Anne Lugon-Moulin, Mismatches between Corruption Perception, Corruption Victimisation, and Anti-Corruption Measures in:

Diana Schmidt-Pfister, Sebastian Wolf (Ed.)

International Anti-Corruption Regimes in Europe, page 123 - 137

Between Corruption, Integration, and Culture

1. Edition 2010, ISBN print: 978-3-8329-5846-6, ISBN online: 978-3-8452-2573-9,

Series: Schriftenreihe des Arbeitskreises Europäische Integration e.V., vol. 70

Bibliographic information
Part III Culture, Perceptions, and Experiences 125 Mismatches between Corruption Perception, Corruption Victimisation and Anti-Corruption Measures Anne Lugon-Moulin 1. Introduction The various dimensions of governance are not easy to understand, capture and measure. Corruption being considered as one of major impediment to good governance or as a symptom of bad governance, it has attracted a lot of attention, at a policy, at a practical and at an academic level. Practitioners and academics connect together when it comes to measuring corruption. In fact, donor agencies and governments need to know the extent and the magnitude of the problem, in order to better adjust their strategies, and are therefore in need of good measurement techniques of the phenomenon, while academics try to better understand the corruption phenomenon also by measuring it more accurately. However, leading to hidden transactions by nature, corruption is not easy to measure and quantify. The perception-based data are well known and widely used; they take perception of corruption, reflecting subjective opinions, as a proxy for victimisation.1 On the contrary, victimisation surveys are less frequent, but provide very interesting insights into the corruption phenomenon. For the majority of them, they demonstrate that perception may not correlate so much with victimisation. This chapter intends to critically review some of the corruption measurement attempts. Starting with a review of studies having operated with victimisation indicators, this chapter will try to analyse and interpret their results. In doing so, we intend to set out several critiques opposed to victimisation-based measurements results and will discuss them. So far, the lack of correlation between perception and victimisation has been researched and acknowledged in several studies, but a detailed practical discussion, from an empirical point of view, on reasons why such discrepancy exists, has been lagging behind. Hence, this chapter will provide new underlying factors explaining the gap between perception and victimisation, in light of what anticorruption projects might have had as consequences so far and how they have impacted on both perception and victimisation. Since more than a decade, anti-corruption actions have been primarily geared to perception results. New explanatory elements regarding the victimisation side of the problem might trigger innovative conclusions on policy recommendations for both 1 Victimisation means having to pay a bribe, with or without the possibility to refuse to do so. 126 governments and donor agencies. This will be covered in the last part, which will be devoted to presenting practical measures. 2. Evidence Provided by Victimisation-Based Surveys The perception-based measures involve collection of information of perceptions of the existence and magnitude of corruption. Recently, the perceptions based approach has been carried out at quite a large scale, generating cross-country and time-series data sets (Banerjee et al. 2009). To mention the most famous ones: Transparency International’s Corruption Perceptions Index (CPI), the World Bank Country Policy and Institutional Assessments (CPIA) Indicators, the Economist Intelligence Unit datasets on corruption risk factors. The CPI, which is the most popular measure, has fundamentally helped to bring the issue of corruption on the international policy agenda. Tools developed recently by Transparency International to complement the CPI are the Bribe Payers’ Index, which assesses the supply side of corruption and ranks corruption by source country and industry, and the Global Corruption Barometer, which is essentially a public opinion survey that assesses the general public’s perception, but also real experience (victimisation) of corruption in specific sectors. The International Crime Victim Surveys (ICVS) data is one of the main instruments to measure corruption victimisation. This method focuses on individuals rather than firms. The survey looks for comparable data on crime across countries on the basis of a combination of computer-assisted telephone interviewing techniques in developed countries and face-to-face surveys in developing countries. According to Svensson (2005), in most developing countries the survey data refer to the experience of urban households, since the surveys are only implemented in the capital (or largest) cities. The largest share of the questions relate to crime issues. On corruption specifically, respondents are asked if government officials required, or expected the respondent, to pay bribes for their service during the previous year. The ICVS is conducted worldwide under the supervision of the United Nations Interregional Crime and Justice Institute (UNICRI) since 1987. In 1996, questions regarding corruption were introduced for the first time. There are not many research papers dealing with victimisation measurement. Most of them are circumscribed to one specific sector or service in a given country and would therefore fall under micro or sectoral level analysis, as opposed to macro-level analysis (Duncan 2006: 143). We would like to point out some of those attempts. Svensson (2003) and Clarke and Colin (2004) used firm-level data to study the incidence (how often) and the level (how much) of bribes paid by private firms in connection with regulations, licenses, public utilities, etc., while Gorodnichenko and Sabirianova (2006) assessed the total monetary compensation received by dishonest officials to evaluate the extent of corruption in Ukraine (both sources cited in Méndez and Sepúlveda 2007). Hunt (2007), cited in Banerjee et al. (2009), utilised the International Crime Victims Surveys and Peruvian Household surveys, both of which contain information on bribes to public officials in key sectors. Other studies 127 cited in Banerjee et al. (2009) include study reports on facilitation payments paid for services that should be free at government health centers (India), the process of obtaining a driving license, extra-level payments and the rules broken in India (Bertrand et al. 2006), and ‘illegal’ payments by truck drivers on regular routes to the traffic police, military officers, and attendants at weigh stations (Barron and Olken 2007). Bonvin (2008) and Bonvin/Lugon-Moulin (2008) used the 2000 ICVS (International Crime Victimisation Survey) data, covering 45’000 respondents in 25 developing and transition countries.2 Seligson (2006) used the Vanderbilt University ‘Latin American Public Opinion Project’ (LAPOP) database which had been utilised by the World Bank as of 1998 (Kaufmann 2008). This dataset covers 6 Latin American countries.3 This paper will focus on the results of these three studies for a variety of reasons: first, the author was very closely associated to the study done by Bonvin (2008); second, these studies are based on datasets which are internationally recognised and used by international organisations; third, they provide cross-country comparisons, as opposed to the above-mentioned list of sectoral and national studies; fourth, both of them include questions about victimisation and perception of corruption, so as to allow correlation analysis between the two dimensions. The results of both studies are striking across countries. In the Bonvin (2008) study, 13.8% of respondents have been victims of corruption in the year preceding the survey, while on the side of perception, 82.3% think that one public sector or more would ask for a bribe in the future. However, diverse patterns can be observed and four groups of countries can be made: high perception /high victimisation; high perception /low victimisation; low perception /low victimisation; and, low perception /high victimisation. In a group of countries, mainly composed of former Soviet Republics and Eastern Europe countries, perception is much higher than victimisation. This result is probably due to a sudden rise of the anti-corruption debate around the accession to the European Union (EU), which has made citizens very much aware of the corruption problem, probably to an extent that they consider their society as entirely corrupt. This debate has been substantially fuelled by the general democracy and good governance promotion in this region since the early 1990s. In the opposite group, namely high victimisation/low perception, the discrepancy could be explained by the fact that respondents of these countries (e.g. Belarus) do not dare to open their voice regarding their political environment, due to strict political control and lack of access to information. In the LAPOP countries (Seligson 2006), the percentage goes in the same order of magnitude: it ranges from 5 points on the scale of victimisation for El Salvador compared to 71 points on the perception side. Bolivia seems to be the country where the closest fit can be observed: 27 against 64 points. Seligson (2006: 389) even finds that 2 Albania, Argentina, Azerbaijan, Belarus, Bulgaria, Cambodia, Croatia, Czech Republic, Estonia, Georgia, Hungary, Latvia, Lithuania, Mongolia, Mozambique, Panama, the Philippines, Poland, South Korea, Romania, Russia, Slovenia, Uganda, Ukraine. 3 Bolivia, Ecuador, El Salvador, Honduras, Nicaragua, Paraguay. 128 ‘those who had experienced the highest frequency of corruption have a lower degree of corruption perception than those who had been victimised only one time’ with respect to El Salvador. At sector level, it is also difficult to identify a correlation between perception and victimisation. The Parliament sector is probably the one with the lowest correlation: in the ICVS analysis, it generates no cases at all, but is regarded as highly corrupt (55%). This can be partially explained by the fact that very few of the respondents have contacts with Parliamentarians and hence, would face corruption facts, but they hear a lot about their politicians being corrupt in the news and media reports. When data are disaggregated, it would be expected that higher social groups would have more contacts with Parliamentarians and hence, would score higher on the victimisation scale. This, however, is not the case: this group of respondents doesn’t behave differently from the entire group. On the other side of the spectrum, the police sector has the strongest correlation between perception and victimisation, probably due to the frequent occurrences of interactions between respondents and policemen. An underlying factors analysis allows a differentiated approach as to how much victimisation is linked to perception. Here, social variables come into play. Both studies come to the conclusion that higher occupational profile/income strata means increased probability of being a corruption victim (Seligson 2006: 394) and a higher perception level of corruption (Bonvin 2008: 31). The same goes with education: the more educated the people are, the more victimised they are and the higher corruption they perceive. This is consistent with the fact that the education level is strongly correlated with the occupational profile. With regard to age, the youngest respondents are less victims of corruption (Seligson) and perceive it less (Bonvin). When crossed-fertilised with criminality data, it appears that crime fearful respondents tend to be less victimised by corruption but perceive it more (Bonvin 2008: 31). Finally, according to Seligson (2006: 397), people living in urban areas are more exposed to corruption than others. The number of government entities in cities being higher than in remote areas, interactions between citizens and government are more numerous in urban areas. A few other available papers demonstrate the same gap between perception and victimisation at national level (Razafindrakoto/Roubaud 2006; Johnson, Kaufman et al. 1998; Abramo 2005). Rose and Mishler (2009) found that contact is most important for paying bribes while political awareness is most important for the perception of corruption in Russia. Such a big convergence of results across studies calls for further reflection and analysis. The gap between victimisation and perception of corruption seems to be by far more important than one could have expected. If it holds true, a breakthrough in the theory of corruption and practice of anti-corruption efforts will have to be acknowledged; taking perception as a proxy for victimisation may not be entirely correct. The following section will discuss a series of arguments which can explain this gap, or, alternatively, challenge the veracity of the discrepancy’s finding by criticising the nature of the data covered by victimisation-based measurements. 129 3. Dialectic Discussion about Critiques of the Victimisation-Based Measurements and the Gap between Perception and Victimisation The initial reaction to acknowledging this wide gap is two-pronged. Both sides of the coin shall be carefully looked at: why would perception be so high and why would victimisation be so low? On the first one, one could argue that fifteen years of anti-corruption battle, mainly focused on sensitisation campaigns and awareness-raising efforts, have reached their maximum effect: people in all countries around the world, and coming from all social strata are aware of the corruption problem, and, hence, tend to automatically perceive it to a rather high degree, regardless of them being victims or not of the phenomenon. This explanation fundamentally contradicts the belief that poor people know insufficiently about corruption and are still in need of more sensitisation campaigns, an argument which can be heard sometimes in the international development aid community. Such an argument, however, is supported by the idea that, if people do not know about corruption, they don’t report on it to the official (police, court) or unofficial channels (civil society organisation, legal help clinics, etc). This, in turn, could explain why the victimisation rate is so law: it’s not because corruption does not happen, but because people do not report on it by ignorance. One part of this argument is certainly correct, namely that people insufficiently report on corruption. However, this is most probably because of lack of trust in the authorities, lack of reliable unofficial channels, fear of reprisal, lack of available official channels, and not because of ignorance. Empirical studies demonstrate however that the ignorance factor is less and less valid. For example, Blundo and de Sardan (2002) provide a very good overview of the corruption facts and knowledge in some West African countries remote areas. But reporting cases to official channels for the sake of prosecution, or reporting to independent organisations collecting information anonymously for conducting surveys, is two different issues. Under that condition, it would be difficult to argue that victimisation is low because of lack of reporting. As a consequence, only the inflated perception effect and, to a minor extent, the ignorance remain as explanations. An optimistic view on the success of anti-corruption efforts to date may offer another pertaining element to why victimisation is so low. If the fight against corruption shall indeed succeed, this would eventually result in a decrease in cases and victims, thereby translating into a lower victimisation rate. The previously explained caveat (lack of reporting mechanisms), the taboo culture surrounding corruption and the lack of sufficient time-series data on corruption victimisation, which would allow real comparisons over a time period including years preceding the start of the anticorruption battle, hamper us to adhering to this idea. Contrary to this view, low victimisation rates could be explained by the lack of law enforcement. If law enforcement with regard to corruption cases is deficient in a country, people will be reluctant to report on these, seeing no positive consequence or even negative ones in disclosing information. This behaviour is even reinforced 130 by the fact that perception of corruption of both the police and the justice sectors is always higher than the reality (Bonvin 2008). Inefficiency and corruption of law enforcement authorities themselves reinforce this vicious circle. Despite the innovative insights provided by the victimisation-based databases, one has to be careful as to which types of corruption are actually addressed here. While perception-based measurements will obtain from respondents answers related to petty corruption as well as grand corruption and political capture (big corruption scandals memories are reflected in those answers), it is likely that victimisationbased measurements answers will be circumscribed to petty corruption only. Very few respondents would have directly faced or been involved in political capture and grand corruption cases. When asked about corruption in general (as it is the case in perception-based measures), people associate corruption to other dubious acts, such as organised crime, clandestine economy, mafia, money-laundering, etc., which are all hidden by nature and larger in scope than corruption per se. As a consequence, this may result in overestimated levels of perceived corruption. This limitation is proven by the sector-level breakdown analysis of Bonvin (2008). If the Parliament collects so little victimisation records, it is not because the Parliament itself is not corrupt, but because the majority of interviewed citizens have probably never had the opportunity to enter into a corrupt deal with Parliamentarians, which would mainly occur at grand corruption or political capture level. Additionally, referring to the above-mentioned argument, reporting on perception of Parliamentarians corruption’s level allows to reflect on all kinds of feelings related to nepotism, trading in influence and grand corruption, misbehaviours of which people have heard, but to which they have actually hardly been exposed. But do petty corruption-related conclusions differ so much from grand corruptionrelated results? In societies which are corrupt at the top, it is very likely that the bottom part of the executive and legislative levels will be of the same nature. Leading by example, rulers of such countries will most probably tolerate petty corruption at lower levels or even encourage it, so as to diminish the probability of being challenged by their own administration and citizens constituencies. Therefore, we could almost derive the conclusion that petty corruption measurement results can be proxies of grand corruption ones, so that potential discrepancies between petty corruption measurement results and grand corruption measurement ones are not so meaningful anymore. The opposite does not hold true: the likelihood that a country would be corrupt at its base (petty corruption), while not being so at higher levels (grand corruption and political capture) is very low. Indeed, how could we explain that rulers of such a country would be clean, but would tolerate petty corruption within their society? There would reasonably be no political or economic incentive for not tackling the issue. Therefore, we cannot say that grand corruption is a proxy for petty corruption. Only petty corruption can be a proxy of grand corruption. Such a conclusion would overrule the argument regarding the different nature of corruption measured by perception and victimisation-based surveys. Another counter-argument against victimisation-based measurements lies in the 131 nature of the respondents. As the victimisation databases contains questions related to payment of bribes for the use of public services, results could potentially be biased by the fact that users of public services only would have to pay extra; in many developing and transition countries, public services tend to be used by poor segments of the population, while private providers are visited by wealthier people. Therefore, one could infer that victimisation-based results do not reflect the entire composition of the society as rich people would rather respond negatively to questions related to paying bribes, because they are not directly confronted to public services officials. This assumption is challenged by Bonvin (2008: 32) who found that in the higher education and income quartile of the ICVS respondents, 25% of the respondents experienced corruption over the last twelve months, compared to only 19% in the overall sample of respondents. Lastly, Seligson (2006: 401) came to a strange but very interesting conclusion on how the functionality of corruption is positively associated with victimisation. In other words, those who have direct experience with corruption are more likely to believe that it gets them what they need. This could be part of the answer, when trying to analyse why perception could be higher than actual victimisation. If people need corruption to ease their way into the system, they are less likely to feel victims of it, but would necessarily consider corruption as a perversity, albeit necessary component, of the system.4 We have reviewed four counter-arguments as to why victimisation-based measurements may not be valid and why the inferring gap between perception and victimisation may not hold true. However, none of them proved to be solid enough. What implications does this have on policy actions? 4. Policy Recommendations From a policy point of view, what matters is to see how the existing gap between perception and victimisation can potentially impact on the population’s expectations and the rule of law in a given country. If perception is higher than victimisation, the country will run two major risks: a reputational one and a social unrest problem. The first one will translate into investors and donors being more likely to leave the country because of its bad corruption stance, and the second problem will be that citizens may become more and more suspicious towards the ruling party and eventually dismiss it. It shall be noted that those risks exist even if perception is not necessarily higher than victimisation, but increase as the gap widens. As a consequence, the government of such country should quickly move away from policies which typically reinforce perception (i.e. awareness-raising campaigns) and increase efforts on the victimisation side, by, for example, increasing the number 4 It is important to note that such opinions are found to be uncorrelated with a belief in the legitimacy of the political system. 132 of contacts points where to report corruption cases, improving the judiciary, with a special focus on law enforcement. An increase in perception, which does not reflect an increase in victimisation, can potentially hamper the rule of law; this is all the more valid, as we know that a high corruption perception will likely have a negative effect on reporting mechanisms (Bonvin 2008: 35) which, in turn, will mean dysfunctional police and justice. One can easily understand why perception-based policies were carried out widely at the expense of victimisation-based policies, by both governments and donor agencies. From a political point of view, it is easier and less risky to launch anti-corruption sensitisation campaigns and contests, instead of trying to reinforce the justice sector by enabling it to undertake effective inquires, run fair trials and enforce decisions, which sometimes target influential political persons or businessmen. From a financial point of view, perception-enhancing policies are also cheaper than, for example, an entire restructuring process of the judiciary, which typically entails capacity-building, human resources, technical and organisational aspects. This conclusion doesn’t come as a surprise. Indeed, it is backed up by recent developments in the anti-corruption community. A few years ago, Transparency International went through an internal turmoil which resulted in two opposing teams: one very much in favour of more awareness-raising actions, and the other one calling for more to be done on the side of law enforcement. Observers of development aid organisations active in anti-corruption would also notice a certain fatigue among donors. After fifteen years of awareness-building efforts by both government entities and the civil society, everybody is now waiting for more concrete results. This is probably why initiatives enabling law enforcement authorities of developing countries to acquire financial investigation skills techniques and asset tracing mechanisms have recently taken shape: those projects directly target criminalisation of corruption cases affecting developing and transition countries. 5 The support to the justice sector has always been, to a rather limited extent, an area of intervention for donors – multilateral agencies in particular (European Union, United Nations Office on Drugs and Crime), but the recent focus on corruption investigations is totally new. However, those new initiatives are limited in scope and size. It is therefore critically important, now, to increase anti-corruption programmes with regard to law enforcement and criminalisation. Unless big fishes are fried for once, no one will eventually continue to believe in the anti-corruption battle, and the gap between perception and victimisation will constantly increase. On the other side of the spectrum, we can interpret the fact that victimisation may be higher than perception in specific countries (Cambodia, Belarus and Colombia) in two ways. On one hand, one could infer that the population is not aware of the extent of the corruption problem it faces. But this hypothesis would contradict the abovementioned argument that overall and worldwide, people tend to know what corruption is and what it is not. If, however, this is not the case, more awareness-raising 5 The International Center for Asset Recovery was launched in 2006; the World Bank StAR (Stolen Asset Recovery) Initiative was launched in 2007. 133 efforts shall be undertaken in such countries. On the other hand, it would probably be more accurate to argue that in contexts where breaches of political and civil rights are common, specifically where freedom of speech is not guaranteed, people tend to remain silent when asked to provide opinions on their political environment. This latter hypothesis calls for more actions on the human rights front. In terms of policymaking, the goal is therefore not only how to tackle corruption, but also how to ensure a more conducive environment to freedom of speech, access to information and the enforcement of civil rights. Besides law enforcement efforts, it would also be of use, both on the sides of governments and on donor agencies’ side, to analyse the magnitude and the different forms of petty corruption in the various sectors of the public administration, with a stronger economic focus. This is valid for both petty and grand corruption. Petty corruption will typically entail facilitation payments for obtaining a service, but also for getting hired, for example. These practices may vary a lot from sector to sector and in terms of geographic coverage (countryside versus cities, etc). Detailed analysis and targeted approaches should be chosen. Grand corruption, however, will not be easy to measure or diagnose. One way to shed more light on grand corruption for donors could be to devote more attention to embezzlements or misallocation of public funds when looking at fiduciary risks assessments within budgetary support.6 Even so, it is striking to note that countries are more and more willing to provide budget support, but hardly dare to apply logical sanctions (delayed disbursement or cuts in budget support) in case a corruption scandal emerges in a country. 5. Concluding Remarks Starting from evidence provided by corruption victimisation-based studies, this paper has reviewed the persisting gap between the perception and the victimisation sides of corruption. A critical discussion has followed, responding to counter-arguments on why victimisation-based surveys may not be valid. It has been found that, from all the critiques exposed, none of them really holds true, so that clear-cut rejection of victimisation-based measurements and neglecting the gap between perception and victimisation would not be justified. Furthermore, such an attitude would lead to inadequate policy choices. Indeed, both the gap itself and the current political reality with regard to corruption show that perception-based policies have reached their maximum effect. It is hardly possible to argue that people don’t know about corruption anymore, and that more awareness-raising campaigns are still needed. However, victimisation-based policies are lagging behind, namely programmes which encourage anti-corruption law enforcement, reporting mechanisms as well as criminalisation and prosecution of corruption cases. Unless strong actions in that sphere are put 6 Budget support refers to the aid modality by which money is being transferred directly to the Treasury of the recipient country. The Paris Declaration, signed in 2005 by many countries, calls for more budget support at the expense of project support. 134 in place, the overall anti-corruption battle will suffer from a lack of credible results and the rule of law in victim countries will not improve. But this is where political willingness will be most needed and we know from the public choice theory that providing the right incentives to politicians across the board will not be an easy task. Those difficulties call for more coherence and more engagement among the international community towards real law enforcement and sanctions in case of misconduct. Bibliography Abramo, C. W. (2005) How Far Go Perceptions?, Transparencia Brasil Working Paper, http:// Banerjee, A./Hanna, R./Mullainathan, S. (2009) Corruption, Blundo, G./de Sardan, J.-P. O. (2002) La corruption au quotidien en Afrique de l’Ouest, approche socio-anthropologique comparative: Bénin, Niger et Sénégal, Marseille. Bonvin, B. (2008) Corruption: Between Perception and Victimisation, in Swiss Agency for Development and Cooperation (ed), Challenging Common Assumptions on Corruption and Democratisation, Bern, 25-39. Bonvin, B./Lugon-Moulin, A. (2008) La Perception Corrompue de la Corruption, in Le Temps, Geneva. Duncan, N. (2006) The Non-Perception Based Measurement of Corruption: A Review of Issues and Methods from a Policy Perspective, in Sampford, C./Schacklock, A./Connors, C./Galtung, F. (eds.) Measuring corruption, Ashgate/Aldershot, 131-160. Johnston, M./Kpundeh, S. J. (2002) The Measurement Problem: A Focus on Governance, Forum on Crime and Society 2 , 33-44. Johnson, S./Kaufman, D. et al. (1998) Corruption, Public Finances and the Unofficial Economy, ECLAC, Santiago. Kaufmann, D./Kraay, A. (2007) On Measuring Governance: Framing Issues for Debate, Issues paper, 0081/2604389-1167941884942/On_Measuring_Governance.pdf . Kaufmann D./Kraay, A. (2003) Governance Indicators: Where are We, Where Should We Be Going?, The World Bank, Policy Research Working Paper 4370. Kaufmann, D./Kraay, A./Mastruzzi, M. (2005) Measuring Governance Using Cross-Country Perceptions Data, GovernancewithPerceptionsData.pdf . Kaufmann, D./Kraay, A./Zoido-Lobatón, P. (2000) Governance Matters: From Measurement to Action, Finance & Development 37, htm. Kaufmann, D. (1998) Corruption Diagnostics: A New Technocratic Framework for the Analysis of Corruption and its Implications for the Design of Action Programs, Miami. Lugon-Moulin, A. (2004) Lutte Contre la Corruption et Coopération au Développement: Synthèse, Enjeux et Perspectives, in Jaggy, B./Balmelli, T. (eds.) Les Traités Internationaux Contre la Corruption, Bern/Lugano/Lausanne. Lugon-Moulin, A. (2007) La Lutte Contre la Corruption : Les Limites de l’aide Internationale, Finances and the Common Good 28/29, 15-17. Méndez, F./Sepúlveda, F. (2007) What do we talk about when we talk about corruption?, http://co 135 Razafindrakoto, M./Roubaud, F. (2006) Are International Databases on Corruption Reliable? A Comparison of Expert Opinion Surveys and Household Surveys in sub-Saharan Africa, DIAL Document de Travail, DT/2006-17, Paris. Rose, R./Mishler, W. (2010) Experience versus Perception of Corruption: Russia as a Test Case, Global Crime 11, 145-163. Seligson, M. A. (2006) The Measurement and Impact of Corruption Victimisation: Survey Evidence from Latin America, World Development 34, 381-404. Stefes, C. H. (2005) Understanding Systemic Corruption. Some Methodological Questions, http:// . Svensson, J. (2005) Eight Questions about Corruption, Journal of Economic Perspectives 19, 19-42. 137 Culture, Corruption and Anti-Corruption Strategies in Post-Communist Europe Åse Berit Grødeland 1. Introduction Since the collapse of communism corruption has been a bone of contention in postcommunist states in East Central Europe (ECE), South East Europe (SEE) and the West Balkans (WB). In recent years, fairly extensive anti-corruption programmes have been introduced in most of them. Such programmes have drawn heavily on anti-corruption strategies developed by international donors. Their strategies tend to be vague, all-inclusive and fairly non-sensitive to the national environments into which they are introduced. Besides, they are rarely accompanied by proper mechanisms to ensure their implementation (Michael, quoted in Rucinschi 2007: 5). This chapter investigates whether national culture (i) is conducive to corrupt behaviour and (ii) obstructs anti-corruption reform. Our purpose, in other words, is to investigate the root causes of corruption. These tend to be overlooked by ‘conventional’ corruption studies and are rarely addressed by anti-corruption programmes. Yet they are of crucial importance for reform.1 The link between national culture and corruption is investigated by examining elite perceptions of (i) national culture and informal practice; (ii) national culture and law; and (iii) how best to do away with culturally ‘informed’ informal behaviour, in seven post-communist states in ECE (Czech Republic, Slovenia), SEE (Bulgaria, Romania) and WB (Serbia, Macedonia, Bosnia and Herzegovina – below referred to as BiH). 2. Culture and Informal Practice Hofstede and Bond define culture as ‘the collective programming of the mind that distinguishes the members of one category of people from those of another.’2 ‘Culture is composed of certain values, which shape behaviour as well as one’s perception of the world.’ Perceptions of the world shared by citizens of a national state may 1 In order to be truly successful, anti-corruption efforts must target not only the manifestations of corruption but also the factors causing them in the first place. Reducing the impact of the former requires a different time span and also a set of different measures than those traditionally employed. What is more, they are sensitive from both a political and cultural point of view and therefore less ‘responsive’ to externally imported anti-corruption initiatives. 2 National culture, then, distinguishes the citizens of one state from those of another.

Chapter Preview



In dieser aktuellen und interdisziplinären Analyse der internationalen Antikorruptionsregime werden mit Schwerpunkt Europa ausgewählte staatenübergreifende Bemühungen der letzten Jahre zur Eindämmung der Korruption einer kritischen Bestandsaufnahme unterzogen. Die Beiträge stammen aus der Politikwissenschaft, Rechtswissenschaft, Soziologie, Wirtschaftswissenschaft und von PraktikerInnen.

Der Band vereinigt sowohl qualitative als auch quantitative Analysen und berücksichtigt darüber hinaus kulturwissenschaftliche Fragestellungen im Rahmen seiner vier Teile: „The European Dimension“, „Political and Legal Instruments“, „Culture, Perceptions, and Experiences” sowie „Practitioners’ Perspectives”.

Mit Beiträgen von: Tanja A. Börzel, Donald Bowser, Ben Elers, Angelos Giannakopoulos, Åse B. Grødeland, Leslie Holmes, Georg Huber-Grabenwarter, Anja P. Jakobi, Anne Lugon-Moulin, Bryane Michael, Holger Moroff, Yasemin Pamuk, Diana Schmidt-Pfister, Gefion Schuler, Andreas Stahn, Dirk Tänzler, Michael H. Wiehen und Sebastian Wolf.