INTRODUCTION
Thanks to investigative journalists, it is increasingly hard to deny that corruption abounds in many autocratic regimes. Explosive reports have uncovered nondemocratic leaders owning private jets, enormous foreign bank accounts, and palaces in many of the world’s most luxurious destinations. Academic work exploiting micro-level data has helped fill out the picture, tracing how elites build illicit fortunes, such as by taking bribes (McMillan and Zoido Reference McMillan and Zoido2004), helping companies evade regulation (Rijkers, Baghdadi, and Raballand Reference Rijkers, Baghdadi and Raballand2017), and profiting from state contracts (Mironov and Zhuravskaya Reference Mironov and Zhuravskaya2016). Yet beyond helping themselves financially, we know little about how opportunities to engage in corruption affect how elected officials carry out the remainder of their official responsibilities. Do corrupt leaders govern differently?
This article develops a simple theoretical framework and exploits new micro-level evidence from a prominent electoral autocracy—Russia—to help shed light on this question. The focus is on legislators, who not only have relatively well-defined, measurable responsibilities, but through their positions also regularly express political preferences, including potential opposition to the regime. First, I argue that the desire to abuse public office for private gain leads elected officials to shirk their responsibilities, devoting more time to making money from their positions than showing up for votes and sponsoring bills. Next, deputies focused on self-enrichment often generate a stockpile of compromising information (or “kompromat”) that regimes can use against them lest they fall out of line politically. Voting more often with regime priorities helps provide cover for opposition deputies to abuse their office for financial gain. Finally, I argue that corrupt officials have shorter careers in legislative office. On the supply side, parties are concerned about the reputational risk of their corrupt members staying too long in office, while the risk of being caught stealing shortens deputies’ time horizons. Rotating out deputies and sharing corruption opportunities across a broader swath of the elite helps ensure party viability over the long term.
To test these arguments, I analyze data on all 1,034 parliamentary deputies working in Russia’s parliament, the State Duma, from 2007 to 2021. To measure individual deputies’ hidden wealth, I use information from annual financial disclosures, which provide surprising details about the income, real estate, and transportation assets of these elected officials. Applying techniques from investigative journalism and forensic economics (Braguinsky, Mityakov, and Liscovich Reference Braguinsky, Mityakov and Liscovich2014), I identify deputies who either failed to disclose income or assets by cross-referencing their disclosures with new datasets on luxury car ownership and usage. Overall, 24% of deputies during the period had either hidden income or assets, as uncovered by this measure. I term this subset “kompromat deputies,” building on the idea that by lying on their financial disclosures, these deputies are not only very corrupt but also compromised politically and distinctly vulnerable to investigations into their corruption.
My analysis uncovers three main findings. First, the measure of kompromat correlates with greater shirking of deputy responsibilities. Kompromat deputies are less likely to show up to roll-call votes, propose legislation, and ask questions during parliamentary debates. Personal enrichment comes at the expense of their normal governing responsibilities.
Second, I find that kompromat deputies display stronger loyalty to the regime, even controlling for party affiliation. Loyalty here is measured as the frequency that deputies support bills initiated by the government, as well as using ideal point analysis. Importantly the correlation between being compromised and regime loyalty is strongest for members of the systemic opposition, a designation given to three parties formally outside the regime. These findings reveal a mechanism that regimes use to co-opt their rivals: opposition deputies exchange self-enrichment opportunities for regime support. The regime looks the other way on personal corruption so long as the opposition supports its legislative priorities. Analyzing potential mechanisms, I find that kompromat deputies are much more likely to lobby on behalf of the security services, rather than corporate interests, suggesting that they are trading political favors for political cover.
Finally, I analyze how engaging in politically compromising corruption affects the career trajectories of deputies. Kompromat deputies are roughly 20% more likely to leave office after a single convocation, and both the ruling party and the opposition see greater turnover among this group of deputies. There is little evidence that deputies themselves fear punishment either by voters in direct elections or by law enforcement. Instead, the results point to how both parties and politicians face incentives to accelerate the revolving door in and out of autocratic institutions.
This article makes several contributions to our understanding of how corruption operates in nondemocracies. First, it is among the first papers to calculate the governance costs of allowing elites to pursue self-enrichment at high levels. Previous work has uncovered the rich payoffs political elites can reap from joining authoritarian institutions, as well as some of corruption’s aggregate economic consequences (Ferraz, Finan, and Moreira Reference Ferraz, Finan and Moreira2012; Olken Reference Olken2007). This article goes further by showing how an interest in corruption changes legislative and voting behavior, adding to work how the exploitation of political office can hamper government functionality (Weaver Reference Weaver2021).
It also provides new micro-level evidence of co-optation under autocracy, showing how members of the opposition trade personal gain for loyalty to the regime (Kavasoglu Reference Kavasoglu2022; Reuter and Robertson Reference Reuter and Robertson2015) and ruling parties use compromising material to maintain elite cohesion (Boix and Svolik Reference Boix and Svolik2013; De Mesquita et al. Reference De Mesquita, Smith, Siverson and Morrow2005; Hollyer and Wantchekon Reference Hollyer and Wantchekon2015). Attempts to weed out corruption may threaten a leader’s support base as well as breathe new life into formerly compliant political institutions. Relatedly, it joins an emerging body of scholarship on how autocratic legislatures work (Gandhi, Noble, and Svolik Reference Gandhi, Noble and Svolik2020), in particular work showing how outside interests affect voting in the Russian Duma (Chaisty Reference Chaisty2013; Dasanaike Reference Dasanaike2022). The results here echo previous work showing how transparency in authoritarian regimes can lead deputies to reduce their participation and become more vulnerable to punishment from above (Malesky, Schuler, and Tran Reference Malesky, Schuler and Tran2012). By demonstrating that political elites under autocracy are concerned with their reputations and accusations of corruption, it also relates to recent work on informational autocrats (Guriev and Treisman Reference Guriev and Treisman2019).
The article finally extends the wealth of studies that use the financial disclosures of public officials to track self-enrichment in office.Footnote 1 Disclosures are the among the most common anti-corruption reforms worldwide (Djankov et al. Reference Djankov, La Porta, Silanes and Shleifer2010), but because officials self-report their wealth, they can still hide or underreport the fruits of their illicit activities. Therefore, relying exclusively on self-reported financial disclosures creates a number of empirical challenges that can bias the measurement of corruption. By applying methods for uncovering hidden earnings as well as validating disclosures against external datasets, this article contributes to the arsenal of forensic economics tools that are critical to identifying the incidence of corruption in hard-to-study political settings (Sequeira Reference Sequeira2012; Zitzewitz Reference Zitzewitz2012). As the Russian case shows, scholars should pay attention to both official and unofficial income earned while in office to understand how corruption operates.
CORRUPTION IN AUTOCRATIC LEGISLATURES
To explore how corrupt officials might govern differently, I narrow the focus to the legislative branch. In contrast to elected bureaucrats, legislators are more likely to enjoy the autonomy to express a wider range of political views and behavior. The nature of their public, individual-level responsibilities allows us to observe not only how they perform their jobs but also how corruption shapes their relations with the broader autocratic regime. Legislators also have ample opportunities to self-enrich while in office. For example, deputies can sell political favors to interest groups and wealthy individuals (Weschle Reference Weschle2022), often in the form of sponsored legislation. In Russia, this selling of access mainly takes the form of special “deputy requests” that can be used to order bureaucrats to investigate and pressure rival economic interests. Other legislators may exploit their political independence to extract spoils from the incumbent government (Reuter and Robertson Reference Reuter and Robertson2015). Co-opting the opposition is thought to be a hallmark of many competitive authoritarian regimes, with the incumbent government organizing lucrative payouts for legislators outside the ruling party in return for their support for the regime. Finally, some legislators may take advantage of their rulemaking powers to pad their own pocketbooks without making agreements with third parties, such as by passing rules to help connected companies, securing employment for family members, or exploiting access to privileged information, key ministers, and government contracts (Blaydes Reference Blaydes2011).
First, I argue that the pursuit of personal financial interests distracts many legislators from their official duties. In other words, corruption leads to shirking. Catering to interest groups can require extensive negotiating and bargaining outside of the physical legislative institutions in order not to arouse suspicions. Legislators pushing their own businesses’ interests may also still have one foot firmly in the private sector and allocate a smaller percentage of their already scarce time and resources to their political responsibilities. Work on the EU, for example, finds that moonlighting politicians, those that work outside jobs, exert less effort on their official duties (Arnold, Kauder, and Potrafke Reference Arnold, Kauder and Potrafke2014; Staat and Kuehnhanss Reference Staat and Kuehnhanss2017). The returns to corruption may be considerably higher than those offered by outside positions, further pulling legislators from their political duties. This results in greater absenteeism and less time devoted to legislating.
Hypothesis 1 Corrupt legislators will miss more votes, propose fewer bills, and participate less actively in parliamentary discussions.
Autocratic regimes regularly punish dissent among elites, especially those in top political institutions. Challengers are repressed and officials that dare criticize the regime often meet at best early exits from office and at worst criminal charges. Several recent anti-corruption reforms are suspected of providing cover for regime efforts to enforce strict loyalty among elites. For example, in China, the so-called Tiger and Flies anti-corruption campaign have allowed the Communist Party to weaponize corruption charges and purge certain factions (Lorentzen and Lu Reference Lorentzen and Lu2018). Similarly in the wake of the 2011–12 anti-regime protests in Russia, opposition-oriented Duma deputies came under political pressure from the Kremlin. In the most striking case, opposition leader Gennady Gudkov lost his legislative seat in 2012 on accusations of continuing to work in the private sector while in office, an illegal practice that is widely ignored for members of the ruling party.Footnote 2
Politicians determined to exploit their positions for personal gain face a trade-off: the more corrupt they become, the more risky it is to oppose the regime. Working outside the regime’s red lines opens up deputies to criminal prosecution, which otherwise might be overlooked if they professed sufficient loyalty. This trade-off lies at the heart of the kompromat strategy for managing elite defections (Darden Reference Darden2008). Regimes investigate, monitor, and sanction corrupt acts committed by elites, using compromising information to threaten those that step out of line politically. In Russia, the government has targeted surveillance systems to monitor elite loyalty. Opposition members interested in politics solely for personal financial gain may ideologically converge toward their ruling party counterparts in order to avoid the wrath of the regime.
Hypothesis 2. Corrupt legislators will be more likely to vote with regime priorities.
It is important to distinguish corrupt legislators from other legislators who directly represent special interests, such as corporations, while in office. In Russia, this latter group has received considerable scholarly attention; for example, Dasanaike (Reference Dasanaike2022) and Noble (Reference Noble2020) show that deputies with business experience more often defect from their parties. Chaisty (Reference Chaisty2013) also finds deputies with ties to specific industries introduce more legislation related to these sectors that they might personally benefit from. Yet corrupt deputies differ in the way they violate established laws in order to self-enrich. Whereas informal lobbying of business interests in Russia is an open secret and often times promoted by the regime, corrupt deputies sell access to politics and therefore must hide their connections and illicit activities. As the results show below, these deputies also transact with interest groups besides large corporations. This illegality creates more significant legal liabilities that shape their legislative behavior.Footnote 3
Finally, the risks involved in selling political access to outside interests result in corrupt legislators leading shorter careers in elected office. The first reason relates to the problems deputies face in convincing parties to maintain their endorsement and affiliation. As gatekeepers to the ballot, parties fear the reputational costs of embarrassing corruption scandals, such as deputies being exposed as living luxurious lifestyles or auctioning off political favors to the highest bidder. Indeed in other settings, there is evidence that corrupt legislators are more likely to see their careers stalled by party elites who block their upward trajectories (Paschall, Sulkin, and Bernhard Reference Paschall, Sulkin and Bernhard2020). Ambitious legislators are wise to keep their corrupt behavior better hidden in order to advance their careers. Although corruption obviously exists at the highest ranks of autocratic regimes, we should expect corruption investigations to hinder rather than accelerate career advancement.
At the same time, corrupt politicians themselves may have shorter time horizons. As argued above, partaking in egregious self-enrichment creates its own set of legal and reputational risks for the individuals involved. Corrupt elites must be mindful that their political behavior sufficiently compensates for their rent-seeking behavior. The intense scrutiny of top politicians thus creates incentives for maximizing enrichment over a shorter term in office, before leaving for safer pastures farther from the eye of law enforcement, the media, and the general public. Legislators, in particular in federal systems, may not see reelection as of primary importance, as subnational office or other posts within the executive branch offer similar opportunities to make money based on previous parliamentary experience (Samuels Reference Samuels2003).
Hypothesis 3. Corrupt legislators will be less likely to remain in elected office.
The result is a type of revolving door under autocracy. The costs of engaging in overt corruption generate rotation in and out of office, as parties want to reduce reputational damage and elites face a trade-off between the financial opportunities of higher office and the scrutiny that accompanies it. This turnover occurs, however, without real accountability, as few elites at the top face actual punishment for their self-enrichment.
DATA AND METHODS
To test whether corrupt legislators govern differently, I examine the case of the Russian Duma, a multiparty, competitive authoritarian parliament with 450 members elected to roughly 5-year terms. I collected data on all 1,034 deputies over three convocations (2007–21). Background data on each individual come from official biographies and the Central Election Commission.
During this period, four parties achieved representation in the Duma, including the ruling party United Russia (UR), and three opposition parties: the Communist Party of the Russian Federation (KPRF), the Liberal-Democratic Party of Russia (LDPR), and Just Russia (SR). Even as President Putin has concentrated power in the executive, the Duma still serves as an “elite battleground” for both party and special interests to bargain, negotiate, and compete for policymaking influence (Noble and Schulmann Reference Noble, Schulmann and Treisman2018). A substantial proportion of bills are amended during the legislative process, even those proposed by federal ministries (Krol Reference Krol2021). In that respect, Russia closely resembles a number of other authoritarian states where special interests mobilize and lobby for policy change (Grömping and Teets Reference Grömping and Teets2023).
However, there are no official laws regulating lobbying, and much of the influence channels operate discretely. A recent civil society report suggests that covert lobbying expenditures in the Duma surpassed $363.4 million in 2010, with the cost of ensuring the passage of a contested law exceeding $1 million (Basmanova, Berezovskaya, and Tel’nova Reference Basmanova, Berezovskaya and Tel’nova2019). Special interests also purchase hundreds of “deputy requests” a year, a powerful tool whereby deputies can direct the activities of bureaucrats. Even though deputies in Russia are formally banned from earning outside income, a host of corruption scandals suggests that these access-selling activities can make a deputy seat very lucrative.Footnote 4 Parliament holds such a financial payoff that candidates have been caught spending millions of dollars in order to secure spots on party lists in advance of elections.Footnote 5
Detecting Corruption Using Financial Disclosures
Under an anti-corruption campaign which begun in 2008, the Russian government began requiring that top officials file extensive financial disclosures each spring. Both elected and appointed officials must declare all income, expenditures, bank accounts, company shares, real properties, liabilities, and transportation assets for themselves and their immediate family members (spouses and dependent children). As Supplementary Table A2 shows, most of this information is classified based on privacy grounds, available only to law enforcement authorities working to combat corruption. But later amendments have required that a small part of every official’s disclosure be released to the general public online (see Supplementary Table A3). Officials who do not comply with the disclosure rules face a number of sanctions, from removal from office up to criminal prosecution (Szakonyi Reference Szakonyi2021). An example disclosure in original Russian and translated into English can be found in Supplementary Figures A1 and A2.
In cooperation with Transparency International-Russia’s (TI-R) Declarator project, I collected all available annual disclosures for Duma deputies (Szakonyi Reference Szakonyi2024).Footnote 6 Most deputies only began filing disclosures in 2010, which I use as the starting year for the sample, up until 2021, the last year of the 7th convocation. Based on each document, I tabulated reported income, the number of real estate assets, and the make and model of all cars for both the deputy and their family members. Deputies are generally compliant with disclosure rules. Of the 5,752 deputy-years in the dataset (deputies enter the dataset each year they were present for at least one vote), disclosures were filed in 4,646 (81%). The main exception were lame duck deputies failing to file in the last year in office.
Based on their disclosures, deputies in Russia are very wealthy. The median deputy earned roughly $103,000 per year (at an exchange rate of 50 rubles to the U.S. dollar), roughly six times the average salary of $18,000 for residents of Moscow. Russia’s parliament has historically been a haven for the rich and famous, boasting elite athletes, movie stars, musicians, and some of the country’s biggest businesspeople (Szakonyi Reference Szakonyi2020).
I create two red flags for identifying corrupt deputies using the disclosure data. Although some deputies may not be telling the truth when filling out their forms, these lapses in accuracy hold real value for investigators trying to identify evidence of corruption. Omissions and inaccuracies on disclosure forms have become a critical anti-corruption tool for journalists and law enforcement around the world, including by Russian oversight agencies.Footnote 7 Indeed, deputies who lie on their forms may be not only hiding illicit activity, but they may be the ones most concerned about their corrupt behavior being exposed. But uncovering false or incomplete information in the disclosures requires cross-referencing them with external registries, a difficult task in most settings but possible in Russia.
The first red flag captures whether a deputy failed to disclose any luxury cars owned or driven. I focus on cars for both practical and theoretical reasons. First, cars are the only asset class where external registries are available to the public; contemporary data on income and real estate assets held by Russians are not available at scale to researchers. Instead, information on car ownership is available from the Russian Union of Auto Insurers (RCA) which hosts an online portal for identifying who owns any car in Russia based on its 17-digit vehicle registration number.Footnote 8 Because querying this database comes at considerable cost, I had to narrow the collection to the top 19 luxury car brands under the assumption that deputies who hide expensive cars are more likely to be earning illicit income through their position.Footnote 9 Using this portal, I built a panel dataset of the individual owners and drivers (and their birth dates) of 2,742,113 luxury cars in Russia from 2011 to 2019. Since most deputies spend the majority of their time in Moscow, I then supplement the insurance data with leaked registration data from the Moscow and Moscow Oblast Traffic Agencies (GIBDD), which cover 2010–21. The GIBDD registry has been used extensively by economists to track hidden earnings, tax evasion, and traffic violations (Braguinsky and Mityakov Reference Braguinsky and Mityakov2015; Braguinsky, Mityakov, and Liscovich Reference Braguinsky, Mityakov and Liscovich2014; Mironov Reference Mironov2015) and is freely available online. More information on how this measure was created is found in Supplementary Section B.
Of the 969 deputies serving from 2010–2021, 28 (3%) owned 38 luxury cars that did not appear in their disclosures. The relatively small number of deputies hiding these luxury assets suggests that they are aware of the relative ease of authorities verifying their car ownership.Footnote 10 Note that the insurance data allow me to measure missing cars both owned and driven by deputies (for example, those leased or registered in a relative or chaffeur’s name), giving a more complete picture of driving activity.Footnote 11
The second red flag captures whether a deputy failed to disclose any income. Registration requirements make cars harder to hide, and Duma deputies may believe that they can more easily shield income from law enforcement authorities (for example, through undeclared or offshore bank accounts). Building off of Braguinsky, Mityakov, and Liscovich (Reference Braguinsky, Mityakov and Liscovich2014), this red flag uncovers hidden wealth by calculating the ratio between the value of cars driven and a deputy’s officially reported income. Deputies may be openly driving cars that on article they are unable to afford. Anti-corruption activists in Russia have used this innovative approach to great success. In 2018, Alexey Navalny’s Anti-Corruption Foundation revealed that Duma deputy Leonid Slutsky drove two Bentleys and a Mercedes-Benz on an official annual income of roughly $30,000.Footnote 12 Deputies who own luxury cars whose value far exceeds their official income may be attempting to hide bribes or illegal side payments.
To calculate the ratio between income and car values, I first assigned make and models to every car owned by a deputy and their family members. I then scraped the for-sale listings on the website of Russia’s largest automobile marketplace (http://www.auto.ru) several times from May to August 2021.Footnote 13 Applying the new car premium and depreciation table calculated by Braguinsky, Mityakov, and Liscovich (Reference Braguinsky, Mityakov and Liscovich2014), I backed out the value of each car at the time it appeared in a deputy’s disclosure.Footnote 14 To give an example, the mean price of a 2012 Honda Civic for sale in 2021 was 827,500 rubles (roughly $12,000). For a deputy who owned that car in 2015, its value would be set at 1,507,803 rubles, or roughly $21,500.
The measure of hidden earnings is the ratio of the imputed market value of all the cars disclosed by the deputy and family divided by the sum of all family income that year. Overall, 207 deputies (20%) and their families drove cars that on average were worth more than their entire family’s annual income. I dichotomize this ratio in order to combine it with the first red flag for a more complete measure of corruption; Supplementary Table D2 shows robustness checks using just the continuous ratio. I also include as a control an indicator for the eight deputies who had taken out a loan to purchase their vehicle; this measure is described in more detail in Supplementary Section B. Not only are loans rare among this wealthy subpopulation, the ratio threshold I use above narrows the focus to deputies who would struggle to pay off loans using their official income. Therefore, I interpret this red flag as capturing deputies living far beyond their officially declared means, not through access to finance but instead using illicit income.
Assessing the Measure of Corruption
In all, 228 deputies, or 24%, failed to disclose a luxury car or had an average hidden earnings ratio of above one during their time in office. I term this subset “kompromat deputies.”Footnote 15 This approach primarily identifies deputies who are highly corrupt, with tens of thousands of dollars missing from disclosures. But the measure also captures more visibly corrupt behavior, and in particular, deputies who are especially vulnerable to investigation by law enforcement and journalists. Duma deputies are high-profile figures whose disclosures attract significant attention. In hiding their corruption in plain sight, kompromat deputies are compromised by their disclosures and may need to alter their legislative behavior.Footnote 16
The comparison group in this case (the non-kompromat deputies) may still be corrupt, but their corruption is less observable, in particular to law enforcement. Deputies without kompromat, for example, may understand how to use proxies or stash money offshore to prevent discrepancies from being easily caught.Footnote 17 The key actors in uncovering the corruption are anti-corruption officials tasked by the regime to verify the disclosures and hold officials accountable. Often the past decade, the Kremlin has used increasingly sophisticated methods to validate the information in disclosures against external registries of real estate, transportation, and banking assets that are not available to the general public (I use the insurance data as a work-around for this data missingness).Footnote 18 Indeed, each year tens of thousands of violations are uncovered (General Reference General2018). Yet the art of money laundering has evolved to such an extent that it is unclear if any law enforcement or financial intelligence unit has the capacity to fully know where Russian officials hide their money abroad.Footnote 19 It is hard to argue that Russian law officials could do 10 years ago what Western investigators are currently struggling with. Moreover, the broader point is that kompromat deputies who purchase domestic luxury assets beyond their means are easier to catch, monitor, and control than those who stash their assets abroad.
Table 1 presents some basic summary statistics about the incidence of kompromat deputies. The Liberal-Democratic Party of Russia (LDPR) contains the largest percentage, with roughly 37% of all deputies labeled as compromised. This aligns with anecdotal evidence of the LDPR getting caught selling seats to the highest bidder. Indeed, if this kompromat measure was just picking up consumption preferences, we should expect to see no variation between political parties, which self-organize around ideology, personal ties, and other shared objectives. There also seems to be a decrease in the ratio in more recent convocations. In response to public criticism of the Duma, speaker Vyacheslav Volodin imposed greater discipline on deputies during the most recent 7th convocation (Noble and Chaisty Reference Noble and Chaisty2022).
Descriptive Statistics
Note: This table calculates the percentage of “kompromat deputies” based on different descriptives. The Dissernet subsetting uses a binary indicator for whether a deputy plagiarized his or her dissertation based on analysis from the Dissernet project (https://www.dissernet.org/). Data on region-level corruption come from the Carnegie Moscow Center. Data on top 10 universities come from the 2022 Forbes Russia ranking of Russia’s best universities. The top-level number for the full sample is slightly larger than that reported in the main text since it measures the percentage of deputy-convocations, rather than the percentage of deputies.
I show three additional validation checks in Table 1. First, deputies who have been caught plagiarizing their dissertations (see Abalkina and Libman Reference Abalkina and Libman2020) are far more likely to be flagged as being compromised. This suggests that the indicator is capturing dishonesty. Second, I break out the percentage of kompromat deputies based on the level of corruption in the region each deputy listed as their place of residence. Data come from a 2010 expert survey conducted by the Carnegie Moscow Center; I code low, medium, and high regions based on their values on this five-point scale. Kompromat deputies more often reside in regions labeled by experts as more corrupt. Finally, drawing on the literature on political selection (Besley, Montalvo, and Reynal-Querol Reference Besley, Montalvo and Reynal-Querol2011; Gulzar Reference Gulzar2021), I coded whether deputies graduated from one of Russia’s top 10 universities as a proxy for their competence. Forbes Russia’s 2022 ranking was used to select the top universities. Less competent deputies with fewer political skills may both shirk their official responsibilities and get involved in less sophisticated corruption schemes. We see, however, that graduates of Russia’s most prestigious institutions are just as likely to become vulnerable to kompromat.
Supplementary Tables B1 and B2 show that the differences between kompromat and other deputies are minimal based on demographic characteristics. Women are less likely to report expensive cars than their incomes cannot afford, in line with other recent work finding that female legislators score better on individual corruption measures (Dollar, Fisman, and Gatti Reference Dollar, Fisman and Gatti2001). Supplementary Table E1 shows robustness checks subsetting on gender, showing that female kompromat deputies behave similarly to their male counterparts. In addition, kompromat deputies tend to be younger and less likely to work in the health care sector. Finally, these red flags are also not correlated with more traditional measures of corruption derived from disclosures data (Fisman, Schulz, and Vig Reference Fisman, Schulz and Vig2012). Columns 4–6 of Supplementary Table B2 show that there is no correlation between deputies’ change on income over their term in office and having kompromat. This suggests that my approach is picking up something different than earning more money in office, which could be explained by many legal activities. Deputies with kompromat are hiding illicitly earned wealth stashed in other asset classes; that obfuscation creates legal vulnerabilities that shape their political behavior.Footnote 20
Measuring Legislative Behavior and Reelection
Outcome measures on voting, session attendance, and other legislative activity come from the official Duma API (http://api.duma.gov.ru). I measure shirking by collecting roll-call votes on 13,086 bills over the three convocations; since bills must pass multiple readings to be sent to the President’s desk, this amounts to 37,391 unique voting events with 16,747,298 votes cast.Footnote 21 Absenteeism is high in the Russian Duma; collectively, deputies missed 35% of votes.Footnote 22 For each deputy, I calculate the percentage of roll-call votes missed each convocation.
Next, I create measures capturing how active deputies are during the sessions that they do attend. One of the deputies’ primary responsibilities is to introduce legislation. Yet in the Russian Duma, only a small number of deputies take the initiative to sponsor bills. As a result, sponsoring legislation is one of the strongest signs that deputies are taking their jobs seriously. For each convocation, I create an IHS-transformed count of the number of bills deputies acted as the sole sponsor.Footnote 23 Legislators also participate by asking questions on the Duma floor during debates. I collect data on all 98,079 questions posed by deputies to bill sponsors or invited experts and calculate an IHS-transformed count of those asked by each deputy in each convocation.
To measure regime loyalty, I follow Shirikov (Reference Shirikov2021) in identifying bills that were initiated by the federal government (ministries, agencies, etc.), noting that these pieces of legislation best reflect the regime’s policymaking goals. I create outcomes to reflect the percentage of times a deputy voted for a federal government bill during each of the three readings; values are captured on a 0–100 scale. These roll-call measures illustrate the differences between the regime and the systemic opposition. Table 2 presents summary statistics by party, first showing that over the period, the ruling party UR held roughly 70% of Duma seats, with the remainder roughly divided among the three systemic opposition parties. All parties rely heavily on businesspeople to fill their ranks, though the Communists, perhaps due to lingering ideology, have far lower numbers. All parties attract celebrities and boast extremely wealthy deputies.
The Regime and Opposition in the State Duma
Note: This table shows summary statistics for the four main political parties in Russia, with United Russia as the ruling party and the other three constituting the systemic opposition.
But the parties differ when it comes to politics. Absenteeism is highest among the LDPR, while rare among those affiliated with the ruling party United Russia (UR). Deputies from Just Russia proposed the largest number of bills, roughly three times more than both the Communists and UR. Finally, the systemic opposition does not unconditionally support legislation sponsored by the regime. Over the period, the Communists sided with the government roughly 82% of the time, matching other work documenting the party’s sometimes uncompromising stance toward the authorities (March Reference March2012). In contrast, LDPR demonstrates much stronger loyalty, voting for government bills almost as often as United Russia (with its steadfast 99.9% support of government legislation).
I next apply roll-call scaling methods using the R package emirt to calculate deputy ideal points along a pro or anti-regime dimension (McCarty, Poole, and Rosenthal Reference McCarty, Poole and Rosenthal2016; Poole et al. Reference Poole, Lewis, Lo and Carroll2008). This procedure fits spatial models to uncover patterns in preferences and ideological voting that might otherwise be missed by simply comparing raw votes. The reference point for each convocation is the leader of the ruling party fraction (Boris Gryzlov, Sergey Naryshkin, or Vyacheslav Volodin). Supplementary Figure B3 plots these deputy-convocation ideal points. Parties vary in their discipline, with the Communists (in red) generally enforcing the most anti-regime (pro-UR) stance of the four parties. Interestingly, United Russia does see some variation in discipline within its voting ranks, with some members at times defecting. I control for party membership in all models.
Finally, I coded whether each deputy was reelected. For the first two convocations, proportional representations and party lists were used to elect deputies, giving political parties control over selection. But in 2016, 50% of the body was elected using single-member districts (SMDs) as candidates competed directly for votes. Overall, roughly 50% of deputies retained their seats in the next convocation.
Empirical Strategy
The unit of analysis in the article is the deputy-convocation. Empirical models use OLS, include convocation fixed effects, and cluster standard errors at the deputy level.Footnote 24 All models also include covariates for each deputy that have been founded to predict political behavior in the Duma, including age (logged), gender, and primary occupation.Footnote 25 I extend the coding scheme of Shirikov (Reference Shirikov2021) to code deputies who have significant business interests or are celebrities (famous athletes, performers, etc.). Together with the occupation dummies, these controls help account for potentially different consumption preferences that could be driving the results. I create a binary indicator if a deputy served as a chair of any committee or fraction during his or her term. High-ranking officials in the Russian government often receive a government car and driver at the public’s expense. Leaked automobile lists suggest a small number of such officials serve in the Duma. I also include an indicator for whether the deputy was elected on the party list or through an SMD, and a running count of their years of experience in the State Duma. Finally, I include an indicator for the 14 deputies who died in office.
Identifying the effect of kompromat on deputy behavior requires that several assumptions hold. The first is that corrupt activities occur prior to the three sets of outcomes being studied: shirking, showing loyalty to the regime, and winning reelection. The structure of the data suggests that this assumption is more strongly upheld for the first and third outcomes rather than regime loyalty. To capture corruption, I examine how deputies spend their corrupt earnings rather than how they acquire them, which is impossible because of the nature of the closed-door, illegal dealings. It is then possible that deputies may exhibit loyalty to the regime first and then be rewarded with side payments or opportunities to earn illicit income. The fact that this reverse relationship is possible does not undermine the central theoretical claims that these deputies are being co-opted by the regime. One of the article’s key aims is to show an exchange of corruption for loyalty, and therefore I am careful not to describe this correlation as causal.
This first assumption, however, holds more strongly for the shirking and turnover hypotheses. It is much harder to argue that engaging in absenteeism or inactivity are lucrative for deputies or that outside interests would pay them not to show up for work. Finally, the turnover measures capture end-of-term outcomes which are measured after all disclosures for the previous term have been submitted. We can be more confident that observed corruption takes place before parties decide on which deputies to retain.
The second identifying assumption is that shirking, loyalty, and turnover are not driven by some unobserved factor that might produce a spurious correlation with the presence of kompromat. This assumption is difficult to directly test, in particular due to the challenge of finding an exogenous instrument for kompromat at the deputy level. We do not observe enough about the lives or activities of these individuals to make a strong case of their corruption being predicated on some other preexisting characteristic. Instead, I show a range of placebo and robustness tests in the Supplementary Material. I also follow the methods proposed by Altonji, Elder, and Taber (Reference Altonji, Elder and Taber2005) and Oster (Reference Oster2019) to investigate whether unobserved variation is likely to be explaining the results. For every model, I report Oster’s
$ \delta $
statistic, indicating how much more important the unobserved characteristics of the deputies would need to be compared to observables to fully explain the results. All results are also shown with an extended set of political and occupational covariates to isolate the effect of the hidden wealth measure.
RESULTS
Table 3 presents models examining legislative shirking. First, we see in columns 1 and 2 that the measure of kompromat is positively associated with absenteeism, as measured by the percentage of roll-call votes a deputy missed over a convocation. That effect is a little more than half that of two common types of deputies who miss votes: those representing SMDs (who may live far from Moscow) and celebrities (who are often used to drive electoral turnout but have little interest in politics). Deputies focusing on making money for themselves show up less often for work and rank among the more absentee members of the institution.Footnote 26
Corruption and Shirking
Note: This table shows results using different measures of legislative shirking as the outcome variables. The unit of analysis is the deputy-convocation. Absenteeism is the percentage of all votes a deputy missed during the convocation. Columns 3 and 4 analyze the weighted number of bills initiated by deputy, and columns 5 and 6 measure the number of questions asked during debates. The reference category for the party member predictors is Just Russia. All models are estimated using OLS with standard errors clustered at the deputy level. ***
$ p<0.01 $
, **
$ p<0.05 $
, *
$ p<0.1 $
.
Not only do kompromat deputies miss roll-call votes, but they are less involved in the sponsorship of legislation. Columns 3 and 4 show that kompromat deputies propose fewer bills. The rightmost columns examine the number of questions asked over the course of a convocation (columns 5 and 6). Kompromat deputies on average ask 20%–25% fewer questions per year, though the results are not as precisely estimated.Footnote 27 Taken together, this table provides evidence of shirking among those deputies found using their office for private gain.
The point estimates on several other characteristics of shirking are also worth mentioning. First, deputies with business activities and celebrities are generally much less active in parliament, as measured by their absenteeism, bill-drafting activity, and interest in asking questions. This contrasts with work by Chaisty (Reference Chaisty2013) who finds that businessperson deputies in earlier convocations put forth more legislation related to their sector. As the authoritarian regime consolidated under United Russia in the 2010s, there may be less scope for individual legislative initiative. In later convocations, these two categories (businesspeople and celebrities) constitute upward of 40% of the chamber, helping ensure a more docile deputy population that delegates legislative responsibilities to only the most active members: those in leadership positions (committees, fractions, etc.) and women.
Returning to Hypothesis 2, Table 4 examines whether kompromat deputies are more likely to support regime priorities. The first outcomes capture the percentage of the bills initiated by the federal government that each deputy voted for, first aggregated across all readings (columns 1 and 2) and then broken out by the three readings (columns 3–8). Deputies who hide income and assets on their disclosures are much more likely to vote with the regime, even controlling for party membership. These findings also come through in columns 9 and 10, where the outcome is each deputy’s ideal point. Kompromat deputies exhibit more pro-regime voting behavior.Footnote 28
Corruption and Regime Loyalty
Note: This table shows results using different measures of loyalty to the regime as the outcome variables. The Govt Bills columns measure the percentage of government-initiated bills that deputies voted for during the convocation, either altogether (columns 1 and 2) or broken out into first, second, or third readings. Ideal points are calculated for each convocation across all readings using the R package emirt. The reference category for the party member predictors is Just Russia. All models are estimated using OLS with standard errors clustered at the deputy level. ***
$ p<0.01 $
, **
$ p<0.05 $
, *
$ p<0.1 $
.
Corruption and Opposition Behavior
Voting against the regime can mean different things based on a deputy’s formal political affiliation. For members of the ruling party, dissenting on a bill might signal discontent with government priorities that could not be resolved behind closed doors but falls short of defecting from the party (Reuter and Szakonyi Reference Reuter and Szakonyi2019). Ruling parties may also be wary of punishing scandal-ridden members for fear of bad publicity. In the case of United Russia, only two deputies over the past two-plus decades have ever been stripped of their deputy immunity (see Supplementary Table A3), with members accused of serious sexual harassment and corrupt activities having kept their seats.
But for members of systemic opposition parties, challenging the regime can carry much greater costs. Opposition deputies who abuse their office for personal gain are much more careful to toe the government line for fear of provoking retribution from the regime. In addition to the case of Gudkov described above, criminal charges have been filed against a handful of opposition deputies for crossing criminal or corrupt red lines, including Ilya Ponomarev (Just Russia), Nikolai Parshin (Communist Party), and Aleksey Mitrofanov (Just Russia).
Table 5 tests whether the correlation between being compromised and regime loyalty differs by party, breaking down the regression models shown in Table 4 into subsets based on United Russia (UR) or systemic opposition parties. Models subset to the opposition include party indicators. Importantly, there are large and statistically significant coefficients on the measure of kompromat across all four outcomes related to regime loyalty, but only for deputies from the systemic opposition. United Russia deputies with hidden income and assets do not appear to change their voting behavior, potentially not fearing that the government will punish them. The Oster sensitivity tests also indicate that it is unlikely that selection on unobservables is driving the results for either ruling party or opposition members.
Corruption and Regime Loyalty, Subset by Party
Note: This table shows results using different measures of loyalty to the regime as the outcome variables, subset by whether the deputy is the member of the ruling party United Russia (odd columns) or a systemic opposition party (even columns). The Govt Bills columns measure the percentage of government-initiated bills that deputies voted for during the convocation, either altogether (columns 1 and 2) or broken out into first, second, or third readings. Ideal points are calculated for each convocation across all readings using the R package emirt. The reference category for the party member predictors is Just Russia. All models are estimated using OLS with standard errors clustered at the deputy level. ***
$ p<0.01 $
, **
$ p<0.05 $
, *
$ p<0.1 $
.
Party discipline is very strong in the Russian Duma, leaving less variation to be explained by demographic characteristics. For the models subset to UR, we see only that fraction leaders consistently vote more in line with the government, while deputies representing SMDs are more likely to oppose it. However, these coefficients are small, given that ruling party deputies vote with government nearly 100% of the time. This level of conformity suggests a limitation of the article’s ability to detect a relationship between kompromat and loyalty among members of the regime.
However, for members of the systemic opposition, the kompromat measure is among the, if not the strongest, predictor of regime loyalty.Footnote 29 Only gender is a more powerful predictor of vote choice after controlling for party. Kompromat deputies (along with women) constitute the most pro-regime faction within the systemic opposition.
Supplementary Table E2 analyzes deputy votes on government bills that turned out to be relatively competitive, excluding any votes that received either less than 10% support or more than 90% support from the chamber. This approach involves subsetting based on the outcome variable, so should be interpreted with caution. The coefficients on the measure of kompromat are five times larger while remaining statistically significant but only for deputies from the systemic opposition. Demonstrating loyalty to the regime is only important as an insurance strategy for those formally outside the regime.
Interestingly, I also find no effect of business deputies defecting from the regime, a result that contrasts with previous work by Dasanaike (Reference Dasanaike2022). This could be explained by a difference in my sample, which covers three convocations and all votes, rather than only budget bills in a single convocation. Businesspeople may not behave differently from their peers across a wider range of policy issues. The results are slightly stronger for two of the three convocations (5th and 7th), suggesting that when United Russia turned up the screws on party discipline and regime loyalty in the wake of the 2011–12 protests, there was less room for kompromat deputies to distinguish themselves. Finally, as a placebo test, I test for party heterogeneity in the shirking results, which should not be affected by opposition deputies trying to curry favor with the regime; Supplementary Table C1 finds no differences.
Corruption and Mechanisms of Influence
What are deputies doing to earn this illicit income? The Russian Duma has long been a preferred political playground for interest groups to seek influence. Although firms working in natural resources dominate the overall Russian economy, Chaisty (Reference Chaisty2013) documents widespread interest in Duma representation across firms working in manufacturing, agriculture, construction, finance and trade. Kompromat deputies may be earning side payments in exchange for proposing legislation and amendments, participating in legislative debates and issuing deputy requests.
Yet firms are not the only interest groups investing in political access. During Putin’s time in power, individuals connected to the security services—the so-called siloviki—have emerged as a powerful political constituency (Taylor Reference Taylor2017). Siloviki dominate the private sector, first through raiding against business and more recently through centralizing control over key economic assets (Rochlitz, Kazun, and Yakovlev Reference Rochlitz, Kazun and Yakovlev2020). Beyond their formal posts in the executive branch, siloviki exert considerable sway on Duma members to ensure little government oversight of their activities (Soldatov and Rochlitz Reference Soldatov, Rochlitz and Treisman2018). Siloviki interests even extend beyond rent-seeking; from 2017 to 2019, the Duma increased spending on health care for members of security services by a factor of eleven, as well as added additional privileges, such as early retirement and housing subsidies (Basmanova, Berezovskaya, and Tel’nova Reference Basmanova, Berezovskaya and Tel’nova2019). Kompromat deputies have multiple suitors to sell access to beyond corporate structures.
To test for the sources of external money, I collect data from TI-R’s Lobbying in the State Duma project which studied lobbying influence in the 7th convocation (Basmanova, Berezovskaya, and Tel’nova Reference Basmanova, Berezovskaya and Tel’nova2019). Researchers analyzed over 48,000 public documents and financial reports to identify lobbying connections between deputies and interest groups, including corporations, federal ministries, security agencies, regional governments, and nonprofits. The result is a dataset of 718 deputy-“interest group” ties for 349 deputies for which such relations were uncovered. I focus the analysis on deputies representing the two main interests above—corporations (53% of lobbying ties) and security services (13% of lobbying ties)—given that they are the two most common relationships.Footnote 30 To the best of my knowledge, this dataset is the only one available for tracking connections between deputies and special interests; unfortunately, only deputies of the 7th convocation (2016–21) are included.
Table 6 shows results predicting whether deputies lobby for corporations or the security services. For each outcome, the first column (columns 1 and 4) shows results for the full sample deputies across all four parties. Kompromat deputies are perhaps slightly less likely to lobby for companies, and slightly more likely to lobby for the security services. However, the results are much more pronounced when subset based on regime (columns 2 and 5) or opposition membership (columns 3 and 6). Kompromat deputies from the opposition are significantly more likely to push for the interests of the security services; this measure is the second most powerful predictor after having celebrity status. On the other hand, these same deputies are less likely to have developed ties with corporations. Importantly, neither trend holds for deputies from the ruling party.
Corruption and Lobbying
Note: This table examines whether deputies in the 7th convocation were classified by TI-Russia as lobbying for the interests of corporations (columns 1–3) or the interests of Russian security services (columns 4–6). For each outcome, results are shown first using all deputies and then broken out by ruling party or systemic opposition. All models are estimated using OLS with standard errors clustered at the deputy level. ***
$ p<0.01 $
, **
$ p<0.05 $
, *
$ p<0.1 $
.
These results provide additional support for the hypothesis that kompromat deputies trade loyalty to the regime in return for financial enrichment. It is possible that deputies feel emboldened to skirt anti-corruption laws precisely because of their ties to the security services. However, signing legal affidavits about one’s personal wealth produces compromising information that the government could later use to punish deputies who step out of line. These individuals sell their legislative autonomy. The overlap between corruption, the systemic opposition, and the security state may help explain why systemic opposition parties mostly fail to constrain, or even oppose, the regime even though they enjoy formal political power. These parties contain a significant number of individuals who have tied their financial fortunes closely to the state and have less incentive to oppose it.
The ideal way for deputies to profit off their political office is to sell deputy requests. However, the Duma has been loathed to share publicly any data on the requests being issued. Only in January 2021 did the Duma chair encourage deputies to post the requests online, citing the need for public transparency based on the potential for corruption.Footnote 31 As of February 2022, only 54 (7.9%) deputies in the 7th convocation (2016–21) had a page available for citizens to view. Supplementary Table F2 shows that kompromat deputies are significantly less likely to have created such webpages to publicly host this request information; in fact, just 2 of these kompromat deputies (out of 71) embraced transparency. Kompromat deputies may be particularly hesitant to allow voters to track their usage of this powerful tool.
Corruption, Career Concerns, and Accountability
Finally, I test the hypothesis about whether kompromat decreases the chances of deputies holding onto their seats. The models in Table 7 first analyze whether deputies ran for reelection (for example, were included on a party list or contested an SMD), and then whether they were indeed reelected. We see that deputies hiding income and assets are approximately 9% less likely to run for reelection. Consequently, columns 3 and 4 show that they were 11% less likely to hold onto their seats, a sharp drop considering a baseline rate of 50% reelection rates. This result is roughly the same for ruling party and opposition deputies (see Supplementary Table F3).
Corruption and Reelection
Note: This table looks at deputy post-convocation career outcomes. Columns 1 and 2 analyze an indicator for whether deputies ran for reelection (either on a party list or in a single-member district). Columns 3 and 4 analyze whether deputies won reelection. All models are estimated using OLS with standard errors clustered at the deputy level. ***
$ p<0.01 $
, **
$ p<0.05 $
, *
$ p<0.1 $
.
In the theory section, I argued that both parties and deputies have reasons to embrace turnover. Below I examine the observational implications of several alternate explanations before arguing in favor of this party and candidate-centric claim. First, we might expect voters to be the ones directly punishing kompromat deputies at the ballot box for shirking their duties and prioritizing self-enrichment (Klašnja Reference Klašnja2015). Instead, Supplementary Table F5 shows kompromat deputies faced no such punishment from voters in SMDs during the 2016 parliamentary elections.Footnote 32 When given the opportunity through direct elections, voters do appear to punish, much less remove deputies, who exhibit more observable corruption. This suggests that kompromat deputies may not be particularly fearful of direct electoral accountability and leave office for other reasons. However, as I discuss further below, parties may anticipate voters’ reaction and remove kompromat deputies themselves.
Next, Russia’s anti-corruption campaign may have incentivized law enforcement to go after kompromat deputies in order to improve the regime’s image (Carothers Reference Carothers2022). Accountability then would operate through the judicial system rather than the ballot box. In Supplementary Table A3, I collected data on all deputies who had their parliamentary immunity stripped from 2008 to 2021; this is the first step in a criminal proceeding against these elected officials. Only 10 deputies were removed from office, and of those just seven in connection to fraud, embezzlement, or corruption charges. Indeed, kompromat deputies were not more likely to be removed and/or prosecuted. Law enforcement activity does not explain the marked turnover in kompromat deputies.Footnote 33
Instead, I argue that the turnover in the Duma is part of a broader dynamic that sees parties rotate out kompromat deputies from office, who themselves also may prefer shorter terms. First, under the proportional representation system, parties are the main gatekeepers to the ballot. Kompromat deputies can inflict potential damage on party brands. Members of the systemic opposition also more often defect from their party leadership, violating the roll-call unity of systemic opposition parties by siding more often with the regime. Conscious of public scandal, the potential loss of votes, and weak member discipline, parties may be rotating corrupt, vulnerable deputies and cutting short legislative careers in order to preserve their reputation. One observable implication of this is that kompromat deputies who leave office should be younger than other deputies who exit at similar times. Indeed, on average, kompromat deputies leave office at 50 years old, compared to 54 years old for those without compromising material.Footnote 34
Another important point is that parties are not selecting out underperforming, lazy deputies. Not only does Table 7 show that showing up to work or proposing bills do not affect reelection chances but that the kompromat deputies still exit at higher rates controlling for these factors. It appears there is something specific about having observable corruption that can make candidates less appealing over the long run to parties.
Yet campaigns still need to be funded. Parties must reserve a certain number of seats for corrupt deputies in order to finance elections. An observable implication of this appears in Supplementary Table F5. Although kompromat deputies are more likely to lose their seats, those who remain still command top spots on the party lists and leadership positions within the body. Kompromat deputies who remain in office actually enjoy lower spots on the subsequent party lists (and therefore will be more likely to enter the next convocation); parties do not completely shun the incidence of corruption but carefully manage its incidence to maximize their chances of gaining power.
Measuring candidate time horizons and individual desire to remain in office is obviously a more challenging task, especially since interviewing these elites is near impossible during Russia’s authoritarian turn. But other evidence from Russia suggests that political connections do not provide ideal long-term protection against repression (Buckley et al. Reference Buckley, Reuter, Rochlitz and Aisin2022). Deputies may see a timely exit from the Duma as their best chance of protecting their financial gains and use their seat as a springboard into lower-profile jobs that still allow for enrichment. Indeed, even though kompromat deputies’ Duma careers are cut short, Supplementary Table F4 draws on data from RuPEP, a publicly available database on Russian elites, to show that kompromat deputies are just as likely to work in government after leaving the parliament. Kompromat deputies often wind up taking seats in the Federation Council, the ceremonial upper house of parliament that plays little role in Russian politics, or working as deputies in regional parliaments. If conditions change, future qualitative research could ask deputies about the reasons behind these post-Duma career paths.
CONCLUSION
Breaking down the numbers, this article indicates substantial governance costs from corruption. On average, deputies with hidden income and assets miss 176 more votes, propose two fewer bills, ask 25 fewer questions, and for those from the opposition, vote more in line with the ruling party. We also see significant clustering of kompromat deputies on committees critical to effective policymaking on economic issues, such as those working on financial services, judicial, small business, and even anti-corruption issues (Supplementary Figures F1 and F2). The results reveal that corruption is indeed systemic within the Russian parliament: just using data on domestic assets uncovers that roughly one quarter of the members of the country’s top legislative body are hiding the true state of their finances from anti-corruption authorities. Why is this degree of corruption sanctioned in the chamber?
One explanation is that the regime prefers not to have a parliament full of ambitious, active members. By rotating the more profit-seeking individuals in and out of office, the regime can more easily ensure the institution does not become a focal point for elite collective action and retains some elements of a rubber-stamping legislative body. Some legislators view their roles more akin to their counterparts in democratic settings: passing legislation not only to extend their party’s hold on power but also to potentially improve societal welfare and protect national security, among other goals. Others shirk their duties while aligning themselves closely with the regime, so as to better exploit their position for personal gain. Challengers’ obedience can be purchased by being provided access to financial spoils without fear of prosecution. The article also finds that kompromat deputies in the opposition push for the interests of powerful security services, helping the regime further co-opt and undermine their potential rivals.
Thus, the key logic of legislative institutions under autocracy may be one of diversity: a regime prefers a body that strikes some balance between ambition and greed, and is willing to overlook some corruption to prevent the development of an autonomous branch that might challenge the executive. Upward-looking deputies may even prize the kompromat they acquire on their corrupt colleagues. While other regimes may design institutions to attract elites based on their ideological commitment rather than financial interest (Hollyer and Wantchekon Reference Hollyer and Wantchekon2015), the case of Russia shows that leaders prefer strategic co-optation. Given the need for external resources and elite cohesion, political parties in autocratic regimes aim for a sweet spot between loyalty and capacity.
There are also good reasons to believe that dynamics in Russia are representative of other competitive authoritarian regimes around the world. Nearly 50% of authoritarian states had multiple parties represented in their legislatures (Simison Reference Simison2022). Russia in that sense is no outlier. Yet the Russian Duma also exhibits the same party discipline coupled with strong pro-government voting behavior as China and Vietnam (Lü, Liu, and Li Reference Lü, Liu and Li2020; Schuler Reference Schuler2021). That degree of centralization moves much of the internal jockeying, negotiations, and co-optation behind closed doors, where opportunities for corruption and rent-seeking are heightened. Finally, concerns have been raised about many autocracies exploiting anti-corruption campaigns to both ensure regime loyalty and stabilize regime dynamics (Lorentzen and Lu Reference Lorentzen and Lu2018). The Russian government’s strategic use of disclosures may fit a larger pattern of regimes of autocracies enabling corruption while also dangling the threat of investigation to more efficiently control it.
SUPPLEMENTARY MATERIAL
To view supplementary material for this article, please visit https://doi.org/10.1017/S0003055424000340.
DATA AVAILABILITY STATEMENT
Research documentation and/or data that support the findings of this study are openly available at the American Political Science Review Dataverse: https://doi.org/10.7910/DVN/P2U84B.
Acknowledgements
Special thanks to the Declarator team at Transparency International—Russia, and in particular Andrei Jvirblis and Alexander Serechenko, for sharing their thoughts on the project. I also thank Natalie Forrat, Adam Scharpf, Abhit Bhandari, Ben Noble, Marko Klašnja, and seminar audiences at BOFIT, Copenhagen Business School, Temple, SPSA, Uppsala University, Georgetown, and GWU for helpful feedback.
FUNDING STATEMENT
This research was funded by the National Council for Eurasian and East European Research and George Washington University.
CONFLICT OF INTEREST
The author declares no ethical issues or conflicts of interest in this research.
ETHICAL STANDARDS
The author affirms this research did not involve human participants.
Open access
Comments
No Comments have been published for this article.