CC BY
34y^K-323
DOI: 10.17072/2218-1067-2022-2-109-122
THE EFFECTS OF CORRUPTION PERCEPTION ON POLITICAL PARTICIPATION:
EVIDENCE FROM AFRICA
S. Toklo
Toklo Sewordor, a Ph.D, Candidate in Political Science and Regional Studies. National Research University Higher School of Economics. E-mail: [email protected] (ORCID: 0000-0003-1914-2706).
Abstract
Scholars have previously examined the relationship between corruption and political participation with mixed results. However, some citizens are aware of corruption but choose to tolerate it. How does this tolerance for corruption influence citizens' political conduct? This research contributes to the literature by examining whether citizens' tolerance for corruption affects their participation in political activities, particularly on voter turnout and protest. The study uses regression analysis to evaluate the relationship between corruption tolerance and these two indicators of political participation. Using round 3 of the Afrobarometer data for 18 Sub-Saharan African countries, I hypothesize that individuals' tolerance of corruption decreases their willingness to vote in elections and participate in protests. The findings corroborate this expectation and demonstrate substantial robustness to various model specifications.
Keywords: corruption perception; corrupt behavior; tolerance; voter turnout; participation; protest; substitution.
Introduction
Transparency International defines corruption as the 'abuse of entrusted power for private gain'. Similarly, in academic literature, corruption is defined as the misuse of public office to attain a private benefit (Rose-Ackerman, 999). Corruption has many adverse effects, such as hindering the economic development of countries (Mauro, 1995), and is considered one of the principal causes of poverty in many developing countries (Otusanya, 2011). Corruption also affects the rate of foreign direct investment (Wei & Shleifer, 2000), and reduces citizens' trust in democratic institutions (Villoria, Ryzin and Lavena, 2013). The acts of engaging in corruption by officials can take many shapes and forms, from petty corruption such as bribes paid to police officers to avoid fines to large-scale corruption, which involves particularly significant and impactful abuses of power.
When public officials who are paid to carry out their duties professionally indulge in any form of corruption, the public is likely to form a negative perception of them. For instance, large-scale corruption may lead to the creation of certain beliefs about the country's political system as being designed to work against them, hence can translate into the way citizens participate in political activities (Gupta, Davoodi and Alonso-Terme, 2002).
Political participation can take many forms, from conventional voting in elections and signing petitions to unconventional forms such as insurgencies and mass uprisings. This study focuses on two forms of political participation: voter turnout and protest. Specifically, the paper addresses the question, does tolerance of corruption influence political participation? These topics are essential because voter turnout reflects whether a person considers voting a meaningful way to demonstrate preferences about how their country is run. Protests are also prevalent when citizens want politicians to solve serious social, economic, or political issues and often serve as the last conventional tool in the hands of citizens that they can use to impose pressure on the government.
Voter turnout and protest have been extensively studied in the previous literature. However, the effect of corruption perceptions on these activities is understudied. This paper investigates the impact of corruption perceptions on voter turnout and protests and attempts to fill this gap in the existing research. I seek
© Toklo S., 2022
to clarify the theoretical mechanisms that govern the hypothesized relationship and provide an extensive empirical evaluation based on the Afrobarometer survey.
This paper makes an important contribution to the body of knowledge on corruption and participation. First, this study contributes to the literature on corruption by demonstrating the detrimental impacts of tolerance of corruption perceptions on people's willingness to vote and protest. Corrupt practices negatively impact the effective functioning of democracies, leading voters to abstain from voting or participating in protests against an underperforming and corrupt government. Second, this article contributes to the broader literature on participation by introducing an important variable, corruption tolerance, which should be considered when estimating voter turnout and protest models in future research.
I concentrate on Africa for a variety of reasons. To begin with, Transparency International has regularly rated most African countries as very corrupt. For example, the 2020 Corruption Perception Index reveals that, except for Botswana, Cabo Verde, Mauritius, Namibia, Sao Tome, and a few other nations that have made a relative success in the fight against corruption, many African countries continue to have high levels of corruption. Similarly, Global Financial Integrity in 2017 found that illegal outflows from Sub-Saharan Africa were the greatest in 2014, ranging from 5.3 percent to 9.9 percent of overall commerce (Global Financial Integrity, 2017). Also, concerns about nepotism and cronyism have turned corruption into a problem that stymies Africa's economic and political growth (Szeftel, 2000). This is also in line with the United States Department of State's Country Reports on Human Rights in 2021, which showed that most African countries' public agencies, including elected officials, engage in corrupt conduct with impunity. According to the report, all levels of government, including the court and security services, were compromised by massive, widespread, and persistent corruption (U.S. Department of State, 2021). Furthermore, political participation, such as voting and protesting, remains low in many African nations (Resnick and Casale, 2011). Having considered these, could there be a particular relation between corruption and political participation in Africa?
The subsequent sections discuss the literature of previous studies that directly address the relationship between corruption and voter turnout and protest. This is followed by an explanation of the arguments advanced in this research. Next, I discussed the methods which present the model estimates and the effects of corruption perception on voter turnout and protest. Finally, I offer a summary of the significant findings as well as potential future research directions.
Theoretical literature
This study focuses on two forms of political participation, voting, and protest. Below, I provide an overview of the relevant body of research on both these topics to identify gaps that this paper seeks to fill.
Voting
One of the long-standing debates among scholars is why individuals vote in elections (Verba, Schlozman and Brady, 1995). In this regard, voter participation is not just one of the most significant markers of democratic legitimacy but is also critical for connecting principals and agents regarding policy representation and public accountability (Franklin, 2004). There have been numerous academic theories for why citizens vote or do not vote. First, citizen voting has long been regarded as a civic responsibility, with some persons believing that if they do not vote in elections, others would think worse of them (Blais, Young and Lapp, 2000). Scholars have cited this line of reasoning as one of the primary causes for voter participation, in which voters see voting as a civic obligation (Riker and Ordeshook, 1968). Indeed, works confirming this have pointed out that individuals are more likely to vote when they see most of their peers voting.
Institutional and contextual factors such as compulsory voting (Blais, 2006), proportional representation (Diaz, 2014), the electoral district's size, voting age, and voting laws (Mishler and Rose, 2005), were also found to be significant predictors of voter turnout. Also, a more competitive election where the margin of victory is small will see a high voter turnout. Thus, when an election is very close, politicians make a concerted effort to reach out to people who are indecisive or who would not ordinarily vote to turn up and participate (Grofman and Selb, 2009).
In the specific case of Africa, there are several reasons why a voter decides to participate in an election. Voter decisions and acceptance of outcomes in Africa like in Ghana are co-partisan based (Toklo, 2021). However, Ninsin (2006) argues that for African citizens to engage in electoral participation, it may involve utilizing coercion, and financial incentives. This line of argument has been criticized by Cheeseman,
Lynch, and Willis (2021), who argue that elections in Africa allow politicians and voters to make moral claims and portray themselves as virtuous agents, which stimulates and strengthens citizen participation.
Compared to psychological, institutional, and contextual factors, the effects of the 'output' side of the political system on voter turnout, such as the effect of good governance on citizens' voting intentions (Ols-son, 2014), received scant scholarly attention. A handful of studies analyzing differences in voter turnout have focused on people's perceptions of corruption, and even those that exist tend to focus mainly on advanced democracies (Karahan, Coats and Shughart, 2006; Escaleras, Calcagno and Shughart, 2012). At the same time, limited attention is given to emerging democracies, especially African countries. Generally, the research examining the relationship between corruption and electoral turnout is divided.
For instance, an empirical study on this relationship was conducted between corruption and voter turnout in established democracies. Results demonstrate that as corruption increases, so does the number of voters who attend the polls (Stockemer, LaMontagne and Scruggs, 2013). This is because public perceptions of corruption will infuriate voters; hence they will turn up on election day and vote for a non-corrupt official (Kostadinova, 2009). Also, corruption may increase the demand for voters on the side of incumbents who want to keep their political power (Karahan, Coats and Shughart, 2006). Hence, corrupt incumbents tend to mobilize electorates, particularly their supporters, to cast votes on election day. Given this, it is claimed that citizens who accept corrupt behavior will turn up to vote in an election (Bonifácio and Paulino, 2015).
On the other hand, certain studies posit that individuals' propensity to participate in the democratic process is reduced by corruption. The quality of government in an area has a favorable effect on regional turnout such that corruption reduces a citizen's chance of voting. When citizens see corruption as a sign that they cannot have confidence in the politicians, they are less inclined to turn out because corruption indicates that the authorities would not respond to their concerns (Sundstrom and Stockemer, 2015; Inman, and Andrews, 2009; Dalton and Weldon, 2005; Mishler and Rose, 2005). The entire political system appears bleak to voters who are disengaged. As a result, citizens have learned that staying at home rather than voting in inconsequential elections is a viable option. This then subsequently affects the entire voter turnout (Kostadi-nova, 2009).
What about citizens who are aware of corruption but choose to accept it? What impact does this tolerance have on their political conduct, such as voting? To the best of my knowledge, there is only one close scholarly research paper devoted to this subject. The authors claim that citizens who accept corrupt behavior will turn up to vote. To add to the literature, this paper investigates the relationship between perceptions of corruption tolerance and voting decisions and departs from the previous argument.
Corruption, in my view, serves as a substitute for voting. People who obtain their objectives through corruption are unlikely to vote for change. As a result, if people are more tolerant of corrupt activities, they do not think alternate methods of preserving their interests are as effective.
According to rational choice theory, individuals use reasonable assessments to make sound decisions and attain results that are consistent with their own individual goals. These outcomes are also connected to optimizing a person's self-interest. Despite the minimal options available, using rational choice theory can offer people the most utility (Blais, Young and Lapp, 2000). In rational-choice theory, agents are frequently believed to have selfish motives. As a result, the act of voting can be rational, with a rational voter deciding to vote based on an evaluation of the projected effects of not voting and the benefits of a corrupt regime.
It is indeed true that some citizens who benefit from corrupt behavior may turn out to vote for the incumbent regime (Karahan, Coats and Shughart, 2006), to continue enjoying the corrupt benefits. However, people who achieve their goals through corruption are unlikely to vote to change the incumbent. Thus, if people are more tolerant of corrupt practices, they do not believe alternative ways of defending their interests are as effective. Hence, those who do not believe corruption is a problem because they benefit from it will not see the need to vote to change the government. In this case, people's willingness to participate in elections is substituted by their willingness to tolerate corrupt behaviors.
Furthermore, a citizen may see voting as a civic obligation (Downs, 1957), or may see friends and co-workers voting (Huckfeldt and Sprague, 1995), or perhaps because he belongs to a particular identity group. However, people who are more tolerant of corrupt behaviors may simply assume that other methods of preserving their interests are ineffective, and hence may be willing to ignore all other factors that influence participation in the election. Voting is not a big issue for such a person, especially if he or she already benefits from the corrupt system. The eagerness of people to engage in voting is substituted by their propensity to tolerate corruption.
H1: People who are more tolerant of corrupt practices exhibit a lower propensity to vote in elections.
Protest
Protests activities aimed at exposing corrupt conduct are becoming more prevalent in many nations with high levels of corruption. Apart from Tunisia, Egypt, and Ukraine, which have recently seen large revolutions, different anti-corruption protests have occurred in Nigeria, Moldova, and India. Protesting is expensive, especially in terms of time, resources, and probable arrests. However, if individuals wish to voice their opinions on politics and policy problems, they continue to demonstrate (Leighley, 1995).
What factors influence why some individuals choose to protest while others do not? Scholars generally agree that group behavior such as identity (Opp, 2009; Van Zomeren, Spears and Leach, 2008), ideological convictions (Matebesi and Botes, 2017), and the desire to improve one's situation influence individuals' willingness to participate in protest actions (Klandermans, 2014).
Other scholars also believed that corruption significantly impacts protest activities. When citizens feel that corruption is to blame for their terrible economic situation, they may be mobilized to protest (Kos-tadinova, 2013; Brunori, 2017; Gimpelson and Treisman, 2018). According to studies, the continuous cost of bribe payments increases citizens' willingness to engage in protests (Gingerich, 2009; Monyake and Hough, 2019). Many individuals may feel compelled to express their dissatisfaction with selfish politicians (Leigh-ley, 1995). Individuals who feel enraged by politicians' corrupt actions (Opp, 2009), or have significant societal incentives, according to collective action theories (Olson, 2012), might join others on the street and demonstrate against the government.
What about citizens who know about corruption yet choose to tolerate it? How does this tolerance affect their willingness to participate in protests? To the best of my knowledge, no systematic studies have provided empirical support for such correlations. In this study, I aim to contribute to the literature by looking at how likely those who are more tolerant or less tolerant of corrupt behaviors are to participate in protests.
The rational actor model at the individual level, explains why individuals cooperate when it benefits them (Chai, 2005). Consequently, joining a demonstration can provide personal advantages and the communal good that the protest aims to achieve (Oberschall, 1994). People's perceptions of corruption, I argue, influence their propensity to participate in protests to the extent that they weigh the benefits of corrupt behavior against the possibility of protesting. According to collective action theories, people are more likely to mobilize and participate in protests when they are dissatisfied and have significant societal incentives (Opp, 1990).
In this case, people who believe corruption is on the rise are more likely to protest if their economic realities deteriorate. Second, even if those who profit from corruption are dissatisfied with some aspects of the present system, engaging in protests increases the risk of losing corruption advantages. As a result, people who accept corruption are unlikely to join demonstrations. Most protests call for economic or political change, and those who profit from corruption are inclined to like the current system. Furthermore, protesting raises the danger of losing corruption advantages, even if individuals who benefit from corruption are dissatisfied with some parts of the current system. Hence, those who accept corruption are less likely to participate in protests.
Again, the substitution argument also holds for people's propensity to participate in a protest. A citizen's willingness to participate in protest is substituted by their willingness to tolerate corrupt behaviors. Those who do not feel corruption is an issue will not see the need to go out and protest the government because most protests seek to influence public opinion or call for laws against bad governance such as corruption. As rational choice arguments indicated earlier - people weigh their choices and choose the best that maximize their interest. For those who do not feel corruption is an issue, for them, corruption is acceptable behavior. This may especially be true if the citizen is a beneficiary of the corrupt regime in place. As a result, citizens' readiness to protest is replaced by their willingness to tolerate corrupt behavior. H2: People who are more tolerant of corrupt practices exhibit a lower propensity to protest participation.
Methodology and Data
The data is from 18 African countries: Benin, Botswana, Cape Verde, Ghana, Kenya, Lesotho, Madagascar, Malawi, Mali, Mozambique, Namibia, Nigeria, Senegal, South Africa, Tanzania, Uganda, Zambia, and Zimbabwe from Round 3 of the Afrobarometer Surveys. The data was collected in 2005 and 2006. While there are recent rounds of Afrobarometer data, none of these rounds asked questions regarding tolerance of corruption. Therefore, I chose this set of data because it contains essential questions that measure corruption tolerance.
The data includes questions that ask respondents about their experiences and opinions on various governance subjects, such as bribery and corruption. It is a cross-national public opinion survey that assesses people's views and actions in response to a wide range of themes. This covers their reactions to different hypothetical instances of public corruption, whether they voted in the last election and their participation in rallies and demonstrations. Additionally, the data uses a standard questionnaire with identical or functionally equivalent terminology, making it appropriate for a cross-national study.
Models
The study employed regression analysis with various model parameters to derive inferences from the data. The dependent variables - vote, and protest are binary indicators; hence the study uses logistic regression with a random effect model to estimate the decision to participate. Like any other regression method, this method explains the relationship between the outcome variable and each of its predictors, making it appropriate for this paper. The results were analyzed using R. Additionally, the study tests these results using the logistic regression without random effect as a robustness check, corroborating the arguments made in this research. The paper also introduces instrumental variable regression analysis to address issues of possible reverse causality.
Dependent Variables
There are two dependent variables specified in the paper. The first dependent variable is voting, where respondents were asked if they had voted in the last general elections. In question (Q30) of the Afro-barometer specifically, respondents were asked whether, concerning the most recent national elections, they had voted or not. The scale point of the questions ranges from 1 to 7. The responses were labeled as; voted in the elections, decided not to vote, could not find the polling station, were prevented from voting, did not have time to vote, did not vote for some other reason, and were not registered. A few additional variables were also coded as; don't know and refuse to answer. Since the idea is to determine whether a citizen partook in the vote (or was willing to vote) versus a citizen who did not vote (or was not ready to vote), the responses were coded into dummy variables as; willing to vote versus not willing to vote (0= not willing to vote; 1= willing to vote).
The category of variables that were considered as willing to vote includes the following: 'You voted in the election, you could not find the polling station, you were prevented from voting' whereas 'you decided not to vote, you did not have time to vote, did not vote for some other reasons, you were not registered, don't know, and refused', were treated as not willing to vote (reference point).
Similarly, the second dependent variable is protest. The study measured propensity to protest using questions where respondents were asked in question (Q31C) to list specific actions that people take, including protest. 'Please say whether you have done any of these things during the past year. If not, would you do this if you had the chance: Attended a demonstration or protest march?'. The scale point for this question ranges from 0 to 4, where higher values imply a willingness to protest, and the lower value indicates not willing to protest. The responses were labeled as 'No but would do if I had the chance, 'Yes, once, or twice', 'Yes, several times', 'Yes, often', and these were considered willing to protest. The other responses: 'No, would never do this', 'do not know, refused' were considered unwilling to protest (reference point).
Independent Variable
The independent variable is tolerance of corruption. In question (Q58), respondents were asked to state whether specific actions were right or wrong. These are: (a) an official of government gave a job to his family member without having the necessary qualification, (b) an official of government sought favor or additional payment for services that are already part of his job, and (c) a public officer locates a developmental project to an area where his friends and supporters reside. The responses to these statements were coded as 'not wrong at all, wrong but understandable, or wrong and punishable'. A corruption index was constructed by summing up all corruption tolerance statements and dividing them by the total number of variables.
The main weakness of using this indicator is the inability to fully take into consideration the cultural dimension of corruption. Corruption in Africa has some cultural foundation that cannot be overstated. Nepotism, tribalism, and misappropriation of public funds are all inevitable outcomes of families' and communities' demands on a public official (De Maria, 2001; Chinweuba, 2018).
Therefore, I acknowledge that this measure may not adequately capture the concept of corruption. As a result, rather than straining the assertion that the measure accurately captures the concept of corruption, I prefer to situate corruption from the general political science definition, as the abuse of public office for private benefit (Rose-Ackerman and Truex, 2012; Nye, 1967). I did this by following Chang and Kerr (2017) that also built on the corruption tolerance index from the Afrobarometer data using the same indicator.
Control variable
Several control variables were added to guard against erroneous correlations. The individual controls include age, gender, education, and employment status. For the country level, the control variables GDP per capita and the democratic status of the selected countries were utilized. Age
A control variable was created for age. The data excludes those who are below the age of 18.
Gender
Respondents were asked to indicate their gender in question (Q101). Gender is coded as a dichoto-mous variable. The responses were coded as (0 = male; 1 = female). Education
The participant's level of education was the topic for questions (Q90). The original responses for education were recoded to 'no formal education (reference point), primary, secondary, and tertiary educa-
Employment
Question (Q94) asked the respondents, 'Do you have a job that pays a cash income?' I combined those who responded 'No' (but were not searching for work) with those who said 'No' (but were looking for work) as unemployed. Those who said 'yes, part-time but looking', 'yes, part-time but not looking', or 'fulltime employees' were categorized as employed. This was coded as (0 = unemployed; 1 = employed).
Democratic status, which tends to be lower in less established democracies, is one influencing element that impacts levels of participation in politics in general (Kostadinova and Power, 2007). Data from the Freedom House Democracy Index is used to control for the democratic status of the countries to account for variations across nations. Based on this index, the countries were labeled free (reference point), partly free, and not free.
Previous research has also shown that more developed nations have higher levels of political involvement than less affluent countries (Inglehart and Welzel, 2010). As a result, I introduced GDP as an additional control variable which was obtained from the World Bank Indicators.
Results
This section presents the regression outcomes by investigating the association between political behavior and tolerance of corruption. Table 1 presents a logistic regression of the corruption models predicting voter turnout.
Model 1 presents the corruption tolerance variable and how it affects turnout. As indicated in model 1, citizens who perceive corruption but tolerate it are less likely to vote. As a result, the log of the probability of voting against the probability of abstaining reduces by 0.31 for every step on the corruption tolerance variable, with a standard error (0.04).
Model 2 estimates the tolerance for corruption variables with the individual-level variables such as age, gender, education, and how they influence turnout. Upon the inclusion of these variables, the result of the estimate shows a robust effect of corruption tolerance on citizens' willingness to vote by reducing the log-likelihood of voting by 0.26 and a standard error (0.05). Models 3 and 4 also included economic and country-level variables, respectively, and the results all show a robust effect (-0.26, SE = 0.05) for both.
The findings also show that female individuals have lower voter turnout compared with individuals who are male. These individual-level effects on voter participation contradict findings (Dahlberg and Solevid, 2016), which found that female citizens exhibit a higher propensity to vote when compared to males.
Table 1
Showing the Effect of Tolerance of corruption on voter turnout
Model 1 Model 2 Model 3 Model 4
Predictors Log-Odds Log-Odds Log-Odds Log-Odds
(Intercept) 1.26 *** (0.11) 0.53 *** (0.12) 0.41 *** (0.12) 0.87 *** (0.25)
Corruption Tolerance -0.31*** (0.04) -0.26 *** (0.05) -0.26 *** (0.05) -0.26 *** (0.05)
Age 31-50 1.48 *** (0.04) 1.43 *** (0.04) 1.43 *** (0.04)
Age 51+ 1.69 *** (0.05) 1.70 *** (0.05) 1.70 *** (0.05)
Gender -0.23 *** (0.03) -0.18*** (0.03) -0.18*** (0.03)
Primary Education 0.13 * (0.05) 0.11 * (0.05) 0.11 * (0.05)
Secondary Education -0.01 (0.05) -0.04 (0.05) -0.04 (0.05)
Tertiary Education 0.39 *** (0.07) 0.29 *** (0.07) 0.29 *** (0.07)
Employment Status 0.43 *** (0.04) 0.43 *** (0.04)
Not Free 0.00 (0.28)
Partly Free -0.30 (0.23)
GDP 0.05 * (0.02)
Random Effects
ICC 0.06 0.06 0.06 0.04
N 18 country 18 country 18 country 18 country
Observations 25,397 25,397 25,397 25,397
AIC 27761.965 25125.451 24987.650 24988.542
log-Likelihood -13877.982 -12553.726 -12483.825 -12481.271
Standard errors in parentheses *p<0.05 **p<0.01 ***p<0.001
Table 2 presents the results of the logistic regression for protest with different model specifications. Model 1 depicts the tolerance of corruption and how it affects people's participation in a protest. The table shows that corruption tolerance is associated with decreased protest participation. Hence, the log of the probability of participating in a protest vis a vis not participating reduces by (0.19, SE=0.04) for each step on the corruption tolerance variable.
Several control variables were included in Models 2, 3, and 4, which guard against spurious effects and ensure a robust check. The effect for the relationship after the inclusion of additional variables in models 2, 3, and 4 also shows a robust outcome. For instance, in model 4, the log odds of citizens' propensity to participate in protest actions decreases (0.16, SE=0.04) when an individual is tolerant of corrupt activities.
Table 2
Showing the Effect of Tolerance of corruption on protest
Model 1 Model 2 Model 3 Model 4
Predictors Log-Odds Log-Odds Log-Odds Log-Odds
Intercept -0.04 (0.12) 0.08 (0.13) 0.04 (0.13) 0.03 (0.30)
Corruption Tolerance -0.19*** (0.04) -0.16*** (0.04) -0.16*** (0.04) -0.16*** (0.04)
Age_31-50 -0.04 (0.03) -0.06 * (0.03) -0.06 * (0.03)
Age_51+ -0.57 *** (0.04) -0.57 *** (0.04) -0.57 *** (0.04)
Gender -0.39 *** (0.03) -0.38 *** (0.03) -0.38 *** (0.03)
Primary Education 0.11** (0.04) 0.11** (0.04) 0.11** (0.04)
Secondary Education 0.31*** (0.04) 0.30 *** (0.04) 0.29 *** (0.04)
Tertiary Education 0.44 *** (0.05) 0.39 *** (0.06) 0.39 *** (0.06)
Employed 0.16*** (0.03) 0.16*** (0.03)
Not Free -0.55 (0.35)
Partly Free 0.03 (0.28)
GDP -0.02 (0.03)
Random Effects
ICC 0.07 0.08 0.08 0.06
N 1 ft 10 country 1 ft country 1 ft 10 country 1 ft 10 country
Observations 25397 25397 25397 25397
AIC 33867.729 33243.702 33216.635 33219.429
log-Likelihood -16930.865 -16612.851 -16598.318 -16596.714
Standard errors in parentheses *p<0.05 **p<0.01 ***p<0.001
Figure 1. The graph shows the mean prediction for voter turnout at a different levels of corruption tolerance
In the graph above, I used a coefficient vector and covariance matrix to draw a sample of a normal distribution. I constructed the data where only corruption tolerance varies while other variables were held constant to see how the actual effect operates. As can be seen, the probability of participating in a vote decreases with a high level of corruption tolerance. From the graph, at '0' corruption tolerance, the probability of voter turnout is high with a mean prediction of '0.597'. However, an increase in corruption tolerance exhibits a lower probability of voter turnout. Specifically, holding all other variables constant, the probability of a voter being highly tolerant of corruption at '1'. This shows a mean prediction of voter turnout to decrease to 0.534.
The graph shows the mean prediction for protest turnout at a different levels of corruption tolerance
0.50
corruption_tolerance
To understand how the actual effect of corruption tolerance affects people's propensity to protest, I created data where corruption tolerance varies while other variables remain constant, (figure 1). At '0' corruption tolerance, the probability of participating in protest is high at a mean prediction of 0.56. Holding all other variables fixed, a high level of corruption tolerance reduces the likelihood of engaging in a demonstration. As can be observed, when corruption tolerance is set to 1, the likelihood of citizens participating in protest falls to 0.52.
There however is probable cause to think that endogeneity makes estimating the influence of corruption tolerance on political participation difficult. Furthermore, corruption tolerance and political participation (vote, and protest) are most likely to have the problem of reverse causality. Thus, those who are less likely to participate in protests and vote are also more likely to tolerate corruption. Instrumental variable estimation is one method for dealing with endogeneity and reverse causality problems. The paper utilized Acemoglu, Johnson and Robinson (2001) settler mortality instrument. Out of the 18 countries that were in the original data from the Afrobarometer, only 9 were captured in the settler mortality instrument developed by Acemo-glu, Johnson and Robinson (2001). See appendix (Table 8) for the countries and settler mortality rates used for the analysis.
The paper takes the view that corruption tolerance is likely a function of institutions. Institutions, in turn, are a function of settler mortality. We can find a causal effect of corruption tolerance on participation by controlling for mortality. That is not to imply that there is a causal relation between settlers' mortality and their tolerance for corruption; rather, the two are correlated. Essentially, all that is required for the instrument to be useful is a correlation. There is a statistically significant association between mortality and corruption tolerance, as shown in Appendix (Table J). Thus, a high mortality rate indicates a lower level of corruption tolerance. Corruption tolerance is first regressed on the instrument mortality and predicted corruption tolerance values are then used to estimate a causal effect of corruption tolerance on participation. This is done in different model specifications. The findings in the instrumental variable regressions (Tables 6, and 7) show that even after introducing the instrument, participation in both protests, and voting decreases as citizens tolerate corrupt behavior.
Discussion
H1: People who are more tolerant of corrupt practices exhibit a lower propensity to vote in elections.
In all four models, the independent variable, tolerance of corruption, was statistically significant in relation to the dependent variable, propensity to vote (Table 1). This is such that the more citizens tolerate corruption, the more likely they will abstain from voting. As argued in this research, corruption can substitute other activities such as voting. This is because people who are more tolerant of corrupt activities do not believe that other methods of safeguarding their interests are as effective. This finding contradicts previous claims that people who tolerate corruption are more likely to vote in elections (Bonifacio and Paulino, 2015). Again, the argument advanced in the study is also true in the graph (Figure 1). As can be seen at every level, the probability of participating in a vote decreases with a high level of corruption tolerance.
Stockemer, LaMontagne and Scruggs (2013) argue that voter turnout increases with corruption rise in advanced democracies, but this may not entirely be the case for many African countries. Citizens turned out to vote for or against a politician will be based on the extent to which such people accept corrupt behavior, especially where many African cultures allow for gifts and reciprocity.
H2: People who are more tolerant of corrupt practices exhibit a lower propensity to protest participation.
The dependent variable, protest is negatively correlated with corruption tolerance in all four models. As a result, the more citizens accept corruption, the less likely they will join protesters in opposing the government. Citizens' perceptions of corruption, as opined in the literature review section, may influence their propensity to participate in protests to the extent that they evaluate the benefits of corrupt behavior against the opportunity to protest. The argument is that most protesters want economic or political change, and those who profit from corruption are likely to be satisfied with the current system. This means the willingness to engage in protest does not necessarily happen simply because a person is identified with a group (Opp, 2009), and ideological beliefs (Matebesi and Botes, 2017).
Also, although bribe payments may influence protests (Gingerich, 2009), however, it may largely depend on how citizens generally accept or are against such behavior in the political system. For instance, African politics have many years been defined by patronage politics—using the state and its resources to stay
in power (Van de Walle, 1994). As a result, African political leaders have used public sector institutions to provide the advantages and patronage necessary to create patron-client support networks for themselves. Top administrators have been appointed almost everywhere based on personal and political relationships with people in positions of power (Mwenda and Tangri, 2005). This is likely to create an environment where persons who benefit from the existing political system do not participate in the protest. Therefore, long ago, Lindberg (2006) noted that the way patrimonial networks dominate and destroy official state institutions is frequently blamed for the problems that plague Africa's political system. Hence, Lindberg (2013) argues that eradicating patrimonialism is necessary for strengthening democratic structures and consolidating democracies.
Juxtaposing these findings and making a reflection on governance issues in many parts of the African countries raises a call for serious attention. Indeed, one of the most pressing issues in Africa for citizens is the issue of corruption. For instance, the 2021 Corruption Perceptions Index (CPI) report for sub-Saharan Africa revealed that amongst all the democratic turbulence, increasing corruption continues to worsen freedoms in the region. In addition, the reports further show that to keep corruption out of the public sphere and for the fear of being exposed, some of the authorities across the region have censored information and repressed critical voices who expose misconduct. This may suggest why some countries in Sub-Saharan Africa exhibit no significant change, making corruption one of the region's most pressing challenges. Similarly, according to a post-election survey report released by the Centre for Democratic Development-Ghana (CDD-Ghana, 2021), 62 percent of Ghanaians believe the administration will not be able to combat corruption. This perception by citizens may by extension play some key role in citizens' willingness to accept corrupt behavior as far as it benefits them and subsequent refusal to participate in the political processes.
The finding of the research highlights two main implications. First, this study establishes the detrimental consequences of tolerance of corruption on political behavior. Corrupt practices do not only impede economic development but also affect the efficient function of democracies, causing citizens to turn away from voting or from participating in a protest against a poorly performing and corrupt administration. Secondly, the common assumption among scholars is that cross-country variations in voting participation are primarily explained by institutional features that make turnout higher in some countries than others and for protests to be more common in some countries. To establish the robustness of the findings observed among developing and mature democracies, future studies should revisit research designs by including other important variables such as tolerance of corruption.
Conclusion
In many African countries, corruption is still a serious issue. While many African scholars agree that widespread corruption hurts economic and political development, there has been little research on the impact of corruption tolerance on citizens' participation in political processes. This research addresses the question of how tolerance of corruption influences political participation?
The findings of this research contribute new insights to the existing literature on voter turnout and protest, particularly in African countries. This study aimed to determine individuals' propensity to vote and protest when they tolerate corruption. The paper put forward a novel theoretical idea that viewed corruption practices as a substitute for participation in voting and protest. The research produced two key hypotheses that expected a negative effect of corruption tolerance on political participation.
First, I explored the argument that those who are more tolerant of corrupt behaviors will be less likely to vote in elections. The study established that tolerance of corruption decreases the log odds of people's willingness to vote in an election. This implies that those who are more tolerant of corruption will be less eager to engage in democratic activities such as voting. Secondly, the study investigates the idea that people who are more tolerant of corrupt behaviors are less likely to participate in protests. The results from the empirical work confirm this hypothesis.
One main concern in this study has to do with generalization. Instead of applying the findings to all African and developing countries, the study is limited to only democratic countries. This research acknowledges the necessity to expand the scope of the study to include a broader sample of other African countries. In Africa, however, there is relatively limited reliable data on public opinion. For example, Afrobarometer mainly focuses on democratic African countries. This suggests that extra caution should be exercised when extrapolating these findings.
References
Blais, A. (2006) 'What affects voter turnout?'', Annu. Rev. Polit. Sci, 9, pp. 111-125.
Blais, A., Young, R. and Lapp, M. (2000) 'The calculus of voting: An empirical test', European Journal of Political Research, 37(2), pp. 181-201.
Bonifácio, R. and Paulino, R. (2015) 'Corruption and Political Participation in the Americas and the Caribbean'', Brazilian political science review, 9, pp. 54-80.
Brunori, P. (2017) 'The perception of inequality of opportunity in Europe'', Review of Income and Wealth, 63(3), pp. 464-491.
CDD-Ghana (2021) URL: https://cddgh.org/62-of-ghanaians -say-akufo -addo -cannot-curb-corruption-cdd-ghana-survey/on 15/03/22.
Chai, S. K. (2005) 'Rational choice: Positive, normative, and interpretive'', Annual Meeting of the American Sociological Association, Marriott Hotel. Loews Philadelphia Hotel, Philadelphia, PA, p. 2005.
Chang, E. C. and Kerr, N. N. (2017) 'An insideroutsider theory of popular tolerance for corrupt politicians', Governance, 30(1), pp. 67-84.
Cheeseman, N., Lynch, G. and Willis, J. (2021) Why Do Elections Matter in Africa? Democracy, Voting and Virtue. Cambridge University Press.
Chinweuba, G. E. (2018) 'Culture and corruption: a critical analysis of the basis of Nigerian depraved experience', UJAH: Unizik Journal of Arts and Humanities, 19(2), pp.112-130.
Dahlberg, S. and Solevid, M. (2016) 'Does corruption suppress voter turnout?', Journal of Elections, Public Opinion and Parties, 26(4), pp. 489-510.
Dalton, R. J. and Weldon, S. A. (2005) 'Public images of political parties: A necessary evil?', West European Politics, 28(5), pp. 931-951.
De Maria, W. (2007) 'Does African corruption exist?'', African Journal of Business Ethics, 2(1), pp. 1-9.
Diaz, F. and Lijphart, A. (2014) Patterns of Democracy: Government Forms and Performance in Thirty-Six Countries. 2. Aufl. New Haven: Yale University Press.
Downs, A. (1957) 'An economic theory of political action in a democracy', Journal of political economy, 65(2), pp. 135-150.
Escaleras, M., Calcagno, P. T. and Shughart, W. F. (2012) 'Corruption and voter participation:
Evidence from the US states', Public Finance Review, 40(6), pp. 789-815.
Franklin, M. N. (2004) Voter turnout and the dynamics of electoral competition in established democracies since 1945. Cambridge University Press.
Gimpelson, V. and Treisman, D. (2018) 'Misper-ceiving inequality', Economics & Politics, 30(1), pp. 27-54.
Gingerich, D. W. (2009) 'Corruption and political decay: Evidence from Bolivia', Quarterly Journal of Political Science, 4(1), pp. 1-34.
Global Financial Integrity 2017 (no date). Available at: http://www.gfintegrity.org/wp-content/uploads/2017/05/GFI-IFF-Report-2017_final.pdf.
Grofman, B. and Selb, P. (2009) 'A fully general index of political competition', Electoral Studies, 28(2), pp. 291-296.
Gupta, S., Davoodi, H. and Alonso-Terme, R. (2002) 'Does corruption affect income inequality and poverty?', Economics of governance, 3(1), pp. 23-45.
Huckfeldt, R. R. and Sprague, J. (1995) Citizens, politics and social communication: Information and influence in an election campaign. Cambridge University Press.
Inglehart, R. and Welzel, C. (2010) 'Changing mass priorities: The link between modernization and democracy', Perspectives on politics, 8(2), pp. 551-567.
Inman, K. and Andrews, J. T. (2009) 'Corruption and political participation in Africa: Evidence from survey and experimental research', Midwest Political Science Association, 3(6).
Karahan, G. R., Coats, R. M. and Shughart, W. F. (2006) 'Corrupt political jurisdictions and voter participation', Public Choice, 126(1), pp. 87-106.
Kostadinova, T. (2009) 'Abstain or rebel: Corruption perceptions and voting in East European elections', Politics & Policy, 37(4), pp. 691-714.
Kostadinova, T. (2013) Corruption and Political Participation: Testing Models of Voting and Protesting.
Kostadinova, T. and Power, T. J. (2007) 'Does democratization depress participation? Voter turnout in the Latin American and Eastern European transitional democracies', Political Research Quarterly, 60(3), pp. 363-377.
Leighley, J. E. (1995) 'Attitudes, opportunities and incentives: A field essay on political
participation', Political research quarterly, 48(1), pp. 181-209.
Lindberg, S. I. (2006) 'Opposition parties and démocratisation in sub-Saharan Africa', Journal of Contemporary African Studies, 24(1), pp.123-138.
Lindberg, S. I. (2013) 'Have the cake and eat it: The rational voter in Africa', Party Politics, 19(6), pp. 945-961.
Matebesi, S. and Botes, L. (2017) 'Party identification and service delivery protests in the Eastern Cape and Northern Cape, South Africa', African Sociological Review/ Revue Africaine de Sociologie, 21(2), pp.81-99.
Mauro, P. (1995) 'Corruption and growth', The quarterly journal of economics, 110(3), pp.681-712.
Mishler, W. and Rose, R. (2005) 'What are the political consequences of trust? A test of cultural and institutional theories in Russia', Comparative Political Studies, 38(9), pp.1050-1078.
Monyake, M. and Hough, D. (2019) 'Citizens, bribery and the propensity to protest', Commonwealth & Comparative Politics, 57(3), pp. 282-302.
Mwenda, A. M. and Tangri, R. (2005) 'Patronage politics, donor reforms, and regime consolidation in Uganda', African affairs, 104(416), pp. 449-467.
Ninsin, K. A., (2006) 'Introduction: the contradictions and ironies of elections in Africa', Africa development, 31(3), pp. 1-10.
Nye, J. S. (1967) 'Corruption and political development: A cost-benefit analysis', American political science review, 61(2), pp. 417-427.
Oberschall, A. R. (1994) 'Rational choice in collective protests', Rationality and Society, 6(1), pp. 79-100.
Olson, M. (2012) 'The logic of collective action
1965', Contemporary Sociological Theory, 124.
Olsson, S. A. (2014) 'Corruption and political participation', QoG Working Paper Series, 11(12), pp. 1-53.
Opp, K. D. (1990) 'Postmaterialism, collective action, and political protest', American Journal of Political Science, pp. 212-235.
Opp, K. D. (2009) Theories of political protest and social movements: A multidiscipli-nary introduction, critique, and syntheses. Routledge.
Otusanya, O. J. (2011) 'Corruption as an obstacle to development in developing countries: a
review of literature', Journal of Money Laundering Control [Preprint].
Resnick, D. and Casale, D. (2011) The political participation of Africa's youth: Turnout. WIDER Working Paper.
Transparency International (2021) URL: https://www.transparency.org/en/news/cpi -2021 -sub-saharan-africa-amid-democra-tic-turbulence-deep-seated-corruption-on 15/03/22.
Riker, W. H. and Ordeshook, P. C. (1968) 'A Theory of the Calculus of Voting', American political science review, 62(1), pp. 25-42.
Rose-Ackerman, S. (1999) 'Political corruption and democracy', Conn. J. Int'l L, 14, p. 363.
Rose-Ackerman, S. and Truex, R. (2012) Corruption and policy reform. Yale Law & Economics Research Paper.
Stockemer, D., LaMontagne, B. and Scruggs, L. (2013) 'Bribes and ballots: The impact of corruption on voter turnout in democracies', International political science review, 34(1), pp. 74-90.
Sundstrom, A. and Stockemer, D. (2015) 'Regional variation in voter turnout in Europe: The impact of corruption perceptions', Electoral Studies, 40, pp. 158-169.
Szeftel, M. et al. (2000) 'Clientelism, corruption & catastrophe', Review of African Political Economy, 27(85).
Teorell, J., Sum, P. and Tobiasen, M. (2007) 'Participation and political equality: An assessment of large-scale democracy', in: Citizenship and involvement in European democracies. Routledge, pp. 408-438.
Toklo, S. (2021) Book review: The moral economy of elections in Africa: democracy, voting and virtue by Nic Cheeseman, Gabrielle Lynch and Justin Willis. LSE Review of Books.
United States Department of State (2021) URL: https://www.state .gov/reports/2021 -country-reports-on-human-rights-practices/on 27/04/2022.
Van de Walle, N. (1994) 'Neopatrimonialism and Democracy in Africa, with an Illustration from Cameroon', Economic change and political liberalization in sub-Saharan Africa, pp. 129-157.
Van Zomeren, M., Spears, R. and Leach, C. W. (2008) 'Exploring psychological mechanisms of collective action: Does relevance of group identity influence how people cope with collective disadvantage?',
British Journal of Social Psychology, 47(2), pp. 353-372. Villoria, M., Van Ryzin, G. G. and Lavena, C. F. (2013) 'Social and political consequences of administrative corruption: A study of
public perceptions in Spain', Public Administration Review, 73(1), pp. 85-94.
Wei, S. J. and Shleifer, A. (2000) 'Local corruption and global capital flows', Brookings papers on economic activity, 2000(2), pp. 3.
Статья поступила в редакцию: 10.03.2022
Статья поступила в редакцию повторно, после доработки: 23.04.2022 Статья принята к печати: 01.05.2022
ЭФФЕКТ ВЛИЯНИЯ ВОСПРИЯТИЯ КОРРУПЦИИ НА ПОЛИТИЧЕСКОЕ УЧАСТИЕ:
КЕЙС АФРИКИ
С. Токло
С. Токло, кандидат политических наук,
Национальный исследовательский университет «Высшая школа экономики». E-mail: [email protected] (ORCID: 0000-0003-1914-2706).
Аннотация
Взаимосвязь коррупции и политического участия является актуальной темой в современных политических исследованиях. Тем не менее, у ученых нет консенсуса относительно характера данной связи и её направленности. В частности, граждане часто осведомлены о фактах коррупции, но предпочитают игнорировать их. В связи с этим возникает вопрос: как толерантность к коррупционным практикам влияет на различные формы политического участия? Данное исследование предполагает, что толерантность к коррупции понижает вероятность голосования на выборах и участия в протестах. Для оценки эффекта используется регрессионный анализ на данных третьего раунда опроса "Афробаро-метр", проводившегося в 18 странах Африки, расположенных к югу от Сахары. Результаты анализа подтверждают основную гипотезу и демонстрируют высокую устойчивость к различным спецификациям моделей.
Ключевые слова: восприятие коррупции; коррупционное поведение; голосование; протест.
