Enter at your own risk......

Please take a few minutes to check out all of my sites.

http://www.desertheatradio.com - No ads - music 24/7 - live dj's that will rock your world!!!

http://www.thebdsmchat.com (don't let the name scare you away - we talk about everything there - come on in make yourself at home!!!)

http://www.ravenskyes.com

http://www.scammersexposedbydjcatlady.com





Sunday, December 13, 2020

The Electoral Consequences of Corruption

 https://www.annualreviews.org/doi/full/10.1146/annurev-polisci-052715-111917


Abstract

Democratic elections have been assumed to play a crucial role in curbing corruption among public officials. Voters, due to their general distaste for corruption, are expected to sanction politicians who misuse public office for private gains. Yet, empirical evidence to date is mixed, and it often suggests that the electoral punishment of corruption is rather mild. Recently, political scientists have made great strides in understanding why corruption might be tolerated by voters. In this review, we identify three key stages—information acquisition, blame attribution, and behavioral response—that underlie a retrospective vote based on corruption. A breakdown of one or more of these stages may lead to a lack of electoral punishment of corruption. We also outline some areas for future progress, particularly highlighting the importance of voter coordination for understanding the extent to which corruption is punished at the ballot box.

Keywords

blame attribution, corruption, elections, political information, voting behavior

INTRODUCTION

Corruption is a complex phenomenon, often deeply rooted in the cultural and political practices of societies. Although the precise costs of corruption are hard to quantify and vary greatly from country to country, research suggests that corruption is bad for economic and social development (e.g., Rothstein 2011). Corruption has been shown to have a detrimental effect on tax revenues (Pani 2010), investments and economic growth (Mauro 1995, Del Monte & Papagni 2001), equality and poverty (Chong & Calderon 2000, Gupta et al. 2002, You & Khagram 2005, Uslaner 2008), and overall subjective well-being and life satisfaction (Tay et al. 2014). Moreover, corruption is said to erode political trust and undermine political legitimacy in a variety of institutional settings (Della Porta 2000, Seligson 2002, Andersen & Tverdova 2003, Chang & Chu 2006).

Although the harmful effects of corruption for society are well documented and largely indisputable, defining which political activities constitute corruption has proven to be an arduous task. The vastly divergent cultural, political, and legal traditions that exist across the globe often lead to context-specific interpretations of corruption that are entrenched in national contexts and local cultures. Attempts to define corruption more generally by international organizations, such as the World Bank (1997), focus on the misuse or the abuse of public office for private gains (see also Rose-Ackerman 1999, p. 91). Although this definition leaves considerable room for interpretation (for example, what exactly constitutes misuse or private gain?), it has become widely accepted within the literature (Rose-Ackerman 1999, Sandholtz & Koetzle 2000, Treisman 2000, Lambsdorff 2002, Kunicová & Rose-Ackerman 2005, Golden 2010, Klašnja et al. 2016, to name just a few). Empirical manifestations of corruption may thus include fraud, misappropriation of public funds, or the acceptance of bribes.

Free, fair, and competitive elections that allow citizens to sanction or reward politicians in a periodic manner are widely believed to have a constraining effect on corruption. Electoral rules may have varying effects on corruption, however (for useful overviews, see Rose-Ackerman 1999, Kunicová 2006, Golden & Mahdavi 2015). For instance, majoritarian electoral systems are associated with lower levels of corruption than proportional ones, because they foster alternation in government and promote political competition (Persson et al. 2003; see also Rudolph & Däubler 2016). The evidence regarding proportional representation suggests that systems using closed lists are associated with more corruption than those employing open lists (Carey & Shugart 1995, Kunicová & Rose‐Ackerman 2005). Finally, district magnitude might also play an important role, especially within proportional systems. Under open‐list proportional systems, for example, corruption increases with district magnitude (Chang & Golden 2007).

A core assumption underlying the research on the effects of electoral rules on corruption is that voters punish corrupt politicians if they have the opportunity to do so. Yet, the empirical evidence on the electoral punishment of corruption is mixed. Although a fair amount of evidence suggests that corrupt activities indeed take a considerable electoral toll on incumbents (e.g., Fackler & Lin 1995, Ferraz & Finan 2008, Krause & Méndez 2009, Winters & Weitz-Shapiro 2013, Klašnja 2016), we often observe empirical deviations from the widespread assumption that voters punish politicians who steal from them. The reelection of corrupt politicians is not merely a trait of developing nations characterized by weak political and economic institutions but is also found in established democracies such as Italy, Japan, and the United States (e.g., Rundquist et al. 1977, Reed 1996, Chang et al. 2010). This raises the question of why voters often fail to punish politicians for corrupt activities. Research has provided a variety of answers to this question. At the individual level, authors have pointed toward the effects of informational asymmetries (e.g., Ferraz & Finan 2008 , Chang et al. 2010), the importance of partisan and other in-group loyalties (e.g., Anduiza et al. 2013, Solaz et al. 2017), or side payments (e.g., Manzetti & Wilson 2007, Fernández-Vázquez et al. 2016). At the contextual level, researchers suggest that the presence of strong economic growth (e.g., Klašnja & Tucker 2013, Zechmeister & Zizumbo-Colunga 2013) or a lack of institutional clarity (e.g., Tavits 2007, Schwindt-Bayer & Tavits 2016) might weaken the electoral punishment of politicians who are known to be corrupt.

This article aims to provide a systematic, although surely not complete, review of a burgeoning line of inquiry exploring the conditions under which voters punish corruption. It mostly reviews the micro-level evidence tackling this issue. The field has experienced somewhat of a boom in recent years, primarily driven by a strong emphasis on research design, especially the use of survey and field experiments that aim to overcome problems of measurement and endogeneity. These recent developments closely mirror research trends in other fields, most notably retrospective voting (for an excellent overview, see Healy & Malhotra 2013). Although these developments occurred in parallel, there seems to be quite limited engagement across the two fields. This is unfortunate, because some of the recent insights from retrospective voting about attribution errors or behavioral biases could really help enlighten the precise individual-level mechanisms underlying the (limited) electoral consequences of corruption. The recent evidence on retrospective voting suggests that holding incumbents accountable for past performance, such as corruption, is not an easy task for voters. Specifically, voters need to acquire all the relevant information about corruption, process it, and weigh it against all other candidate information they deem important when deciding whom to vote for. Most research to date suggests that voters often do not hold politicians accountable for corruption and thus deals with the breakdown of one of these steps, although not explicitly. Existing research pays less attention to the implications of electoral outcomes for candidate behavior.

This article reviews the growing literature on voter responses to corruption, but it does not discuss the vast and very important work studying the causes of corruption (for useful overviews, see Rose-Ackerman 2006, Fisman & Golden 2017) or developments in vote buying and clientelism (for an excellent review, see Mares & Young 2016). Rather, its aim is to provide an overview of the conditions under which voters punish corruption. First, we discuss issues of measurement. Second, we theorize the individual-level mechanism underlying a retrospective vote based on corruption and outline three important stages: information acquisition, blame attribution, and behavioral response. A breakdown of one or more of these stages can result in corruption not being punished at the ballot box. The ensuing sections summarize existing research based on which of these three stages it predominantly explores. Finally, in the concluding section we highlight four themes in need of further research.

MEASURING CORRUPTION

Corruption is hard to measure, because the subject of study is carefully hidden from the researcher's eyes. Corruption is also hard to define, because it is often entrenched in context-specific meanings. Even if we agree political corruption to mean the misuse of public office for private gain, important questions arise. Does misuse signify that the act is a deviation from legal standards, from moral standards, or perhaps from both? Do private gains relate only to public officials themselves and their family or also to their party? These issues are important, because when exploring the electoral consequences of corruption scholars often assume that respondents in surveys or participants in experiments share their research definition of corruption. Yet, this might not always be the case. Evidence suggests that norms about when the misuse of public office constitutes corruption vary greatly among societies. Based on experiments run in Australia, India, Indonesia, and Singapore, Cameron et al. (2009) show that great variation in the propensities to punish corrupt behavior exists across cultures. Subjects residing in high-corruption countries, determined by existing corruption indices, exhibit a higher tolerance of corruption than residents in low-corruption countries. Similarly, laboratory experimental evidence provided by Barr & Serra (2010) demonstrates that the level of corruption in subjects’ home countries helps to predict who acts corruptly in the laboratory. Different societal norms about which type of behaviors constitute corruption among public officials and which do not surely affect how we examine the electoral punishment of corruption, especially cross-nationally.1

People's corruption perceptions may also vary extensively within one country. Some studies have found that the wealthy and educated perceive more corruption; others find that the wealthy and educated perceive less corruption (Davis et al. 2004, Redlawsk & McCann 2005, Tverdova 2011). Maeda & Ziegfeld (2015) show that these conflicting findings are resolved when economic development is taken into consideration. Whereas in richer countries the poor and uneducated tend to perceive higher levels of corruption, in poorer countries they perceive corruption to be lower. The reported within-country heterogeneity may lead to a situation where some voters in a specific election punish corruption while others tolerate it, and at the aggregate level these competing responses might cancel out. Against this backdrop, carefully constructed micro-level analyses of the extent to which different groups of voters might punish corruption seem crucial in future work (see, e.g., De Vries et al. 2017, Solaz et al. 2017).

In addition to individual heterogeneity, people's perceptions of corruption might not always be perfectly correlated with more objective measures. The vast majority of work measures corruption indirectly through subjective perceptions. Surveys and other forms of corruption measurement might tap into different things. Figure 1 plots people's perceptions of how widespread corruption is in their country, based on the second module of the Comparative Study of Electoral Systems (CSES), against the level of corruption measured by the Corruption Perceptions Index (CPI). CPI scores merge data on corruption in a country from 11 different institutions (for more information, see http://www.transparency.org/research/cpi/overview). Figure 1 shows that people's average corruption perceptions are positively correlated with more objective summary measures of corruption.

figure
Figure 1  Corruption perceptions and corruption perception index measurements. The dots in the figure represent the average corruption perception based on the Comparative Study of Electoral Systems (CSES) module 2 (2001–2006) and corruption measured through the Corruption Perceptions Index (CPI) for 38 countries. We matched the CPI and the CSES scores for each election year; if two elections were covered, we selected the CPI of the last election covered.
Download Full-ResolutionDownload PPT

Measurement of corruption clearly matters, and the subjective perceptions that have gained prominence in the literature may not always be useful when examining the electoral consequences of corruption, even if in aggregate they correlate with more objective measures. For example, corruption perceptions of incumbent party supporters may be tainted by their partisanship. Indeed, studies of economic voting suggest that the relationship between party choice and perceptions of incumbent performance is highly endogenous (e.g., Evans & Andersen 2006, Evans & Pickup 2010). Researchers increasingly aim to overcome problems relating to measurement and endogeneity by relying on surveys and field or laboratory experimental evidence (see, e.g., Anduiza et al. 2013, Botero et al. 2015, Chong et al. 2015, Solaz et al. 2017). The experimental approach allows researchers to reduce the variation in how people understand corruption by carefully controlling the setting and the information subjects are exposed to. The majority of the evidence we survey in the next sections employs experimental methods.

CONCEPTUALIZING THE ELECTORAL CONSEQUENCES OF CORRUPTION

To date, much of political science research on corruption has focused on how institutional frameworks create specific incentive structures that facilitate (or hamper) corruption among public officials. Although considerable disagreement exists about which institutional rules (proportional versus plurality or parliamentary versus presidential systems, for example) facilitate the occurrence of corruption (for an excellent overview, see Golden & Mahdavi 2015), the literature seems to agree on one important aspect: Voters punish corrupt politicians if they have the opportunity to do so. This idea is based on the classical interpretation of the retrospective voting model as a means to either sanction low-performing politicians or select high-performing ones (e.g., Key 1966, Fearon 1999). Voters aim to incentivize good behavior on the part of their representatives by rewarding good and punishing bad performance. If these insights apply to corruption, then voters, because of their distaste for corruption, would be expected to sanction corrupt politicians and select honest ones. If voters fail to punish corrupt politicians, this is largely interpreted as a product of the institutional structures in which voters reside. This expectation directly links voters’ corruption evaluations to their vote choices (Figure 2a).

figure
Figure 2  Relationships between corruption evaluations and vote choice: (a) the traditional model versus (b) the revisionist model.
Download Full-ResolutionDownload PPT

Casting a retrospective vote based on corruption is most likely quite challenging, especially compared to voting based on other outcomes, such as economic performance. As highlighted in the previous section, corruption is hard for voters to detect, because politicians go to substantial lengths to conceal their corrupt activities. What is more, knowing that voters disapprove of corruption, candidates have an incentive to actively mislead the public by accusing competitors of corruption, even if this information is false. Voters might become aware of this and perceive most corruption information not to be credible. Furthermore, parties can provide explanations about corruption allegations or force individual members to resign and take the blame. This makes it difficult for voters to decide whether to blame parties for corruption of individuals. Finally, in countries where corruption is not commonplace, voters can punish political malfeasance because clean alternatives exist; in countries where corruption is widespread, however, voters may lack the incentives to punish corrupt activities because opposition parties will most likely be equally corrupt while in office.

Indeed, the empirical support for the traditional model as depicted in Figure 2a is mixed. Some studies demonstrate that corrupt activities take a considerable electoral toll on incumbent politicians (see, e.g., Fackler & Lin 1995, Ferraz & Finan 2008, Krause & Méndez 2009, Winters & Weitz-Shapiro 2013, Klašnja 2016). Krause & Méndez (2009, p. 179), for example, suggest that “corruption in public office is effectively punished by voters.” The authors have constructed a data set registering changes in incumbent vote shares, economic conditions, and corruption data from the CPI for 35 countries over a 12-year period (1995–2007). They find that voters reward good performance based on improving economic conditions and sanction bad performance due to an increase in corruption. Their findings suggest that this process might not be uniform across countries, however, because it hinges on the country's political system and democratic experience. Other work suggests that the electoral retribution of corrupt behavior in office is rather mild (for an excellent overview, see Golden 2010). For example, studies on Italy (Chang et al. 2010), the United Kingdom (Vivyan et al. 2012), Spain (Rivero & Fernández-Vázquez 2011, Fernández-Vázquez et al. 2016), Japan (Reed 1996), and the United States (Rundquist et al. 1977, Peters & Welch 1980) suggest that politicians implicated in corruption are often reelected. What is more, recent work demonstrates the effect of specific individual or contextual factors—such as information asymmetries (e.g., Chang et al. 2010, Botero et al. 2015), partisanship (e.g., Anduiza et al. 2013, Muñoz et al. 2016), or the state of the economy (Klašnja & Tucker 2013, Zechmeister & Zizumbo-Colunga 2013)—to explain the conditions that lead to a lack of electoral punishment of corruption. The idea that voters punish a politician they know to be corrupt is thus becoming increasingly contested.

On closer inspection, current findings may be less conflicting than they seem at first sight. Recent advances in the study of retrospective voting suggest that holding politicians accountable for past performance is more laborious than often assumed. In their excellent review of this literature, Healy & Malhotra (2013, p. 289) suggest that retrospective voting can be considered a four-step process. First, voters observe a change in their own or societal welfare, for example due to an event, political action, or policy outcome. Second, they attribute responsibility for this change in welfare to particular elected office holders and adjust their evaluations of the performance of these officials accordingly. Third, these attributions lead voters to adjust their voting decisions. Fourth, their votes are translated into specific election results that incentivize office holders to adjust their behavior. Casting a ballot based on corruption evaluations can be seen to constitute a retrospective vote and thus will also involve these steps. When one or more of these steps break down, the electoral punishment of corrupt politicians might not happen. The research we review in the ensuing sections provides specific insights into this larger process. These studies focus more on steps one through three and less on step four. The understanding of the different stages involved in the electoral punishment of corruption may account for the fact that some studies find that corruption takes a clear electoral toll on incumbents whereas others do not.

If we revise the traditional understanding of corruption voting as presented in Figure 2a, we obtain a more complex model, presented in Figure 2b, which involves three important stages: information acquisition, blame attribution, and behavioral response. In the first stage of information acquisition, voters need to observe corruption. This is not always easy, because elected public officials engaged in corruption have an incentive to hide information or may try to influence the media or the judiciary not to report on or prosecute their corrupt activities. What is more, in some countries corruption is so common that one more corruption scandal will make very little difference for individual voting decisions. The literature suggests that people learn about corruption through personal experience or indirect perception, mostly by way of media reporting (e.g., Seligson 2002, Ferraz & Finan 2008, Gingerich 2009, Chang et al. 2010, Klašnja et al. 2016).

In the second stage, blame attribution, people attribute responsibility for corruption and adjust their performance evaluations of elected office holders accordingly. Blame attribution is the process by which voters allocate blame (or credit) for certain political events or outcomes to public officials. Research has demonstrated that blame attribution is far from straightforward and is often plagued by group-serving biases (Taylor & Doria 1981). Whereas voters tend to attribute positive events or outcomes to their preferred in-group, they absolve the in-group of blame for negative events or outcomes. With a few recent exceptions (e.g., Anduiza et al. 2013), blame attribution has received little attention in the literature compared to the effects of information acquisition. This is unfortunate because attribution biases may be crucial to understanding the conditions under which corruption is punished electorally. Even if voters are informed about a corruption scandal involving an elected public official, they may not assign responsibility correctly when the scandal involves a politician from their preferred party or ethnic group, for example.

Even if information is sufficient and of high quality, and blame is attributed correctly, voters may not punish corruption. The third stage involves behavioral responses. Voters have to consider a large array of factors when deciding to cast their ballot, and corruption may not be a salient issue in a specific election campaign (Chang et al. 2010, Klašnja et al. 2016); even if voters find corruption important, they might be willing to trade it off against some other benefit. Voters might explicitly trade corruption against some material gains (Manzetti & Wilson 2007, Fernández-Vázquez et al. 2016), or engage in an implicit trade by turning a blind eye in favor of other benefits such as ideological proximity (Rundquist et al. 1977, Peters & Welch 1980) or strong economic performance (Klašnja & Tucker 2013, Zechmeister & Zizumbo-Colunga 2013).

Finally, the last step, namely the behavioral responses of office holders based on the (lack of) electoral punishment of corruption, has received rather scant empirical attention. The idea of retrospective voting is that election results should incentivize office holders to adjust their actions. Against this backdrop, the electoral punishment of corruption should lead public officials to engage in less corruption as they fear the wrath of the electorate. A variety of theoretical models provide predictions about candidate behavior (e.g., Ferejohn 1986, Fearon 1999, Besley 2006, Svolik 2013, Klašnja et al. 2017). Solaz & De Vries (2017) designed laboratory experiments to examine if a lack of electoral punishment makes candidates engage in greater corruption. Their evidence suggests that when candidates anticipate low or no electoral costs for bad performance—due to the presence of strong in-group loyalties, for example—they engage in more corrupt activities. This evidence supports the idea that a lack of electoral competition may breed corruption.

In this section, we have developed a schematic understanding of the processes by which voters can punish corruption at the ballot box. The conflicting evidence reported in the literature regarding the extent to which voters hold elected public officials accountable for corruption may no longer be so surprising. In some cases the three stages of corruption occur, but in others they do not. Although the literature to date has not explicitly engaged with the three stages we outlined here (information acquisition, causal attribution, and behavioral response), we review existing work in light of which of these stages it predominantly examines.

INFORMATION ACQUISITION

One reason voters may fail to punish corrupt politicians, formalized in political agency models, is that corruption is hard to detect (Ferejohn 1986, Fearon 1999, Besley 2006). Political accountability works imperfectly when informational asymmetries exist. This raises the question of how voters get informed about corruption. Klašnja et al. (2016) distinguish between two different channels through which voters gather information about corruption: direct experience and indirect perception. Direct experience “could include being asked to pay a bribe by a police inspector to avoid a ticket for a fictional offense or by a doctor to be seen at what is supposed to be a free public medical clinic” (Klašnja et al. 2016, p. 69; see also Seligson 2002, Gingerich 2009). Indirect perception is based on media reporting or prominent court cases, and thus it most likely involves higher-level political corruption such as the misappropriation of public funds (e.g., Ferraz & Finan 2008, Chang et al. 2010).2

Some studies suggest that the lack of electoral punishment of corruption is mostly a function of the quantity of information available to voters (e.g., Ferraz & Finan 2008, Chang et al. 2010, Costas-Pérez et al. 2012). Information dissemination could be limited by a lack of media reporting, political control of the media, high levels of illiteracy, or a lack of prosecution of corruption cases, for example. In their study of the effects of publicly released audit reports about corruption practices in Brazilian municipalities, Ferraz & Finan (2008) show that information dissemination increases the electoral punishment of corruption. The authors use the fact that the Brazilian government randomly selected municipalities for audit reports. Their evidence suggests that audit outcomes had a significant effect on the electoral performance of incumbents and that these effects were more pronounced in municipalities where a local radio station was present to disclose this information (for similar findings, see Costas-Pérez et al. 2012). Similarly, in a study of legislators in Italy's lower house between 1948 and 1994 and of candidates of the two largest parties (Christian Democracy and the Italian Communist Party), Chang et al. (2010) show that only in the 1992–1994 legislature did corrupt legislators or candidates face a serious electoral penalty. The authors suggest that this structural break in the response to corruption by Italian voters was due to the media coverage that grew out of the Clean Hands operation in the early 1990s. The authors argue that a wider dissemination of information about corruption in the media not only makes corruption more salient in voters’ minds but also provides a coordination signal to voters, who realize that others will most likely punish corruption as well.

Some work, however, suggests that providing information about corruption does not always increase electoral accountability. For example, Klašnja & Tucker (2013) compare experimental evidence from Sweden and Moldova and demonstrate that voters in Sweden, a low-corruption country, respond more to information about corruption than do voters in Moldova, a high-corruption country. Moreover, they show that in the high-corruption country, voters react to corruption information only when economic conditions are poor. This suggests that voters in high-corruption countries may at times experience corruption fatigue; that is, corruption may be so widespread that information about one additional case of corruption may make little difference to people's overall corruption evaluations.

Furthermore, in a field experiment in Mexican municipalities, Chong et al. (2015) find that voters do respond to audit information about excessive corruption presented in flyers, but they do so by staying away from the voting booth. These findings suggest that remedying information asymmetries may not always lead to the results scholars or practitioners hope for (see also Vivyan et al. 2012). Although supplying information may ultimately result in a loss in incumbent vote share, it may do so by depressing active participation in elections overall (see also De Figueiredo et al. 2011).

The heterogeneity in findings might be due to the fact that not only the quantity but also the quality of the information provided matters. If media news stories and/or judicial reports lack credibility, then people might be less inclined to respond to the information by updating their corruption evaluations. Winters & Weitz-Shapiro (2013) and Botero et al. (2015) suggest that the credibility of the informational source is indeed crucial. Presenting evidence from a survey experiment conducted in Brazil, Winters & Weitz-Shapiro (2013) show that respondents are more likely to punish a candidate for corruption when the allegations come from a more credible source (from a federal audit rather than from the opposition party). Botero et al. (2015) find that Colombian respondents react more strongly when information about corruption is provided by a trusted source, in particular when accusations are made by the leading national newspaper rather than a nongovernmental organization or the judiciary. The source credibility effect is most pronounced among the less educated. Overall, these findings suggest that both the quantity and quality of the information voters receive about corruption matter crucially for electoral accountability.

CAUSAL ATTRIBUTION

Information about corruption alone is not a sufficient condition to adjust voting preferences. Voters also need to assign blame correctly. Attribution describes the process through which people allocate blame or credit for certain political events or outcomes to public officials. Blame attribution is made difficult by the fact that people do not process information objectively. Psychologists have demonstrated that attributions are affected by group-serving biases (Taylor & Doria 1981). People tend to make internal attributions (to their own in-group) for positive events or outcomes and make external attributions (to an out-group) for negative events or outcomes. Retrospective voting studies, largely inspired by work in social psychology, suggest that voters have strong group-serving biases when it comes to responsibility judgments (Conover et al. 1987, Bartels 2002, Rudolph 2003, Evans & Andersen 2006, Evans & Pickup 2010). To our knowledge, no study explicitly examines blame attributions for corruption as a dependent variable. Recent work has aimed to uncover the effects of priming in- and out-groups more generally.

A recent study by Anduiza et al. (2013) highlights the importance of partisan in-groups. Citing survey experimental evidence from Spain, the authors find that the use of partisan labels affects voters’ judgments of which act is considered to constitute corruption. In their study, actions taken by the in-group party were viewed as less corrupt when party labels were provided. These partisan effects were most pronounced among less politically sophisticated citizens, who were less likely to view the actions of their own party as corrupt. The importance of partisanship may not be universal, however. Konstantinidis & Xezonakis (2013) report survey experimental evidence from Greece to show that partisanship does not moderate the electoral punishment of corruption. The authors suggest that partisanship played a crucial role in explaining voters’ tolerance to corruption in Greece in the past, but by 2013, the year their experiment was conducted, this was no longer the case. The media exposure of corruption in the context of the Greek debt crisis and the condemnation of corruption by the challenger party Syriza broke the partisan tolerance of corruption. The Spanish evidence presented by Anduiza et al. (2013) stems from 2010, before the politicization of corruption by the challenger party Podemos. Cross-national work by Bågenholm (2013) shows that when party elites blame opposition party officials for corruption, the electoral punishment of the accused party increases compared to when no party exploits corruption. The actions of party elites and the mass media might indeed be crucial for correct blame attribution. At the same time, however, politicians are rational agents who want to be reelected and may thus actively play the blame game to distort information about responsibility, as they do in any other political domain (e.g., Hood 2010).

Erroneous blame attribution may be the result not only of group-serving biases, but also of institutional complexity. For example, direct experience with corruption might often involve petty bureaucracy, such as being asked for a bribe by a policeman, and people may not always be able to attribute this behavior to the elected public officials who enable it (Abramo 2008, Klašnja et al. 2016). Tavits (2007) suggests that institutional clarity of responsibility matters for the ability of voters to hold officials accountable for corruption (for a more comprehensive discussion and examination of this relationship, see Schwindt-Bayer & Tavits 2016). The concept of clarity of responsibility is widely used in the literature on economic voting to explain its variation across contexts (Powell & Whitten 1993, Anderson 2000, De Vries et al. 2010). The basic contention is that political systems that diffuse power among multiple actors (parties in particular) obscure lines of responsibility, making it difficult for voters to evaluate and sanction the government in power for economic performance. Tavits (2007) shows that the same mechanism can be applied to the question of corruption in a sample of OECD and Eastern European democracies. She operationalizes clarity of responsibility by examining majority status of the government, duration of cabinets, opposition influence, and party system fragmentation, and finds that in countries where it is more difficult for voters to establish lines of accountability, levels of political corruption tend to be higher (for an individual-level analysis, see Xezonakis et al. 2015). The fact that corruption at the local level tends to be severely punished at the ballot box, whereas the evidence at the national level is more conflicting, seems to support these findings, given that blame attribution is more proximate at the local level (e.g., Ferraz & Finan 2008).

BEHAVIORAL RESPONSE

Information acquisition and correct blame attribution are necessary but surely not sufficient conditions for the punishment of corruption. The third stage in our model of corruption voting focuses on behavioral responses. When voters are sufficiently informed about corruption and assign causal responsibility correctly, they have several behavioral options: switching, abstaining, or sticking to their previous vote. We discuss each of these options in turn. The viability of alternatives is a core factor that conditions voters’ ability to switch party choice. Only when other parties perceived as credible and clean alternatives exist can voters punish corruption by switching. Schleiter & Voznaya (2016) argue that party system competitiveness plays a critical role in conditioning the electoral punishment of corruption. They argue that only when voters can coordinate on credible alternatives to underperforming and corrupt incumbents will corruption be punished electorally and decrease as a result. They present evidence from 70 democracies around the globe to support this intuition. Employing data for US House incumbents running for reelection between 1978 and 2008, Hirano & Snyder (2012) show that incumbents in safe seats who are involved in scandals are more likely to be punished electorally because they exhibit a much higher likelihood of facing multiple challengers in a primary relative to ordinary incumbents. Bågenholm & Charron (2015) also underscore the importance of the viability of alternatives. They show that voters who place themselves on the extremes of the ideological spectrum, and thus consider fewer parties as viable, are more likely to neglect corruption.

When a viable alternative is lacking, voters might react to corruption by abstaining. The literature on the relation between corruption and turnout is relatively small but provides a rather consistent result: Corruption is negatively associated with turnout. This finding is demonstrated by evidence from Latin America (McCann & Dominguez 1998, Davis et al. 2004) and Eastern Europe (Slomczynski & Shabad 2012) and has been replicated in a large-scale cross-country analysis including 70 different democracies (Stockemer et al. 2012). These cross-national results are robust to an instrumental variable approach that addresses the concern that turnout and corruption may be endogenous because voters’ active participation in the democratic process might reduce corruption. Recent field experimental evidence from Mexico underlines the potential causal relationship between corruption and turnout. Chong et al. (2015) randomly assigned the extent and type of information voters received about corruption in several electoral precincts and show that voters responded to corruption information by staying away from the voting booth.

In addition to switching and abstention, voters have the option of tolerating corruption and casting a ballot for a party they know to be corrupt. Even when voters are informed about corruption and they assign blame correctly, they might be willing to trade it off against some other benefit. For example, voters might engage in a material trade whereby they tolerate corruption because they directly benefit from it;3 Mares & Young (2016) provide an excellent overview of the extensive and important literature on vote buying. Yet, the exchange of specific material benefits is difficult and often costly both in organizational and monetary terms. Moreover, “in numerous real-world situations people seem to support corrupt politicians even when material inducements appear absent” (Rundquist et al. 1977, p. 956).

Voters might be also willing to turn a blind eye to corruption based on an implicit trade. It may well be the case that poor performance in one area, in this case corruption, is compensated for performance in some other area, such as economic growth (Klašnja & Tucker 2013, Zechmeister & Zizumbo-Colunga 2013). Moreover, voters might care about other attributes of candidates, such as ideological proximity (Rundquist et al. 1977), partisanship or some other shared-group considerations (Solaz et al. 2017). Even though voters generally do not approve of corrupt conduct, they need to weigh their corruption evaluations against other internal calculations. This may translate into an overall positive balance in favor of reelection.

Citing survey evidence from 19 presidential systems in Latin America, Zechmeister & Zizumbo-Colunga (2013) show that individuals faced with bad economic conditions apply a higher penalty to presidential approval for perceived corruption. Similarly, by presenting experimental evidence from Moldova, Klašnja & Tucker (2013) suggest that in high-corruption countries, corruption is punished only in dire economic conditions. Winters & Weitz-Shapiro (2013) and Muñoz et al. (2016) conducted survey experiments to explore whether voters are willing to tolerate corruption in exchange for good performance in another domain. This empirical evidence is not conclusive. Whereas Winters & Weitz-Shapiro (2013) find no evidence of implicit trade, subjects in the experiment conducted by Muñoz et al. (2016) were willing to accept corruption in exchange for sound economic management.

Voters may be willing to overlook corruption if they receive representational benefits. Rundquist et al. (1977) argue that voters might prefer a corrupt candidate who represents their policy and issue positions to a clean candidate who does not. In line with well-established models of political competition (Downs 1957), the authors argue that the key commodity for trade in politics is ideological representation. They provide experimental evidence for this intuition by showing that voters were indifferent to charges of corruption as long as a candidate shared their stance on the Vietnam War. This effect was stronger than the effects exerted by the disclosure of party information (see also Peters & Welch 1980, Welch & Hibbing 1997). A variant of this argument is put forward by Eggers (2014), who shows that voter willingness to punish UK Members of Parliament implicated in an expenses scandal was dependent on strategic considerations about marginality. Electoral punishment of corruption was lower in marginal districts, that is to say in areas where the partisan stakes were high.

Not only partisanship or ideology but in-group considerations more generally may lower the electoral punishment of corruption. Banerjee & Pande (2007) present a theoretical model to suggest that as a society becomes more ethnically polarized, and citizens vote following ethnic identity rather than any other marker, corrupt activities of candidates of the same ethnic group are less likely to be punished electorally. This is because voters “feel an instinctive pull towards their coethnics” (Banerjee & Pande 2007, p. 5). Solaz et al. (2017) provide empirical evidence for this type of in-group intuition. By utilizing the fact that a large-scale corruption scandal involving the ruling party fell in the fieldwork period of the European Social Survey, they are able to demonstrate that in-group loyalty conditioned the electoral punishment of Spanish voters. Moreover, they examine the robustness of these findings in the laboratory where they can carefully control the information and benefits voters receive. They show that priming group identities of both voters and candidates reduces the electoral punishment of corruption even when clean alternatives exist and voters are fully informed about and receive no benefits from corruption. Although in some circumstances in-group considerations may lead to efficient decision making, in the case of corruption group identities may have perverse side effects, and lead to a lack of electoral punishment of corruption. The evidence presented by Solaz et al (2017) is important as it suggests that even when people know their own leaders to be corrupt and accurately assign blame, they may still fail to sanction because they place greater weight on other benefits they will receive from a candidate, namely in-group status, than they do on that individual's malfeasance. The reason for this is that the voter's own utility is still higher with diversion of rents by the corrupt politician than the voter anticipates it would be if a candidate from an out-group were elected.4

Of the four steps of corruption voting we outline in Figure 2b, the behavioral response of office holders based on election outcomes has received the least empirical scholarly attention. The idea is that the electoral punishment of corruption (or the lack thereof) should incentivize office holders to adjust their behavior accordingly. A variety of theoretical models provide predictions about candidate behavior (e.g., Ferejohn 1986, Fearon 1999, Besley 2006, Svolik 2013, Klašnja et al. 2017). A formal model by Klašnja et al. (2017) shows that if voters receive some benefit from corruption, even a very small one, this can lead to a vicious cycle: Voters fail to punish corruption in one constituency, which may encourage politicians in other constituencies to engage in more corruption, and as the general level of corruption grows, corrupt candidates are more likely to enter the political realm, further discouraging voters from punishing corruption. Countries can get stuck in high-corruption equilibria, so-called corruption traps, despite the presence of elections. The authors comment that escaping such a corruption trap may require a huge amount of coordination among voters, entrants, and sitting politicians. This is often very unlikely. Recently, empirical researchers have started to explore the role of voter coordination by conducting survey experiments in a series of European countries that differ in the extent to which they are stuck in high-corruption equilibria (De Vries et al. 2017).

Banerjee & Pande (2007) present a theoretical model to suggest that as a society becomes more ethnically polarized and citizens vote along ethnic lines rather than any other factor, criminal activities of candidates of the same ethnic group are less likely to be punished electorally. If political polarization occurs among ethnic groups, corruption is less likely to be punished as voters “feel an instinctive pull towards their coethnics” (Banerjee & Pande 2007, p. 5). The by-product of this might be a selection effect. Parties alter their candidate choices in line with ethnic polarization as the probability of winning an election increases with ethnic similiarity rather than quality, and thus the quality threshold for candidates is lowered. Following this intuition, Solaz & De Vries (2017) present laboratory experimental evidence to show that a lack of electoral punishment due to in-group loyalties indeed increases corruption. Their evidence suggests that when in-group loyalties are primed and in-group versus out-group polarization increases, candidates anticipate that they will face low electoral costs for bad performance and subsequently engage in more corrupt activities. Finally, Eggers (2014) demonstrates that members of the UK Parliament in nonmarginal seats, where partisan stakes were low, were more likely to be implicated by the 2009 expenses scandal than were members in marginal seats. Eggers's (2014, p. 463) evidence suggests that “politicians filing expenses claims (or parties placing politicians in constituencies) took calculated risks based partly on the electoral punishment they were likely to suffer if improper behavior was brought to light.” Overall, these studies suggest that in situations where candidates can anticipate lower electoral penalties for corrupt activities, corruption might become more widespread as a result.

CONCLUSIONS AND NEW DIRECTIONS

Conventional wisdom suggests that democratic elections through which citizens can periodically sanction or reward politicians have a constraining effect on corruption, because voters are expected to punish corrupt activities. Yet, we often observe empirical deviations from this expectation as voters keep reelecting politicians who steal from them. We have suggested that punishing politicians for corrupt activities at the ballot box is indeed more difficult than often assumed. Recent evidence suggests that this is not merely a function of a lack (or incompleteness) of information, as formal political agency models have suggested (Ferejohn 1986, Fearon 1999, Besley 2006); it is also due to attribution errors and the fact that voters have to balance many different considerations when casting their ballot. The study of the electoral consequences of corruption has made great strides in recent years, partly owing to a reliance on experimental methods. The latest evidence suggests that even though voters have a general distaste for corruption, this often fails to translate into the punishment of corrupt politicians.

Political facts about corruption are hard to come by, because politicians aiming to sway public sentiment in their favor in order to secure reelection often have incentives to spin and engage in a blame game. In addition, voters’ group-serving biases make correct blame attribution difficult. Moreover, the costs of getting informed about corruption and making correct blame attributions may not always outweigh the benefits associated with electorally sanctioning corrupt politicians, especially in contexts where corruption is prevalent. Finally, voters may lack viable alternatives or be uncertain about how others will behave. Regarding the latter, sanctioning politicians for corrupt actions requires voter coordination (see also De Vries et al. 2017, Klašnja et al. 2017). If correctly informed voters expect that others will fail to punish corruption, they might base their voting decision on other considerations rather than their corruption evaluations.

Based on this review of the extensive literature on the electoral consequences of corruption, we can highlight four themes that deserve more attention in future research. First, as we just suggested, we need to consider that the relationship between corruption and accountability might be affected by the extent to which people are able to coordinate and punish corruption collectively. Coordination is of crucial importance to trigger a behavioral response among voters. Chang et al. (2010) suggest that in order to coordinate electoral retaliation for ongoing corrupt activities, voters need a coordination signal. They show that corruption went largely unpunished in Italy prior to “the Clean Hands operation, in which judicial inquiries originating in Milan ultimately implicated thousands of politicians, including five ministers” (Chang et al. 2010, p. 180). The extensive media coverage following the Clean Hands operation reduced voters’ uncertainty about the actions of fellow citizens, allowing Italian voters to coordinate and to break from years of tolerating corruption. How these findings generalize beyond the Italian case and how coordination signals matter for the electoral punishment of corruption are important areas for future research (see, e.g., De Vries et al. 2017).

Second, we need deeper insights into the role of blame attribution. Studies on retrospective voting have gone a long way in unpacking the black box of attribution. Work has focused on individual-level factors, such as group-serving biases (e.g., Rudolph 2006, Malhotra & Kuo 2008), citizens’ (in)ability to benchmark (Kayser & Peress 2012), and excessive weighing of recent events (Healy & Lenz 2014), as well as contextual factors, such as institutional context or vertical and horizontal clarity of responsibility (e.g., Powell & Whitten 1993, Whitten & Palmer 1999). It seems reasonable to accept that these factors also condition retrospective voting based on corruption.

Third, we currently lack a thorough understanding of how the viability of alternatives matters. Voters can be expected to punish corruption electorally only when viable alternative options are available. The literature thus far has focused mostly on the presence of anticorruption or ideological alternatives, but surely voters do not care only about corruption or ideology, so viability should be more broadly defined. The presence of viable alternatives is surely endogenous to the overall degree of corruption in the system (Klašnja et al. 2017) and therefore perhaps difficult to study in a causal way. Finally, and perhaps most challenging, we need to understand how candidate behavior conditions voter decision making. Corrupt politicians, as strategic agents, may for example anticipate a low electoral toll of corruption, aim to circumvent electoral punishment by withholding or manipulating information, or emphasize other dimensions that are relevant to voters’ decision making to distract their attention from bad performance (e.g., Eggers 2014, Solaz & De Vries 2017). This constitutes an important avenue of further research.

disclosure statement

The authors are not aware of any affiliations, memberships, funding, or financial holdings that might be perceived as affecting the objectivity of this review.

literature cited

Posted by at

No comments:

Post a Comment

Your message has been sent to the moderator for approval and shall appear very soon!

Subscribe to: Post Comments (Atom)