Public Perceptions of Electoral Fairness in Russia

Russians could scarcely be said to have ‘chosen’ their electoral system. The 1993 constitution (Konstitutsiya 1993), itself adopted in controversial circumstances, set out no more than a few basic principles: citizens had the right to vote and to be elected to the organs of state power and local government (article 32), including the State Duma, which was to be elected every four (later five) years at a time (articles 96 and 97). The manner in which they were to elect their representatives, however, was to be established by subsequent legislation (article 96), and the result was a complex and changing set of provisions that set out both the general framework for elections at all levels (the ‘framework law’, last adopted in 2002 (‘Ob osnovnykh’ 2002)) and the specific arrangements that would govern elections to the State Duma. So far, a different set of arrangements has governed each successive election, up to and including the law of 2005 that governed the parliamentary elections of 2007. Every law has been longer than its predecessor, in an attempt to provide for every eventuality. (Even so, the 2005 law had been amended 14 times by the spring of 2010.) The other constant—not, of course, unique to Russia—is that all of these changes have generally been conceived by the ruling group of the time as a means of advancing its own interests.1

Elections, however, even in a Russian environment, also have a popular dimension. Turnout, until the law was amended in 2007, had to reach a certain minimum: 25% in the case of the Duma, and 50% in the case of presidential elections. The vote ‘against all’ could not exceed the vote for any of the individual candidates in a single-member constituency (until another amendment in 2006) or the election would have to be repeated (as in three of the constituency seats in the 2003 parliamentary election). Formal rules apart, elections had to confer a degree of legitimacy on each successive government such that it could take decisions that would be accepted and implemented, even if they were not in the immediate interests of those who were affected by them. The ‘coloured revolutions’ in other post-Soviet republics were one form of reaction to the conduct of elections in a manner that was thought to misrepresent the wishes of the electorate (Lane & White 2010; Ó Beacháin & Polese 2010). More generally, a lack of confidence in the electoral system as a means of expressing the preferences of the public at large was typically associated with lower levels of turnout and the vote for minor parties, and lower levels of confidence in government itself (Bélanger & Nadeau 2005; Birch 2010).

Russian elections, for such reasons, raise most if not all of the issues that arise in other contexts about the relationship between regime and society, and about the most important of the mechanisms that connect them. Russians, for instance, voted in relatively large numbers in the post-communist years: nearly 64% cast a ballot in December 2007 and 70% in March 2008, which compared well with the larger Western democracies; but they had voted in even larger numbers in the communist period, and there were various forms of administrative influence that helped to raise turnout in subsequent years, in some parts of the country to entirely ‘Soviet’ levels. For example in Chechnya, in December 2007, it reached a record-breaking 99.5%, comfortably exceeding the 87% for the Russian Republic as a whole that had been recorded at the last Soviet elections in 1989.2

Aggregate turnout figures, in such circumstances, are of limited value as a means of assessing popular orientations. We also need more sensitive, individual-level indicators of a kind that are most readily provided by nationally representative surveys. In this essay we draw on an exercise of this kind conducted shortly after the December 2007 Duma election, and on earlier exercises of the same kind, in order to explore how Russians perceived the contest in which they had just engaged, and in particular, how ‘fair’ they thought it had been. In the later parts of the essay we consider these findings in a broader cross-national context. The comparative evidence can help us ‘place’ Russia more securely. The Russian case can, in turn, make its contribution to a comparative literature that has perhaps been overly preoccupied with the wholly competitive elections that take place in the established Western democracies, which continue to account for a small minority of the world's political systems.3

How ‘fair’ are Russian elections?

Standard definitions of democracy assume, as a minimum requirement, the existence of free, competitive elections. However, as elections have proliferated around the world in many diverse settings, it has become apparent that while elections may often be ‘free’, ruling parties and incumbent elites may still exercise such an influence on the results that they can hardly be considered ‘fair’. This can be achieved in a variety of indirect ways, such as stifling debate or restricting media freedom, so that an election, while nominally competitive and open to no serious objection in the way in which the vote is counted, may all the same be heavily weighted in favour of the governing authorities.

Schedler has identified seven requirements that must be satisfied before an election can be considered fully ‘free and fair’. It must be the means by which those who hold power are selected; there must be an unrestricted choice of alternatives; voters must be able to form their preferences; there must be universal suffrage; citizens must be free to express their preferences; their votes must be fairly aggregated; and they must provide a mandate for the government that is formed as a result. These, for Schedler, were ‘coherent configurations of essential attributes’, the violation of any one of which would disqualify the entire exercise (2002, pp. 40–41). We use ‘fair’ in this context to identify an election that fails to meet one or more of these criteria. Moreover, like Rose and Mishler (2009, p. 118), by placing the emphasis on ‘fair’ we avoid the catchphrase ‘free and fair’, in which ‘multiple desiderata are indiscriminately lumped together’.

This new focus on the fairness of elections has shifted attention away from formal procedures and towards a more complex examination of the extent to which arrangements of this kind give ordinary citizens a real measure of influence over those who govern them. First-hand evaluations of the nature of the contest as it is experienced by those who actually engage in it have accordingly become increasingly common in many of the post-communist and post-authoritarian countries, conducted by a range of agencies acting on behalf of national governments and the international community as a whole (Elklit & Reynolds 2005; Kelley 2008). A particularly important monitoring framework is operated by the Office of Democratic Institutions and Human Rights, an agency of the Organization for Security and Co-operation in Europe (OSCE), which has been reporting on the quality of Russian elections since the presidential contest in the summer of 1996 (OSCE-ODIHR 1996). The European Institute of the Media, based in Düsseldorf, began its different monitoring exercises even earlier, in December 1993 (Lange 1994), in both cases as part of a much larger international programme. Exercises of this kind have generated a mass of country-specific reports, based on increasingly detailed guidelines collected in a variety of handbooks and manuals.4 Agencies of this kind regularly pronounce not just on whether an election can be considered to be ‘free’, but whether it can also be considered to be ‘fair’.

The concept of ‘fairness’ is a qualitative judgment that voters themselves are in an ideal position to make, above and beyond the strict definition used by Schedler (2002). It is voters who experience the election campaign, directly through contact with election candidates and parties, and indirectly through the mass media. Unlike outside observers, they will also cast a ballot and therefore have a degree of ‘ownership’ over the election outcome. Voters' perceptions of fairness, in turn, will have consequences for their evaluation of the political system; if they regard the system as fair, they will normally be more likely to vote, and democratic participation has in turn been shown to lead to greater regime legitimacy as well as greater satisfaction with democracy in general (Anderson & Guillory 1997; Anderson et al. 2005; Birch 2010). Conversely, when voters regard the electoral process as unfair, they will be more likely to withhold their support from the regime; one consequence of this may be a decline in turnout, a phenomenon that has been especially marked in the established democracies in recent years (Dalton & Wattenberg 2000; Franklin 2004). In newly democratising systems a lack of trust can, in extreme cases, lead to a challenge to the legitimacy of the system itself, through mass protests. It was certainly a common characteristic of the ‘coloured revolutions’ throughout the post-Soviet countries that they followed elections that had effectively been ‘stolen’, and stability was not restored until a new contest had produced a result that was generally regarded as legitimate (Tucker 2007; Kuntz & Thompson 2009).

Since the end of communist rule there have been four presidential and five parliamentary elections in Russia, at all of which there was a choice of candidates and parties, but there has been increasing international concern that the context within which the competition takes place has privileged the Kremlin and its chosen contenders. This has certainly been clear from the successive reports of the OSCE's monitoring missions. The 1999 parliamentary election, according to the OSCE's final assessment, had been a ‘benchmark in the Federation's advancement toward representative democracy’ in spite of ‘episodic challenges that could have undermined the general integrity of the process as a whole’ (OSCE-ODIHR 2000). The 2003 report was much less complimentary, noting that ‘while well-administered’ the Duma election of that year had ‘failed to meet a number of OSCE commitments for democratic elections, most notably those pertaining to: unimpeded access to the media on a non-discriminatory basis, a clear separation between the state and political parties, and guarantees to enable political parties to compete on the basis of equal treatment’ (OSCE-ODIHR 2004). In 2007 and 2008, after a disagreement with the Central Electoral Commission on the basis on which they could operate, there were no OSCE observers at all.5

Survey-based investigations have also reached a variety of conclusions, and not simply because of differences in sample size or question wording. Not surprisingly, the Russian authorities themselves have focused most attention on the generally supportive conclusions of the All-Russian Centre for the Study of Public Opinion (VTsIOM), which has been under effective state control since 2003 and which received a contract from the Central Electoral Commission itself to monitor changes in public attitudes over the two most recent election periods (Vybory 2008a, p. 368; 2008b, p. 227). Regular surveys were also conducted by the Public Opinion Foundation (FOM) and by the Levada Centre, which has generally been considered the most independent of the major agencies (it was one of the winners of the traditional competition for the most accurate prediction in December 2007, although most other estimates were close to the final outcome (Vybory 2008a, p. 370)). The findings of these various agencies were not always directly comparable, for technical and other reasons.6 All the same, it was clear that they had reached rather different conclusions about the nature of the electoral process, ranging from the generally supportive results that were reported by state-sponsored VTsIOM to the much less enthusiastic findings of the Levada Centre.

There was, in fact, a generally shared consensus that the higher the level of government, the more likely the corresponding elections would be properly conducted (see Table 1). National parliamentary elections, on the whole, were regarded as more likely to be honestly conducted than regional or local ones; elections to the presidency were thought to be more honest than either of them. There was also some indication that elections were more likely to be thought to have been honestly conducted after they had taken place than beforehand; in the case of the parliamentary elections, for instance, 39% thought they were likely to be honestly conducted in September 2007, but 52% the following January. On VTsIOM's evidence, younger respondents were more likely to believe in the honesty of the electoral process than older ones, but all age-groups were strongly supportive; so were more affluent citizens, and those who intended to vote for Dmitri Medvedev in the presidential contest. Supporters of the parties that had been successful in the Duma election, not surprisingly, were also more likely to believe in its integrity: United Russia (Edinaya Rossiya) and A Just Russia (Spravedlivaya Rossiya) supporters were the most positive, Communists (Kommunisticheskaya partiya Rossiiskoi Federatsii) and habitual non-voters rather more sceptical (‘O chestnosti’ 2008).

TABLE 1 
Perceptions of Election Fairness at Different Levels of Government 2007–2008 (VTsIOM data )

	Date of survey	Entirely honestly	Somewhat honestly	Somewhat dishonestly	Entirely dishonestly	Hard to say
President	May 2007	17	45	18	6	13
	Sept 2007	15	42	19	7	17
	Jan 2008	15	49	15	4	17
Duma	May 2007	7	33	34	10	17
	Sept 2007	7	32	32	12	18
	Jan 2008	9	43	22	5	20
Regional assemblies	May 2007	6	29	35	11	19
	Sept 2007	7	28	33	13	19
	Jan 2008	8	40	24	5	23
Heads of towns and districts	May 2007	6	28	33	13	20
	Sept 2007	7	27	33	14	20
	Jan 2008	8	38	24	7	24
The question was: ‘How honestly, in your view, do election campaigns usually take place to the various organs of government?’.
Source: adapted from ‘O chestnosti vyborov v Rossii’, Press-vypusk, No. 877, 21 February 2008, available at: http://wciom.ru/index.php?id=268&uid=9667, last accessed 23 January 2011; rounded percentages.

It was also clear, on VTsIOM figures, that ordinary Russians were increasingly inclined to believe their elections had taken place ‘openly, democratically and honestly’, and that the results could be regarded as reliable. This was certainly true of Duma elections (Table 2), and the same was true of presidential elections, according to the same source: more than half (53%) in both March 2004 and March 2008 thought they had been an authentic expression of the wishes of the electorate; just 8% or 10%, respectively, thought they could not be trusted (‘Tsentrizbirkom’ 2008). As in the Duma election, there were more who thought the election had been conducted honestly after it had taken place than who expected it would be conducted honestly when they were asked two months beforehand, when just 44% thought the results would be reliable and the same proportion thought they would be largely or entirely falsified (Buzin & Lyubarev 2008, p. 46). In the event of any disagreement it was the Central Electoral Commission, a state body, whose views were generally regarded as the most authoritative, followed by independent observers from Russian public organisations; very few (3–4%) mentioned the OSCE, although the presence of international observers of this kind was generally welcomed (Buzin & Lyubarev 2008, p. 48).

TABLE 2 
Perceptions of Fairness in the 2003 and 2007 Duma Elections (VTsIOM data )

	2003	2007
How would you assess the recent election campaign for the State Duma?
‘The elections took place in an open, democratic and honest political struggle’	49	53
‘These elections cannot be considered wholly free and democratic’	26	19
‘Hard to say’	25	28
To what extent do you think you can trust the results of the parliamentary election?
‘I think the results are reliable and correspond to the expressed wishes of the electorate’	39	45
‘There might have been some falsifications in local areas but they did not affect the results of the voting in the country as a whole’	33	29
‘I think the results of the elections should not be trusted’	14	12
‘Hard to say’	14	14
Source: adapted from ‘Vybory-2007: byli oni chestnymi, svobodnymi i demokratichnymi?’, Press-vypusk, No. 838, 17 December 2007, available at: http://wciom.ru/index.php?id=268&uid=9384, accessed 23 January 2011; rounded percentages.

The Public Opinion Foundation, established in 1992 and closely associated with the Yel'tsin and Putin election campaigns, was rather less reassuring. The counting of votes had certainly been conducted honestly, for a majority, and more so than in previous years (Table 3). Indeed, not a single respondent had voted for a friend or relative, or had seen anyone doing so; about two-thirds reported the presence of observers. There were relatively few (about 9% of voters) who had been dissatisfied with arrangements at the polling station itself, in some cases because the atmosphere had simply been too ‘workaday’. There were objections, for instance, to the ‘lack of curtains in the voting booth’, to ‘long queues’ and the refusal of members of the electoral commission to ‘respond to questions’. In some cases the ‘space was too small’, and there had been an ‘unpleasant, musty smell’; there were occasional problems with names that had been missing from the electoral register, and a few were simply disappointed by the lack of spectacle: they had been ‘boring—elections used to be a festival, and now there's just a buffet, not even a concert’, was one complaint; others reported that ‘they gave out free tickets to a discotheque, but not to me’, or that ‘the price of beer was too high’ (‘Protsedura’ 2007).

TABLE 3 
Perceptions of Fairness in Duma Elections, 1999–2007 (FOM data )

	1999	2003	2007
In your view, did the counting of votes in the parliamentary election take place honestly, or were the results tampered with or falsified?
It took place honestly	38	52	53
It was falsified	32	18	11
Hard to say	30	30	36
Are you personally satisfied or dissatisfied with the results of the parliamentary elections?
Satisfied	39	49	65
Indifferent	20	19	14
Dissatisfied	25	19	11
Hard to say	16	14	10
Source: adapted from Vybory (2008a, p. 373); rounded percentages.

Reflecting on the longer term, however, there were more who thought elections conducted between 1999 and 2007 did not reflect public opinion than who took the opposite view; and there were more who thought the elections to the Duma over the previous 15 years had been corrupt than who thought they had been fair (Table 4). About 6% of all respondents had engaged themselves in some kind of violation at one of the elections that had taken place over this time, or knew of a friend or relative who had done so. In some cases, according to the responses to an open question, it had been straightforward bribery (‘they offered R250 [about $10] to vote for their candidate’; ‘they handed out money and the whole courtyard got drunk for a week’). In other cases it had been personation (‘you come to vote and it turns out you have already voted’; ‘people voted for their relatives even though they were already dead’), or direct falsification of the votes cast, or pressure to vote for the ruling party or risk dismissal, or improper attempts to influence the vote on polling day itself. The best way of dealing with these various problems, according to FOM's respondents, was to increase the level of transparency of the entire exercise, with greater use of observers and ‘strict control of the officials who organise the elections’ (Sedov 2007c).

TABLE 4 
The Honesty of Elections in 2003 and 2007 (FOM data )

	2003	2007
Do elections reflect public opinion?
Yes	38	35
No	43	50
Hard to say	21	15
Were elections to the Duma over the past 15 years generally honest or dishonest?
Honest	n.d	26
Dishonest	n.d.	34
Hard to say	n.d.	39
The questions were: ‘Some people think that election results reflect public opinion. Others think that election results do not reflect public opinion. With which view, the first or the second, do you agree?’; ‘Speaking of the elections to the State Duma that have taken place over the past 15 years, would you say that on the whole they were honestly or dishonestly conducted?’.
Source: adapted from Public Opinion Foundation (FOM) (2007); rounded percentages.

The Levada Centre's conclusions were still more sceptical. In September 2007, for instance, just 16% thought Russian elections took place ‘in complete accordance with the law and [reflected] the views of the majority of electors’; a more substantial 43% thought ‘various forms of manipulation and falsification of the results [were] possible in the process of voting and counting’, and another 26% thought the authorities used their administrative influence ‘systematically and constantly’ to manipulate the results. Levels of distrust, on this evidence, had increased substantially since the previous parliamentary elections in 2003 (Sedov 2007a). Some 17% had themselves or through their family encountered cases in which the ‘representatives of the local administration, the management at their workplace, electoral commission or other official figures had bribed or threatened them to take part in elections’ (‘Chestnost'’ 2007). For a majority, surveyed a month beforehand, there would be ‘only the appearance of a contest’ in March 2008 when the Kremlin's favoured candidate, Dmitri Medvedev, was obviously the leading contender (Table 5).

TABLE 5 
Expectations of the Presidential Elections, 2005–2008 (Levada Centre Data )

	April 2005	April 2006	April 2007	October 2007	January 2008	February 2008
‘There will be a real contest among the candidates’	29	33	34	32	37	30
‘There will only be the appearance of a contest: the president will be the one Putin selects/the distribution of the vote will be decided by the authorities’	57	51	49	54	48	58
‘Hard to say’	14	17	17	14	15	13
The question wording was: ‘What do you think, at the presidential election of 2 March of this year, will there be a real competition of candidates or only an imitation of it, and the distribution of the vote will be decided by the authorities?’
Source: adapted from ‘Kak vy schitaete … ’, n.d. [February 2008], available at: http://www.levada.ru/vybory20080.html, last accessed 24 January 2011; rounded percentages.

In a separate exercise in the summer of 2007, the Levada Centre found that only 28% thought the Duma elections that were due to take place the following December would be honestly conducted; more than twice as many (58%) took the opposite view. The most likely departures from electoral propriety were expected to include ‘dirty tricks’ such as the leaking of compromising material (35%); the bribery of electors by the parties (30%); falsification of the results in electoral commissions at various levels (28%); the manipulation of party lists and ballot papers in polling stations (27%); different degrees of media access for the parties (26%); pressure on electors by local authorities to persuade them to vote for a particular party (22%); biased coverage of the campaign on the main television channels (the same); and bias by electoral commissions in registering—or otherwise—a national list of candidates (20%). Only 12% expected ‘no violations or manipulations’ of any kind; and it was the Kremlin's favoured party, United Russia, that was expected to be the main beneficiary (Sedov 2007b).

Would these various abuses make a difference? Nearly half of those who were asked by the Levada Centre in the summer of 2007 thought they would have a ‘significant’ or ‘very significant influence’ on the results, more than twice as many as those who took a different view. Only 15% thought it would depend on the ‘will of the electorate’ which parties won seats in the Duma election; another 28% thought it would depend on the efforts of the parties themselves, but nearly half thought it would depend on President Putin (23%) or his senior officials (22%). Just a third (32%) thought the various forms of monitoring by political parties, journalists and foreign observers would ‘significantly reduce these abuses’. Taking everything into account, would the December elections ‘reflect the will of the people of Russia?’. For the largest group of respondents, the answer was ‘definitely’ or ‘probably not’ (45%); rather fewer (40%) took a more optimistic view (Sedov 2007c).

The relatively close fit between survey predictions, exit-poll results and the results themselves was noted by the Chairman of the Central Electoral Commission, Vladimir Churov, in his report to the first session of the new Duma. As he told the deputies,

Preliminary sociological investigations conducted by ten independent organisations, the results of the exit polls conducted on polling day by two independent sociological campaigns [sic], and the parallel count conducted on the basis of copies of 42% of the protocols by one of the parties, almost completely coincide with the official results of the count at 100% of the polling stations in the Russian Federation, a total of almost 96,000. (‘Vystuplenie’ 2007)

Putin himself expressed particular satisfaction that the parties that had won seats accounted for more than 90% of the popular vote, rather than the previous 70%, which was a result that was likely to improve the legitimacy of the Duma within which they were represented.7

The counting of ballots or indeed the fit between the preferences of the mass electorate and the distribution of the vote, in fact, were hardly the point. As two Russian scholars pointed out,

The main shortcomings of at least the December 2007 Duma elections [were] not direct falsifications … but the lack of political competition, of equality among the contenders, and the massive use of administrative resources. These kinds of violations are not easily identified and documented but they can be easily appreciated at an intuitive level. (Buzin & Lyubarev 2008, p. 50)

The fact that so many falsifications had been identified, in spite of the fact that every effort had been made to avoid reporting them in the press, gave every reason to believe they had been widely distributed. On this basis, Buzin and Lyubarev suggested, the December 2007 Duma elections had been more unfair than the elections that had preceded them at the end of the 1990s and the early years of the new century, and they had been the first elections in which such abuses had directly affected the outcome (Buzin & Lyubarev 2008, pp. 50, 176).

Examining ‘fairness’ in Russian elections

We seek in what follows to advance these discussions further by drawing on a series of nationwide surveys conducted since 2000 and normally based on a larger sample size than the VTsIOM, FOM or Levada surveys described in the preceding section.8 The surveys on which we draw were conducted not only in the immediate post-election period, when political interest is traditionally high, but also in the inter-election periods. In addition, the surveys asked questions that related to engagement in a wide range of political actions and feelings of efficacy, or the perceived capacity to influence political outcomes. This enables us to test not only the main explanations of perceptions of fairness in elections and how they are related to political attitudes, but also to examine their consequences for the political system more generally. We know, for example, that electoral participation leads to stronger feelings of efficacy and trust in the political system, as well as ‘satisfaction with democracy’ (Banducci & Karp 2003; Karp & Banducci 2008). However, there is much less research on how perceptions of electoral fairness affect such attitudes.9

Mirroring the results presented in the preceding section, the five surveys we have conducted since 2000 show that the Russian public has mixed views about the fairness of Duma elections and that there has been relatively little change in their views over the entire period (Figure 1). In terms of the counting of the vote, in 2001 almost seven out of every 10 respondents thought the process was fair ‘to a significant extent’ or at least ‘to some extent’; by 2010 the corresponding figure was 64%, demonstrating that there had been relatively little change over almost a decade. There had, however, been a noticeable drop in the proportion who believed in the basic fairness of the electoral system in the middle of the period and especially in 2005, when only 56% were prepared to take this view. This short-term change in public perceptions is likely to have been influenced by the abolition of the direct election of governors at the end of 2004, and of single-member constituencies in the new Duma election law that was adopted in May 2005—both of which were radical and, for most Russians, unwelcome moves.

FIGURE 1 Perceptions of Fairness of Duma Elections, 2001–2010

The other aspects of the election campaign covered in Figure 1—fairness in the television coverage of the election campaign and fairness in the conduct of the local election campaign—show a similar pattern to the counting of the votes, with around half or more seeing them as fair over the course of the decade. The main exception was again in 2005, when just under half saw the local election campaign as fair. Perceptions of media coverage of elections is particularly important in new democracies, as voters often lack stable partisan cues and are more likely to use the information they pick up through other sources to guide their choices (White & McAllister 2006; White et al. 2005). Nevertheless, what the results in Figure 1 suggest is that a sizeable minority of the Russian public are sceptical about the fairness of elections, and that their reservations extend not just to the counting of votes but to the conduct of the election campaign as a whole, ranging from media coverage of events and personalities to what occurs at the local level.

Attempts can be made to influence voters in a variety of ways, from direct threats or bribes to indirect inducements of one sort or another. Three of the surveys—those conducted in 2000, 2005 and 2008—asked the respondents if their vote had been cast under any form of pressure, and if so, whether or not that pressure had influenced their choice at the ballot box (Table 6). In 2000, just over 3% of those who cast a ballot reported that they had been subjected to some form of pressure, a figure which increased to 4% in 2005, and then doubled to 8% in 2008. Those who said that the pressure had changed how they voted were a small minority—just 1% of all voters in 2008. Nevertheless, the results do suggest that an increasing number of voters feel that they vote under some kind of external influence.10

TABLE 6 
Votes Cast Under Pressure, 2000–2008

	2000	2005	2008
Yes, vote made under pressure	0.3	0.5	1
Yes, vote made under pressure but made no difference to my choice	3	4	8
No pressure	97	96	91
Total	100	100	100
(N)	(1,157)	(1,790)	(1,723)
The question wording was: ‘Could you please tell whether anyone made any attempts through giving presents to you, threatening you or in any other way to influence your decision on your participation/non-participation in the election or on your choice how to vote in the election to the State Duma?’
Source: authors' surveys (see note 8); percentages.

Did the belief that the election might be corrupt deter respondents from voting? The 2003 and 2008 surveys asked respondents who had reported abstaining to give the reasons for their decision.11 This was done in the form of an open-ended question, which was subsequently coded into a wide range of broad categories and then further refined into the seven categories used in Table 7. As most previous studies have demonstrated, for example Marsh (1991), abstention generally occurs by accident rather than design; most citizens intend to vote, or at least do not have a principled objection to voting, but events and circumstances intervene. While personal circumstances are important in explaining non-voting in Russia—in both surveys around one in three non-voters said that they had simply been unable to get to the polling booth—it is notable that the most frequently mentioned reason for non-voting was a general sense of alienation from or distrust of the political system.12 In 2008, 44% of those who had not voted mentioned alienation or distrust as the reason for abstaining, compared to 32% who indicated that they had simply been unable to get to the polling booth. Most pertinent from our perspective is the relatively small number who explained their abstention by a fear that the election result would be falsified; in 2003 and 2008 just 4% gave this as a reason for abstaining.13 Perceptions of electoral corruption, then, do not appear to significantly deter Russian citizens from attending the polls.

TABLE 7 
Reasons for Abstention, 2003 and 2008

	2003	2008
Could not get to polling station	35	32
Indecision, lack of choice	9	7
Alienation, distrust	39	44
Results will be falsified	4	4
Duma irrelevant	3	3
No political interest	8	4
Other	2	6
Total	100	100
(N)	(508)	(425)
The question wording was: ‘Why didn't you participate in the last State Duma/Presidential election?’
Source: authors' surveys (see note 8); rounded percentages.

In summary, the results suggest that a sizeable minority of Russian citizens have some concern about the fairness of Duma or presidential elections, and that there has been relatively little change in these proportions over the last decade. However, these patterns are very much perceptions; those who report some form of pressure aimed at influencing their vote represent a very small (though increasing) proportion of the electorate. Perhaps more interesting is the relative stability of these results; most studies of new democracies, such as Norris (1999), find that as competitive politics develop and become more embedded within the attitudes and values of the electorate, confidence and trust in electoral institutions increase in something like the same proportion. That does not appear to have occurred in Russia, perhaps because what would otherwise have been an increasing confidence in competitive politics has been undermined by the establishment of what President Medvedev himself conceded was effectively a one-party state that allowed few opportunities to influence it through the ballot box, and which was developing signs of a ‘stagnation’ that would be ‘just as fatal for the ruling party as for oppositional forces’ (Medvedev 2010).

Explaining attitudes towards election corruption

The results presented so far represent something of an anomaly. There is a sizeable minority who regard the conduct of Russian elections as essentially unfair, ranging from the counting of votes to the coverage of the campaign on television. On the other hand, our surveys suggest that only a very small proportion of citizens have any direct and personal experience of electoral irregularity. It follows that this widely shared sense of unfairness does not have its origin in that personal experience, but must rather derive from other aspects of the political system. In other words, in line with Buzin and Lyubarev (2008), it is not the casting of ballots that Russians regard as unfair so much as the wider context within which the electoral contest takes place. In what follows we test five explanations to account for these perceptions.

The first explanation sees the origins of perceptions of electoral unfairness in the socioeconomic position of citizens. For instance, those with tertiary education may have a broader worldview than many of their fellow citizens and, as a consequence, may view the political system unfavourably when they compare it with those in the Western democracies; this may give them a heightened sense of unfairness when they are asked about the conduct of elections. Similarly, older citizens who had their formative political experiences in the Soviet years may be more likely to regard the new post-communist system as unfair, either because it is different to the one that had existed in earlier years or because it seems more chaotic and less satisfactorily regulated (Mishler & Rose 2002). We take the social and socioeconomic position of citizens into account in this study by using measures of age, gender and tertiary education.

The remaining explanations all focus on political attitudes. Attitudes towards ‘democracy’, for instance, may influence the way in which elections are perceived. We examine this second hypothesis through two questions: the extent to which the respondent sees the country as moving towards democracy, and whether they believe that democracy is the best form of political system. A third hypothesis identifies turnout as itself the critical variable, such that those who vote will be more likely to have a favourable view of elections than those who abstain. Turnout has generally declined in the post-communist countries after the excitement of their ‘founding elections’, and this may account for at least some of the increasing propensity to regard elections as unfair (Kostadinova 2003). The fourth explanation is concerned with partisanship: the more respondents support the political changes that have taken place under the Putin leadership, we might suggest, the more they will support the changes that have affected elections in particular. We test this through measures of support for the government, and approval of Putin in person. The final explanation hypothesises that broader attitudes towards bribery and corruption will be of overriding importance, and that citizens will be more likely to take an unfavourable view of the electoral process the more they believe that corruption has increased across the entire society.

In order to test the net effect of these various explanations, we conduct a regression analysis predicting how fair the respondent thought the election was (Table 8). Since we are fortunate in having four surveys—in 2003, 2005, 2008 and 2010—in which all of the questions were asked in the same way, we are able to replicate the analysis for each of the four elections. This enables us to identify the effects that are most consistently important, as well as observing their differential effects over time. The results are reported in the form of standardised regression coefficients, which show the importance of each of the independent variables judged against the other independent variables in the equation.

TABLE 8 
Explanations for Electoral Fairness, 2003–201 0

	Standardised coefficients
	2003	2005	2008	2010
Social background
Gender (male)	−0.02	−0.05*	−0.02	−0.02
Age	−0.02	0.02	0.00	−0.01
Tertiary education	0.00	−0.04	−0.05	−0.02
Democratic orientation
Country moving to democracy	0.04	−0.04	0.11**	−0.01
Democracy best system	0.04	0.03	0.03	0.07**
Voted last election	0.14**	0.06*	0.06*	0.07**
Partisanship
Supports government	0.08**	0.11**	0.21**	0.19**
Approves of Putin	0.24**	0.18**	0.25**	0.12**
Perceptions of corruption
Most officials corrupt	−0.08**	−0.15**	−0.06*	−0.11*
Bribery increasing	−0.07*	−0.04	−0.12**	−0.07**
Adjusted R-squared	0.15	0.11	0.26	0.12
(N)	(2,000)	(2,000)	(2,000)	(2,017)
Notes: *, statistically significant at p<0.05, **, p<0.01. Ordinary least squares regression analysis showing standardised (beta) coefficients predicting the probably of seeing the election as fair.
Source: authors' surveys (see note 8).

The results in Table 8 show that by far the most important and consistent predictor of fairness is partisanship. Those who are more supportive of the government and more likely to approve of Putin are much more likely, all other things being equal, to see an election as being fair. Moreover, the effect of partisanship in predicting that an election will be perceived to be fair appears to be increasing; in 2008, for example, the impact of partisanship was greater than all the other independent variables combined. It is also notable that while approval of Putin is consistently important across the four elections, government support is increasingly important, and indeed it exceeds approval of Putin, by this time prime minister, in 2010. This demonstrates the degree of overlap between the public's view of Putin and of the government, even after Putin stood down from his second term as president in 2008.

The other effects in Table 8 are of lesser importance. General perceptions of corruption across the society as a whole are consistently important across the four elections, and those who saw corruption as generally in decline were also more likely to extend this positive view to elections, net of a wide range of other factors. Voters are also consistently more likely than non-voters to see elections as fair, although the magnitude of the effect was greater in 2003 than in any of the later elections. There is little support for the other two explanations. Views of democracy matter little, and although there was some effect for the country moving to democracy in 2008 and for democracy being the best system in 2010, this was not the case in other years. Similarly, social background has no effect, apart from a minor gender effect in 2005.

How Russians view the fairness of their elections is, then, very much associated with how they view their government as a whole. This is a level of association between the state itself and the popularity of the government which would be considered unusual in the established democracies, but which may be more common in countries with a less extended experience of competitive politics where citizens may not always distinguish the institutions through which government is conducted from the individuals that direct it. The institutions through which government is conducted, such as elections, are also more likely to be manipulated by the government in countries that have a less extended experience of competitive politics.

The question remains as to whether or not attitudes towards the fairness of elections affect more general attitudes towards competitive politics.14 This is an area in which there has been relatively little research (Birch 2008, pp. 315–16), but where it is suspected that long term lack of confidence in the electoral system produces a range of negative consequences that pervade the political system. In order to address this question we examine how perceptions of fairness in the 2010 election influenced respondents' satisfaction with democracy, and their sense of political efficacy. Both of these attitudes are often seen as an essential ingredient of representative democracy: satisfaction with democracy taps a diffuse sense of citizen support for democratic institutions (Anderson & Guillory 1997; Cannache et al. 2001; Karp et al. 2003), while political efficacy measures the belief that the system will work in the citizen's interests (Hayes & Bean 1993; Karp & Banducci 2008). To test this hypothesis, we regress perceptions of the fairness of the election on satisfaction with democracy and political efficacy, also controlling for the wide range of variables included in Table 9. For this analysis, we use the 2010 survey.

TABLE 9 
The Consequences of Electoral Fairness for Political Attitudes, 2010

	Standardised coefficients
	Satisfaction with democracy	Political efficacy
Elections fair	0.11**	0.06*
Social background
Gender (male)	−0.02	0.05*
Age	−0.06*	−0.06*
Tertiary education	−0.04	0.04
Democratic orientation
Country moving to democracy	0.22**	0.10**
Democracy best system	0.06*	−0.05
Voted last election	0.03	0.08*
Partisanship
Supports government	−0.01	−0.05
Approves of Putin	0.05	0.02
Perceptions of corruption
Most officials corrupt	−0.08**	−0.22**
Bribery increasing	−0.07**	−0.13**
Adjusted R-squared	0.17	0.13
(N)	(2,017)	(2,017)
Notes: *, statistically significant at p<0.05, **, p<0.01. Ordinary least squares regression analysis showing standardised (beta) coefficients predicting the probably of seeing the election as fair.
Source: authors' 2010 survey (see note 8).

The results in Table 9 show that perceptions of the fairness of the election do indeed have significant implications for broader attitudes towards the democratic system, net of a wide range of other factors. In predicting satisfaction with democracy, election fairness, with a coefficient of 0.11, is second in importance only to the belief that the country is moving towards democracy (with a coefficient of 0.22), and is more important than more general perceptions of corruption. Perceptions of election fairness are less important in predicting political efficacy, but nevertheless remain statistically significant. Most important in this equation are beliefs about the general corruption of officials; as we would expect, a perception that corruption is widespread among public officials reduces the expectation that a citizen will be treated fairly. These results suggest that for Russians, at least in 2010, their beliefs about the fairness of their elections were an important influence on their general attitudes towards politics, net of a wide variety of other circumstances.

Conclusion

It has long been known that popular confidence in the integrity of electoral processes is a key component of an effective, stable and widely supported democracy. The various reforms that have been introduced over two centuries to make elections fairer—from the secret ballot and the abolition of property qualifications in the nineteenth century to the public funding of political parties in the twentieth—have also implicitly recognised the importance of citizens viewing their electoral processes as fair.15 The relatively short experience of competitive politics in Russia—the decade before the First World War, and then on a fully inclusive franchise with a choice of parties and candidates since 1993—has placed an added emphasis on the importance of citizens viewing the process of completing and counting the ballots as honest.

The results presented here, using a wide variety of surveys conducted over more than a decade, show mixed results. Few citizens report being pressured to cast their ballot in a certain way, although the proportions mentioning that pressure had been placed on them increased threefold between 2000 and 2008, to almost one in 10 of all voters. Equally, however, relatively few non-voters see the perceived unfairness of the elections themselves as a reason for abstention; more important is a broader sense of alienation from the political process. Nevertheless, while the integrity of the electoral process has come to be viewed more favourably over the 2000s, a significant minority still see it as essentially dishonest. There are various explanations for this distrust of the electoral process, the most important being the association between the Putin regime and the conduct of elections. The public perception that aspects of the electoral system have been manipulated by the authorities themselves and to their own advantage has obviously had some effect, but other aspects are also important, notably perceptions about corruption across Russian society as a whole.

Our results confirm and extend the findings of Rose and Mishler (2009) in several important ways. First, using a wider range of surveys and a longer time series, we confirm the partisan element in public perceptions of electoral fairness, and the extent to which that has increased in the years since Putin gained power. Any change in the views about the conduct of elections in Russia will only come about when there is a significant change in the centralised forms of government that became established over the two terms of the Putin presidency. We are also close to Rose and Mishler in their identification of the various ways in which perceptions of unfairness undermine trust and confidence in political institutions. We confirm this finding and also demonstrate how these views of electoral fairness influence more general attitudes towards democracy and the legitimacy of the system. In addition, like Rose and Mishler, we show a significant link between perceptions of fairness and more general views about corruption in Russian society. Electoral corruption is but one dimension of a broader problem, and any remedial action to deal with the problem must treat it as such.

Finally, how do Russian elections compare with other post-communist societies in their perceived fairness among the public? Table 10addresses this question using the Comparative Study of Electoral Systems survey, which asked a question about the fairness of elections across nine post-communist countries using a 1–5 scale. Judged against these other countries, which have undergone similar democratic transitions, Russians are second only to Ukrainians in regarding the fairness of their electoral processes unfavourably. While the Russian figures relate to 1999, and our results show some improvement since that time, Russia remains far behind such countries as Romania, where 62% in 1996 regarded their most recent election as ‘completely fair’, or Hungary in 1998, where 59% took the same view. Clearly, Russia has a long way to go to improve the standing of its elections, compared to what would be regarded as the obvious comparator countries.

TABLE 10 
Perceptions of Electoral Fairness in Nine Post-communist Countries

Rank	Country and survey year	Completely fair	2	3	4	Completely unfair	Total	(N)
1.	Romania (1996)	62	19	9	4	6	100	(1,123)
2.	Hungary (1998)	59	23	13	3	2	100	(1,436)
3.	Poland (1997)	47	25	18	6	4	100	(1,615)
4.	Czech Republic (1996)	47	33	16	3	1	100	(1,212)
5.	Slovenia (1996)	46	22	21	5	6	100	(1,878)
6.	Belarus (2001)	45	14	13	12	16	100	(851)
7.	Lithuania (1996)	31	25	18	12	14	100	(965)
8.	Russia (1999)	25	19	28	13	15	100	(1,537)
9.	Ukraine (1998)	23	14	28	12	23	100	(972)
The question wording was: ‘In some countries, people believe their elections are conducted fairly. In other countries, people believe that their elections are conducted unfairly. Thinking of the last election in [country], where would you place it on this scale of 1–5 where ONE means that the last election was conducted fairly and FIVE means that the last election was conducted unfairly?’ Rank is based on responses to the first option.
Source: Module I of the Comparative Study of Electoral Systems (CSES).

The level of confidence that citizens express in their electoral processes has consequences for their views of the political system as a whole. We have demonstrated that there are consequences for the extent to which citizens have satisfaction in democracy, and in their general sense of political efficacy, even after a wide variety of other factors have been taken into account. These are the underlying values that provide the reservoir of public support that holds a competitive political system together, and ensures its ongoing ability to withstand the crises that will inevitably challenge it. Across the established democracies, Birch (2008) has shown that institutional factors such as proportional representation and the public funding of political parties are most important in ensuring popular confidence in electoral processes. In Russia, the close association with the character of the political system is a more pertinent consideration. Only when the entire process is seen to be free from government manipulation will Russian elections come to be regarded with the same degree of confidence as their counterparts elsewhere.

Notes

The support of the Australian Research Council under grant LX0883137 to Ian McAllister and of the UK Economic and Social Council to Stephen White under grants RES-000–22–2532 and RES-062–23–1378 is gratefully acknowledged, as is the support of the Leverhulme Foundation to Stephen White through its Major Research Fellowship F/00 179/AR.

¹For a recent review see Luchin et al. (2010).

² Pravda, 5 April 1989, p. 1.

³See, however, Schedler (2006, 2009) and Birch (2009). The Economist Intelligence Unit's ‘Democracy Index 2010’ found just 26 ‘full democracies’ out of 167 states worldwide in 2010—fewer than 16% of all states, and an even smaller proportion (just over 12%) of the world's population (see http://graphics.eiu.com/PDF/Democracy_Index_2010_web.pdf, last accessed 24 January 2011).

⁴See for instance OSCE (2010); on the wider experience see Bjornlund (2004) and Alvarez et al. (2008).

⁵There were ‘almost 300,000 observers from the political parties’ at the December 2007 Duma election, but just 299 from other countries (Vybory 2008a, p. 455); there were ‘almost 160,000 from the candidates and political parties’ at the March 2008 presidential election, but just 235 from other countries (Vybory 2008b, pp. 287–88).

⁶VTsIOM, for instance, based its results on a sample of 3,000 distributed across 46 different regions, and also carried out exit-polls (Vybory 2008a, pp. 368, 371); FOM used 1,500 respondents in 44 different regions (‘Protsedura golosovaniya’ 2007), and the Levada Centre a sample of 1,600 with a smaller number of primary sampling units (see for instance ‘Chestnost’’ 2007). There have been several retrospective comparisons of survey forecasts and electoral outcomes; see for instance Volkov and Grazhdankin (2008).

⁷ Izvestiya, 4 December 2007, p. 3.

⁸All of the surveys were conducted by personal interview and are nationwide, stratified surveys representative of the adult population aged 18 years and over. The 2000 survey was conducted by VTsIOM between 19 and 29 January 2000, with an N of 1,940 respondents; all of the remaining surveys were conducted by Russian Research and have an N of 2,000 respondents. The fieldwork dates were: (2001) 10–26 April 2001; (2003) 21 December 2003–16 January 2004; (2005) 25 March–24 May 2005; (2008) 30 January–27 February 2008; and (2010) 12 February–1 March 2010.

⁹See however Birch (2008, 2010) and Rose and Mishler (2009).

¹⁰The 2000 survey also asked the respondents if they had seen or heard of any electoral irregularity. Just 1% said they had seen an irregularity themselves, while 24% had heard about such an irregularity (0.4% from a family member, 7% from friends and neighbours, and 19% through the television or radio).

¹¹The question was also asked in the 2000 survey, but in that case permitted multiple choices so is not exactly comparable with the 2003 and 2008 surveys.

¹²This category combines a variety of reasons such as ‘there is no party which reflects my political views’, ‘I don't trust any of the politicians’, and ‘I believed my vote wouldn't change anything’.

¹³In 2000 it was mentioned by just 11%.

¹⁴We follow the convention of causality here, assuming that attitudes towards the electoral system influence more general attitudes towards democracy (see, for example, Anderson & Guillory 1997; Karp et al. 2003). However, there is undoubtedly some element of reciprocal causation here, in that attitudes towards democracy may influence views of the electoral system.

¹⁵See for example Elklit (1999) and Mozaffar and Schedler (2002).