Business deserts politics under weak institutions: evidence from Russia, 2003–2010

Abstract

Businesspeople are expected to invest in political connections when the institutions are weak. Using an original dataset from Russia in 2003–2010, we document changes in political connections of the richest businesspeople and show that within the institutional environment of a fledgling autocracy political engagement stops paying off, and the businesspeople retreat from politics. The businesspeople’s political disengagement reveals their insider assessment of the quality of Russian political institutions, indicating that as autocracy consolidated in Russia, its political institutional structure was in decay. This finding contributes to our understanding of authoritarian institutions, suggesting that even though autocracies nowadays might be institutionalized to a much higher degree, in the short-run autocracy is still detrimental to institutionalization.

Introduction

Extensive literature suggests that the businesspeople political engagement occurs primarily under conditions of poor institutionalization of rule of law and property rights protection: ¹ Not being able to achieve their goals through institutional means, the businesspeople turn to politics to protect their property rights and gain a competitive edge.

The literature does not ask what happens if the political institutions mobilized by the businesspeople are equally poor, as is the case under electoral authoritarianism—a type of political regime “that permits certain institutions normally associated with democracy, such as elections and political parties, to exist, while remaining authoritarian in the basic patterns of power distribution and reproduction.” ² As follows from this definition, the formal political institutions (that serve as an avenue for businesses to exert political influence), albeit nominally present under electoral autocracy, are at the same time effectively reduced to a mere façade only disguising the actual way the regime works, and virtually inconsequential on their own. ³

This poor state of political institutions should make it less rewarding for the businesses to engage in politics for the sake of protecting their property, even where no proper economic institutions to achieve such protection exist. Strikingly, although the poor quality of economic institutions should push businesses to invest in political engagement, when paired with the equally poor quality of political institutions, it would result in businesses retracting from politics altogether.

In this paper we observe and register this curious phenomenon by focusing on the case of Russia. The property rights protection in Russia has remained rather problematic throughout the 1990s and early 2000s. ⁴ It was furthermore dealt a heavy blow with the Yukos affair, when in 2003–2004 the state attacked, bankrupted, and nationalized Russia’s largest private company and sent its owners and managers to prison or in exile. ⁵ Far from being an isolated event, it was followed by a number of other high-profile nationalizations and sometimes even raider attacks by the state, ⁶ which by the late 2000s led analysts to rate protection of private property rights in Russia rather low. ⁷

The poor state of property rights protection was matched in the 2000s with a steady decay in Russia’s political institutions as the country slid into autocracy. ⁸ We hypothesize that following this process, the existing political institutions gradually lost their efficacy, and as a result the efficiency of political engagement as a private mechanism of property rights protection (and the return on investment in such engagement) should have decreased quite significantly over the 2000s. This in turn must have pushed the Russian businesspeople, rather active politically in the early 2000s, ⁹ out of politics, ¹⁰ even though the institutions of property rights protection and the rule of law did not improve.

To test this hypothesis, we assemble and analyze a unique dataset on political connectedness of the richest Russians for years 2003–2010, the period when the Russian political regime admittedly shifted from defective democracy to electoral authoritarianism. ¹¹ We contribute to literature on businesspeople political connections by describing and analyzing a case of complete institutional failure whereby both economic institutions (those securing property rights protection as a public good) and political institutions (allowing the businesses to acquire political connections privately if they need to) are rendered ineffectual over time as autocracy consolidates.

The paper starts by theorizing businesspeople’s political engagement under poor institutions. We then follow by describing the business–state relations in Russia in the 2000s in the context of decay in Russian political institutions. We present our data and the results of our analysis that show the businesspeople retreating from politics during the observed period. Methodologically, our analysis also provides an approach to treat the problems of observing the effects of institutions where these cannot be taken for granted or inferred from the mere fact of the institutions’ existence, as is the case under electoral autocracy. We return to this point in the conclusion.

Paradox of businesspeople’s political de-participation under poor institutions

In this article we provide an empirical illustration of a phenomenon of businesspeople’s retreat from politics under conditions of poor institutionalization. We focus on those institutions that allow for access to power (e.g., legislatures, elected and non-elected executives) and observe temporal dynamics of their mobilization by the businesspeople. Following the existent literature on businesspeople’s political engagement ¹² we assume that under conditions of poor protection of property rights the businesspeople will in principle be interested in using these institutions. But we only expect that to be the case to the extent they are effective as a means of property rights protection.

The richest businesspeople stand out as a research object for three reasons. First, their lives ¹³ are relatively well reported, both in terms of quality of the information and the number of observations available. Second, the richest businesspeople can be assumed to have reliable private information about the value of the political institutions they are engaged with or might contemplate mobilizing. This partly stems from their belonging to the establishment, thus having insider knowledge about the actual state of politics. But more importantly, even if they have no insider access to observe the state of political institutions directly, they should be able to infer it indirectly from the returns they and their competitors get on investment in political connectedness in the short run.

Third, the richest businesspeople can be expected to actually want and be able to engage in politics as a matter of non-market business strategy. ¹⁴ This is especially the case in poorly institutionalized environments with no reliable institutions of rule of law. Indeed, a vast body of research into business strategies in under-institutionalized contexts documents this phenomenon ¹⁵ and shows this strategy actually pays off: where economic institutions do not protect property rights, political connections do. ¹⁶

This literature treats being politically connected as a substitute for using courts and other formal mechanisms of property rights protection. ¹⁷ The key distinction between the two is that, whereas formal institutions (such as courts, and the rule of law generally) function as a public mechanism of property rights protection, political engagement allows to obtain such protection privately. ¹⁸ It would seem, as a corollary, that whenever the former is present, it should crowd out the latter: political engagement as a substitute should only arise when the institutional environment is dysfunctional. This, however, is not the case, and even with the functioning institutions of property rights protection a firm might be interested in gaining a competitive edge over other firms on the market by engaging in politics. Indeed, research into business–state relations in well institutionalized environments supports this proposition, ¹⁹ showing that businesses might use political connectedness simply as a non-market strategy. ²⁰

Note, however, that regardless of the quality of economic institutions, going political to protect one’s property rights is conditioned by the very existence of an opportunity to do so. Whether such an option exists in turn depends on the quality of political institutions. Indeed, just as the firms might not be supplied with public mechanisms of property rights protection, so, too, there can prove to be no institutional structure allowing businesses to develop political connectedness despite all aspiration to do so.

One particular institutional setting where this effect should prove especially visible is electoral authoritarianism—a type of political regime “that permits certain institutions normally associated with democracy, such as elections and political parties, to exist, while remaining authoritarian in the basic patterns of power distribution and reproduction.” ²¹ As follows from this definition, the formal political institutions (that could potentially serve as an avenue for businesses to exert political influence), albeit nominally present under electoral autocracy, are at the same time effectively reduced to a mere façade disguising the actual way the regime works, and virtually inconsequential on their own. ²²

As the regime transitions toward electoral authoritarianism, the existent formal political institutions gradually lose their potency. This in turn should make it less rewarding for the businesses to participate in these political institutions for the sake of protecting their property, even where no proper economic institutions to achieve such protection exist. Strikingly, although the poor quality of economic institutions should push businesses to invest in political engagement, when paired with the equally poor quality of political institutions, it would result in businesses retracting from politics altogether.

The implication that we are testing in this paper is therefore as follows: If the general trend under autocracy is toward reducing the institutions to their semblance while preserving the institutional façade, then the businesses would be able either to observe that latent incremental institutional decay directly (through their insider access to information), or to infer that change from diminishing returns on investment they made into political connectedness earlier. As political engagement becomes less attractive as a non-market strategy, the businesspeople will start retreating from politics. We would then observe a decrease in the businesspeople’s political participation and could conclude that if political institutions are demobilized by the groups that have access to reliable information about their real quality, this should serve as a signal of institutional decay, even if the institutional façade remains intact nominally.

Here lies the paradox: under conditions of institutional deficiency businesspeople should invest in political connections to compensate for that, but they do not. To the contrary, as the case of Russia in the 2000s illustrates, as the institutional environment got sparser, businesspeople got even more disengaged from politics. That happened, because it was not only the institutions of property rights protection that deteriorated, but also the political institutions one would need to mobilize to engage in politics.

Political institutional dynamics in Russia in the 2000s

Conventional history of business–state relations in post-Soviet Russia invokes the metaphor of a pendulum swinging between the two extremes of the business capture, when the government dominates the private business, and state capture, when the businesspeople gain control over the government. ²³ The 1990s are characterized as a period when, as the state weakened, the business expanded its control over the operations of government while also managing to escape its regulation at all levels. ²⁴ The pinnacle of this process was a brief period in the late 1990s. After providing the much needed financial and media support to Russia’s unpopular incumbent Boris Yeltsin during the 1996 presidential elections, the richest businesspeople got the upper hand. ²⁵ They managed to establish a stronger influence over state policies, secured access to the most lucrative state-owned assets, and even briefly held some important positions in the government.

In the early 2000s the oligarchs started losing their influence—a development attributed to the new president Vladimir Putin’s successful attempts at restoring state autonomy and, generally, strengthening the Russian state. ²⁶ With respect to the oligarchs this new policy took the form of their so-called “equidistancing.” It was informally sealed during their meeting with the president in 2000 when, in return for the government pledging not to revise the 1990s privatization, which made the oligarchs rich, the government demanded that they stay out of politics (an arrangement known afterward as the “Shashlik agreement” due to allegedly being negotiated over a barbeque). The government took some steps to demonstrate its good will. It significantly simplified the tax system, ²⁷ while at the same time making the legislation more business-friendly. ²⁸ A new streamlined taxation scheme was further developed specifically for the oil extracting sector. ²⁹

All of these measures were supposed to establish a more balanced business–state relationship, but in 2003 the largest private company of Russia, Yukos, was bankrupted and nationalized, its owners and top management sent to prison. Another big oil company, Sibneft, was nationalized almost immediately afterward, and by the late 2000s the pendulum had clearly swung back. The Yukos affair thus marked the beginning of a period of business capture in business–state relationship that, judging by the growing share of state ownership in Russia’s economy, lasts to date. ³⁰

Note that as the government sought to limit and regulate the businesspeople’s political involvement, this need not have resulted in their complete (or even systematic) withdrawal from politics. First, the policy of equidistancing only targeted the oligarchs—the extra rich businesspeople whose fortunes were big enough to allow them to substantially derange the functioning of the government or undermine the ruling group’s authority. This policy was not supposed to apply to the smaller businesses, apparently deemed less dangerous for the government since their political engagement would probably even not be noticeable for the Kremlin.

Second, as was already clear from the Yukos affair, the oligarchs’ political engagement was only punishable on a case-by-case basis. Some of those extra rich businesspeople who were part to the Shashlik agreement were allowed, if not encouraged, to engage in political activity, as long as such engagement was approved by the Kremlin. ³¹ The decisive point was therefore not political engagement as such, but rather the motivation behind it: as long as the businessperson’s political engagement was not considered detrimental to the federal government’s vital interests, they could invest in political connectedness if they needed to.

The new policy thus did not mean to completely eliminate the businesspeople’s demand for political connectedness, but rather to regulate it informally. Furthermore, since the quality of property rights protection and other economic institutional features stayed the same (and with the abovementioned reservation of limiting the oligarchs’ direct involvement on a case-by-case basis), the demand for political connectedness must have remained high. At the same time, what changes significantly during this period are the supply-side factors—in particular, the quality of those political institutions that the businesspeople could mobilize to build up their political connectedness. ³² This change manifests itself most visibly in the series of political reforms conducted in 2001–2005 and aiming at shifting the balance of power (both in terms of its vertical and horizontal separation) in favor of the federal executive.

In the vertical dimension, the reforms started as early as in 2000 when president Putin complemented the existing regional division of Russia into 89 federal units ³³ with a new superstructure of seven federal districts, each headed by a so-called polpred—the president’s envoy responsible for ensuring that the policies enacted by the federal center are implemented uniformly in her respective district, and that the governor does not meddle with the selection and placement of the federal agency personnel in their region. ³⁴ The governors’ powers were further reduced as they lost their seats at the Federation Council (the upper house of the Russian parliament) as of 2002, and thus lost the direct political influence they had in the matters of national policymaking. Finally, the governors were dealt a particularly heavy blow as the gubernatorial elections were abolished in favor of the governors’ nomination by the president. The legal requirement for a regional legislature’s consent proved to be formal: the president could both depose a governor and nominate a new one as he pleased. As a corollary, this quite obviously reduced the post’s attractiveness for a businessperson seeking to invest in a political connection, since such investment could be easily nullified if the president dismissed the governor.

In parallel with these developments the balance was also shifting in the executive–legislative relationship at the federal level. ³⁵ The political opportunity structure began getting sparser as the Kremlin first micro-managed the creation of a party of power, the United Russia, that became a majority party in late 2001, ³⁶ and obtained a two-third majority after the 2003 State Duma (lower chamber of Russia’s Federal Assembly) election by negotiating with independent candidates running in the single member districts. ³⁷ The Kremlin further launched a campaign to reduce the number of officially registered parties, and between 2003 and 2009 this number collapsed from 46 to just six parties formally allowed to run their lists in legislative elections. The few parties that persisted were put under informal control of the presidential administration, ³⁸ so even if a businessperson would consider running on one of these parties’ tickets she would still have to be informally vetted by the Kremlin and the safety of her Duma seat would further depend on it. An opportunity remained to get elected by plurality vote in a single member district since half of the Duma seats were still filled this way. Winning in these districts was not as much a matter of negotiating with the parties as one of the candidate’s personal effort, which granted such candidates bigger autonomy and security of their posts. ³⁹ However, in 2005 the electoral system was reformed, and as of 2007 the elections were conducted by pure proportional representation. The opportunities to become a Duma deputy (member) therefore shrank, both in terms of all seats now being distributed by the parties, the registered parties’ very number being reduced very significantly, and their being controlled directly by the Kremlin. ⁴⁰

Thus, as the government succeeded at weakening all alternative centers of power and extending its control over them, and as autocracy consolidated in Russia throughout the 2000s, the opportunities political institutions could effectively provide businesspeople with grew poorer. We hypothesize that as a result of this change in the political opportunity structure the businesspeople would gradually become less inclined to invest into such formal political connectedness simply because it would not pay off, thus testifying to the formal institutions’ decay over time. To do that we first test the hypothesis that there was no economic return from political connections during the studied period (H1). After testing it we can move on to test our second hypothesis that businesspeople’s propensity to possess political connections decreased through the studied period (H2). However, as we describe above this might have been a nuanced process. In particular, the decrease in political connectedness was more pronounced among the richer businesspeople (H3). Indeed, it is our expectation that in this new state of things the oligarchs would rather deal directly (and informally) with the Kremlin and the presidential administration because securing formal political engagement with the federal government, the parliament or the regional governments would at best imply idle costs, and at worst would bring additional existential risks.

Data

To test these hypotheses, we put together an original dataset containing various information about the richest Russians. It consisted of three basic units. The first one contained information on the fortunes of these people and was compiled from the annual rankings published by the Finans magazine between 2004 and 2011, when the magazine went out of business. Rankings were published yearly in February containing information on each preceding year, so our dataset covers the years 2003–2010. The magazine’s methodology was to invite experts to assess the value of the industrial assets and the real estate that the participants own, as well as their other incomes. On average each of Finans’s annual rankings included 500 businesspeople, most of them also present in the next years’ rankings (the panel, however, is not entirely balanced due to occasional dropouts). The panel data collected also provided for a significantly broader individual coverage as compared to the other reputable ranking compiled by the Forbes magazine (only providing data on 100 richest Russians annually). Forbes’ ranking furthermore was only launched somewhat later: its coverage starts with 2004 thus missing the important pre-Yukos affair data completely.

Aktion-media, the publisher of Finans, mostly deals in specialized professional and business journals: their portfolio includes such titles as Glavbukh (Accountant general), Kadrovoie Delo (The HR), Finansovyi Direktor (CFO), and General’nyi Direktor (CEO). Unlike Forbes, a household name in Russia, these titles are known mostly within the business community. However, because of this narrow specialization, as well as embeddedness in the business community and the expertise they command, Aktion-media is considered one of the most reputable business publishers in Russia. Still, to double-check the reliability of the asset data they provide, we ran several comparisons between Finans and Forbes estimates. Results indicated a rather strong comparability between the two sources. First, the percentage overlap between the Forbes list and the top 100 businesspeople on the Finans ranking ranged between 70% and 88% across years. Second, correlations between the two estimates of assets (log-transformed) were .85 on the level of observations (person–years) and .82 on the level of persons. ⁴¹

The second unit in our dataset was the information on the industries that the participants of the ranking were active in. This information was coded into a categorical variable covering twelve industries: retail trade, petroleum and gas extraction, construction and development, finances, engineering, agriculture, metallurgy and mining, chemical industry and perfumery, transport, food industry, IT and telecommunications, and the energy sector. The data on these sectors was also made available by the Finans magazine for 2008–2011. The years 2004–2007 were thus coded manually with the information collected from various sources (such as, for instance, the factbook compiled by the Kommersant, one of Russia’s most respected daily newspapers). For the participants present in the ranking both in 2008 or later, and in the years before that, the manual coding was checked for consistency with the coding by Finans, with the priority given to the information supplied by Finans as of 2008.

The third unit contained information on the participants’ political office and their ties with government officials and members of legislatures. Political connectedness was coded in four distinct categories: Duma deputies and members of the Federation Council; members of regional legislatures; and high-ranking officials in either the federal or the regional executive (which includes federal and regional ministers, deputy ministers, heads of all kinds of executive agencies and services, governors, and city mayors). Overall, 437 out of 3,497 person–year observations (or 12.5% of the sample) were coded as having a political connection, either holding a political office directly or being informally connected to an office-holder (and therefore engaging in politics indirectly).

There was no single source where the information on such ties would all be conveniently collected, so we relied on a variety of sources, including the Kommersant factbook for the more official information and the Compromat.ru website for the rumors about the informal ties. (Note that most of the time Compromat.ru only reposts information from other sources—normally, respectable and trustworthy newspapers. The information available through Compromat.ru was used whenever the source it referenced was deemed trustworthy.) Another valuable source of information was Russia’s largest news aggregator run by Yandex that automatically aggregates information from various sources, including the regional and local media (which was particularly useful for revealing ties to the regional government, the mayors, etc.). Whenever a formal connection was established for any individual for any particular year, we also double-checked for the duration of that connection to make sure adjacent years were coded accordingly, including cross-referencing our data with the official sources.

A special effort was made to find out about the businesspeople’s possible informal connections to high-ranking officials (both federal and regional, executive and legislative). Informal connections were defined conservatively as either being a close relative (which includes spouses, children, parents, and siblings), or being reported as a close friend of an official where such friendship dates back before the official assumed their position. While not excluding such possibility, our coding did not register close informal connections that would have arisen after an influential officeholder was in office. On the one hand, such information is often speculative and hard to pin down conclusively (as opposed to information about a relationship that predates the time when the officeholder assumes their position). On the other hand, high-ranking officials would often have to deal with businesspeople while in office on a regular basis, providing for some sort of connection. However, we assumed the connections from before they assumed their office to be inherently stronger.

Unlike a formal connection (a businessperson holds a political office herself), an informal connection (familial relationship or long-term friendship with a politician) is not an asset that businesspeople can freely acquire or dispose of. ⁴² Although the informal connection might have been working to a businessperson’s benefit, neither having such informal connection, nor losing one would normally be considered a strategic choice for a businessperson, at least not in the short run. All such informally connected businesspeople (188 observations, 5.1% of the sample) were therefore excluded from the analysis, ⁴³ so the analysis presented below focuses on the connections that could be acquired or disposed of strategically. The analysis only focuses on the businesspeople who do not hold a preexisting informal connection to influential politicians and must make a conscious choice to either engage in politics by holding a political office themselves directly, or to disengage from it.

Results

We start from estimating returns on political connections over the 2003–2010 period and testing the corresponding hypothesis (H1). To do so, we predict businesspeople’s fortunes (log-transformed to account for decreasing marginal utility) with a full interaction between the connection status and time using a panel regression with businessperson-level fixed effects. This analysis allows us to estimate the differences in asset changes for businesspeople with and without connections. Figure 1 compares estimated trends in asset growth by connectedness on both log and original scales. The results show strong support for H1. Specifically, even though connected businesspeople, on average, had more assets, they were not growing faster from 2003 to 2010 compared with unconnected businesspeople. ⁴⁴ The interactive effect of time and connectedness is not significant on the 95% confidence level (b=−0.04, p=.769). In other words, having a political connection (that likely incurs both material and non-material costs for businesspeople) is not found to provide any tangible economic benefits. ⁴⁵

Figure 1. Trends in log assets for businesspeople with and without political connections.

Note: See Table S1 in Supplementary Material for regression coefficients.

Evidence presented above has important implications for businesspeople’s strategies with respect to investing in political connectedness. We find that, throughout the studied period, returns on connections were effectively absent. Therefore, those contemplating making such investments should in hindsight have preferred to not invest in new political connections or abandon existing ones. But even not having the benefit of hindsight, businesspeople could still anticipate such development from the ongoing institutional changes. To the degree that they expected such changes to weaken the institutions they considered investing in, this should have resulted in their gradual retraction from politics. Following this lead, we turn to exploring whether and to what extent businesspeople’s propensity to invest in political connections has actually decreased in response to the institutional changes.

To test this hypothesis (H2), we run a model that predicts the probability of a businessperson’s political participation in the studied period (2003–2010). For our analysis the most interesting factor is time since we seek to establish if the businesspeople’s formal political engagement decreased between 2003 and 2010. To control for the fact that those businesspeople already having connections in the previous year are more likely to have such connections in any given year, we introduce an element of autocorrelation (the dependent variable is added as a predictor with a one-year lag). Other control variables are a businessperson’s fortune and industry.

Table 1 presents results from two panel logistic regression models. Model 1 uses a random-effects estimator with standard errors clustered on the level of individuals (businesspeople). ⁴⁶ Model 2 uses a fixed-effects estimator and thus includes only individuals for whom change in political connections is observed within the analyzed time period. In the two models, explanatory variables are normalized to range from 0 (smallest observed value) to 1 (greatest observed value). Both models give results in favor of H2: over time, as the negative and significant time trend shows, businesspeople gradually retreat from politics.

Since estimated coefficients from logistic regressions are difficult to interpret in substantive terms, we also present the results translated into the key quantities of interest, predicted probabilities. The estimates are presented in Figure 2 as probabilities of having a connection for each year given that such a connection was present in the previous year (i.e., probabilities of keeping a connection). Left-hand panel presents results based on the logistic regression estimating a time trend (Model 1 in Table 1). It shows that for a businessperson with a political connection in 2002, the probability of having the same connection in 2003 is slightly higher than 90%. The corresponding estimate for 2009 and 2010 is only about 75%.

Figure 2. Predicted probabilities of keeping a connection.

Note: See Table S4 in Supplementary Material for the model with year-specific effects.

Table 1. Panel logistic regression results predicting having a political connection

Note: Standard errors are in parentheses. Industry varies on the level of individuals. See Table S2 in Supplementary Material for full results.

We also investigate to what extent the estimated time trend may be driven by the specific years included in the study. The right-hand panel of Figure 2 plots estimated probabilities of keeping a political connection using a regression with year-specific effects (rather than a single time trend). After staying at approximately the same level in 2003–2005, businesspeople’s probability to keep connections decreases in 2006 and 2007. Then, there is a slight increase in probability in 2008. Its sources are not entirely clear. For instance, it might have happened as businesspeople’s response to the financial crisis and search for state protection at the times of hardship. Still, in 2009 and 2010 the probability of keeping a political connection goes down again to the lowest levels observed. Therefore, even with the 2008 outlier, the overall negative time trend is evident independently of the model specification.

Finally, we explore whether the gradual withdrawal of businesspeople from politics in 2003–2010 has been largely driven by the richest ones induced to cut their political engagement in response to the government’s equidistancing policies after the Shashlik agreement and, especially, the Yukos affair (H3). To test this hypothesis, we estimate an interactive effect of time and assets on probability to keep political connections in order to see whether the retreat from politics was stronger among the wealthier businesspeople. Following the guidelines on the presentation of interactive effects, ⁴⁷ we show our results graphically in Figure 3. The estimates demonstrate that more affluent businesspeople start with slightly higher probabilities to keep connections but, as time goes, lose their connections at a somewhat higher rate. The difference in trends, however, is not significant (b=−0.08, p=.206).

Figure 3. Predicted probabilities of having a political connection by year for businesspeople having connection previous year. Note: See Table S5 in Supplementary Material for regression coefficients.

This result is important for our theoretical argument. One alternative explanation for businesspeople’s gradual loss of political connections in 2003–2010 is conscious efforts by the government to limit their political participation. However, such efforts should have been directed first and foremost against the richest and the most influential businesspeople—the so-called oligarchs. The absence of a significant difference in the speed of losing connections between more and less affluent businesspeople suggests that government efforts could not have been the primary reason why they have been gradually leaving politics. We believe that the reason this happened has been the decline in profitability of political connectedness over time conditioned by weakening of Russia’s political institutions. The businesspeople’s strategies only reflect this process: as political institutions get less efficient in allowing them to achieve their interests, they retreat from politics.

Discussion and conclusion

In this paper, we have attempted to trace the institutional decay in the context of a consolidating autocracy using Russia in 2003–2010 as the case study. We have focused on a particular aspect of institutions’ effectiveness: their ability to protect the interests of the richest businesspeople bypassing public political process. Our empirical strategy has been based on a conjecture that, as insiders, the most affluent members of the business community have knowledge of, or relatively good prognostic capabilities with respect to the institutions’ economic value. Therefore, their actions can be seen as an indicator of institutions’ quality in cases when it cannot be readily inferred. Specifically, we have hypothesized that a consolidating autocracy leads to weakening of formal institutions and, as a result, to businesspeople abandoning political connections as they would do with any other ineffective investment.

In our empirical analysis, we have reported three essential findings. First, having political connections in 2003–2010 was not associated with increased asset growth among Russia’s richest businesspeople, indicating low efficiency of institutions in advancing economic interests. Second, businesspeople were gradually abandoning their political connections over the same time period—likely, in response to low returns on investments in political institutional connections. Third, political withdrawal was happening at largely similar rates for businesspeople with different wealth levels, thus suggesting that this process was not primarily caused by the government’s crackdown on political ambitions of the oligarchs. Altogether, we have confirmed the methodological value of relying on insiders’ actions as an indicator of institutions’ value when such value cannot be taken for granted. We have also demonstrated that the consolidating autocracies have negative consequences for capabilities of political institutions—even when they continue to formally exist.

Our research also poses new questions related to businesspeople’s political engagement strategies. When faced with a low level of property rights protection through the rule of law and, at the same time, unable to compensate for that by acquiring political standing, where do the businesspeople go? The richest businesspeople may opt to externalize justice and enforcement through relocating into a more secure jurisdiction or choose external courts for arbitration with each other. ⁴⁸ Members of the poorer strata within the business community, in turn, can engage with the formal institutions of property rights protection where possible: for instance, go to the generally less corrupt and more independent commercial (arbitrazh) courts. Another important non-market strategy might be networking with informally connected businesses via joint ventures—essentially, acquiring an informal political connection by proxy. Existing research shows proliferation of this strategy in the 2010s. ⁴⁹ Going political may also remain a feasible option, ⁵⁰ as long as other businesspeople do the same. Although in light of the results of this research, one should expect it to become less efficient and therefore less attractive in the longer run.

Importantly, this possibility does not impact our principal findings: as autocracy consolidates in Russia, the quasi-democratic institutions it retains are gradually robbed of their meaning. Social groups that should in principle be interested in mobilizing these institutions prefer to disengage from them, thus signaling their increasing irrelevance and impotence. This suggests a correction to the current discussion on autocratic institutions that seems to place too much of an emphasis on the existence of quasi-democratic institutions in the post-third-wave autocracies and the role they play in these autocracies’ political development. It may be an important macro-historical trend that autocracies nowadays are institutionalized to a much higher degree than previously. ⁵¹ On a meso-historical level, however, autocracy is still detrimental to institutionalization.