In November 2012, China saw a new team of national political leaders assume office with Xi Jinping at the helm as the general secretary of the Communist Party. Earlier in the same year, Bo Xilai, who had been considered to be a promising candidate for the upcoming new political leadership, was expelled from the Party.
What explains such different fates of Chinese politicians in the absence of democratic elections? Despite it being the world's most populous country and increasingly influential in international politics, we know little about how top politicians in China are selected from among a pool of candidates. The past few decades of spectacular economic performance make this question even more important. How has the Chinese non-democratic political system selected the rulers who have been conducive to economic growth, or at least not detrimental to the development of the Chinese economy?
The previous academic research on the selection of politicians in China is divided into two camps. On the one hand, scholars such as Li and Zhou (2005) examine the determinants of the promotion of Chinese provincial leaders, a pool of candidates for future top political offices in Beijing, and find that provincial economic growth is positively associated with the promotion chance of provincial leaders. This evidence suggests that the selection of top politicians in China is based on meritocracy. On the other hand, Shih et al. (2012) provide evidence against the meritocracy view and instead find that officials are ranked higher in the Communist Party hierarchy if their connections are among top leaders, indicating that nepotism is a key determinant of the selection process.
New research: Social connections and impeccable track records
Our recent research (Jia et al 2013) attempts to reconcile these two opposing views. We analyse the biographical data of Chinese politicians together with economic statistics for 1993-2009 and find that top politicians in China are those who have both social connections to the previous generation of top politicians and an excellent track record as lower-ranked political-office holders.
Social connections to the current top leaders alone do not secure the bright future as long as the performance as a politician is not satisfactory. Excelling at the political job alone does not ensure the entry to the top echelon of politicians without social connections to them. In other words, social connections and performance are complementary as the determinants of political promotion in China.
In our study, social connections among Chinese politicians are measured by using their publicly available CVs. If a pair of politicians used to work in the same branch of the Chinese government or the Communist Party during the same period, we code them as connected. It is well known that both Jiang Zemin and Hu Jintao, the two previous general secretaries, promoted their respective former colleagues once they became the top leader of China. Our data indeed captures these connections: those formed in the Shanghai municipality government for Jiang and in the Communist Youth League for Hu. With this data, we measure whether each politician is connected to any member of the Politburo Standing Committee, the highest decision-making body in the Communist Party in charge of the promotion to top political offices in China.
To measure the performance of politicians, we follow the literature by focusing on the political leaders of 31 provinces of China (including five autonomous regions such as Tibet and four municipalities including Shanghai). For these politicians, their province’s economic growth can be used as a measure of their performance. Many of the top politicians in Beijing in the past two decades used to run the provincial governments. Indeed, Xi Jinping used to be the provincial leader in Fujian and Zhejiang, Hu Jintao in Guizhou and Tibet, and Jiang Zemin in Shanghai. Given that the legitimacy of the Communist Party as the sole ruling party of China hinges on the achievement of economic growth, it is conceivable that provincial GDP growth is the most important indicator of provincial leaders' performance among others.
We then analyse how social connections and provincial economic growth associate with the promotion of provincial leaders over the period of 1993-2009. Figure 1 shows our key finding. The horizontal axis indicates the relative growth performance after taking into account the inherent differences across provinces (e.g. coastal provinces grow faster irrespective of who runs the province) and across years (e.g. economic growth slowed down country-wide after the Asian financial crisis of 1997). Number 1 refers to the lowest third, 2 the middle, and 3 the highest. For each group, we plot the promotion rate that is also conditional on differences across provinces and years. The zero on the right vertical axis means that the promotion rate is the average. The solid line indicates the conditional promotion rate for provincial leaders connected to the Politburo Standing Committee; the dashed line for those unconnected.
Figure 1. The complementarity of connections and performance in the promotion of Chinese provincial leaders
The difference between those connected and unconnected is stark among the top-third performers. If they are unconnected to the incumbent top leaders, the chance of promotion is about the same as those who have achieved lower growth. Among those connected, however, they are a lot more likely to be promoted to top political positions in Beijing than those whose performance is the bottom two-thirds. The figure shows that the promotion rate is about nine percentage points higher for connected top-third performers than the rest. Since the average promotion rate is around 7%, it is a big difference.
You might wonder whether connected provincial leaders are assigned to fast-growing provinces. The bar graph in Figure 1 shows the frequency distribution of provincial leaders who are connected and fall into each category of growth performance (with the left vertical axis indicating the ratio to all provincial leaders including those unconnected). The figure clearly indicates that being connected does not necessarily imply high economic growth. It is true that connected and unconnected provincial leaders differ in several dimensions of observable characteristics. These differences, however, do not explain the correlation pattern observed in Figure 1. Most importantly, we do not find similar patterns of promotion for those connected to the past and future members of the Politburo Standing Committee or for those who used to work at the government branch where the current members of the Politburo Standing Committee also used to work but in a different period. Instead, it is having worked together with those who currently have the decision-making power that makes a difference.
Another possible explanation of Figure 1 is that provincial economic growth might be an indication of real connections. We only observe whether provincial leaders used to work together with the Politburo Standing Committee members. Working together does not necessarily imply a good relationship. It might be that only those actually connected are promised to be promoted and, as a window-dressing, they obtain the support from the central government to boost the provincial economy. As far as the observable strength of connections (the number of years working together, the age difference) is concerned, evidence does not support this possibility. That is, economic growth of the provinces ruled by connected politicians does not differ significantly by these measures of the strength of connections.
The remaining issue is why we see the complementarity of connections and performance in the selection of top politicians in China. We argue that connections play a role of fostering the loyalty of subordinates to top leaders. To run the national government efficiently, top politicians in Beijing are likely to prefer promoting those whose talent is proved by their performance. On the other hand, they should also prefer those who do not pose a threat to their political survival. Social connections may ensure the latter. Thus, only among those connected, performance does predict the promotion outcome. We find evidence in support of this hypothesis. The complementarity between connections and performance is stronger for those provincial leaders whose connected Politburo Standing Committee member is much older than them. Since the 1990s, China has undergone generational changes of top political leadership every ten years (with the last change being in 2012). Politicians of the same generation compete for higher office in China while different generations of officials do not. Consequently, it is plausible that provincial leaders show more loyalty to older connected top leaders.
Rulers in non-democratic countries often face a trade-off between competence and loyalty when they select officials to high office (Egorov and Sonin 2011). If competent politicians are likely to pose a threat to the power of the incumbent leaders, loyal but incompetent officials are preferred, resulting in a poor quality of government in autocracy. Our research might suggest the way China has avoided this trap. A system of job rotation and promotion within the Communist Party might have helped pairs of officials build trust by working together, allowing top politicians to choose the most competent among a pool of trustworthy subordinates with such connections.
Egorov, Georgy, and Konstantin Sonin (2011), “Dictators and Their Viziers: Endogenizing the Loyalty–competence Trade-off”, Journal of the European Economic Association 9(5), 903-930.
Jia, Ruixue, Masayuki Kudamatsu and David Seim (2013), “Complementary Roles of Connections and Performance in the Political Selection in China”, CEPR Discussion Paper, no 9523.
Li, Hongbin, and Li-An Zhou (2005), “Political Turnover and Economic Performance: the Incentive Role of Personnel Control in China”, Journal of Public Economics 89(9-10), 1743-1762.
Shih, Victor, Christoper Adolph and Mingxing Liu (2012), “Getting Ahead in the Communist Party: Explaining the Advancement of Central Committee Members in China”, American Political Science Review 106, 166-187.