The Public Perception of Police Corruption in Venezuela and its Effect on National Government

It is not surprising that many citizens of Latin American countries believe public officials are corrupt. Decades of authoritarian rule, crime, and highly publicized corruption scandals have reduced confidence in many, if not all, government institutions. The public perception of corruption in Venezuela however, presents an intriguing case. Traditional explanations—economic factors, regime stability, and quality of life—are unable to account for the high levels of perceived corruption in government. This author argues, the public perception of police corruption strongly influences the perception of corruption in national government in Venezuela. Using OLS regression, the author statistical analyzes national-level Latinobarómetro data from 2002-2010. The results indicate a strong, significant relationship between perception of police and government corruption. Key-words: corruption; Venezuela; Latin America; law enforcement. La Percepción Pública de la Corrupción Policial en Venezuela y su Efecto en el Gobierno Nacional Resumen. No es sorprendente que muchos ciudadanos de los países de América Latina creen que los funcionarios públicos son corruptos. Las décadas de gobierno autoritario, la delincuencia y los escándalos de corrupción muy publicitados han reducido la confianza en muchas, si no todas, las instituciones gubernamentales. La percepción pública de la corrupción en Venezuela, sin embargo, presenta un caso interesante. Factores tradicionales explicaciones económicas, la estabilidad del régimen y la calidad de vida son incapaces de explicar los altos niveles de percepción de corrupción en el gobierno. Este autor sostiene que la percepción pública de la corrupción policial influye fuertemente en la percepción de la corrupción en el gobierno nacional en Venezuela. Utilizando la regresión OLS, el autor analiza los datos estadísticos Latinobarómetro, a nivel nacional, 2002 a 2010. Los resultados indican una relación fuerte y significativa entre la percepción de la policía y la corrupción gubernamental. Palabras clave: corrupción; Venezuela; América Latina; la policía.


Introduction
Rampant political corruption throughout Latin America dates back to colonial times.For centuries, authoritarian regimes tolerated widespread abuses of power.These same countries adopted democratic institutions and market-style economies in the late 20th century.The centralized exercise of power was replaced by a combination of decentralized structures and formalized legal codes.In this new political environment, the potential existed for citizens to establish new, and more positive, dealings with their government and other public officials.However, this potential was never realized as annual corruption indices show 54% of Latin American citizens consistently perceive little, or no, integrity in their governments (Latinobarómetro, 2010).
Scholars have presented a wide range of theories on the potential factors contributing to public perception of corruption.Arguably the most widely-held theory for explaining the perception of corruption relates to economic factors-general economic conditions of country, personal economic situation, or liberalization reforms.If a country's economy is thriving, the public will have a more optimistic view of the country's political institutions.Conversely, in a weak economy, the public will be less tolerant of the country's public officials.Other literature has also suggested regime stability and the degree of democratization explain differing degrees of perceived corruption (Di John, 2005;Seligson, 2006;Gatti et al, 2003).
Venezuela is an anomaly among Latin American countries.According to most political corruption theories, Venezuela should rank fairly high (lower levels of perceived corruption).In reality however, Venezuela's rankings are consistently near the bottom (higher levels of perceived corruption), of published corruption indices 1 .According to Transparency International's Corruption Perception Index (CPI), Venezuela's score has consistently declined since the latter part of the 1990s-2.77 in 1997 down to 2.4 in 2003-and the most recent score of 1.9 (2011). 2 This remarkably low CPI score of 1.9 earned Venezuela the distinction of being the second most (publicly perceived) corrupt country in the Americas (with only a 0.1 point difference, Haiti captured the lowest rank in 2011 (Transparency International, 2011)).
These developments present an interesting case, as the traditional causal factors do not seem to account for the exceedingly high levels of perceived corruption.Venezuela possesses substantial oil reserves and, as a result, its per capita GDP is $12,767 (CIA World Factbook, 2010).Furthermore, Venezuela has enjoyed a relatively stable constitutional democracy for more than three decades (see Table 1).One factor that does seem to explain this situation is the public perception of law enforcement corruption and its correlation to political corruption in government.Survey data indicates a widespread distrust of law enforcement in Venezuela.In fact, the perception of corruption in Venezuela's law enforcement community exceeds that of all other major governmental functions.Its negative rating also exceeds that reported in most other Latin American countries (see Table 2).
Law enforcement is one branch of government that can (and does) have a significant impact on public perception of corruption.The 2010 Latinobarómetro survey asked respondents in Venezuela 1 These indices gauge perception of corruption in all public sector activities."Broadly speaking, the surveys and assessments used to compile the index include questions relating to bribery of public officials, kickbacks in public procurement, embezzlement of public funds, and questions that probe the strength and effectiveness of public sector anticorruption efforts" (Transparency International, 2011, CPI Methodology webpage). 2See Table 5 in Appendix, 1980Appendix, -2007 data. data.about their levels of confidence in the police-one measure of perception of corruption.Over 27% said "no confidence at all" and another 41.6% replied "little confidence" (Latinobarómetro, 2009).In 2009, a news article in The Guardian, discussed public confidence in the Venezuelan police force, "the public has little confidence in the force.In one poll 70% (of respondents) said, "police and criminals are practically the same" (The Guardian, 2009).The purpose of this study aims to demonstrate how both real and perceived police corruption effect the public's perception of national government corruption in Venezuela.Specifically, I argue public perception of police corruption strongly influences public perception of government corruption.Four critical factors contribute to the present circumstances.First, regular interaction with government officials, such as governors and mayors, is limited or nonexistent.Even in densely populated urban areas, regular encounters with government officials may be scarce (Blake and Morris, 2009).In contrast, the police force is one division of the government that interacts with the public much more regularly.Second, decentralization reforms in the late 1980s and early 1990s created a broad, fragmented law enforcement structure.Authority is dispersed throughout the various tiers of government resulting in a lack of coordination between agencies and also a disproportionate ratio of police to citizens.Third, there have been numerous incidences of police brutality and scandalized criminal cases involving the police, fueling public perceptions of corruption (Birkbeck, 2009, Antillano, 2009, Portal, 2010).Additionally, between January 2008 and March 2009, police were implicated in 755 homicide cases."With some cases including multiple killings, the number of dead is likely to be significantly higher" (Caroll, 2009).Lastly, law enforcement is highly politicized; police officers at municipal, metropolitan, state, and federal levels are thought to have strong ties to their respective government officials (Winslow, 2010).
The preceding four factors provide the foundation for the following hypothesis: H1: The perception of corruption in law enforcement influences the perception of corruption in Venezuela's national government.
The rest of the paper is as follows.Section 2 presents a review of the current literature on public perception of corruption.A general overview of many political corruption theories are discussed, but the primary focus of section is Arnold J. Heidenheimer's theory on the perception of political corruption.This argues the patron-client-based system, and the norms embedded within it, influence political exchange relations between the state and society.In this system, protection is sought through personal relationships with the patron; trust in the state institutions-i.e. law enforcement-is severely lacking.Section 3 explains the historical background of Venezuela coupled with a description of the current law enforcement structure.A discussion of data, methodology and statistical techniques follows in Section 4. Section 5 discusses the empirical results in detail.The final section concludes with policy implications and suggestions for future research.

Literature Review
Literature on the depths of political corruption-both real and perceived-is immense.Scholars have also analyzed the relationship between several governance indicators to provide a more comprehensive analysis on corruption.Some of the most prominent governance indicators include: economic factors-degree of economic freedom, economic growth, and liberalization reforms; quality of government institutions; and micro-level factors-age, gender, and occupation (Gatti and Fisman, 2002;Gatti, Paternostro, and Rigolini, 2003;Mocan, 2004;Di John, 2005;Rose and Mishler, 2010).This section provides a description of several relevant theories that attempt to explain variation and causal factors of public perception of corruption.
Economic factors such as privatization of public goods or growth in the country's economy can contribute to perceptions and actual incidences of corruption (Rose and Mishler, 2010).If a country has a high economic growth rate, citizens will be better off financially and thus less likely to believe that their government is corruption, since they are benefitting from whatever the government is doing.In the 1980s, Venezuela experienced a debt crisis (1982); employment and economic growth were adversely affected.However, as Little and Herrera point out, most Venezuelans citizens were still not concerned with political corruption (1996, p. 268).Contrary to economic theories on perception of corruption, the adverse economic conditions did not increase the salience or perception of corruption during the debt crisis of the early 1980s.Venezuela's CPI score was actually the highest it has even been (at 3.19, see Table 5).
In a recent and innovative study, Rose and Mishler found Russian citizens tend to have strong negative perceptions of public officials, whether the actual incidence of bribery is high, medium, or low (2010, p. 15).Specifically, they argue, "the more individuals perceive governments as corrupt […], the less likely they are to support the political regime" (2010, p. 13).Using data from the New Russia Barometer, Rose and Mishler report 89% of respondents perceived Russian police to be corrupt, yet only 5% of them had actually experienced corruption (been asked to pay a bribe).This can be attributed to a variety of contextual factors including the reports of corruption in the media, second-hand information, and elite perceptions of corruption that are disseminated to the rest of society (Rose and Mishler, 2010).In separate study conducted in 2005 by Canache and Allison, they find that "citizens perceive corruption, and they connect those perceptions to their evaluations of incumbent leaders and political institutions" (p.106).
The quality of state institutions has also been found to be a significant determinant of the perception of corruption (Mocan, 2004).Mocan argues, "an increase in the quality of the institutions in the country and the existence of uninterrupted democracy are associated with reduced corruption propensity" (2004, p. 2).While this was true for the 49 countries examined in her study, the existence of uninterrupted democracy is not a sufficient explanatory variable for perception of corruption levels in Venezuela.Venezuela has been a comparatively successful, uninterrupted constitutional democracy since 1958, yet the public still believes there are high levels of corruption among the police and the government-which has contributed to the consistently low CPI scores for nearly two decades3 .
Levels of trust also influence perceptions of corruption, both interpersonal trust and trust in public institutions.Wallace and Latcheva examine Central and Eastern European countries in their 2006 study and find a strong relationship between loss of trust in public institutions and increased perceptions of corruption (2006, p. 82)."Those respondents who think of corruption in 1998 as decreasing compared to the communist period tend to trust political and legal institutions" (Wallace and Latcheva, 2006, p. 82).In a similar respect, Rothstein and Uslaner suggest, "At the individual level, people who believe that in general most other people in their society [including the police] can be trusted are also more inclined to have a positive view of their democratic institutions" (2005, p. 41).Trust and confidence in the state institutions and the public officials that work in them will alleviate some concerns regarding corruption in a country.However, if that trust is compromised-by alleged incidences of corruption or the widely accepted belief that all police are arrogant and violent for example-then perceptions of corruption can be expected to rise.
Gatti, Paternostro, and Rigolini ( 2003) examine other individual level factors that are thought to influence perceptions of corruption.The authors argue this information provides evidence, "that the social environment has a strong influence on the individual attitudes towards corruption.Individuals living in regions where people are on average relatively less averse to corruption tend to also be more forgiving of corruption" (2003, p. 3).Their results indicate women, employed and less wealthy individuals are more averse to corruption (p.2).
Traditional theories on corruption provide only a partial explanation of the case of Venezuela; much is left unexplained.Arnold Heidenheimer's perception of corruption theory provides the missing theoretical tools for understanding public perception of police corruption and its association with political corruption within the national government.

Perception of Corruption
Perceptions of corruption vary considerably, as they are influenced by a multitude of factors.Two of the most important determinants of the perception of corruption are the relations between the state and society-political exchange relations-and the strength of established social norms (Heidenheimer, 2002, p. 139).Historical factors play a prominent role in both the existence and perception of corruption.Historians argue, "Paternalism, weak rule of law permeating both colonial and postcolonial Latin America under civilian and military-led governments" (Blake and Morris, 2009, p. 3), and traditional patron-client relationships are just some of the roots causes of public perception of corruption.Before proceeding into a thorough discussion of perceptions of corruption, it is important to specify which definition of corruption is being used.For the purposes of this paper, I refer to corruption as, "the abuse of public roles invested with power that is impersonal, limited, and held in trust" (Johnston, 2005, p. 64).
Political scientists have a slightly different perspective on the origins of this phenomenon.Arnold J. Heidenheimer's theory embraces historically conditioned relationships, but he expands this explanation to include crucial social factors like norms and community values.In this theory on perception of corruption, Heidenheimer identifies four prototype societies and discusses the pattern of political exchange relations in each.This theoretical framework is buttressed by his infamous blackgray-white spectrum of corruption-from intolerable to tolerable forms of behavior.
The perception of corruption theory elucidates the moral standards of different societies, which determines those activities a community condemns as corrupt or accepts as tolerable.I present four prototypical societies and argue Venezuela most closely resembles the traditional patron-client based system, given its history of paternalism and authoritarian rule.
In the traditional patron-client system, citizens have strong ties to powerful protectors and their identification with the general community is rather weak.Ties to upper-class patrons develop into a reciprocal relationship; the client receives protection and the patron receives the client's vote (Heidenheimer, 2002).The patron-client relationship is important because it demonstrates the lack of trust in the state and state institutions (i.e. the law enforcement system).If the community-wide belief is that the only avenue for protection is to engage in a personal relationship with the patron, then the lack of public confidence in the police will be very low.This is exemplified in Venezuela where 'politics of privilege' characterizes the political and economic sectors (Di John, 2005, p. 114).For three decades  Venezuela was a stable, democracy society; which was maintained through political pacts between the two ruling political parties and corporatist bargaining (Di John, 2005) 4 .While the traditional patron-client based system discussed here is a prototype, a similar relationship can be discerned in the communities throughout Venezuela and Latin America.
In the traditional familist (kinship) based system, loyalty to immediate family members is of paramount importance.All others relationships-to friends, community, and the state-are not as important; people have very little trust in anyone outside of the nuclear family.The modern bossfollower-based system is one that thrives in large urban communities.Bureaucrats and public officials who obtain their positions through legitimate methods do not have as much influence over the public as those who are associated with the 'political machine'.Examples of this prototypical community are illustrated best by modern American cities, such as New York or Chicago.Finally, the civic-culturebased system is the least corrupt of the four prototypes.Strong, developed community-regarding norms exist and political exchange relations are diversified-patron-client or boss-follower relations no longer exist; the leader is answerable to the community at large (Heidenheimer, 2002).
Not only do corrupt practices exist, but also exist to varying degrees.Heidenheimer identifies three categories of corrupt behavior: 'petty', "bending official rules for friends"; 'routine', which involves patronage, nepotism, and regular gift-giving; and 'aggravated', very frequent incidences of various forms of corruption (2002, p. 147-151).For the traditional patron-client-based system, routine and aggravated corruption behaviors are the most prevalent.
The last aspect of Heidenheimer's theory describes a monochromatic scale of corruption, the black-gray-white scale.Tied to the prototypes of political obligation relations are norm patterns within those communities that evaluate corrupt practices.'Black' corruption is "one in which a majority consensus of both elite and mass opinion would condemn the practice and would want to see punished on grounds of principle (Heidenheimer, 2002, p. 151).In Venezuela, this occurs at the highest levels of government.For example, in 1993 ex-president Carlos Andrés Pérez was forced to resign due to charges of corruption-he allegedly misappropriated $17 million in public funds (Little and Herrera, 1996, p. 268).'Gray' corruption refers to those practices that only a segment of society feels should be punished (typically elites), but the vast majority does not.Lastly, 'white' corruption refers to activities that both elites and the masses view as tolerable (Heidenheimer, 2002, p. 152-153).In Venezuela 'speed money'-a bribe used to increase economic efficiency, usually by reducing bureaucratic 'red tape'-is generally considered an acceptable form of corruption (Kaufmann and Wei, 1999).This was especially true during the economic liberalization reform period during the late 1980s and 1990s (Little and Herrera, 1996).For the traditional patron-client community, Heidenheimer argues that these shades of corruption are relatively 'lighter' than the other societies (2002, p. 153).The strong, informal relationships that exist in this community relax the intolerance of otherwise 'black' or 'gray' corrupt activities.The formal and informal relationships present in society as well as the established social norms provide the foundation understanding current public perceptions of corruption.

Venezuelan Political Exchange Relations
Venezuela's political environment embodies Heidenheimer's traditional patron-client based system.It is one that has been characterized by clientelism and nepotism for centuries, despite democratic reforms (Little and Herrera, 1996;Di John, 2005;Gates, 2009).These lingering practices influence political exchange relations in modern Venezuelan society.
For the first half of the 20 th century, Venezuelan leadership was characterized by an authoritarian regime-from 1908-1935and again 1950-1958(Little and Herrera, 1996;Di John, 2005).In 1969 Rafael Caldera became the first president of Venezuela.Although political conditions were relatively stable in this new political environment, the system was 'duopolized' by a two-party system for the next several decades (1958-1993(Little and Herrera, 1996, p. 267)).Until the 1998 elections, the Democratic Action (AD) and the Christian Democratic (COPEI) parties dominated the political environment at both the state and federal level (US State Department, 2010).
The last few decades of the 20 th century generated a whirlwind of political and economic changes.The elections in 1998 ended the duopoly of political office by AD and COPEI and marked the beginning of Hugo Chavez's reign as president (1999)(2000)(2001)(2002)(2003)(2004)(2005)(2006)(2007)(2008)(2009)(2010)(2011)(2012)(2013).Later that year a new version of the Venezuelan Constitution was adopted, allowing Chavez to institutionalize some of his campaign promises (Birkbeck, 2009).One of the most important changes to occur was decentralization (1989)(1990)-as part of broader political reforms.State, local, and municipal locales now had control over police agencies-how many police to hire, what types of agencies to establish, duties of those officers-in their jurisdictions.

Venezuelan Law Enforcement
The police are probably the most visible and least known state agencies.Such opacity is intensified in the Venezuelan police forces in which secretiveness, corporativism, and the lack of openness towards society reinforce their authoritarian attitude (Antillano, 2009, p. 105).
The structure of law enforcement in Venezuela is quite different in comparison to other Latin American countries, in that a coherent structure does not exist."At the moment, there is no police system in Venezuela; there are just lots of different police bodies" (Janicke, 2009, p. 1).Specifically, authoritarianism is still highly prevalent in the police forces; "authority is exercised by individuals, rather than by roles" (Birkbeck, 2009, p. 302).Some of law enforcement's defining characteristics include an authoritarian and militarized nature and use of ineffective law enforcement tactics.Political ties are also evident in the assigned responsibilities of various police forces, enhancing the rather suspicious relationship between the police and government officials.Close relations promote the notion that corrupt practices in the police equate corrupt practices in the national government.
The Venezuelan police forces consist of municipal, metropolitan, state, and national agencies and (as of 2009) the new Bolivarian National Police (PNB) force.Lower levels of law enforcement operate under the discretion of local mayors and governors, while others agencies-the Caracas Metropolitan Police (PM)-are overseen by the Ministry of Justice and the Interior (U.S. State Department, 2010).
At the municipal, metropolitan, and state levels, there are currently 99 different forces with about 18,000 officers.Police officers at each of these levels are poorly trained and inadequately supplied.The municipal forces are concentrated in the wealthier cities, since the police are dependent upon tax revenue (Birkbeck and Galabrón, 2001;Birkbeck, 2009).Local mayors oversee with functioning of municipal law enforcement agencies and their functions mirror those of state police forces (Birkbeck, 2009).
There are 23 state police agencies as well as a 24 th force solely for Capital District of Caracas."PM Caracas is known for its excessive use of force against the general population, and is widely accused of corruption and misuse of the public's trust" (Venezuelaanalysis.com, 2011, p. 1).The governor controls state forces, but final authority rests in the Ministry of Justice and the Interior.It is also important to note that high-ranking state police officials have almost always been chosen from National Guard forces.This reinforces the already militarized nature of law enforcement and the nepotistic relations between the various police agencies and the state (Venezuelaanalysis.com, 2011).
Finally, there are four main agencies that comprise the national level police forces: the Judicial Police-the primary investigative agency; the Political Police-responsible for investigating political crimes; the National Guard-primary duties include counternarcotics control and environmental crime; and lastly the Traffic Police-responsible for investigating traffic crimes (Venezuelaanalysis.com, 2011).

Decentralization and the Police
From 1989 to 1990, far-reaching political, administrative, and financial reforms aimed at overhauling Venezuela's political and economic systems were initiated.One of the most important changes during the reform era was decentralization.This brought about the largest increase in the number of police agencies since the initial creation of the Venezuelan law enforcement system.The redistribution of power "meant a realignment of police functions, previously dependent both directly and indirectly from central government" (Antillano, 2009, p. 116).Efficient coordination mechanisms and an explicit law enforcement structure did not accompany the changes; clear jurisdictional guidelines for public policing and criminal investigation procedures remain fairly absent."The lack of a national structure poses challenges for regulation and accountability of the policing system" (Antillano, 2009, p. 112).Thus, law enforcement in Venezuela today is fragmented, convoluted, and there is widespread lack of coordination between each level of law enforcement.
Decentralization created many more opportunities for citizens to interact with the police through the creation of myriad state, local, and municipal agencies.The police became the most visible branch of national government in Venezuelan society in most, if not all, areas.Of the 126 total police agencies, there are 24 state forces, 99 local forces and 3 national forces, for a total of (approximately) 115,997 police officers (Antillano, 2009, p. 112).Venezuela is one country with the highest proportion of police in the region, with a rate of 429 officers per 100,000 inhabitants; significantly higher than international standards (350 officers per 100,000 inhabitants) as well as most Latin American countries (see Table 3).Police are a part of the government; they are the "state made of flesh" (Punch, 2000, p. 322).More interaction with (perceived) corrupt police officers will increase the likelihood that citizen perceives the entire government as corrupt.

Public Perception of Police Corruption
Four primary factors contribute to the negative perceptions of the police among the public.First, many citizens believe police officers partake in criminal activity (Gates, 2009).In a 2009 INE survey, 18% of respondents said they believed most crimes were committed by police officers.There points to a general lack of confidence and distrust of the police amongst the local residents.Additionally, this survey reported over 40% of local respondents perceives the local and state police agencies very poorly (INE, 2009).
Second, citizens perceive the police force to be highly politicized; the police, at all levels, are perceived as the 'private forces' of the governing authority.Law enforcement is held accountable to those occupying elected office (Winslow, 2010).For example, municipal police are largely responsible for the personal security of the local mayor.Various state and national public officials have also used the police to either suppress of protect protest marches, depending on their personal political partisanship (Birkbeck, 2009).
Third, there is a high pervasiveness of police brutality."A particularly pernicious feature of the Venezuelan police is the extension and intensity of use of physical force.In recent years there has been growth in the number of episodes involving police brutality against citizens.According to PROVEA the number of known acts against the personal integrity of police officers (including torture, degrading punishment, hurt by indiscriminate use physical force or control public disorder, etc.) is doubled between 1999and 2005" (Provea, 2000-2006;;Antillano, 2009, p. 124).There have been numerous reports of kidnappings, human rights violations, and other acts of violence committed by the police.As of July 2009, 1,800 of the nearly 9,000 MP officials were under investigation for various criminal acts-arbitrary arrests, torture, and unlawful detention (U.S. State Department, 2009Department, -2010)).
Finally, for many Venezuelans, direct interaction with police officers is their only personal experience with government officials-or the one that occurs most often.Therefore, it is likely that citizens will equate perceptions of the police with perceptions of the national government.Law enforcement plays a pivotal role in influencing public opinions about the rule of law-and about government more generally (Blake, 2009).
It has been acknowledged repeatedly by National Commission for Police Reform (CONAREPOL), that the police are perceived by the public as "fragmented, dislocated and inconsistent, which has involved irregularities in the provision of services, inconsistent and arbitrary security policies and inadequate police performance in multiple law enforcement agencies" (CONAREPOL Recomendaciones Generales, 2007, p. 3).National victimization surveys also consistently corroborate CONAREPOL reports; citizens feel the police are violent, arrogant, and negligent (INE, 2006 and2009).And in 2008, over 61% of citizens believed there was a high to very high probability of bribing a policeman (see Table 4a, p. 18).Perceived corruption among politicians is remarkably similar (Table 4b).In 2008, 60% of respondents thought there was more corruption among politicians than the previous year5 .

Response Total
There  Each of these factors foster preexisting perceptions of corruption among the police.Since the police are an extension of the national government and also have close, politicized ties with member of government, public perceptions will associate police corruption with political corruption in the national government.

Data and Methodology
The perception of corruption is a subjective concept; therefore it is imperative that several different measures are used to ensure the most accurate analysis.The primary source of data for this analysis is Latinobarómetro survey data, which includes: confidence in the police and government as well as other relevant control variables (complete list of questions in Appendix).
Direct questions regarding corruption in public institutions are scarce.The Latinobarómetro survey does ask a battery of questions on corruption, but most the majority of these questions are broad and were also asked in select years (making comparisons between police and government difficult).Instead, I use "confidence in government/police" to measure corruption perceptions in the two public institutions of interest.As discussed in the previous sections, confidence and trust are interrelated concepts and can be used as indicators for the public perception of corruption in government institutions.Respondents with higher levels of confidence in law enforcement should perceive low levels of corruption and vice versa.
Survey data has been one of the primary analytical tools used to quantify the immeasurable concept of corruption.Corruption indices from these institutions use surveys responses from members of the business community, local residents, and other analysts to provide insight into this illusive field (Transparency International, 2010;World Values Survey 1996, 2000).The CPI aggregates data from 13 sources by 10 independent institutions (Transparency International, 2010, p. 1)6 .All sources measure the overall extent of corruption (frequency and/or size of bribes) in the public and political sectors, and include an assessment of multiple countries.Latinobarómetro is a non-profit, private corporation that conducts national-level surveys to gauge public opinion on issue areas such as democracy, economic conditions, and corruption.To obtain a representative sample of the entire population, intensive interviews of 1,200 citizens are conducted each year of the study (Latinobarómetro, 2009b, p. 26).
A number of limitations arise when using the CPI or other public opinion surveys.Corruption indices are based almost entirely on elite perceptions of corruption in a country: business leaders, country experts, and other scholars."They are asked to assess the level of corruption in national governments familiar to them, daring on their own undisclosed experience, what they are told by others whom they know in policy-making circles, and on the government's reputation" (Rose and Mishler, 2010, p. 4).CPI analysts are asked to evaluate the extent of corruption and to assess countries based on a predetermined definition of corruption.Determining the 'extent of corruption' is a highly subjective task; views on how extensive is in a country is likely to vary depending on the personal views and backgrounds of the different analysts.Additionally, the definition of corruption used by TI focuses solely on quid pro quo transactions-bribery, embezzlement, misuse of public funds-thus, other forms of corruption such as nepotism or clientelism, is not explicitly evaluated (Transparency International, 2010b, p. 1-5).This can lead to biased evaluations-and thus a biased score-for a country, one that does not accurately reflect the true real or perceived levels of corruption.Despite these methodological shortcomings, organizations like TI and Latinobarómetro provide valuable insight into the elusive realm of corruption.
Lastly, to evaluate the hypothesis, I analyze Latinobarómetro data (2002-2010) using OLS and multivariate regression techniques.The dependent variable used in this study is confidence in government.The independent variable of interest is confidence in police.I include several other variables that could influence an individual's confidence in the government-evaluation of Hugo Chavez, interpersonal trust, crime victimization, and actual experience with corruption.Gender, education, age and occupation are four individual-level variables typically included in social science research, as they have some of the most significant effects individual perceptions (Mocan, 2004;Rose and Mishler, 2010).

Discussion
In accordance with the strong correlations visible in a simple cross-tabulation of Latinobarómetro data, the results of the OLS regression analyses for 2002-2010 indicate a strong, significant relationship between the perception of corruption in law enforcement and perception of corruption in government.The economic indicator, current economic evaluation of the country, was the only other factor that also maintained a significant relationship with the perception of corruption in government over time.
A strong, significant relationship between perceived corruption in law enforcement and government are evident in the aforementioned analyses (Tables 5a and 5b, display regression results from 2002-2010).For all but three of the years in the analysis, there was a strong (p= .000***)significant and positive relationship between corruption perceptions in law enforcement and government, controlling for all other variables; as confidence in law enforcement decreased, confidence in the government decreased as well (perception of corruption increased).Below are two regression equations to illustrate the effects of all of the independent variables.

Selected Regression Equations: 2002 2005
If all other variables are held constant, then for every one unit increase in lack of confidence in police, we can expect a .304unit increase in the lack of confidence in government (2002 data); the more citizens perceive police to be corrupt, the more likely they are to perceive the government as corrupt.
The adjusted r-square ranged from 0.153-0.634(2002)(2003)(2004)(2005)(2006)(2007)(2008)(2009)(2010); the variables included in the study explained an increasing portion on the variance as the years progressed.I attribute this to the growing number of police corruption scandals and also the government's relative lack of action to curb police corruption.These factors augmented citizen's lack of confidence in police and therefore the government.Demographic variables had either weak and statistically significant or minimal (and insignificant) effects on perceptions of corruption7 .The economic variable included-current economic situation of the country-maintained a strong, significant effect on the perception of corruption for all years in the analysis (p=.000***).This is to be expected.Indicated in myriad political science studies on the determinants of corruption, economic factors influence an individual's corruption perceptions.
Other variables that were included in the analysis, victim of a crime, interpersonal trust and personal evaluations of Hugo Chavez, had mixed effects on the dependent variable.Surprisingly, if respondents knew of a corrupt act, they were less likely to perceive the government as corrupt, though the strength of this relationship was not statistically significant (except 2002, 2009, 2010, see Tables 5a  and 5b for exact significance values).Crime victimization had mixed effects on the perception of corruption in government, with both positive and negative relationships recorded, all of which were statistically insignificant.The relationship between an individual's level of interpersonal trust and perceptions of corruption were also statistically insignificant.The results of the final independent variable, rating of Hugo Chavez, indicated a negative relationship that was statistically significant in 2008-2010; more negative evaluations of Hugo Chavez corresponded with higher levels of perceived government corruption.
This was the case for all years of the analysis except 2005, 2008, and 2009.In these years, there was a strong, negative relationship between perception of corruption in law enforcement and government.I attribute this negative relationship to the enactment of anti-corruption measures and high-profile prosecutions of corrupt public officials.In 2005, the government passed a multitude of anti-corruption reforms, including the foundation of the National Justice System Commission, to help alleviate some of the issues with bribery and other corrupt acts in the judicial system (Suggett, 2010).In 2007, "the former president of two banks, on the charge of stealing $27 million from the state's foreign currency administrative agency, CADIVI, through a false import contract" (Suggett, 2010).Lastly, in 2009, President Chavez founded the new Bolivarian National Police (PNB) to help alleviate rampant corruption among the country's law enforcement system (Birkbeck, 2009).
This surge in anti-corruption measures throughout the latter half of the 2000s, could account for the negative relationship that is evident in some of the years of analysis.However, as previously stated, each year of analysis reported a strong, significant relationship between confidence in police and government, holding all other variables constant.In sum, sufficient empirical support is obtained in favor of the hypothesis; the perception of corruption in government is strongly influenced by the perception of corruption in law enforcement.The effect of political corruption on the governability of a state depends in large part on what the public perceives to be corrupt.The police are an important, and highly visible, branch of the national government.For this reason, citizen's evaluations of law enforcement significantly influence their evaluations of the government.Bayley stresses this correlation in his 1985 study, stating, "a government is recognized as being authoritarian if its police are repressive, democratic if its police are restrained.It is not an accident that dictatorial regimes are referred to as police states" (p.129).
Determining shades of corruption is important for the legitimacy and credibility of government institutions.If citizens no longer believe that certain practices are tolerable, it is imperative that government recognize this and make the necessary legal revisions or risk widespread public opposition or worse.Adverse perceptions of the police force have negative reverberations on the government, as the police as a direct extension of the national government and of national leadership.
Despite rigorous statistical analysis, this study is not without its limitations.Other factors not included in this study-such as crime rates-could potentially provide a better understanding of the relationship between corruption perceptions in law enforcement and government.Finally, further testing using a more complete dataset with responses from direct questions on corruption could help improve the reliability and validity of results.
In the case study, additional testing could examine perceptions of corruption in countries that present a similar case-but lack intensive scandalization in the media-to determine if the effects of anti-corruption efforts on perceived corruption are upheld.Also, the inclusion of other variables-for example, quality of institutions or religious traditions-could potentially provide a more complete explanation for the determinants of perceptions of corruption.Lastly, the inclusion of superior indicators for perceptions of anti-corruption efforts could improve the validity and reliability of these conclusions.

Conclusion
"If man defines a situation as real it is real in its consequences" (Moroff, 2004, p. 71).Police corruption in Venezuela may occur to a much less extent than is believed by much of the population.However, the sheer ratio of police to citizens, the fragmented, decentralized law enforcement system, and the actual incidences of police brutality have augmented Venezuelans' belief that the police are corruption.This has correlated with the perception of corruption in the national government as well.In this way, the public defines the situation of police corruption as real.As an extension of the state, this has translated into perceptions of corruption among the national government as well.The fundamental implication for such a negative public perception of police corruption is its effect on the government leadership; potentially resulting in protests, riots, or even coup attempts.The national government must be responsive to public perceptions of corruption in law enforcement, even if reality does not necessarily mirror perceived practices.

Recommendations for Future Research
In this study, I analyzed a single aspect of a much larger issue, perceptions of corruption.There are several prospects for future research that would provide greater support for this study and other similar work.First, I suggest expanding the scope of the empirical analysis to other countries in Latin America.If cross-national evidence of this relationship is found, this could have important implications for how governments choose to go about eradicating corruption (or at least the perception of corruption) from their respective countries.A media content analysis to monitor the frequency of reports regarding law enforcement (both positive and negative reports) could provide greater insight into the public's perception of police corruption.

Appendix Table 6. Historical Data on Perceptions of Corruption in Venezuela (1980-2007).
Source: Internet Center for Corruption Research.
*Historical data provides the assessment of countries in retrospective.Conclusions concerning time trends should be based on the comparison between the 1996 score and the historical data.Comparisons with the 1995 ranking may be less precise.Index Type 20072006200520042003200220012000199919981997199619951988-19921980-1985 Venezuela