simultaneously principal of their subordinates at lower levels; and the most important incentive driving the local agents in the multi-layered hierarchy is their concern for career advancement.
I argue that the middle level administrators' monitoring of and involvement in corruption is strongly influenced by their prospect of further promotion, or their "promotion likelihood", which is mainly determined by officials' age, education level, local performance, and personal connections with the superiors. Their monitoring effort is lowest and corruption is most likely to be the worst when they have a mediocre likelihood of further promotion. This is because the rising stars often have some distinct advantage that others don't; and the laggards know that their chances of further promotion are nil without much advantage. On the contrary, the mediocre officials believe they have a chance if only some "extra advantage" can be introduced. It becomes worthwhile for them to generate more achievements by all means including corruption and to cultivate more personal connections, possibly by bribes, to obtain a promotion. In brief, there is an "inverse-U" relationship between officials' promotion likelihood and the degree of local corruption. Degree of corruption varies across regions according to the local administrators' career prospects.
The causal mechanism is applied to explain the real estate corruption and office-buying-and-selling cases in recent years, utilizing interview notes of fieldwork in China. I also construct an original dataset of estimated corruption based on provincial level excessive real estate investment between 1995 and 2004. Combining this data with proxies of provincial administrators' promotion likelihood, I statistically test the proposed hypotheses and some existing explanations to corruption.
