What Explains Corruption Perceptions?

The Dark Side of Political Competition in Russia’s Regions

Gulnaz Sharafutdinova

Abstract Most empirical studies of corruption rely on data using perceptions of cor-
ruption as a proxy for actual corruption. While this approach might be appropriate for
advanced democracies, it is less effective for hybrid regimes. In these regimes corrup-
tion allegations are often used in political battles, raising public perceptions of corrup-
tion and thus reflecting the degree of political competition rather than actual corruption.
The data on public perceptions of corruption in Russian regions produced by Transpar-
ency International and the Information for Democracy Foundation (INDEM) shows that
higher levels of political competition and press freedom along with lower economic
development appear as the key variables contributing to higher public perceptions of
corruption in Russian regions.

In his May 2006 presidential address to the State Duma, Russia’s former president Putin
criticized the government for not being able to “remove one of the greatest obstacles
facing our development, that of corruption.”1 The anticorruption agenda of the Russian
government was reaffirmed in earnest by Russia’s new president, Dmitry Medvedev,
who characterized corruption as a systemic problem requiring a “systemic answer”
and launched a large-scale, anticorruption campaign.2 Ironically, despite the widespread
perceptions of entrenched and rising corruption, it is not high on people’s list of issues
requiring government attention. Although more than half the population believes that
the majority of high-echelon officials are corrupt,3 in 2001 only 12 per cent of Russians
selected corruption as one of the top three societal problems.4 In a 2006 poll conducted
by VTsIOM, 21 per cent of respondents selected corruption as one of the top three
problems.5 In 2008 corruption was the eighth most important problem in the eyes of
the Russians, who apparently view it more as a norm rather than an exception.6 Noting
the mismatch between the extent of corruption and its public perception as a low prior-
ity among societal problems, the director of the Information for Democracy (INDEM)
Foundation, a respected Russian think tank, prescribed wider public campaigns to raise
the significance of corruption in the eyes of the public.7 A paradoxical situation seems
to have emerged: the government, experts, and president seem to view corruption as an
important problem, while the population does not view it as a government priority.
147
Comparative Politics January 2010

Strangely enough, the perpetrator of the crime moralizes about the crime, while the
victim appears indifferent.
The reason for this paradox might be simple. The low priority assigned by the pub-
lic to fighting corruption might reflect disbelief in government’s ability and commitment
to clean up its act. In 2008 the Public Opinion Foundation (FOM) reported that more
than half the population did not believe in the possibility of uprooting corruption in
Russia.8 People may have simply lost hope of ending corruption, or there could be
some less obvious factors at play influencing public perceptions of corruption. As noted
by various analysts, perceptions of corruption are “not the same thing as corruption
itself.”9 This has been confirmed by more recent studies that tested whether the well-
known corruption indices (produced by Transparency International or the World Bank)
measure corruption.10 Beliefs about corruption might be driven by various factors not
linked to actual amounts of corruption. The study of these factors may suggest an an-
swer to why people do not view the problem of corruption as a government priority.

Corruption Perceptions: Good or Evil?

The issue of corruption has received much attention among political scientists in general
and those interested in postcommunism in particular.11 The causes and consequences of
corruption are well studied and widely discussed.12 At the very least, corruption hinders
economic growth, retards institutional development, erodes the state, and carries signifi-
cant political costs.13 Corruption takes various forms in different parts of the world.14
The analytical approaches used to study it reflect this diversity. The rise of oligarchic
capitalism in some postcommunist countries drew attention to flawed privatization of
the 1990s and the resultant state capture by the self-interested elite.15 Corruption patterns
reflected in the predominance of informal institutions on the social level have been
analyzed using cultural and institutional approaches.16 Regardless of analytical perspec-
tive, measuring corruption has proved problematic.17 Most widely publicized studies
and measures (such as that of Transparency International) rely on public perceptions
of corruption as a proxy for its actual levels. It might be a sensible strategy, when the
objective indicators of corruption are hard to develop. At the same time, an overreliance
on perceptions-based data creates significant problems in the analysis of corruption. Cor-
ruption “wears different clothes” and even if the phenomenon of bribery is universal, the
elite-level corruption might range from private rent-seeking to a more systematic corrup-
tion when economic resources are exploited for political goals.18 Analyzing corruption
based on perceptions might obfuscate crucial differences between types of corruption.
What is even more important, the formation of perceptions might be driven by factors
other than corruption reality.19
While most analysts agree on the economic, social, and political costs of actual
corruption, few share an opinion on the effects of the perceptions of corruption. In fact,
there is a huge gap between the two major views on the effects of perceptions. One view
expects that heightened perceptions and increased public awareness of corruption must
148
 Gulnaz Sharafutdinova

help citizens to hold their governments accountable through democratic mechanisms

and, specifically, the electoral process.20 This argument assumes that public perceptions
of corruption correspond to its actual levels and that the public has access to free and
independent media that accurately reports on the quality of governance and on govern-
mental corruption.21 It therefore applies to functioning democracies in which people
have the mechanisms to keep the government accountable, such as legislatures with
investigative powers, independent courts, and free and fair elections.
The other view points to the “moral externalities” of corruption: heightened percep-
tions of corruption can promote the view of corruption as a normal activity and therefore
“lower the threshold of inhibition for others to engage in the same kind of behavior.”22
In this second camp are also those who argue that heightened perceptions of corruption and,
especially, numerous anticorruption campaigns, undermine state legitimacy and “create a
profound problem of confidence in the state structures.”23 The main effect of anticorruption
campaigns in the institutional environment characterized by structural corruption is, they
argue, to weaken state capacity, thus creating a vicious circle, complicating the tasks of gov-
ernance even further.24 In a similar vein, some studies of corruption perceptions in Latin
American countries have demonstrated that perceived corruption results in withdrawal from
politics and undermines government legitimacy.25 The proponents in this second camp are
more agnostic about whether the actual level of corruption corresponds to public percep-
tions. They focus on the negative consequences of high perceptions of corruption.
The stakes in this debate are high. Each argument presupposes certain policy im-
plications for the government and advocacy groups dealing with corruption. Should
advocacy groups aim to uncover, publicize, and scandalize cases of corruption, or should
they adopt other strategies to promote the view that not everybody is involved in cor-
ruption and corruption is not a norm?26 Should the government encourage public anti-
corruption campaigns or try to deal with the issue in a more insulated manner? These
questions do not have obvious answers, especially in hybrid regimes that combine demo-
cratic and authoritarian elements and that are more prone to exploiting corruption issues
for political purposes.27
The issue of public perceptions of corruption is distinct from corruption per se.
Therefore, the causes and consequences of public perceptions need to be addressed
separately. In this article, using empirical data from the Russian regions, hypotheses
about the factors that influence public perceptions of corruption are tested. To illustrate
the significance of corruption perceptions, the same dataset is used to explore the relation-
ship between perceptions of corruption and public trust in the government on different
levels, testing whether citizens assign blame in accordance with their perceptions. The
findings suggest that perceptions matter regardless of their source.

Why Should We Care?

There are several important reasons to care about public perceptions of corruption as a
phenomenon separate from the reality of corruption. As noted earlier, various authors
149
Comparative Politics January 2010

link public perceptions of corruption to state delegitimization. It appears that popular

beliefs about corruption undermine the legitimacy of the government and state institu-
tions. The regional corruption dataset (described below) allows for exploring the link
between public perceptions of corruption on the regional and federal levels and public
trust in regional and federal authorities. Figure 1 presents the results of a simple correla-
tion between public perceptions of corruption on the regional level (aggregating percep-
tions data at the regional level) and public trust in regional government, and Figure 2
shows public perceptions of corruption on the federal level (aggregating perceptions
data at the regional level) and public trust in federal government.

Figure 1 Perceptions of Regional Corruption vs. Trust in Regional Authorities

Figures 1 and 2 reveal a strong correlation between the two variables—the higher
the perceptions of corruption, the lower the public trust in government, both at regional
and federal levels. This corroborates earlier studies that demonstrated a significant
negative correlation between corruption indices and regime support in postcommunist
countries, adding urgency to the task of examining the sources of public perceptions
of corruption.28
The second, potentially debilitating effect of heightened perceptions of corruption
might be the moral externalities produced by the widespread beliefs about corruption.
When corruption is believed to be everywhere, it becomes viewed as a normal activity,
thus creating a convenient justification for all subjects involved in it. “The system made
me do it” syndrome might be poisonous,29 exonerating those involved in corrupt activities
150
 Gulnaz Sharafutdinova

Figure 2 Perceptions of Federal Corruption vs. Trust in Federal Authorities

of individual responsibility. This pertains acutely to the members of the elite that can
blame the accepted “norms” of behavior for their possibly illegal actions. When corruption
becomes acceptable as a norm, any attempt to break the vicious cycle—of corruption per-
ceptions regenerating corruption practices—will have to be informed by an understanding
of the sources of corruption perceptions.

Hypotheses

Various factors might explain differing levels of public perceptions of corruption. Be-
cause of the widespread use of corruption perceptions as a proxy for actual corruption,
the first argument that should be tested is whether people base their perceptions on their
own experience and that of their family and friends. Unofficial transactions, bribes, and
other forms of corruption are widespread in postcommunist regions in general and in
Russia more specifically.30 Therefore, the first hypothesis (H1) is that public percep-
tions of corruption are higher in those regions where people have greater experience
with corruption.
At the same time, it could be expected that a personal encounter with corrupt acts of
the government, especially at the higher levels, is a rare occurrence for each individual
and therefore might not be the most influential factor in forming public perceptions of
corruption.31 The studies of public opinion formation highlight the importance of in-
formation transmitted through the mass media.32 Accordingly, various means of mass
151
Comparative Politics January 2010

communication might be responsible for fueling corruption perceptions. Several surveys

in postcommunist countries have demonstrated that most people learn about corruption
from newspapers, radio, or television.33 This yields the second hypothesis (H2), that pub-
lic perceptions of corruption are higher in regions with higher degrees of press freedom.
If corruption perceptions are a good proxy for corruption itself, however, this logic
would have to be turned upside-down. Strengthening press freedom is frequently men-
tioned as a main strategy for fighting corruption.34 Free press is expected to play the role
of government watchdog and to provide a check on governmental misconduct by ex-
posing and undermining corrupt officials. Democratic theory allocates great importance
to press freedom, and countries with freer press are usually considered less corrupt. If
perceptions of corruption reflect actual corruption, then the rival hypothesis (H2.1) is
that public perceptions of corruption are higher in regions with less press freedom
and lower in regions with greater press freedom.
When the concept of press freedom is explored in the postcommunist context, par-
ticularly Russia, things become even more complicated because there are additional
factors related to the degree of exposure of corruption in the media. As noted by var-
ious observers, the press has become free but not independent.35 Similarly, what has
usually been perceived as free press in Russia is often not autonomous and free of
particularistic influence.36 While free of state control, the print media as well as the
television networks fell under the domination of the oligarchs—the owners of vast
financial-industrial groups that emerged in Russia in the 1990s. Privately controlled,
the media resources were exploited for the sake of promoting oligarchic interests, whether
in the political or economic sphere. These patterns repeated on the national as well as the
regional level. One of the widely used informal practices that emerged was the use of media
to expose corruption in order to undermine and discredit political and economic opponents.37
In effect, the corruption issue was used as a tool for political struggle, similar to negative
campaigning in advanced democracies. Scholars of U.S. electoral campaigns have found
that the amount of negativity is determined by the degree of electoral contestation—the
closer the race, the meaner the campaign.38 Anecdotal evidence from the Russian regions
confirms that the harsher the political clashes, the more intense the “mud-slinging” and
information wars.39 Thus, the degree of political competition might influence the use of
corruption exposure as a tool of political struggle, yielding the third hypothesis (H3) that
public perceptions of corruption are higher in more politically competitive regions.
One of the frequently observed, yet still controversial, relationships with regard to
corruption is its link to economic development. Various studies have reported a strong
correlation between GDP per capita and corruption, demonstrating that corruption un-
dermines development.40 At the same time, there is no widespread agreement on the
causality between these two variables. Less-developed countries might be incapable of
fighting corruption, so the causal link might run the other way too.41 More recently, a
study that carefully distinguished experienced corruption from perceived corruption also
established a statistical link between economic development and perceptions of corrup-
tion.42 Therefore, a fourth hypothesis (H4) is that public perceptions of corruption are
higher in less economically developed regions.
152
 Gulnaz Sharafutdinova

One plausible conjecture about the nature of causality in this relationship is that
corruption perceptions might be related to the economic conditions in which people
find themselves. In economically depressed regions, people are more likely to be more
dissatisfied with their lives and therefore to have higher perceptions of corruption as a
universal blame factor. Corruption in such circumstances serves as one of the more con-
venient, all-fit explanations. This reasoning—that corruption perceptions might reflect
general dissatisfaction with living conditions—could be tested further using a demo-
graphic indicator such as the proportion of retired people in the population. Numerous
studies of Russian transitional society have highlighted the plight and deprivation of
Russia’s pensioners after the collapse of communism.43 The retirees are among the
most underprivileged strata in Russia. They also represent the most restless and alert
group of the population.44 It is plausible therefore to expect that public perceptions of
corruption are higher in the regions with higher proportion of pensioners (H5).

Data, Operationalization, and Measurement

Most empirical corruption studies are cross-country or small-N case studies. This article
is based on a cross-regional comparison of Russia’s regions, which allows for a com-
parison of units within a similar cultural environment. Hence, the cultural factor—often
considered significant in explaining corruption—is controlled. The subnational focus
also allows for exploring the linkage between corruption perceptions of both regional
and federal levels of government.
The dataset is a product of a unique, systematic study of corruption in Russian
regions undertaken by Transparency International in collaboration with INDEM.
Released in 2002, the results from this study have already been used to examine the
causes of corruption across Russia’s regions.45 This dataset, covering forty of the
eighty-three regions of the Russian Federation, is employed to examine factors influencing
public perceptions of corruption in the regions and, more particularly, of how corrupt
regional and federal authorities are. It combines perception-based and experience-based
data, thus allowing a test of correspondence between perceived and experienced corrup-
tion.46 It also includes the indicators of public perceptions of corruption on the local,
regional, and federal levels, as well as across the various branches of power.
The data are based on individual responses aggregated at the regional level. The
aggregated character of the data does not allow for incorporating in the analysis some
individual-level traits that undoubtedly influence corruption perceptions. But in the ab-
sence of alternative data, this dataset still permits the study of how regional variation
shapes public perceptions of corruption. The data are representative, including regions
from each of the seven federal districts, and account for differences in federal status and
levels of economic development. The dataset is based on a stratified random sample
drawn from the regional population.47 The conceptualization of corruption relied upon
by the analysts included two aspects—everyday corruption (viewed mostly as bribes
paid to officials) and business corruption (involving state capture, business capture,
153
Comparative Politics January 2010

and administrative corruption). The survey questionnaire included two types of

questions that inquired about respondents’ assessment of corruption in their region
and also tried to get at personal corruption practice.48 The integrated index of corruption
perceptions for each region was created based on both aspects of corruption men-
tioned above.49
The data on press freedom necessary for testing H2 were produced by a nongovern-
mental think tank, Obshchestvennaia Expertiza (Public Expertise).50 This think tank
assessed press freedom across eighty-seven Russian regions in 1999 along three spe-
cific dimensions—freedom of access to information, freedom of production of infor-
mation, and freedom of dissemination of information. I relied on the index of the
freedom of information dissemination based on the investigation of local information
dissemination and the extent to which the government promotes and encourages dis-
tribution of printed materials.51
The independent variable tested in H3 is political competition. Assessing the level
of political competitiveness in a region is a complicated task. McMann and Petrov
surveyed experts on Russian regional politics to gauge levels of regional democracy;52
Marsh adopted Vanhanen’s “index of democratization;”53 Moraski and Reisinger
looked at the competition for seats in the federal legislature, the number of candidates
for governor on the regional ballot, and the margin of victory in gubernatorial elections.54
Here, two measures of political competitiveness are used, adopted from a compre-
hensive “Democratic Audit of Regions” conducted by Obshchestvennaia Expertiza.55
A proxy for political competition (or lack of it) in the regions that is specific to Russia
under President Putin is the degree to which the new party in power, United Russia, has
monopolized the regional parliaments. The success of United Russia in the regional
legislature reflects the degree to which the regional authorities—for the most part allied
with the party in power—control the political space and the administrative resources.56
Even the most autonomy-minded governors under Putin preferred a bandwagon strat-
egy, joining and supporting United Russia.57 The anecdotal evidence from such central-
ized and tightly controlled regional regimes as Tatarstan and Bashkortostan confirms
that the domination of United Russia reflects the political monopolization in the region.
These two republics are among the top five regions with the highest percentage of
United Russia members in their regional parliaments. These percentages do not reflect
any special ideological appeal of United Russia in these regions, but rather the effec-
tive control over administrative resources by the regional governments and the absence of
opposition. Hence, the higher the proportion of United Russia deputies in the regional
legislature, the lower is the degree of political competition in the region. The Democratic
Audit of Regions contains the data on the share of seats controlled by United Russia in
the regional legislatures in 2003. The indicator of political competition used in this study
was constructed by subtracting these numbers from 100.
The second indicator is based on the results of a broader assessment of regional
democracy conducted by Public Expertise. This measure was constructed using eight
different indicators of democracy that relied on longer term electoral statistics and such
indicators as political competition (measured through the margin of victory and the
154
 Gulnaz Sharafutdinova

number of candidates running in the elections), length of incumbency, degree of party

monopolization in the regional legislature, vote ‘against all,’ election threshold, the use
of administrative resources (measured by voter turnout), and the measure of distortion in
translating a party’s vote share into seats in the legislature.58 Based on these indicators,
the analysts working on the Democratic Audit of Regions divided eighty-eight regions
into five categories (from 1 5 least democratic to 5 5 most democratic).59
The economic development variable in H4 was constructed using the data on gross
regional product per capita produced by Goskomstat.60 The level of unemployment is
used as an additional measure of regional economic health, which might be relevant for
corruption perceptions because unemployed people are more likely to be dissatisfied with
the government. The data on age and the percentage of retired people as well as the
unemployment data were taken from Goskomstat. Table 1 presents the summary sta-
tistics for the variables used in the analysis.

Table 1 Summary Statistics

Variables Mean Std. Dev-n Min Max

Regional corruption 0.67 0.06 0.45 0.75
Federal corruption 0.61 0.04 0.47 0.67
Experience 0.56 0.13 0.31 0.81
Free press 0.19 0.16 0 0.894
Political competition 0.45 0.19 0.05 0.97
GRP (in rubles) 60412.03 46107.64 28801.2 294550.9
Unemployment rate 0.1 0.02 0.07 0.14
Pensioners 0.25 0.04 0.14 0.32
Dem1 0.075 0.27 0 1
Dem2 0.1 0.3 0 1
Dem3 0.175 0.38 0 1
Dem5 0.175 0.38 0 1

Results

To test the main hypotheses, I estimated two multivariate linear regression models by
the ordinary least squares (OLS).61 The first model explaining perceptions of corruption
in Russian regions included, as predictors, indicators of corruption experience, gross
regional product,62 freedom of the press, political competition, unemployment rate,
and the proportion of pensioners. The results of this test are reported in Table 2. The
overall result of the model is robust; together the explanatory variables account for
38 per cent of the variation in corruption perceptions across Russia’s regions. The
associated p-value of the F statistic for the joint significance of explanatory variables
in the model (.01) allows for rejecting the null hypothesis that together the indepen-
dent variables have zero impact. For none of the explanatory variables the variance
155
Comparative Politics January 2010

inflation factor was greater than 5; multicollinearity therefore was not significant in
this model (and other models below).63

Table 2 Model 1

Regional Corruption 1 Federal Corruption 1

Experience -.09 (.07) -.06 (.05)
Free press .15 (.07)* .09 (.05)
Political competition .09 (.04)* .04 (.03)
LogGRP -.09 (.03)** -.04 (.02)
Unemployment rate -.17 (.55) -.3 (.4)
Pensioners -.10 (.28) -.09 (.2)
Constant 1.73 (.46) 1.18 (.33)
N 39 39
R-squared .38 .21
F-statistics 3.3 1.41
Prob>F .01 .2
Note: N539 because the value on political competition for one region was missing.
* - significant at less than 5 per cent.
** - significant at less than 1 per cent.

The main findings from testing this model bear on H2, H3, and H4. Perceptions of
corruption are higher in regions with freer press (H2), higher political competition levels
(H3), and lower levels of economic development (H4). The estimated coefficients for
these variables have the expected signs and are statistically significant. The experience
with corruption (as reported by the respondents) was not found to affect public percep-
tions about corruption; to the contrary, the sign associated with the estimated coefficient
for this variable shows the opposite relationship, if insignificant. The support for H2
reflects the role of the mass media in shaping public opinion and, specifically, public
perceptions about corruption. Tight control over information in the regions indeed limits
perceived corruption. The results from this model indicate that 1 unit decrease in the
freedom of the press index is associated with 1.5 unit decrease of corruption perceptions
on the regional level.64 This effect does not appear at the federal level. It may be ex-
pected that public perceptions of corruption at the federal level in Russia are heavily
influenced by state-controlled television coverage that represents the most popular
medium of information nationwide.65
While the discovery of the media effect might not be very surprising, the link of
public perceptions of corruption to political competition is probably the most interesting
and novel finding. The results of this regression model suggest that a unit increase in
political competition corresponds to .9 unit increase in corruption perceptions. This
finding is not obvious. In fact, it goes against a plausible expectation that competitive
politics might decrease perceptions of corruption by influencing the degree of actual
corruption (at least when the link between political competition and accountability
156
 Gulnaz Sharafutdinova

holds). This finding, however, supports studies conducted in Latin America that bring
attention to the role of political factors influencing corruption perceptions.66
As revealed in the earlier study of regional corruption in Russia, the perceptions of
corruption depend strongly on the gross regional product. People in the poorer regions
tend to think that corruption is higher than those living in the richer regions.67 The estimates
show that for every unit increase in the level of gross regional product, the level of corrup-
tion perceptions decreases by .9 units. At the same time, the second economic indicator—
unemployment rates—is not correlated with corruption perceptions. The demographic
variable measured using the percentage of pensioners in the regional population is also
insignificant in accounting for corruption perceptions. This finding undermines the pro-
posed conjecture about corruption perceptions reflecting general dissatisfaction with liv-
ing conditions and economic underdevelopment. Further research is needed to accurately
account for the causal link between economic development and corruption perceptions.
The same predictors were also tested with the dependent variable of public percep-
tions of federal level corruption (see Table 2). The results manifest that the regional
level predictors are not correlated with public perceptions of corruption on the federal
level. The associated p-value of the F statistic for the joint significance of explanatory
variables in the model (.2) also does not allow for rejecting the null hypothesis that
together these independent variables have zero impact.68
The second model (see Table 3) uses an indicator of democracy instead of political
competition. This indicator is a categorical variable ranging from 1 to 5, used in the
regression model as dummy variables.69 Having a limited number of observations, I ex-
cluded the unemployment rate and the number of pensioners (not related to corruption
perceptions in the first model) in order to maintain the model parsimonious. Hence, this
model included four dummy variables of democracy, gross regional product, and press
freedom indicator.

Table 3 Model 2

Regional Corruption 2 Federal Corruption 2

Dem1 -.0998 (.0266)** -.0421(.0208)*
Dem2 .0052 (.024) -.0029 (.0188)
Dem3 .0323 (0204) .0269 (.0159)
Dem5 .0207 (.0187) .0083 (.0146)
LogGRP -.0692 (.02)** -.0342 (.0156)*
Press freedom .08 (.06) .07 (.04)
Constant 1.4036 (.2105) .9694 (.1643)
N 40 40
R-squared .54 .30
F-statistics 6.35 2.40
Prob>F .0002 .05
* - significant at less than 5 per cent.
** - significant at less than 1 per cent.
157
Comparative Politics January 2010

This model also produces robust results. Together, the predictors account for
54 percent of the variation in the public perceptions of corruption across the regions.
The p-value of the F statistic for the joint significance of explanatory variables in this
model (.0002) allows for rejecting the null hypothesis that together the independent
variables have zero impact. The results are similar to the first model. The estimated
coefficients show that perceptions of corruption in the regions increase with the level
of democracy and decrease with the level of economic development. At the same time,
the coefficients are significant only at the lowest level of democracy, thereby highlighting
that perceptions and public opinion can be effectively controlled in the absence of any
democratic features in the regime at the lowest possible degree of democracy. Any degree
of regime liberalization away from a tightly controlled, nondemocratic regime, results in
government losing the capacity to influence public opinion and corruption perceptions.
This finding resonates with recent studies of “color revolutions” in the postcommunist
region that brought attention to the instability of competitive authoritarian regimes even
with limited degree of pluralism.70
Diverging from the first model, the variable of press freedom was not statistically
significant in this model, although the estimated coefficient has the sign expected by H2:
the freer the press, the greater are the perceptions of corruption. It is likely that the
various indicators used to form an indicator of democracy override the press freedom
indicator. The same model ran against the indicator of federal corruption perceptions
and yielded the results that also differed from the first model. This model shows sta-
tistically significant correlation between regional level indicators of democracy and
economic development and public perceptions of corruption on the federal level. In less
democratic and more developed regions, people tend to think better not only of the
regional level government but also of the federal level government.

Discussion

The two factors that influence public perceptions of corruption most robustly are regional
economic development and political competition. What sense could we make out of the cor-
relations found? When using these subjective indicators as a proxy for corruption, Dininio
and Orttung hypothesized that “higher levels of development reduce incentives to public
officials to resort to corruption because more resources make possible better-paid jobs
and longer time horizons due to a more stable environment.”71 This argument could be
located within theories that link institutional quality to economic factors.72 Eliminating the
assumption that perceptions of corruption reflect actual corruption and, therefore, institu-
tional quality, necessitates reconsidering this argument when viewing the above-mentioned
correlation as a link between perceptions of corruption and regional economic develop-
ment. It becomes more important then to consider how and why economic development
influences mass opinion rather than institutional quality and incentives to public officials.
As posited earlier, one potential explanation of the link between regional economic
development and perceptions of corruption might lie in the overall public dissatisfaction
158
 Gulnaz Sharafutdinova

with regional underdevelopment and the logic of blame employed by the discontented
public. Living in less-developed regions and facing problems with employment and
poverty, people might be more prone to blame something for their economic difficulties.
Because socially the word corruption is often used “as a metaphor for everything that
is not right,”73 corruption might be the most available, common blame strategy. How-
ever, the fact that the proportion of unemployed people and pensioners in the regional
population—who are likely to be most dissatisfied—is not found to be correlated with
corruption perceptions, raises serious doubts with regard to this explanation. Appar-
ently, corruption perceptions do not stand as a proxy for dissatisfaction and an easily
available accusation.
Another link that should be considered in future research is the extent to which the
relationship between economic development and social trust could parallel the link be-
tween perceptions of corruption and economic development. As shown in recent studies,
trust is positively related to economic development, and at the same time is likely to be
inversely linked to public perceptions of corruption.74 Public perceptions of corruption
could then be seen as a proxy for and a reflection of the lacking social trust in lesser
developed economies.
The impact of regional political competition on public perceptions of corruption is
the most controversial and noteworthy finding of this study. It is controversial when
considered along with the argument suggesting that increased political competition re-
duces corruption.75 While agnostic about the impact of political competition on actual
corruption, this article shows that political competition promotes popular beliefs about
corruption. Even if and when political competition works to decrease actual corruption,
on the level of perceptions the situation might be worsening. The short-term negative
effect from how political competition plays out in hybrid regimes might work to under-
mine the potential longer term positive effect from political competition on actual cor-
ruption. In countries with weak democratic institutions, even such short-term effect of
heightened perceptions of corruption might result in a popular disappointment with
democratic institutions, undermining them in the short run and precluding any potential
of strengthening them in the long run.
Caution is warranted in interpreting this finding. Could it be determined by a re-
gional perception bias driven by respondents’ fear or unwillingness to be sincere in an
authoritarian context? Could the local survey implementers simply “doctor” the results?
There are important reasons to believe that any potential data errors are random and that
there is no systematic regional perception bias in less competitive regions. The lack of
political competition in the most authoritarian regimes inside Russia do not amount to
total control over the population and to regional regimes’ capacity to punish individuals
for “wrong” responses. The most authoritarian regional regimes lack anything that
would resemble an intrusive secret police that could monitor individuals. Even in the
least competitive regions people have experienced some degree of civil society revival
and benefitted from the work of nongovernmental organizations and opposition news-
papers. Furthermore, all regions experienced and witnessed the more pluralist and com-
petitive national-level politics of the 1990s that created a common political space with
159
Comparative Politics January 2010

shared expectations and attitudes. In short, a certain degree of pluralism present across
Russia (even if more limited in some regions) rules out any mass expectations of indi-
vidual punishment for incorrect personal answers to a survey. Therefore, even if partic-
ular individuals might have been insincere in their responses, the sources of insincerity
would be random and would not produce any systemic regional perception bias.
These findings are supported by other studies, such as the Business Environment
and Enterprise Performance Survey (BEEPS) administered by the World Bank among
twenty-eight postcommunist countries. One of the surprising findings of this study was
that corruption measures were lower in the least competitive states (such as Kazakhstan
and Uzbekistan) and higher in more competitive post-Soviet regimes (such as Kyrgyzstan,
Ukraine, and Russia).76 These findings provide additional empirical support for this
study, making it highly unlikely that the local survey implementers might have some-
how affected the results. A similar relationship uncovered by a study administered by
different researches and in different contexts alleviates potential concerns about the
dataset used in this article.
This finding is further corroborated by close observers of regional politics in Russia
that have long noticed the “dark side” of political competition as it played out in numer-
ous electoral campaigns in the regions.77 The wide range of dirty tricks employed in
regional elections included the use of kompromat (compromising materials) and nega-
tive advertising referred to as “black PR,” falsely attributed letters and newspapers
widely distributed in the public, clone candidates changing their real names and pass-
ports to steal votes from opposing candidates, provocation schemes with disappearing
journalists implicating particular politicians, and many other manipulative techniques.
The more uncertainty and competition political actors faced, the more they were prone
to employing political consultants and relying on dirty tricks and manipulation. The
electoral ploys became so common in the regions that the media started referring to
them by the name of the region where a particular trick was employed first—Belgorod,
Kursk, Bryansk or Orel alternatives.78 The dirty tricks aimed at manipulating the voters
and their preferences. The unintended result of these techniques, however, was to bury
the public in the piles of compromising materials, thus raising public perceptions of
corruption and undermining trust in all elites involved in these political battles.79
The first model revealed the link between increased public perceptions of corrup-
tion and press freedom. On the surface, a free press functions in Russia as expected by
classic democratic theory: it provides greater information about governmental miscon-
duct, thus educating the public about government corruption. Logically, the people must
react by “throwing the rascals out” in the next elections and the new politicians must
presumably be less corrupt. That is why generally, in older democracies, freer press is
associated with less corruption and is frequently prescribed as a tool for fighting corrup-
tion. This logic fails in the Russian context, however, where free press does not mean
independent press and where media outlets have been instrumentalized for advancing
political and economic interests of their owners and sponsors. As noted earlier, during
the 1990s media resources became an important asset coveted by the national and
regional-level oligarchs (business magnates). This asset was employed in political
160
 Gulnaz Sharafutdinova

and economic struggles between the oligarchs and vis-à-vis the state.80 Therefore, the
propensity of the press to publicize and focus on corruption issues differed in the re-
gions depending on the degree of political and economic competition. The freedom of
the press, as expressed in the presence of various owners and lack of governmental
monopoly, did not ensure the functioning of the press as the “fourth estate”—a guardian
of the public interest and watchdog on the governmental activities. In brief, analytical
lenses other than a liberal model of press and media are necessary to fully capture the
role of media in the Russian context. Anthropologists, more sensitive to differences in
context, advanced the concept of “media-political clientelism” to characterize such ap-
parently non-Western workings of the media.81 Political scientists should integrate this
understanding to modify and improve their theories as well.
Although in the first model regional indicators did not influence public perceptions
of corruption on the federal level, thus demonstrating that people rely on different
sources of information to form their judgments about corruption on various govern-
mental levels, the second model that relied on composite regional democracy rankings
revealed the link between corruption perceptions on regional and federal levels. In richer
and less democratic regions people have lower perceptions of corruption on the fed-
eral level. This finding once again points to the independent importance of percep-
tions as such. Obviously, factors other than actual corruption on the federal level are
at play in shaping perceptions, if regional-level variables influence how people perceive
federal authorities.
Finally, an important issue is the limits of potential generalizations available from
this study. The discussion of why and how political competition in Russia’s regions
affect corruption perceptions brings attention to how competition plays out, the un-
restricted nature of political struggles reflected in electoral machinations, instrumentali-
zation of the media, and the use of false, compromising materials against the opponents
that inadvertently undermines public trust in elites. If these are the main mechanisms
that translate political competition into rising corruption perceptions, then the findings
travel to those countries and those political systems that feature such warfare-like elec-
toral campaigns and “media-political clientelism” not restricted by the rule of law. Among
the postcommunist countries that come closest to such description are pre-2004 Ukraine,
pre-2001 Moldova, and Russia under Yeltsin.82 The dark side of political competition
is however not limited to the postcommunist region. Similar tendencies could prob-
ably be uncovered in hybrid regimes in Africa, Latin America, and Asia, adding to
our understanding of unstable democratic regimes and repeated authoritarian reversals
in these regions.
Alas, this database does not contain the data on public trust in political institutions
that would allow us to explore whether heightened perceptions of corruption are nega-
tively related to trust in specific political institutions. This link—that popular beliefs about
corruption would taint public evaluation and therefore trust in institutions—appears as
plausible. Although there are convincing studies that link a high degree of distrust in
political institutions in the postcommunist region to institutional performance,83 the
degree to which perceptions of corruption could also be linked to the levels of trust
161
Comparative Politics January 2010

in institutions should also be seriously considered. After all, this pattern holds on the
national level in Russia. While public perceptions of corruption have been rising during
the 1990s, the degree of trust in political institutions has been continuously eroding.84
That trust in political institutions is crucial for democracy has become an accepted
proposition by now. Its role in new democracies, however, is even more critical. The
declining trust in political institutions might be detrimental in countries with fragile
democratic institutions, where citizens have not yet developed a value commitment
towards specific political institutions. Russia has already experienced the devastating
effect of such lack of commitment in its political institutions. In September 2004,
using the pretext of the hostage-taking tragedy in Beslan, Russia’s president abolished
regional elections, thus depriving the citizens their right to elect the government on
the regional level. Paradoxically, public opinion polls revealed considerable support for
Putin’s decision, highlighting the fact that people do not value the electoral mechanism
as such.85 This trend of curtailing democratic institutions has since continued in Russia
and, intriguingly, has not affected the rating polls of former president Putin or spoiled
his plans of having his close associate Medvedev elected as Russia’s new president in
March 2008.

Conclusion

Diverging from and modifying the scholarly trend in corruption research based on the
subjective evaluations of corruption, this article sheds light on public perceptions of
corruption as a phenomenon different from corruption per se. Based on the data from
the Russian regions, popular beliefs about corruption have high political costs in terms
of public trust in government, thus establishing that the formation of these beliefs re-
quires special scholarly consideration. The analysis of factors influencing public percep-
tions of corruption in the Russian regions reveals that people tend to see more corruption
in the context of lower regional economic development, freer (but not autonomous)
press, and more contentious politics.
These findings are pertinent in the context of the authoritarian turn in part of the
postcommunist region, helping to explain the underlying reasons behind the strategies
undertaken by the leaders of Russia, Belarus, and Central Asian states. To improve
control over population and public opinion, political elites in these states found it use-
ful to simply do away with contentious politics, political opposition, and free press
altogether. These policies could find public support (as in Russia, for example) only
because the reasoning provided for authoritarian change echoed the popular beliefs
about corruption and instability associated with pluralism, competition, and free press.
By highlighting the inconvenient link between free (but not autonomous) press, politi-
cal competition, and public perceptions of corruption in the Russian setting, this article
cautions against hasty acceptance of liberal Western assumptions about the nature of
free press and political competition for the analysis of competitive authoritarian re-
gimes as exist in Russia.
162
 Gulnaz Sharafutdinova

NOTES

I would like to thank Jevgenijs Steinbuks, Alexei Troshev, and the anonymous reviewers for their generous
and helpful comments and suggestions.
1. An English version of the speech is available at http:www.kremlim.ru/eng/speeches/2006/05/10/
1823_type70029type82912_105566.shtml. (All electronic references in these notes were last accessed on
December 14, 2009.)
2. “Sistemnyi otvet,” Rossiiskaia gazeta, July 4, 2008.
3. From the preliminary report on the Diagnostics of Russian Corruption: 2001-2005 (INDEM Foundation),
available at http://indem.ru/en/publicat/2005diag_engV.htm.
4. Georgii Satarov, ed., “Anticorruption Diagnostics in Russia: Sociological Analysis,” (Moscow, INDEM
Foundation, 2001), available at http://www.anti-corr.ru/awbreport/index.htm.
5. “Korruptsia – norma zhizni,” http://wciom.ru/arkhiv/tematicheskii-arkhiv/item/single/2861.html).
6. Ibid.; WCIOM Press-vypusk # 950 from May 7, 2008, entitled “Glavnye ugrozy dlia Rossii” (http://
wciom.ru/arkhiv/tematicheskii-arkhiv/item/single/10110).
7. Satarov, “Anticorruption Diagnostics in Russia.”
8. http://bd.fom.ru/report/cat/power/corr/d081021.
9. Quoted in Stephen D. Morris, “Political Corruption in Latin America: A Research Note,” unpublished
manuscript, 2004, p. 8, available at www.southalabama.edu/.../sdmweb/crptn%20in%20LA%20note.pdf.
10. Dilyan Donchev and Gergely Ujhelyi, “What Do Corruption Indices Measure?” unpublished
manuscript, 2009, available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id51124066.
11. Leslie Holmes, Rotten States? Corruption, Post-communism, and Neoliberalism (Durham, NC: Duke
University Press, 2006); Ivan Krastev, Shifting Obsessions: Three Essays on the Politics of Anticorruption
(Budapest: CEU Press, 2004); William L. Miller, Ase B. Grødeland, and Tatyana Koshechkina, eds., A
Culture of Corruption: Coping with Government in Post-communist Europe (Budapest: CEU Press, 2001);
Rasma Karklins, The System Made Me Do It: Corruption in Post-Communist Societies (Armonk, NY: M.E.
Sharpe, 2005); Anna Grzymala-Busse, “The Discreet Charm of Formal Institutions: Post-communist Party
Competition and State Corruption,” Comparative Political Studies, 39 (April 2006): 271–300; and Steven
Kotkin and Andras Sajo, eds., Political Corruption in Transition: A Skeptic’s Handbook (Budapest: Central
European University Press, 2002).
12. For an excellent review, see Johann Lambsdorff, “Corruption in Empirical Research – A Review,”
Transparency International Working Paper, 1999.
13. Susan Rose-Ackerman, Corruption and Government: Causes, Consequences, and Reform (New York:
Cambridge University, 1999); Arnold Heidenheimer, Michael Johnston, and Victor T. LeVine, eds., Political
Corruption: A Handbook (New Brunswick, NJ: Transaction Press, 1989); and Michael Johnston, “The
Political Consequences of Corruption: A Reassessment,” Comparative Politics, 18 (July 1986): 459–77.
14. Michael Johnston, Syndromes of Corruption: Power, Wealth and Democracy (New York: Cambridge
University Press, 2006).
15. David Hoffman, The Oligarchs: Wealth, Power, and the New Russia (New York: Public Affairs, 2002);
Chrystia Freeland, Sale of the Century: Russia’s Wild Ride from Communism to Capitalism (New York:
Crown Publishers, 2001); and Joel Hellman, “Winners Take All: The Politics of Partial Reforms in Post-
Communist Transitions,” World Politics, 50 (January 1998): 203–34.
16. Karklins, System Made Me Do It; Alena Ledeneva, How Russia Really Works: The Informal Practices
That Shaped Post-Soviet Politics and Business (Ithaca, NY: Cornell University Press, 2006); and Igor Kliamkin
and Lev Timofeev, Tenevaia Rossiia: ekonomiko-sotsiologicheskoe issledovanie (Moskva: RGGU, 2000).
17. Charles Sampford, Arthur Shacklock, Carmel Connors, Fredrik Galtung, eds., Measuring Corruption
(Aldershot, UK: Ashgate 2006).
18. On systematic corruption, see John J. Wallis, “The Concept of Systematic Corruption in American
History” in Corruption and Reform: Lessons from America’s Economic History, ed. Edward L. Glaeser and
Claudia Goldin (Chicago: University of Chicago Press, 2006).
19. Naci Mocan, “What Determines Corruption? International Evidence from Micro Data,” NBER Working
Paper No. 10460, May 2004; Roman P. Zyla, “Corruption in Ukraine: Between Perceptions and Realities,”
in State and Institution Building in Ukraine, ed. Taras Kuzio, Robert. S. Kravchuk, and Paul D’Anieri (St.
Martin’s Press, 1999).
20. Charles L. Davis, Roderic Ai Camp, and Kenneth M. Coleman, “The Influence of Party Systems on
Citizens’ Perceptions of Corruption and Electoral Response in Latin America,” Comparative Political Studies,
163
Comparative Politics January 2010

37 (August 2004), p. 678; Barbara Geddes and Artur Ribiero Neto, “Institutional Sources of Corruption in Brazil,”
Third World Quarterly, 13 (December 1992), p. 657; see also, Satarov, “Anticorruption Diagnostics in Russia.”
21. Geddes and Neto, “Institutional Sources,” p. 657.
22. Klaus Offe, “Political Corruption: Conceptual and Practical Issues,” in Building a Trustworthy State in
Post-socialist Transition, ed. Janos Kornai and Susan Rose-Ackerman (New York: Palgrave Macmillan, 2004),
pp. 86–87.
23. Virginie Coulloudon, “Russia’s Distorted Anticorruption Campaigns,” in Political Corruption in
Transition, ed. Stephen Kotkin and Andras Sajo (Budapest: Central European University Press, 2002), p. 188.
24. Ibid.
25. Davis, Camp, and Coleman, “Influence of Party Systems,” p. 700; and Mitchell Seligson, “The Impact
of Corruption on Regime Legitimacy: A Comparative Study of Four Latin American Countries,” Journal of
Politics, 64 (May 2002): 408–33.
26. As suggested by the senior economist of the World Bank (James Anderson, “Corruption in Latvia:
Survey Evidence,” report commissioned by the World Bank, http://www.politika.lv/en/index.php?id515662).
27. Krastev, Shifting Obsessions.
28. Richard Rose, William Mishler, and Christian Haerpfer, Democracy and Its Alternatives: Understanding
Postcommunist Societies (Baltimore, MD: Johns Hopkins University Press, 1998), p. 188.
29. Used as the title of Karklin’s 2005 book on corruption.
30. Ledeneva, How Russia Really Works; and Karklins, System Made Me Do It.
31. Miller et al., Culture of Corruption, pp. 61–62.
32. James Curran and Michael Gurevitch, eds., Mass Media and Society (New York: Oxford University
Press, 1991).
33. Karklins, System Made Me Do It, p. 51.
34. Shyamal K. Chowdhury, “The Effect of Democracy and Press Freedom on Corruption: An Empirical
Test,” Economic Letters, 85 (October 2004): 93–101; and Rudiger Ahrend, “Press Freedom, Human Capital
and Corruption,” DELTA Working Paper 2002–11, 2002, available at http://ideas.repec.org/p/del/abcdef/
2002-11.html.
35. Remzi Lani and Frokk Cupi, “The Difficult Road to the Independent Media: Is the Post-Communist
Transition Over?” Journal of Southeast European and Black Sea Studies, 2 (January 2002): 75–90; and Colin
Sparks, Communism, Capitalism and Mass Media (Tirana: AMI, 2000).
36. Laura Belin, “Political Bias and Self-Censorship in the Russian Media,” in Contemporary Russian
Politics: A Reader, ed. Archie Brown (New York: Oxford University Press, 2001); and Chris Groner, “The
Post-Communist Media: Comparative ‘Oligarchisation’ and After,” Slovo, 18 (January 2006): 5–41.
37. Ledeneva, How Russia Really Works.
38. Stephen Ansolabehere and Shanto Iyengar, Going Negative: How Political Advertisement Shrink and
Polarize the Electorate (New York: Free Press, 1997).
39. Gulnaz Sharafutdinova, “Why Was Democracy Lost in Russia’s Regions? Lessons from Nizhnii
Novgorod,” Communist and Post-Communist Studies, 40 (September 2007): 1–20; and Joel Moses, “Political-
Economic Elites and Russian Regional Elections 1999–2000: Democratic Tendencies in Kaliningrad, Perm
and Volgograd,” Europe-Asia Studies, 54 (September 2002): 905–31.
40. Daniel Kaufmann, Aart Kraay, and Pablo Zoido-Lobaton, “Governance Matters,” World Bank Policy
Research Paper No.2196, 1999.
41. As suggested by Lambsdorff, “Corruption in Empirical Research.”
42. Donchev and Ujhelyi, “Corruption Indices,” p. 14.
43. Natalya V. Tchernina and Efim A. Tchernin, “Older people in Russia’s transitional society: multiple
deprivation and coping responses,” Ageing & Society, 22 (2002): 543–62; Vladimir Mikhalev, “Social
Security in Russia Under Economic Transformation,” Europe-Asia Studies, 48 (January 1996): 5–25.
44. The protests most threatening to Putin’s regime were initiated by pensioners in January–February
of 2005, when the government replaced the system of benefits to retired and disabled people by a petty
monetary compensation.
45. See Phyllis Dininio and Robert W. Orttung, “Explaining Patterns of Corruption in the Russian
Regions,” World Politics, 57 (July 2005): 500–29.
46. Dininio and Orttung have already noted a very low level of correlation between the two.
47. The sample size is about 7500 households and business representatives. The more detailed information
about the data and methodology is available from the author upon request.
48. The description of the project’s methodology includes a discussion of the technique used in this study,
available at http://www.transparency.org.ru/proj_index.asp.
164
 Gulnaz Sharafutdinova

49. The description of the methodology used for creating indices is available at http://www.transparency.
org.ru/proj_index.asp.
50. I also tested the model using the regional press freedom rating produced by the Glasnost Defense
Foundation (http://www.gdf.ru/map). Their indicator is a categorical variable, taking values from 1 (least
free) to 4 (most free). The results were generally consistent with those presented here.
51. The other two dimensions were constructed based on the analysis of the legislative regulation of
the press in the regions as well as the responsiveness of regional governments. The preliminary tests
showed that these two dimensions (access to and production of information) were not correlated with
the dependent variable.
52. Kelly M. McMann and Nikolai V. Petrov, “A Survey of Democracy in Russia’s Regions,” Post-Soviet
Geography and Economics, 41 (April–May 2000): 155–82.
53. Christopher Marsh, “Social Capital and Democracy in Russia,” Communist and Post-Communist
Studies, 33 (June 2000): 183–99; and Tatu Vanhanen, Prospects of Democracy: A Study of 172 Countries
(New York: Routledge, 1997).
54. Bryon Moraski and William M. Reisinger, “Explaining Electoral Competition across Russia’s Regions,”
Slavic Review, 62 (Summer 2003): 278–301.
55. http://www.freepress.ru/publish/publish043.shtml.
56. Nikolai Petrov, “Russia’s ‘Part of Power’ Takes Shape,” Russia and Eurasia Review, Jamestown
Foundation, 2 (16), 2003.
57. In 2003 thirty out of Russia’s eighty-nine regional leaders were members of this party (Caroline
McGregor, “United Russia Betting on Governors,” Moscow Times, October 15, 2003).
58. Reported in the project description available at http://www.freepress.ru/publish/publish043.shtml.
59. This project was a result of collaboration of three respected Russian think tanks (INDEM, Public
Expertise, and Merkator) and is arguably one of the most accurate systematic assessments of regional-level
democracy in Russia under Putin.
60. www.gks.ru.
61. In doing that, I assume that regional-level variables are strictly exogenous with respect to individual
perceptions of corruption. Under this assumption, OLS yields unbiased and consistent estimates. Though there
are statistical methods that may improve the efficiency of the estimates (Thomas M. Stoker, “Empirical
Approaches to the Problem of Aggregation Over Individuals,” Journal of Economic Literature, 31 [1993]:
1827–74), the results are already statistically significant. Therefore more precise statistical method will only
strengthen the results of this paper.
62. To even the measurement scales, I used the logarithms of gross regional product indicator.
63. In the earlier specification of this model I included the variable of ethnic heterogeneity (measured by the
proportion of non-ethnic Russian population in the regions) and found it to be highly correlated with political
competition measure. I excluded this variable from the final model because my earlier research suggested a
causal relationship between these two variables (Sharafutdinova, “When Do Elites Compete? The Determinants
of Political Competition in Russian Regions,” Comparative Politics, 38 [April 2006]: 273–93).
64. I tested for a potential nonlinear effect of press freedom on corruption perceptions (hypothesized
by Dininio and Orttung, p. 526) by creating dummy variables for the upper and lower segments of the
press freedom variable and estimating regression with these dummies. The results were consistent with the
linear analysis.
65. On the role of television in Russia, see Ellen Mickiewicz, Changing Channels: Television and the
Struggle for Power in Russia (Durham, NC: Duke University Press, 1999).
66. Davis, Camp, and Coleman, “Influence of Party Systems.”
67. Dininio and Orttung, “Explaining Patterns of Corruption.”
68. I tested this model removing the insignificant variables. The results highlighted the significance of
political competition and economic development variables revealing that the impact of political competition
is more robust than the press freedom variable.
69. The ranges between the categories are arbitrary; therefore, to avoid the bias in model specification I
decomposed this variable into dummy variables.
70. Vitalii Silitski, “Has the Age of Revolutions Ended?” Transitions Online, January 13, 2005; and Lucan
Way, “Kuchma’s Failed Authoritarianism,” Journal of Democracy, 16 (April 2005): 131–45.
71. Dininio and Orttung, “Explaining Patterns of Corruption,” p. 504.
72. Edward L. Glaeser, Rafael La Porta, Florencio Lopez-de-Silanes, and Andrei Shleifer, “Do Institutions
Cause Growth?” Journal of Economic Growth, 9 (June 2004): 271–303.
73. Karklins, “System Made Me Do It,” p. 5.
165
Comparative Politics January 2010

74. Janos Kornai and Susan Rose-Ackerman, eds., Building a Trustworthy State in Post-Socialist Transition
(New York: Palgrave Macmillan, 2004).
75. Grzymala-Busse, “Discreet Charm”; and Rebecca Bill Chavez, “The Construction of the Rule of Law in
Argentina: A Tale of Two Provinces,” Comparative Politics, 35 (July 2003): 417–37.
76. Joel S. Hellman, Geraint Jones, Daniel Kaufmann, and Mark Schankerman, “Measuring Governance,
Corruption and State Capture: How Firms and Bureaucrats Shape the Business Environment in Transition
Economies,” Policy Research Working Paper 2312, 2000.
77. Sharafutdinova, “Why Was Democracy Lost”; Moses, “Political-Economic Elites”; and Rustem
Bikmetov, “Rol’ sredstv massovoi informatsii v vyborakh gubernatora Nizhegorodskoi oblasti 2001” in
Regionalnye vybory i problemy grazhdanskogo obshchestva v Povolzhie, Carnegie Moscow Center, Working
Papers (3): p. 137.
78. Moses, “Political-Economic Elites,” p. 907.
79. Sharafutdinova, “Why Was Democracy Lost.”
80. Ivan Zassoursky, Media and Power in Post-Soviet Russia (Armonk, NY: M.E. Sharpe, 2004); Belin,
“Political Bias.”
81. Natalia Roudakova, “Media-Political Clientelism: Lessons from Anthropology,” Media, Culture and
Society 30 (1), 2008, pp. 41–59.
82. See, for example, Andrew Wilson, Virtual Politics: Faking Democracy in the Post-Soviet World
(New Haven, CT: Yale University Press, 2005).
83. William Mishler and Richard Rose, “What Are the Origins of Political Trust? Testing Institutional and
Cultural Theories in Post-Communist Societies,” Comparative Political Studies, 34 (February 2001): 30–62.
84. This has been illustrated by Levada polls. See also, Mishler and Rose, “Origins of Political Trust.”
85. http://www.levada.ru/press/2004092902.html.

166