Scoring Anti-Corruption Efforts

This study focuses on major, national-level reforms and campaigns that depart from routine low-level enforcement. These episodes represent the most challenging reforms for autocrats to undertake and have the clearest effect on a country’s overall level of corruption. It is not necessary for reforms to be exclusively or even primarily about corruption; in many cases, political leaders pursue corruption control within a larger reform agenda.

The first task, then, is to define and identify major anti-corruption efforts. To qualify, cases should meet three criteria. First, there should be an announced reform push with corruption control as a stated goal. Second, there should be a surge of at least 50% in corruption-related investigations from one year to the next. This can either be a surge in investigations into public officials and bureaucrats generally or, in an elite-focused campaign, a surge in just the number of elites and high-ranking officials investigated. The reasoning behind requiring a surge is that the reform effort should be distinct from routine anti-corruption enforcement, which varies dramatically across countries. Third, there should be at least .01% of all public officials and bureaucrats investigated. Or, in an elite-focused campaign, at least one hundred elites and high-ranking officials should be disciplined.Footnote ¹ If the number of total public officials and bureaucrats is unavailable for a country in a certain timeframe, I use the standard of .001 percent of the national population. If a regime had been doing nothing against corruption, investigating a small number of low-level officials might not signal a major effort, despite being a quantifiable surge. For that reason, I set minimum thresholds on investigations.

With these standards, I identified twenty-five authoritarian anti-corruption efforts globally between 1950 and 2019. Numerous other campaigns and reform measures assessed did not qualify. For example, Russia’s president Dmitry Medvedev (2008–2012) repeatedly vowed to address the country’s well-known corruption problems, and his administration established an Anti-Corruption Council (Holmes Reference Holmes2012, 235). But anti-corruption prosecutions actually declined during his presidency (Krastev and Inozemtsev Reference Krastev and Inozemtsev2013). Medvedev himself admitted in 2011 that there were “very few successes in this direction” (Reuters 2011). This study draws on a wide array of sources to score cases, including government reports and data, analyses by international organizations and U.S. intelligence, interviews with country experts, and existing scholarship (table 1; figure 1). This list of twenty-five cases is incomplete but shows that significant reform efforts are not uncommon in authoritarian regimes.

The cutoff for a successful reform is four out of eight points because in order to score at least four points, an anti-corruption effort must have successfully implemented at least one institutional reform, taking it beyond just arresting offenders and into checking or disincentivizing future corrupt behavior. In a failed effort, an anti-corruption effort takes place, but investigations are either backtracked on, abandoned, or left unsupported by the institutionalization of new or strengthened anti-corruption rules. In a successful reform, by contrast, offenders are widely investigated and disciplined and new or strengthened rules are successfully enforced. In the most successful reforms, new or strengthened rules systematically constrain even corrupt behaviors by elites and high-level officials.

How Autocrats Curb Corruption

In the nine cases in which authoritarian regimes substantially curbed corruption, I find that autocrats largely used authoritarian methods at odds with the democratic approach, and that there were some commonalities in these methods across regimes. Rather than following conventional recommendations to strengthen the rule of law, checks and balances, or media freedoms, autocrats controlled and directed campaigns from above, taking extra-legal measures and quashing grassroots anti-corruption activism as necessary. Though specific anti-corruption laws and law enforcement tactics of course varied across cases, there were overarching commonalities: the use of power centralization, top-down control and penetration, and regime propaganda.

Moreover, anti-corruption campaigns in these cases did not coincide with political liberalization or the greater use of quasi-democratic institutions by regimes overall. This does not mean that a regime being fully authoritarian or being a personalized regime predicts effective corruption control—far from it. But among successful reforms in nondemocracies, anti-corruption efforts often coincide with a rollback of liberalization, as in Xi’s case, or a weakening or elimination of quasi-democratic institutions, as in the South Korean case. One partial exception is the case in Taiwan when premier Chiang Ching-kuo began a gradual political liberalization alongside anti-corruption efforts in the early 1970s by promoting “Taiwanization” and “youthification.”Footnote ² Even so, Chiang smoothly succeeded his father and consolidated personal power over the Kuomintang regime.

Table 2 shows that many successful anti-corruption reforms led by authoritarian regimes share similar features. However, it also shows that Singapore was a deviant case. Corruption control under prime minister Lee Kuan Yew and his People’s Action Party involved a mixture of democratic and authoritarian approaches (see Vogel Reference Vogel, Sandhu, Wheatley and Alatas1989; Shang Reference Shang2002; Tan Reference Tan, Tay and Seda2003).

Figure 1 Select scoring of anti-corruption efforts

Table 2 Common features of successful authoritarian anti-corruption reform

Going beyond East Asia

Because the majority of success cases examined in this study are in East Asian countries, some observers might be tempted to see authoritarian anti-corruption as a product of Confucianism or East Asian cultural traditions more generally. It could be argued, for example, that Confucian precepts of benevolent autocracy continue to steer some East Asian leaders into taking hardline, moralistic stances against corruption. Scholars have noted that East Asian cultural traditions may shape the perception of corruption and therefore the politics of it in the region in different ways (Sun Reference Sun2001; Cameron et al. Reference Cameron, Chaudhuri, Erkal and Gangadharan2005; Mensah Reference Mensah2014).

While how corruption is combated may be influenced by culture, authoritarian anti-corruption is not a culturally determined phenomenon. First, as a relatively fixed factor, culture does not help to explain the variation in corruption control outcomes within East Asian countries over time. China, South Korea, and Taiwan all experienced sharp variation in outcomes in the twentieth century, including sometimes even under the same autocrat. Second, several autocrats in the region who led anti-corruption reforms—including Mao Zedong, Chiang Kai-shek, and Park Chung-hee—portrayed themselves as modernizers daring to break with cultural traditions in order to improve their countries, whether in a communist, nationalist, or militarist direction. And third, there are similarities in the authoritarian approaches utilized in non-Asian success cases, such as in Cuba and Rwanda, as described later.

Instead, the concentration of cases in East Asia reflects the fact that successful authoritarian anti-corruption is more likely to occur in regimes that are relatively high-performing in terms of governance and economic outcomes. Scholars have long recognized the high variance in good governance and economic growth among authoritarian regimes—e.g., South Korea versus the Philippines or Rwanda versus Somalia (Bates Reference Bates1981; Haggard Reference Haggard1990; Wintrobe Reference Wintrobe1998). Several East Asian countries, including autocracies, rose into the high-performing category ahead of many other developing nations in the twentieth century. A great deal of scholarship has sought to explain this regional “miracle,” often citing factors such as the historical strength of East Asian states, certain legacies of Japanese and Western colonialism, and the rise of broad-based revolutionary movements (Johnson Reference Johnson1962; Kohli Reference Kohli1994; Wade Reference Wade1999; Woo-Cumings Reference Woo-Cumings1999; Hui Reference Hui2005). Since East Asian cultural traditions have shaped state development and other historical factors, culture may contribute to anti-corruption outcomes indirectly in this way (Fukuyama Reference Fukuyama1995, Reference Fukuyama2011). Nevertheless, high-performing authoritarian regimes have of course also emerged in other developing regions.

Anti-corruption reforms in Rwanda have been successful under autocratic president Paul Kagame. Kagame came to power after leading the Rwandan Patriotic Front to victory in 1994 and ending the Rwandan genocide. Given deference as a revolutionary hero, Kagame brooks no opposition and rules with relatively few constraints (Khisa Reference Khisa2016). Kagame has used his power to become an “illiberal state-builder” (Jones, DeOlveira, and Verhoeven Reference Jones, De Oliveira and Verhoeven2012). Citing Singapore as a model, he has sought to establish a strong political order, create rapid growth, and clean up rampant corruption. In October 1999, Kagame began a series of high-level anti-corruption purges, including of cabinet members and longtime political allies (Waugh Reference Waugh2004, 153). To institutionalize anti-corruption norms, Kagame oversaw the establishment of the “Office of the Ombudsman, the Anti-Corruption Unit in the Rwanda Revenue Authority, the Auditor General, and the National Tender Board” (World Bank 2015). International aid and development organizations, international press, and expert analysts have praised Kagame’s “rigid ethical standards” and anti-corruption measures (Kinzer Reference Kinzer2008, 236; Global Integrity 2009; The Mo Ibrahim Foundation 2016).

Scoring Xi’s Campaign

Despite numerous studies of this campaign, a systematic assessment of its corruption control outcomes at the national level is still lacking. The authoritarian nature of the CCP regime and the political sensitivity of this campaign make it exceedingly difficult to get a full picture. Nevertheless, my proposed scoring system can help organize the available evidence about this campaign’s accomplishments.

One sign of the campaign’s effectiveness is that discipline enforcement, though imperfect, has been undeniably far-reaching both horizontally and vertically within the party and state. Top officials from every province have been disciplined, along with more than seventy SOE executives and sixty-three generals between 2012 and 2017 (Li Reference Li2019, 55). In the first six years after the Eight-point Code was announced, the party has dealt with some 250,000 suspected violations of the anti-austerity and anti-extravagance rules, disciplining nearly 350,000 people (Xinhua 2018a). There has been no shortage of low-level officials disciplined—1.34 million cadres at the township level and below and more than .64 million rural cadres in Xi’s first term (Xinhua 2018b). As a result of investigations conducted all across the country, the CCDI confiscated 20.1 billion yuan in allegedly corrupt funds between the start of the campaign in late 2012 and June 2015 (Sina 2016a). Unlike in some previous Chinese campaigns, enforcement has not suffered significant backtracking. The conviction rate is over 99%, and virtually no high-ranking official publicly implicated has been exonerated or released after being imprisoned (RFI 2016). The unprecedented length of the campaign reflects strong political will at the top and makes it increasingly unlikely that its verdicts will be overturned. In sum, the campaign earns three points in the discipline enforcement category: scope of enforcement (1); vertical reach, meaning targeting elites (1); and permanence of punishments (1).

Anti-corruption bodies, procedures, and organizational arrangements that were established or enhanced in the campaign have become institutionalized. Under Xi, the CCDI’s ad hoc inspection teams have been “strengthened and enhanced in terms of their scope, intensity, and frequency of their inspections” (Guo Reference Guo2014, 613). More than half of the “leading cadres” brought down in the campaign, the CCDI reported in 2016, were initially discovered by inspection teams (Sina 2016b). The Xi administration also created central inspection groups—a new institution—to augment the existing inspection system. Because they report to the Leading Small Group on Central Inspection Work (中央巡视工作领导小组), which was headed by Xi’s deputy and CCDI chief Wang Qishan during Xi’s first term, these groups (teams) provide the “most direct channel for central supervision of leaders in both local government and key SOEs” (Yeo Reference Yeo2016, 60, emphasis original). The CCDI has also installed disciplinary inspection teams in central state and party institutions on a more permanent basis; it reported in January 2016 that it had successfully installed 47 such teams in all 139 central party and state institutions (CCDI 2016).

The latest and greatest in corruption control institution-building is the National Supervision Commission (国家监察委员会), an overarching anti-corruption body created in March 2018 by the National People’s Congress. The NSC can police not only the wrong doing of China’s roughly 90 million party members, as the CCDI does, but also non-party state employees. It has absorbed the anti-corruption functions of several governmental agencies, including the State Council Ministry of Supervision and the Supreme People’s Procuratorate. Discipline inspection committees under the NSC can absorb 20% of procuratorates’ workforce nationwide—a massive personnel transfer (Li Reference Li2019). While the full impact of the NSC’s formation remains to be seen, this scale of institutionalization as a result of an anti-corruption campaign is unprecedented in the reform era (Keliher and Wu Reference Keliher and Wu2016).

It is clear that the campaign’s myriad anti-corruption laws and norms have substantially changed the behavior of Chinese officials and bureaucrats. As Elizabeth Economy writes, “Xi has sought to eliminate through regulation even the smallest opportunities for officials to abuse their position. Regulations now govern how many cars officials may own, the size of their homes, and whether they are permitted secretaries” (Economy Reference Economy2018, 30). Economic studies analyzing land transactions, luxury imports, car sales, and new business registration conclude that there has been success in curbing corruption in those areas (Qin and Wen Reference Qin and Wen2015; Chen and Zhong Reference Chen and (Zachary) Zhong2017; Chen and Kung Reference Chen and Kung2018). The same is true of studies that use the price discounts officials receive on housing and corporate spending on “entertainment and travel costs” as partial proxies for corruption (Chu, Kuang, and Zhao Reference Chu, Weida and Daxuan2019; Yi and Patten Reference Yi and Patten2019). Market research shows that party regulations against extravagant spending by officials in restaurants—“corruption on the tongue tips”—strongly affected the high-end dining industry starting in 2013, as well as the hotel industry (Jing Daily 2013). Government statistics show that the number of cases of misuse of public funds on dining, presents, and travel was initially large in 2013 but has been declining sharply year-on-year through 2017. If this trend is accurate, it suggests that enforcement of new rules has succeeded in deterring these behaviors (People’s Daily 2017). Overall, in terms of rulemaking, the campaign earns three points for institutionalizing reforms for at least a five-year period: reform of organs tasked with anti-corruption work (1), elimination or reform of governmental practices plagued with corruption (1), and revised laws and party rules that specifically address corrupt practices (1).

A major limitation of the campaign has been the failure to systematically enforce reforms that could address high-level corruption, such as among elites. The party’s “Rules for Disciplinary Action” (中国共产党纪律处分条例) put restrictions on the business activities of officials’ family members, which is seen as a crucial reform because so much high-level corruption is hidden in this manner. But, as Pei (Reference Pei2018, 220) explains, the party’s rules on this and many other proscribed activities remain so “vaguely defined” that they can be easily skirted, even after revisions to the “Rules for Disciplinary Action” in 2015.Footnote ⁴ Regulations have been similarly weak with regard to the country’s stock market, where corruption is common and China’s rich are getting richer. The CCDI signaled in November 2018 that it was working on new measures to address the problem. In an analysis of leaked documents, the International Consortium of Investigative Journalists (ICIJ) found “nearly 22,000 offshore clients with addresses in mainland China and Hong Kong … Among them are some of China’s most powerful men and women—including at least 15 of China’s richest, members of the National People’s Congress and executives from state-owned companies” (ICIJ 2014). Some leaked records point to problematic holdings by family members of current and past leaders, including Xi’s brother-in-law Deng Jiagui and former premier Wen Jiabao’s son Wen Yunsong (ICIJ 2014). Xi’s campaign proposes to prevent capital flight, tax evasion, and international money laundering, but international document leaks have revealed how troubled and contradictory these tasks are for China’s leaders.

A second limitation of the campaign is uneven implementation of anti-corruption reforms at lower levels of the state, such as at the township and village levels, especially in rural areas.Footnote ⁵ As China’s state media acknowledges, institutional change is “more difficult” in rural areas than elsewhere (Xinhua 2016). Corruption related to land use and expropriation continues to be common, and the party’s vast anti-poverty campaign still lends itself to misallocation of funds by rural cadres, despite some improvements to its administration (CECC 2017). The campaign’s challenges go beyond the vastness of the Chinese countryside and variation in local conditions. One key problem, as Juan Wang explains, is that the Xi administration inherited a situation of weakened state capacity at the lowest levels because of rising discontent and disaffection among village cadres since at least the early 2000s (Wang Reference Wang2017). Moreover, local governments under massive debt burdens and with weak capacity have in some cases been overwhelmed by the intrusive anti-corruption campaign and ground to a halt in “fear” and “paralysis” (Ahlers and Stepan Reference Ahlers, Stepan, Heilmann and Stepan2016).

Overall, Xi’s anti-corruption campaign involves substantial disciplinary action (3) and has enforced various institutional reforms (3), resulting in a score of six out of eight possible points. The campaign’s main limitations, which cost it two points, are failures of systematic enforcement of reforms at the extremes—at the high end among elites, and at the low end among rural officials. These problems are also cited by experts in their assessments of the campaign (Zhu Reference Zhu2015).

How do this score and the conclusion that the campaign has successfully carried out reforms stack up against existing expert opinion? Scholars who have taken on the question are split between those who conclude that the campaign has had a real if limited effect on corruption (e.g., Manion Reference Manion2016; Fang Reference Fang2017; Sun and Yuan Reference Sun and Yuan2017; Bertelsmann Transformation Index 2018; Economy Reference Economy2018) and those who do not see substantial improvements (e.g., Griffin, Liu, and Shu Reference Griffin, Liu and Shu2016; Pei Reference Pei2016, Reference Pei2018). This latter camp includes the World Bank’s control of corruption indicator and Transparency International’s CPI, both of which show only slight improvement in corruption control in China between 2011 and 2018. In the last estimate, therefore, my scoring of the campaign yields a somewhat more positive result than the majority of expert analyses.

Xi’s Authoritarian Methods

While some analysts have argued that China needs to take a more democratic approach or that the political liberalization of the pre-Xi era was promising for corruption control (Manion Reference Manion2004, 201; Pei Reference Pei2016; Ko and Weng Reference Ko and Weng2012; Fu Reference Fu, Trevaskes, Nesossi, Sapio and Biddulph2014), I find that it actually took an authoritarian approach to reduce corruption under Xi. This approach can be seen in at least three characteristics of Xi’s campaign: power centralization, including both personalization and institutional consolidation; top-down control and penetration, including bypassing certain government organs and even employing Maoist tactics; and regime propaganda, which has shut out independent media and public activism.

Power centralization in the campaign can be seen in both the personalization around Xi and the institutional consolidation under him. On personal power, official media highlights Xi’s leadership, Xi’s reforms, and Xi’s instructions for combating corruption. Outside analysts often refer to “Xi Jinping’s anti-corruption campaign,” which was not common usage for previous campaigns under presidents Jiang Zemin or Hu Jintao. Xi has led the norm-breaking prosecution of several “big tigers” since 2013, though he reportedly consulted party elders before starting the first investigation against Zhou. Some analysts speculate—reasonably—that Xi’s status as one of the “princelings” (descendants of influential party officials) gave him more personal authority to challenge other elites (Hui Reference Hui2019). Xi is also seen as more charismatic than his predecessor Hu, with his minor cult of personality and regular use of folksy aphorisms (Lam Reference Lam2015b). These personal factors are likely important, but it is also true that they are to some degree a function of Xi already having accumulated power. After all, this cult of personality was not organic, but rather created at least in part through state-controlled media.

In terms of institutional centralization, the campaign bypasses governmental and legal organs that theoretically should be indispensable in anti-corruption work, like the judiciary, to consolidate power in the party center (Minzner Reference Minzner2015, 7). The law governing the NSC, the Supervision Law, does not put the NSC within China's legal system, but rather makes it “ultimately accountable only to the CCP” (Horsley Reference Horsley2018). In the military, Xi’s campaign has punished Central Military Commission-level leaders at an unprecedented scale and enacted a reorganization of military regions that entrenched interests had blocked under past administrations (Char Reference Char, Bitzinger and Char2018). While Chinese experts who are supportive of Xi’s anti-corruption efforts reject comparisons to Mao-era political campaigns, they accept that a party center-led “struggle” against corruption represents a return to CCP “tradition.”Footnote ⁶ As a professor at a prominent Party School explained in a frank interview: “Naturally, the center has to be powerful for our political system to work.”Footnote ⁷ This comment speaks to the logic that in such a large state ruling such a large country, there may be many veto players on any national reforms.

Alongside the centralization of the campaign, the Xi administration has increased top-down control. Greater top-down control allows the campaign to penetrate organizations in which corruption is being protected and to reduce bureaucratic resistance (Fabre Reference Fabre2017). One concern in the campaign has been to penetrate and reform corruption-sheltering “independent kingdoms” (独立王国) that had taken over parts of the state (People’s Daily 2018). The party has taken several measures to enhance the CCDI’s control over provincial and lower-level discipline inspection committees, such as advancing the “dual leadership system” (Guo and Li Reference Guo and Songfeng2015). Ad hoc inspections help avoid obstructionist bureaucracies and local governments. The obstruction-evading power of anti-corruption investigators is reflected in the use of the term “paratroopers”—meaning investigators who are “dropped” from the central government into lower-level party or government units to fight corruption (Caixin 2016).

At the individual level, while enforcement has been far less violent than in some previous campaigns and torture is theoretically banned, reports have come out detailing instances of torture, coerced confessions, and suicides by officials suspected of corruption (Human Rights Watch 2016). The campaign has taught the world the Chinese abbreviation shuanggui, which is a disciplinary procedure in which suspects are held indefinitely, incommunicado, and without the presumption of innocence. As one former CCDI member explained in an interview: “Sometimes there’s no other way. I used [shuanggui] sparingly because it was extralegal … history may judge us for it.”Footnote ⁸ While the new National Supervision Law appears to do away with the practice, it in effect means that shuanggui has been normalized and even strengthened.

Moreover, Xi has promoted the use of harsh Maoist tactics in the campaign. With its inspection teams, self-criticism sessions, and public political theater, Xi’s campaign “displays the unmistakable imprint of earlier Maoist anticorruption campaigns” (Duara and Perry Reference Duara and Perry2018, 24). Though this trend has alarmed many Chinese people, including some party members, it is part of Xi’s return to “seemingly anachronistic tools” of governance (Ahlers and Stepan Reference Ahlers, Stepan, Heilmann and Stepan2016, 37). While few believe that Xi can become a second Mao, his power consolidation, party-first agenda, ideological and anti-Western rhetoric, and wide-ranging crackdown on dissent are all reminiscent of Mao (Zhao Reference Zhao2016).

Finally, Xi’s administration has used propaganda and its control over the media to build the party’s narrative about the campaign and drum up public support. A flood of official media coverage of the campaign has sought to instill several messages: that anti-corruption efforts must be led by the party, that corruption is a personal moral failing rather than a result of perverse structural incentives, and that resisting the party’s campaign is futile. The extensively-promoted propaganda campaign “Remain true to our original aspiration and keep our mission firmly in mind” (不忘初心, 牢记使命) aims to revive in party members a spirit of public service and a sense of historical mission, which theoretically will deter them from corrupt activities. To drive home the idea that corruption is wrong and corrupt officials are bad people, sentencing documents often include salacious details about mistresses and extravagant spending.Footnote ⁹ In addition, the propaganda department has helped create several relatively high-quality television shows about the issue of corruption that have been commercially successful in recent years. The miniseries called Always on The Road (永远在路上) (2016), for example, trumpeted the party’s anti-corruption successes and aired emotional confessions by officials under arrest. Propaganda can also get personal—Xi has revived the party tradition of “democratic life meetings,” which are criticism and self-criticism sessions that psychologically pressure people to confess their crimes and to report others.

Any possibility of a more democratic approach to corruption control, such as through grassroots organizing against government corruption, has been blocked. Citizen activists with the New Citizens Movement, which has campaigned for governmental transparency over officials’ assets, have been detained and harassed (Human Rights Watch 2016). Chinese journalists report a “closing space” to cover corruption, explaining that the government is reporting less information on cases, defense lawyers are warier of talking to independent journalists, and independent journalists are more careful about what they write.Footnote ¹⁰ As one Chinese anti-corruption activist explained to me, the regime has “made anti-corruption more convenient” for itself by undermining its own legal procedures with suspects. In the Hu Jintao era, high-profile defendants would appeal their cases; now, they tearily confess their crimes for documentaries used in regime propaganda. One journalist I interviewed claimed to have solid leads on corruption by more than one Central Committee member, but that she was not allowed by her superiors to investigate further.