Using data-science to evaluate whether Xi Jinping's anti-corruption sweeps were really about consolidating power

China-watchers observed the rise-and-rise of Chinese premier Xi Jinping with caution and sometimes alarm, but also held out some hope that despite his authoritarian tendencies and thin skin, Xi was genuinely committed to rooting out the rampant corruption that has plagued the country since its rapid industrialization under Deng Xiaoping: the creation of an untouchable elite and a hereditary princeling class immune to civil justice; looting by respected members of the business community; and a sense that the looters are exfiltrating their money, bypassing currency controls, and stashing the booty in apartments overseas, fueling both the Bitcoin and real-estate bubbles worldwide.

Xi, after all, has pretty impeccable revolutionary credentials, having marched with Mao and then serving seven years in a forced-labor camp during the Cultural Revolution, sleeping on a bed of bricks in a cave, shitting in a bucket, and living on thin rice-gruel.

But when Xi secured a second term and embarked on a purge of 'corrupt' elements, many worried that he'd learned the wrong lessons from the Cultural Revolution, and had seized on the pretense of corruption to clear the political ranks of his challengers. The fact that an early arrest included Bo Xilai, the charismatic (and very, very corrupt) official seen as a potential successor to Xi didn't assuage the fears -- and then when Xi delighted President Trump by effectively declaring himself Chinese President For Life, people really started to worry.

Were they right to worry?

A new paper called Personal Ties, Meritocracy, and China’s Anti-Corruptionby USF economist Peter Lorentzen and NUS political scientist Xi Lu (presented at the University of Chicago's East Asia Workshop on Politics, Economy and Society) takes a fascinating, data-driven approach to answering this question.

The Chinese Communist Party (which conducted the purges) has been extraordinarily forthcoming in the data they published on the corruption cases they prosecuted, and the Chinese press has taken the convictions as leave to do deep and excellent investigative work on the bosses who have been brought low, departing from the received wisdom of, 'it’s okay to swat flies, but don’t hit a tiger.'

The authors cross-reference the legal data with the investigative data and then again with economic data from each region, reasoning that officials who were promoted after their areas of responsibility had declined economically were likely the recipients of patronage appointments, while those who were promoted after gains in their spheres of responsibility were likely the recipients of bona fide promotions.

From all this data, the authors ask some really interesting questions, like 'Were the close associates of the party leadership spared from investigation?' and 'When an official is convicted or corruption, did the data from the areas they oversaw support the conclusion that they were not good at their jobs?' and 'When a big boss was targeted by investigators, was that boss's immediate circle also targeted (which would suggest a decapitation strike against a political rival), or was the purge limited to officials against whom the data supports a charge of corruption).

The conclusion is mixed. Xi's own personal associates and those in his power structure were reportedly completely spared from any investigation (so either he protected them or they were uniformly honest and upright), but the rest of the politburo was not so lucky, with investigations ranging almost uniformly through China's party power structure. High profile corruption convictions, like Bo Xilai, appear to have been legitimate anti-corruption actions and not attacks on a political rival, and overall, the anti-corruption effort mostly targeted genuinely corrupt officials.

So maybe a summary would be, 'Xi took on an honest effort to purge the part of corruption wherever it was to be found, unless it was to be found in his immediate circle; the anti-corruption effort therefore did remove a large number of corrupt officials and mostly spared innocents from politically motivated attacks, but it also allowed Xi to consolidate power by removing key rivals' lieutenants, while leaving Xi's social graph undisturbed.'

Our analysis helps us evaluate two competing interpretations of Xi Jinping’s wide-ranging corruption crackdown. One interpretation is that it is largely a sham exercise masking an attempt to destroy the support networks of political competitors and consolidate power in one set of hands, and undermining the increasingly meritocratic and pragmatic governance introduced under Deng. The other interpretation views Xi’s crackdown as reaffirming the party’s commitment to improving governance within the framework of a Leninist authoritarian hierarchy. In this interpretation, the increasingly egregious and extravagantly public corruption of the post-reform era had eroded the regime’s support and hampered its ability to achieve its policy goals. In a time when China could no longer rely on the easy gains to be had from its initial process of economic reform and opening up this change had become imperative.

This study finds support for both claims. In section 2 we showed that local governments perceived as corrupt by private firms tended to have higher rates of indictments. We also showed that the largest network of indicted officials was not centered around Xi’s most likely competitor for party leadership, but rather around three other high-level leaders who were already on their way out of power. In section 3 we offer an explanation for this targeting. In the provinces where these three tigers were most influential, promotions appeared substantially less meritocratic than elsewhere, and instead appeared to be a reward for corruption. Moreover, we find that individuals who appear to have been promoted for the wrong reasons were much more likely to have been were more likely to later be indicted in the crackdown. Finally, in section 4, we find evidence that Xi Jinping protected individuals close to him, while his fellow members of the Politburo Standing Committee were unable to do the same.

Personal Ties, Meritocracy, and China’s Anti-Corruption [Peter Lorentzen and Xi Lu]

(via Naked Capitalism)