http://www.vanderbilt.edu/lapo...ghts/IO872_en_v2.pdf

AmericasBarometer Insights: 2012
Number 72

To Bribe or Not to Bribe
By Juan Camilo Plata  
juan.c.plata@vanderbilt.edu
Vanderbilt University
Executive Summary.

This Insights report explores public opinion in Latin America and the Caribbean regarding whether bribery is justifiable. Initial analysis reveals that male, wealthy, young and urban citizens are more likely to report that paying a bribe sometimes is justified. In a second set of analyses, three possible motivations for this attitude are explored. First, I argue that if people have negative perceptions of the political and economic context (lower evaluations of the economy, concern about the impact of crime, and perception of corruption as widespread), they will have a pessimistic view of government and thus find bribery to be a comparatively appropriate way to interact with public authorities. Second, if people have low trust in public authorities, I suggest they will be more prone to using bribes in order to secure more certain outcomes.  And, third, people interested in politics should be more likely to gather information about the public sector and the best means to deal with it; this knowledge gives them an advantage if they try bribery and a sense that it is nothing more than “politics as usual.” 

Bribery, as a form of corruption, distorts interactions between citizens and the state. When a public bureaucrat takes a bribe, he is using his control over public resources for private gain (Warren 2004). While any particular instance of bribery is localized, pervasive bribe‐taking can have broader negative impacts on society and politics. For example, it can negatively affect national economies and investment levels (Mauro 1995) and erode trust in the fairness of institutions (Seligson 2006). And finally, it can reduce levels of diffuse support for the government and levels of interpersonal trust (Seligson

2002). In this Insights report 1 I explore who across the Americas finds it sometimes justifiable to pay a bribe, and why. I focus the analysis on the following question from the 2010 AmericasBarometer 2 survey by LAPOP, in which 40,990 survey respondents from 24 countries were asked to respond “Yes” or “No” to the following statement:  

EXC18. Do you think given the way things are, sometimes paying a bribe is justified?

Figure 1 shows the percentage of respondents answering “Yes” to the question in each country. Rates of agreeing that bribery can be justified under current circumstances range from between 7.6% and 7.8% in Guatemala, Chile and Brazil, to more than 30% in Haiti and Guyana. More than 20%of individuals in Mexico, Belize, and Trinidad & Tobago respond affirmatively. About 15 of the

24 countries have    between 10 to 20% of the population agreeing that bribery is sometimes acceptable.  The variation presented above suggests that
attitudes toward bribery vary depending on national context. Yet, while explaining this cross‐national variation is worthy of additional study, this Insights report focuses on individual‐level predictors of this attitude. In what follows, using sociodemographic variables, I first explore if the likelihood of finding bribery justified isrelated to being part of particular sectors within 

the population. Next, I test three additional, non‐rival expectations. First, people evaluating the social and economic context poorly will be more likely to accept that it may be justified to pay a bribe. Second, people lacking trust in the public authorities will agree that bribery is justifiable to compensate for the associated uncertainty. And, third, people interested in politics will be more likely to find bribery justifiable. 4 I will further explain the logic of each of these three expectations in what follows, but first I examine a very basic model for bribe justification.

A Simple Profile of Bribe Justifiers
Extant research supports the expectation that some variation in responses to the question of whether bribes are justified can be attributed to sociodemographic factors (e.g., Redlawsk and McCann 2005). 5 I assess the predictive power of socioeconomic and demographic factors in a simple model, in which the dependent variable is respondents’ yes (1) / no (0) answers to a
question about whether, given the current circumstances, paying a bribe is sometimes justified. 6 Given the dichotomous nature of the dependent variable, I use logit analysis. Among the independent variables, I include wealth7 and education (coded in four categories: none, primary, secondary, and higher). Assuming that those who are wealthier are more likely to have the means, motives, and opportunity to offer bribes, I expect that wealthier people will be

more likely to condone bribery (see also Redlawsk and McCann 2005). Additionally, I suggest that the more educated know better how politics actually works and abandon “naïve” visions of altruistic politicians, perceiving more benignly public “favoritism,” including bribery (Johnston 1986: 385‐387; Jong‐Sung and Khagram 2005).  I also include an urban versus rural locale
variable (coded 0 for rural and 1 for urban). Seligson (2006) suggests that the urban milieu may be related to a higher acceptance of paying a bribe; state offices are more common in those contexts, thus creating more opportunities for bribery and, by extension, greater tolerance for this act. Age (coded in years) is expected to be related to bribery. Seligson (2006) finds that the
relationship between age and corruption victimization is not linear, suggesting it is associated to the life cycle and the need to deal with state institutions; I tested, but found no evidence for a non‐linear relationship between age and perceived acceptability of bribe‐paying. And finally, I include a measure of gender (coded 0 for male and 1 for female). The rationale for including this variable is research on gender role differences, which finds that women tend to condemn the violation of public norms than men (Redlawsk and McCann 2005:
265).  
Figure 2 shows the results of the logit regression analysis by reporting the effects of each predictor in the model using standardized coefficients, so that the relative effect of each can be more easily discerned. The estimated effect of each independent variable on the dependent variable is represented by a dot. If the dot and its corresponding bars, which indicate the 95% confidence interval, fall to the left of the 0 line, then the relationship is considered both negative and statistically significant; if the dot and bars fall to the right of the 0 line, the relationship is considered positive and statistically significant.

The results in Figure 2 show that education does not significantly predict one’s tendency to justify bribery, considering current conditions. As expected, however, women are less likely to condone paying bribes. Also as expected, wealth and living in an urban setting are positively associated with the likelihood of considering paying a bribe justified. Finally, as age increases,
the probability that someone will agree that bribery is sometimes justified decreases. 8
Interestingly, the effect of age is quite strong. Some (Torgler and Valev 2006) suggest this effect can be attributed to the increasing stakes of older people in maintaining a fair system in which they are deeply invested. Nonetheless, if that were the case, wealth level should capture some of this effect. Some of this may be related to an individual’s decreased tendency, over time, to interact with public officials in arenas where bribe‐taking is common, though it is unclear if this would explain completely what is a very strong effect. The next section explores the impact of additional factors to develop a more extensive explanation of when and why bribery is likely to be considered justified.  

Why Might Individuals Condone Bribery?
While the above tells us something about who is more likely to report that paying a bribe is justified, it does not tell us much about why. I assert that to understand when individuals find paying a bribe justified, one needs to take into consideration three general sets of criteria: perceptions of social and economic conditions; perceptions related to uncertainty when dealing with public officials (measured via trust in political institutions); and, awareness of “politics as usual” (measured via political interest).  

With respect to perceptions of economic and social conditions, the expectation is that negative assessments will make individuals more likely to find bribery appropriate; the reason is that poor circumstances will motivate individuals to
accept unattractive means in order to secure better outcomes (Mocan 2008). This mechanism is consistent with what others have argued: opinions about corruption are informed by perceptions of other salient issues in a given country, such as violence, poverty, or protection of rights (Abramo 2007).  
Respondents’ perceptions of perceptions of economic and social conditions are assessed through retrospective evaluations of the national economy, 9 perceptions of the impact of crime on the country’s future, 10 and beliefs that

corruption is widespread among public officials. 11
Uncertainty in the interaction with public employees should also help explain who finds bribery justified. Husted (1999) found that, at least for risk‐averse people, paying a bribe is a means for producing more certain results in environments where the arbitrariness of justice and bad economic conditions lead to uncertainty. The same reasoning could be extended if we assume that low levels of trust in public authorities produce uncertainty when
dealing with them. Therefore, when institutional trust is low, the likelihood of finding the use of corruption justified to compensate for uncertain  outcomes will be higher. This expectation is consistent with the findings of Morris and
Klesner (2010: 1273), according to whom tolerance of corruption is associated with lowerlevels of trust in political institutions.  
The following variables are included to capture the effect of trust in institutions on the belief that it is justifiable to pay a bribe: trust in the
municipal government 12 and trust in the national police. 13 I selected these trust measures because they refer to comparatively proximate interactions with public officials. However, to gauge the relevance of more distant ties as well,
I include trust in the national government 14 and trust in the justice system. 15 Generally speaking, the expectation is that lower levels of trust will be associated with a higher propensity to find bribe‐taking sometimes justifiable, under the circumstances.

  
Finally, dealing with the state is costly. In addition to the economic costs, it requires having information about public procedures and motivation to deal with it in spite of its rigidities. People interested in politics have an
advantage in this realm because they monitor more closely the public world, learning about its processes, its timing, and the most effective
means to deal with it. 16 These are all important skills when negotiating with state officials and securing their complicity in illegal transactions (Collins et al. 2009: 93). Along these lines, I expect that people are more likely to find paying
a bribe justified if they are interested in politics, 17 because they have a better understanding of the public sector and they are more alert to the best means to get things done, even if that implies paying bribes (Huntington
2006).  
The results in Figure 3 support expectations regarding the relationships between negative perceptions of one’s context and finding paying
bribes to be a justifiable behavior. If respondents perceive that the national economy has gotten worse in the past 12 months, that crime is a
threat for their future, or that corruption is very common among public officials, they are more likely to report that paying a bribe is justified.
Additionally, trust in the more immediate authorities (police and the municipal
government) is negatively related to the likelihood of believing that paying bribes is justified if these variables are included in the model independently of the other trust measures (in analyses not presented here). But, when trust in the justice system and the national government are added, as is the case in Figure 3, those variables are statistically significant and trust in the police and the municipal authorities cease being statistically significant. This finding

suggests the underlying dimension behind the initial results is not the institutions’ proximity to individuals, but rather their association  with the
State more generally. If the national government or the justice system are not perceived as trustworthy, the likelihood of finding paying a bribe appropriate increases. 18
  
Finally, the more interested in politics someone is, the more likely she is to find paying a bribe justifiable. Although the model does not include a direct measure of the contexts where bribery could be used, interest in public affairs certainly 

favors the acquisition of information relevant when considering paying a bribe.

Conclusions

Paying a bribe is believed to be justified when there are negative expectations
about one’s context. Thus, if crime is perceived as a threat for one’s future, if
the economy is perceived as having gotten worse in recent months, if corruption is thought to be common among public officials, or if public institutions are notperceived as trustworthy, it is understood that a reasonable way to attempt to secure a better outcome is through bribery. Additionally, when there is a high interest in political affairs people have a better understanding of the public sector and are more alert to the best means to get
things done, even if that implies paying bribes. 19
   
Although corruption might be seen as an appropriate response to overcome bad
conditions in a country, we must consider possible ways to circumvent this logic. Increasing the general public’s interest in public affairs so that one particular group is no longer advantaged when dealing with the state may
have the desirable properties of promoting a more realistic evaluation of how common corruption is (Allison and Canache 2005) and increasing public oversight. Additionally, higher  interest in politics and better oversight might

increase the levels of trust in public institutions, also leading to greater disapproval of bribery. However, it may be an uphill battle to instill high levels of political interest in all citizens; and, moreover, it could have the reverse and
unfortunate consequence of spreading tolerance of bribery and distrust of institutions. A more direct route, then, would be to work to increase trust in institutions; the more that people perceive the authorities as playing a fair game, the more likely they will be to approach politics with norms of fairness.

In conclusion, corruption is not just the product of unscrupulous people, but for many citizens bribery is a rational adaptation to their situation and the available means. Therefore, solutions also ought to be found in the improvement of people’s living conditions and in facilitating effective and bribe‐free interactions with their state.

Rev numbers don't lie. You do.

Do you think given the way things are, sometimes paying a bribe is justified?

More than 30% in Guyana said YES!

He he, even the people logging on to the tiefs.