Content

Leslie Templeman Holmes, International Anti-Corruption Regimes and Corruption Levels in European and Eurasian Post-Communist States in:

Diana Schmidt-Pfister, Sebastian Wolf (Ed.)

International Anti-Corruption Regimes in Europe, page 23 - 47

Between Corruption, Integration, and Culture

1. Edition 2010, ISBN print: 978-3-8329-5846-6, ISBN online: 978-3-8452-2573-9, https://doi.org/10.5771/9783845225739-23

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

Bibliographic information
Part I The European Dimension 25 International Anti-Corruption Regimes and Corruption Levels in European and Eurasian Post-Communist States Leslie Holmes 1. Introduction It is impossible to be certain that what Naim (1995) has called the global ‘corruption eruption’ of the 1990s really occurred and was not largely a function of altered perceptions.1 In the particular case of the post-communist states, altered perceptions may in turn have been a function of the dramatically changed nature of the mass media in many countries; suddenly, journalists were much freer to investigate and report their findings, while the sometimes quasi-anarchic nature of partisan politics often resulted in wild, unsubstantiated allegations of corruption in the media. Nevertheless, few would dispute that the post-communist world really did experience a significant increase in corruption levels. One of the aims of this paper is to attempt to determine whether or not there has been any improvement in recent years – and, in particular, whether or not it appears that international organisations (IOs) are exerting a positive influence in this area. According to the man who was until late-2008 ‘Mr. Anti- Corruption’ (and Mr. Governance) at the World Bank (WB), Daniel Kaufmann, the world has made very little progress in a decade of anti-corruption efforts (cited in Naim 2005: 96), and the approach to the corruption challenge is currently ‘tepid’ (Kaufmann 2009: 28); these pessimistic conclusions will be tested here, to determine whether or not they pertain in the case of the post-communist states of Central and Eastern Europe (CEE) and the Commonwealth of Independent States (CIS). It will be argued that the corruption situation is better in the former than in the latter, and that the influence of the European Union (EU) might well be one of the reasons for this. However, it must be acknowledged at the outset that profound methodological problems arise when attempting to measure both corruption and influence. The former will be considered below. One problem in measuring the latter is that it is usually difficult to isolate factors and to determine how best to weight these. Assuming that trends can be demonstrated – and such a claim is tentatively made here – how do we know that these have been influenced by IOs, rather than other variables? Al- 1 I wish to thank the Australian Research Council for funding that facilitated research for this project (Large Grant No. A79930728 and Discovery Grant No. DP0558453); the Faculty of Arts at the University of Melbourne and the University of Konstanz for providing sufficient funding to make my participation at the conference possible; and Diana Schmidt-Pfister and Sebastian Wolf for invaluable feedback on the original paper. 26 though it will be argued that there is circumstantial evidence that the EU, in particular, has been exerting a positive influence on corruption levels, the evidence can ultimately only be suggestive. A second problem, related to the first, is that it is not possible to unpack the influence of individual IOs in any quantitatively satisfactory way; the evidence can only justify informed and reasoned inferences. Third, many of the sources that must be used in attempting to assess the impact of IOs on corruption levels are (understandably) biased. In general, IOs wish to emphasise their successes more than their failures – until and unless countries they have been attempting to influence positively career so significantly off track that a given IO then wants to blame domestic elites, publicise their failings, and distance itself from them. Many nongovernmental organisations (NGOs), especially domestic ones, also have biases. If they are externally (i.e. by foreign agencies) funded and wish to continue receiving this support, it will often be in their vested interest to emphasise – even exaggerate – the corruption problem in their country. This had led analysts such as András Sajó (1998; 2003) and Ivan Krastev (2004) to criticise the so-called corruption ‘industry’ generated by Western funding. Finally, surveys and interviews reveal that both citizenries and elites are typically divided on what causes the rise and demise of corruption, and on the impact of external agencies. In sum, the most that can be expected from this type of research is informed, research-based inferences. 2. International Anti-Corruption Regimes Most major IOs have by now adopted programmes and policies designed to combat corruption. But the development of these is relatively new, dating essentially from the mid-1990s. Moreover, the early versions for the most part lacked bite; they were merely guidelines or recommendations. And some IOs, notably the World Trade Organization, have been criticised for adopting too weak a stance on corruption anyway (Eigen 2003). For the limited purposes of this paper, the focus will be on just four IOs – the United Nations (UN), the Organisation for Economic Co-operation and Development (OECD), the Council of Europe (CoE) and the EU. It was not until 2003 that the UN adopted its Convention against Corruption (UNCAC), and it is even more recently that member states have signed and ratified it; by January 2008, more than 140 states had signed up, while almost 110 had ratified it. UNCAC officially entered into force in December 2005 – just four years ago. While being hailed as ‘the crowning event in the war on corruption’ (Naim 2005: 96), its very newness means that it is still too early to assess its overall impact; indeed, while gap (compliance) analyses have been conducted in a small number of countries (e.g. Bangladesh, Indonesia), the process of fully reviewing the implementation of UNCAC is not due to commence until 2010 at the earliest. But UNCAC was not the first anti-corruption measure adopted by the UN. The first significant move was the adoption by the General Assembly in 1996 of a Declaration against Corruption and Bribery in International Commercial Transactions. This urged governments to pass legislation that would not only deny tax deductibility to companies 27 paying bribes overseas, but would also make such payments a criminal offence. In 1998, the Assembly strengthened this commitment by adopting a resolution that encouraged governments to criminalise the bribery of state officials in other countries. This was more explicitly directed at corruption – in the traditional sense of the private abuse of public office – than the earlier declaration, since the latter was perceived by some to refer primarily to bribery within the private sector. Various UN agencies have also adopted anti-corruption measures. One of particular relevance to the current analysis is the Programme for Accountability and Transparency (PACT) adopted by the UNDP (UN Development Programme) in 1997 (UNDP 2008: 7). Like the World Bank, the UNDP sees the fight against corruption largely in terms of poverty reduction and the need to promote sustainable development. In 1998, the UNDP published a policy paper, ‘Fighting Corruption to Improve Governance’, and in 2004 an ‘Anti-Corruption Practice Note’; the latter has since been revised to take account of the entry into force of UNCAC (UNDP 2008). Most of these measures focus on what the UNDP can and should do to promote better governance, particularly in countries with high levels of corruption; it sees the promotion of better governance, including greater transparency and accountability, as the key to reducing corruption levels. The UNDP treats CEE and the CIS as an important area in which to fight corruption, and has established a regional office in Bratislava. Until the adoption of UNCAC, many analysts cited the OECD’s anti-corruption measures as the most significant, and the earliest example of an IO addressing this issue. Thus Transparency International (TI) described the OECD’s May 1994 ‘Recommendations on Bribery in International Business Transactions’ as ‘the first multilateral agreement among governments to combat the bribery of foreign officials’ (Transparency International 1996: 15). In the same year as the UN did (1996), OECD member states agreed to end tax regimes that permitted companies to claim relief for bribes paid overseas. In the following year, the OECD adopted the Convention on Combating Bribery of Foreign Public Officials that became effective in early-1999; this has been described as the ‘first international legal instrument addressing the issue [of involving the supply side of corruption, the private sector – LTH]’ (UNDP 2008: 22). A number of post-communist states subsequently signed up to this, including Bulgaria, Czechia and Slovakia; while research indicates that the OECD Convention proved to be far less effective than many had hoped (see our Conclusions), the fact that a number of CEE states subscribed to it suggests that, symbolically at least, they considered it important to demonstrate commitment to such documents. Although the OECD Convention has had limited impact, it must be acknowledged that many states (inc. Germany, the Netherlands and Australia) did remove the taxbreak loophole for companies investing overseas. There is also evidence that some of the OECD’s activities and agencies have been more successful. One of these is the Financial Action Task Force (FATF), established in 1989 primarily to combat money-laundering. During the Yeltsin presidency, Moscow took little notice of the fact that the FATF frequently identified Russia as one of the world’s non-cooperative 28 states in combating money-laundering. But while his overall record on addressing the serious corruption problem in his country is mixed, President Putin did push through anti-money-laundering laws in 2001 – Russia had had no such law until then – in response to FATF criticisms. His country was rewarded not only by being removed from the blacklist of non-cooperative states and territories in 2002, but was then granted full membership of FATF in the following year. Although the impact of FATF blacklisting should not be exaggerated – in all but the most extreme cases, the FATF only encourages international and national financial institutions to be wary in their business and financial transactions with economic entities within a country on the list – it does at least identify states that need to address corruption more seriously. The Russian case constitutes a prime example of an IO using both the stick and the carrot, even if the stick is rather thin and brittle!2 Only very rarely does the FATF go beyond these measures and seek to exert more direct influence. For instance, it has taken more serious action against only one CEE or CIS state (Ukraine) – and that was for less than three months (December 2002 – February 2003), after FATF had identified Ukraine as a major money-laundering centre. Focusing now on purely European IOs, the CoE adopted a set of 20 ‘Guiding Principles’ for combating corruption in 1997, which was followed by a Criminal Law Convention on Corruption in late-1998, and a Civil Law Convention less than a year later. In May 1999, the CoE formally established the Group of States against Corruption (GRECO) to monitor implementation of its various measures designed to combat corruption. Member states of GRECO are required to participate fully in the evaluation process and to agree to be evaluated themselves. The Group originally comprised just 17 CoE states (6 of them post-communist), but this number had doubled to 34 by 2002 (including 10 more post-communist states) and had increased to 46 countries by 2008 (including six additional post-communist states; Belarus is the only European post-communist state not to have joined). An evaluation of compliance with the 20 Guiding Principles was conducted over a three-year period at the beginning of the 2000s, and was followed by production of the actual reports in 2003-2005; a number of states initially refused to allow publication of their reports, but all have since permitted this. The second evaluation round ran 2003-2006, while the third began in 2007 and is due to be completed in December 2009. The first two evaluation rounds produced a number of general proposals for improving compliance with the CoE’s principles, from which it is clear that much still needed to be done by signatory states in areas such as legislation, whistleblower protection, conflict of interest, and mechanisms for dealing with money-laundering (for an overview see Dunga 2008). Nevertheless, the fact that even countries that have been criticised in 2 While space limitations preclude explicit consideration here of the World Bank’s role in combating corruption, note that the World Bank also sometimes uses the carrot. For instance, both Latvia and Romania have been praised for particular anti-corruption measures. However, the World Bank praise for Romania in 2002 was implicitly challenged by the EU in its mid-2008 assessment of the corruption situation in Romania, and had been for some years before Romania’s accession (Pridham 2007). Thus IOs sometimes disagree in their assessment of a particular corruption situation. 29 compliance reports remain in GRECO suggests some success for this approach, at least in terms of securing formal commitment to anti-corruption measures and forcing elites to address sensitive issues. The CoE, with funding from the EU, has also produced targeted programmes for the post-communist states of CEE and the CIS. One set is the Octopus Programmes, initiated by the Council in 1996; following what the CoE perceived as the success (but not completion) of Octopus 1 (1996-1998), Octopus II was formally launched in February 1999 and finished at the end of 2000. These programmes were targeted at corruption – as well as organised crime – in 16 CEE and CIS states. Much of the funding was for training seminars for police officers, judges, public servants and other state officials involved in the fight against corruption. Assessing the success of this is more appropriately considered later in this paper. All that will be noted at this juncture is that, while the programme was welcome, the fact that the relatively small sum of 2.45 mn. Euros was devoted to it, to be spread among 16 countries over two years, suggests that it was not as high a priority as the fanfare announcing it might indicate.3 More recently, the CoE (often in collaboration with the EU or individual West European states) has produced even more targeted – country-specific – anticorruption programmes, such as PACO in Serbia and other SEE states (2004-2006), Rucola I and II in Russia (2005-2007), MOLICO in Moldova (2006-2009) and GEPAC in Georgia (2008) (see Schmidt-Pfister 2010). While it is too early to assess the long-term impact of such projects, the soft data currently available (see next section) do not in general indicate any noteworthy early successes. In addition to funding some of the CoE projects, the EU has since 1995 introduced various anti-corruption initiatives in its own right. One of the most significant was the release of Agenda 2000 in July 1997. This was designed inter alia to prepare the EU for further enlargement. One ramification of Agenda 2000 was the publication of individual Commission Opinions on each of the applicant states. Each country Opinion contained a summary of tasks that the applicant state was to fulfil. Only one factor was contained in the political summary of every post-communist applicant state – the need to bring corruption under control. The EU had thus stipulated that evidence of attempts to address the corruption issue was a precondition of membership. It is sometimes claimed that the EU loses its leverage over erstwhile applicant states once they become member states (e.g. Mungiu-Pippidi 2007). While there may in the past have been little persuasive evidence to counter such claims vis-à-vis corruption, there has been proof since the EU’s two newest members joined in 2007 of the EU’s determination to deal firmly now with members that are perceived to be doing too little to combat corruption. Thus, following an investigation by the Office de Lutte Anti-Fraude (OLAF), the EU produced a report apparently described by some EU officials as ‘the most scathing ever written about a member state’ (Deutsche Welle 2008). And the EU did more than simply criticise; it first froze at least €500 mn. (figure cited in Frankfurter Allgemeine 2008; it was €825 mn. ac- 3 The Octopus Programme continues on a much smaller scale, mainly in the form of conferences (e.g. regarding cyber crime in 2008), though little is publicly reported about it. 30 cording to Deutsche Welle 2008) in July 2008,4 then announced in November that Bulgaria would be permanently deprived of €220 mn. because of mismanagement and corruption relating to EU funding granted in the pre-accession period (Spiegel Online 2008). This action was described by the Financial Times (17 December 2008) as the ‘first time the European Commission has stripped a member state of funds in this way’. In sum, IOs have adopted a number of measures over the past 15 years designed to reduce corruption. While many of these programmes have been global, some have specifically targeted CEE and the CIS. Circumstantial evidence on the impact of these various measures can now be considered. 3. Assessing the Level of Corruption in CEE and CIS States Anyone researching corruption knows that it is impossible to measure it with any high degree of precision or confidence. Official statistics – from those countries that produce any at all – usually cite some or all of the following numbers: reported or suspected cases of corruption; investigations; prosecutions; convictions. Typically, the figure for each of these is lower than for the preceding variable – and the number of convictions in most states is very low indeed. This in part reflects the difficulties of investigating and proving corruption. It also means that official statistics are not very useful in assessing levels of corruption, other than as a starting point. Researchers have therefore turned to other methods for estimating corruption levels, notably perceptual and experiential surveys (see too June 2008: esp. 8-14). The perceptions-based TI Corruption Perceptions Index (CPI) has many faults and potential pitfalls (see e.g. Galtung 2006) and must be treated warily. Nevertheless, it remains the most frequently cited source of data on corruption in most countries of the world. It is also still the single best source for cross-polity and longitudinal data, even if TI itself warns against using its scores – and even more so its rankings – for such purposes. If scores are treated with caution (not fetishised) and as merely indicative, rather than definitive, the CPI can still be a useful starting point for analysing corruption levels across polities and years. For all the criticisms, perception is a form of reality, and influences political and investment decisions. And, unfortunately, there is little else available! Unless all attempts at measurement and compari- 4 The official EU press release on this in July 2008 acknowledged that the Commission ‘has provisionally suspended EU funding because of irregularities found through our control and auditing system’, but did not provide any Euro sums (http://europa.eu/rapid/pressReleasesAction.do?reference=IP/08/1195&format=HTML&aged=0&language=EN&guiLanguage=en). Nor were these included in the February 2009 Commission report on the Bulgarian corruption and organised crime situation, though it did make it clear that the Bulgarian authorities’ treatment of both phenomena appeared to have deteriorated in the latter half of 2008, since there were fewer convictions than in earlier periods (Commission of the European Communities 2009b: esp. 5-7). The February 2009 report included six benchmarks (conditionalities) the Bulgarian authorities needed to address in 2009. 31 son are to be abandoned – which plays into the hands of corrupt elites – cautious steps can be taken using the CPI. TI only began to produce CPIs in 1995, when it assessed only one postcommunist state (Hungary). Hence, it is not possible to test Naim’s ‘corruption eruption’ claim for the period preceding the years in which it was published, other than by measuring media coverage (which does reveal far more reporting, but still does not prove a substantial increase in corruption). However, we can examine the apparent trend lines since the mid-1990s, and even more so in the 2000s, since the methodology has gradually become more standardised and the number of countries assessed has increased. Since the primary interest here is in attempting to evaluate the impact of external forces on corruption levels, the data can be analysed to identify patterns (if any) – even if it is important not to make unsubstantiated causal inferences. Given that the EU appears by many criteria to have been able to exert the most leverage over many post-communist states – certainly over those that have applied for EU membership – a suggestive way of presenting the data is to separate successful applicant from non-applicant states. This produces the following tables (in which the lower the figure, the higher the perceived level of corruption): Table 1: TI CPI scores of post-communist EU member states 1996 1998 2000 2002 2004 2006 2007 2008 Trend Bulgaria 2.9 3.5 4.0 4.1 4.0 4.1 3.6 Improvement until 2007 Czechia 5.4 4.8 4.3 3.7 4.2 4.8 5.2 5.2 Substantial deterioration, then improvement Estonia 5.7 5.7 5.6 6.0 6.7 6.5 6.6 Improvement Hungary 4.9 5.0 5.2 4.9 4.8 5.2 5.3 5.1 Steady at relatively good level* Latvia 2.7 3.4 3.7 4.0 4.7 4.8 5.0 Improvement Lithuania 4.1 4.8 4.6 4.8 4.8 4.6 Initial improvement, then steady at relatively good level Poland 5.6 4.6 4.1 4.0 3.5 3.7 4.2 4.6 Substantial ‘long-term’ deterioration, now improving Romania 3.0 2.9 2.6 2.9 3.1 3.7 3.8 Initial deterioration, then improvement Slovakia 3.9 3.5 3.7 4.0 4.7 4.9 5.0 Improvement since 2000 Slovenia 5.5 6.0 6.0 6.4 6.6 6.7 Improvement Mean Average 5.3 4.1 4.2 4.3 4.4 4.8 5.0 5.0 Improvement (1996 excluded – too few cases) Source: Adapted from Holmes 2009: 279. – Notes: TI assesses levels of corruption on a 0-10 scale. Empty cells indicate that the particular state was not assessed that year. * The term relatively in this table means in comparison with other transition and developing states. 32 Table 2: TI CPI scores of post-communist states that are not EU member states 1996 1998 2000 2002 2004 2006 2007 2008 Trend Albania 2.5 2.5 2.6 2.9 3.4 Improvement Armenia 2.5 3.1 2.9 3.0 2.9 Steady Azerbaijan 1.5 2.0 1.9 2.4 2.1 1.9 Mixed, but overall poor performance Belarus 3.9 4.1 4.8 3.3 2.1 2.1 2.0 Initial improvement, then marked deterioration Bosnia and Herzegovina (BiH) 3.1 2.9 3.3 3.2 Steady Croatia 3.7 3.8 3.5 3.4 4.1 4.4 Deterioration, then improvement Georgia 2.4 2.0 2.8 3.4 3.9 Improvement Kazakhstan 3.0 2.3 2.2 2.6 2.1 2.2 Mixed, but overall poor performance Kyrgyzstan 2.2 2.2 2.1 1.8 Deterioration, overall very poor performance Republic Macedonia 2.7 2.7 3.3 3.6 Improvement Moldova 2.6 2.1 2.3 3.2 2.8 2.9 Mixed Montenegro 3.3 3.4 Too few scores to comment Russia 2.9 2.4 2.1 2.7 2.8 2.5 2.3 2.1 Mixed, but recently deteriorating Serbia 3.0* 1.3* 2.7* 3.0 3.4 3.4 Recently improving Tajikistan 2.0 2.2 2.1 2.0 Steady, but overall poor performance Turkmenistan 2.0 2.2 2.0 1.8 Recent mild deterioration Ukraine 2.8 1.5 2.4 2.2 2.8 2.7 2.5 Mixed, then recent mild deterioration Uzbekistan 2.4 2.9 2.3 2.1 1.7 1.8 Deteriorating Mean Average 2.9 3.0 2.5 2.8 2.5 2.6 2.7 2.7 Essentially steady state Source: As for Table 1. Note: * Serbia and Montenegro; while it is not possible to determine whether or not the Montenegrin component skewed the Serbian, the two states received almost identical scores in 2007 and 2008, suggesting that this was not necessarily the case. 33 Dividing groups of states in this way produces a clear picture. In general, the new EU member states appear to have been improving (Hungary is an exception, but has not deteriorated to any meaningful extent; Bulgaria’s apparent decline is too recent to be certain about any trend line – while some have implicitly questioned Romania’s performance),5 and have had a lower level of corruption than the other CEE and CIS states that have either not yet been admitted to the EU or else have not applied; these non-EU states have mostly been either more or less steady state or else deteriorating. But there is one clear exception to this general pattern; those countries that are already candidate (formally negotiating) members (Croatia; Rep. of Macedonia), have formally applied for membership (Albania – April 2009; Montenegro – December 2008), or else have publicly stated their intention to apply (Georgia; Serbia), have all been either improving or at least not deteriorating in recent years. Removing these states from Table 2 results in mean average scores of 2.45 in 2004, 2.5 in 2006, 2.4 in 2007 and 2.3 in 2008; these are not impressive figures, and suggest that the post-communist states with no connection to the EU are in general making no progress in their putative attempts to reduce corruption. By way of comparison, the mean averages for the five or six (because of Montenegro) would-be member states were 2.7 in 2004, 2.9 in 2006, 3.4 in 2007 and 3.7 in 2008 – still well below the mean average for the post-communist EU member states, but significantly above that for the other non-member states, and clearly improving lately. A correlation does not prove causality; but it is noteworthy that, according to the CPI, post-communist states that have either joined the EU, or else are seeking to, are perceived to have improved on the corruption front in recent years – whereas most of those that have shown little or no interest in EU membership have not. Given the various problems associated with the CPI, it is worth comparing its results with others generated by alternative methods (i.e. using multi-angulation). Whereas the CPI is a perceptions-based survey, another of TI’s surveys (see below), as well as some of those conducted by agencies such as the WB and teams of scholars (notably Miller, Grødeland and Koshechkina – see e.g. 2001), include experiential questions. Unfortunately, none of these alternatives is as comprehensive as the CPI. Nevertheless, some can be used tentatively to test the hypothesis that corruption levels appear to be improving in most EU-oriented post-communist states, but not in the non-EU-oriented ones. One of the most useful of the experiential surveys is the International Crime Victim Survey (ICVS) that has been conducted on several occasions since 1989. Unfortunately for our purposes, the ICVS has two significant drawbacks. First, the sample varies from survey to survey. Second, it sometimes surveys countries, at other times only capitals. This renders it difficult to use for the type of comparative and longitudinal analysis being attempted here. Nevertheless, it is worth exploiting the data to the maximum permissible extent. 5 Belgian prosecutor Willem de Pauw claimed in late-2007 that the corruption situation in Romania was worsening, and he was much more critical than the EU was at that time (Economist, 3 July 2008). 34 Table 3: ICVS – Percentage of respondents experiencing bribe-seeking by public officials in previous 12 months 1992 1996/7 2000/1* 2000/1 2004/5 1996/7 and 2004/5 compared Albania 13.8 59 Azerbaijan 21 Belarus 12.0 21 Bulgaria 19.1 16 8.4 ↓ Croatia 15.2 10 Czechia 2.4** 8.8 6 Estonia 3.8 9 5.2 3.1 ↓ Georgia 21 29.9 17 Hungary 3.9 10 4.9 ↑ Kyrgyzstan 21.3 Latvia 14.0 14 Lithuania 13.4 23 Rep. Macedonia 7.7 Poland 5 7.4 16 5.1 4.4 ↓ Romania 11.9 19 Russia 12* 18.7 17 Slovakia 13.9 Slovenia 0.6* 1.5 2 Ukraine 12.6 16 Yugoslavia 17.4 Sources: 1992: Siemaszko (1993: 92) and Valkova (1993: 498); 1996-1997: Zvekic (1998: 48-50); 2000 – capital cities: Alvazzi del Frate/van Kesteren (2004: 25-28); 2000 – country-studies: Van Dijk et al. (2007: 90); 2004-2005: Van Dijk et al. (2007: 90). Notes: * Capital cities only; ** Czechoslovakia It will be obvious from the numerous empty cells in Table 3 that the relevant data from the ICVS are very patchy. The 2nd ICVS (1992) preceded the first CPI, so that no direct comparison with the latter is possible. But the survey results revealed that considerably more citizens in Georgia and Moscow had been asked to pay a bribe in the previous twelve months than in Poland or Czechoslovakia, which in turn were more corruption-prone than the Slovene capital. The third ICVS (1996-1997) was far more comprehensive, and revealed that the worst post-communist states (i.e. those in which the highest percentage of citizens had paid a bribe in the previous year) were Georgia, Kyrgyzstan and Bulgaria; the best were Hungary, Estonia and Slovenia. These results are broadly compatible with the CPI ones for the late-1990s. The fourth ICVS (2000-2001) suggested that the most corrupt post-communist capitals, judged by the percentage of citizens asked or expected to pay a bribe in the twelve months before the survey, were Tirana (worst), Vilnius, Baku and Minsk, while the least were Prague and Ljubljana. While the Lithuanian results were surprising in light of the CPI results and others considered below, the rest were more or less in 35 line with the latter. Unfortunately, the 5th ICVS (2004-2005) included only four postcommunist states. Of these, Bulgarian citizens had experienced more bribery in the previous year than Hungarians, who in turn had experienced slightly more than Poles; Estonia emerged as the least corrupt by this criterion. Yet again, the results resonate well with the CPI’s. Indeed, with the exception of the Lithuanian outlier, the ICVS and CPI results are – while not identical – very similar; if states (or capitals) were to be classified into ‘good’, ‘bad’ and ‘middling’ on the basis of these two sets of surveys, the lists would look almost the same. This said, the 5th ICVS cannot be used to test our hypothesis about EU influence, since three of the four states were already member states, while the fourth was soon to become one – so that there are no negative cases against which to test the hypothesis – and is a single case anyway. Earlier, reference was made to the fact that TI does conduct a survey that includes experiential questions. This is the Global Corruption Barometer (GCB), which was first trialled in 2002-2003. The following tables show the results of a question about actual experience of bribery in those CEE and CIS states for which there are results in the years selected earlier for the CPI analysis:6 once again, the data can be presented so as to distinguish those states that are members of the EU from those that are not to test our hypothesis: Table 4: TI GCB – Percentage of respondents who admitted paying a bribe in the previous 12 months, post-communist EU member states 2004 2006 2007 2009 Trend Bulgaria 6 8 7 9 Slight deterioration, but low level Czechia 21 17 13 11 Steady improvement Estonia 6 - - - Cannot assess Hungary - - - 14 Cannot assess Latvia 18 - - - Cannot assess Lithuania 32 - 29 30 Steady state at high level Poland 5 5 - 4 Steady state at low level Romania 25 20 33 14 Very mixed Sources: Hodess and Wolkers (2004: 22); Hodess and Lavers (2006: 17-18); Transparency International (2007: 21); Riaño (2009: 32). 6 The one exception is that the GCB results for 2009 have been used, since there was no GCB 2008; conversely, the 2009 CPI had not been published at the time of writing. 36 Table 5: TI GCB – Percentage of respondents who admitted paying a bribe in the previous 12 months, post-communist states not members of the EU 2004 2006 2007 2009 Trend Albania 30 66 71 - Serious deterioration to very high level Armenia - - - 43 Cannot assess Azerbaijan - - - 46 Cannot assess Belarus - - - 13 Cannot assess BiH 13 - 5 9 Mixed Croatia 9 7 8 4 Steady state, then improvement Georgia 6 - - 2 Insufficient data – but very low level Rep. Macedonia 9 9 44 4 Suspiciously mixed Moldova 32 27 30 28 Steady state at high level Russia 21 8 17 31 Suspiciously mixed Serbia* - 13 21 20 Deterioration Ukraine 25 23 30 21 Mixed at rather high level Sources: As for Table 4. Note: * Excludes Kosovo. This time, it is difficult to identify a pattern. This is partly because of insufficient results – the GCB country mix is frustratingly diverse from year to year – and partly because some results, notably those for the Republic of Macedonia and Russia, are so variable that their validity must be questioned. Overall, the GCB cannot yet be used for making any inferences about trends; the most they can be used for is in terms of classifying countries as having high, low or middling levels of corruption based on citizen experience of bribe-paying. In terms of this criterion, and the EU’s recent approach to Bulgaria and Romania, it is interesting that Bulgaria fares much better than its neighbour – and considerably better than Lithuania, which was admitted to the EU in the May 2004 enlargement. The extremely low rates in Georgia and high rates in Albania are also noteworthy. To this point, the experiential results cited focus on citizen responses. But another method for measuring corruption levels is to survey firms about the percentage of income they pay in bribes, and how frequently such bribes are paid. The best-known survey of this kind is the Business Environment and Economic Performance Survey (BEEPS) of the European Bank for Reconstruction and Development and the WB, which has so far been conducted every three years since 1999, and which combines – sometimes blurs – experiential and perceptual questions. In the first (1999), businesspeople were likely to have paid the highest percentage of income on bribes in Ukraine, Armenia, Kyrgyzstan and Azerbaijan – and the smallest proportions in Poland, Slovenia, Latvia and Croatia (Pradhan et al. 2000: xvii-xviii). These findings are broadly in line with many of the other survey data cited. However, a useful comparative analysis of the 2002 and 2005 BEEPS results for CEE and the CIS appears to throw a spanner in the works! The results in the ‘Europe and Central Asia’ (basically CEE and the CIS) summary report are aggregated into EU8 (i.e. postcommunist states that acceded to the EU in 2004), SEE (South Eastern Europe – 37 which, somewhat frustratingly, includes Turkey) and the CIS. While the EU8 emerge as less corrupt than either the CIS or SEE – as would be expected on the basis of most of the other surveys cited here – all three regions appeared to improve corruption-wise between 2002 and 2005 (World Bank 2006b: 3-4). This said, the frequency of bribe paying declined proportionately much more in the EU8 than in either the CIS or SEE. At the time of writing, no results from the 4th BEEPS were publicly available. Moreover, the BEEPS website makes it clear that the CEE and CIS results from the 4th BEEPS will not be directly comparable with earlier findings. Corruption measurement remains a tricky and constantly evolving business! Having provided aggregated findings from BEEPS, the results from individual countries can be considered and compared with the other data cited here. Table 6: Enterprise perceptions and experience of corruption, post-communist EU member states, according to BEEPS7 1999 (A) 1999 (B) 2002 (A) 2005 (A) Trend 2002-5 (A) 2002 (B) 2005 (B) Trend 2002-5 (B) Bulgaria 2.6 25 53 39 ↓ 32 16 ↓ Czechia 2.2 28 36 50 ↑ 13 10 ↓ Estonia 1.9 14 14 22 ↑ 12 6 ↓ Hungary 2.0 31 22 28 ↑ 22 10 ↓ Latvia 30 26 ↓ 17 7 ↓ Lithuania 2.5 17 48 46 ↓ 21 24 ↑ Poland 2.5 33 50 47 ↓ 18 15 ↓ Romania 2.8 53 56 48 ↓ 37 22 ↓ Slovakia 2.4 33 50 25 ↓ 36 11 ↓ Slovenia 1.7 9 18 14 ↓ 7 4 ↓ Sources: 1999: ACNielsen (1999: 41 and 82); 2002, 2005: World Bank (2006b). Notes: 1999 (A): Corruption scaled as a problem to doing business 1-4 (1 = no obstacle; 4 = major obstacle); 1999 (B): Need for irregular payments to get things done – frequent, usual or always; 2002, 2005: Refers to the percentage of firms identifying corruption as a problem of doing business; Trend 2002-2005 (A): Refers to whether more firms (↑), fewer firms (↓) or the same percentage of firms (→) identify corruption as a problem in doing business in 2005 compared with 2002; Trend 2002-2005 (B): Refers to whether more firms (↑), fewer firms (↓) or the same percentage of firms (→) claim that bribe payments are frequent. 7 Unfortunately, the WB provides detailed BEEPS data only in very expensive software that is not readily accessible to many academics. Because of this, the figures in tables 6 and 7 are based on WB bar-graphs, and may consequently be one percent out in some cases. This would not affect the overall picture, however. 38 Table 7: Enterprise perceptions and experience of corruption, post-communist states that are not EU member states, according to BEEPS 1999 (A) 1999 (B) 2002 (A) 2005 (A) Trend 2002- 2005 (A) 2002 (B) 2005 (B) Trend 2002- 2005 (B) Albania 60 60 → 35 46 ↑ Armenia 1.9 41 36 42 ↑ 14 10 ↓ Azerbaijan 2.8 54 37 50 ↑ 27 27 → Belarus 1.8 17 37 14 ↓ 23 21 ↓ BiH 59 46 ↓ 22 20 ↓ Croatia 2.7 19 44 34 ↓ 13 11 ↓ Georgia 3.0 33 60 48 ↓ 37 7 ↓ Kazakhstan 2.7 25 33 27 ↓ 30 23 ↓ Kyrgyzstan 3.3 26 44 60 ↑ 43 52 ↑ Rep. Macedonia 49 53 ↑ 23 26 ↑ Moldova 3.0 32 60 50 ↓ 34 22 ↓ Russia 2.4 33 29 40 ↑ 37 39 ↑ Serbia- Montenegro 32 51 ↑ 14 33 ↑ Tajikistan 48 42 ↓ 35 21 ↓ Ukraine 2.4 28 53 41 ↓ 35 27 ↓ Uzbekistan 2.2 46 22 19 ↓ 21 22 ↑ Sources and Notes: As for Table 6. Tables 6 and 7 do correlate reasonably well with the CPI, though it would be dishonest not to highlight important differences. First, there are inconsistencies within the BEEPS results themselves. Thus, whereas the results suggest that the frequency of bribe-paying had declined 2002-2005 in all but one of the EU actual (from 2004) and prospective (to 2007) member states, corruption was seen as more of a problem in doing business in three of the ten states. There were similar discrepancies in four of the sixteen non-EU post-communist states. Problems also arise when the results from both tables are compared with those from other surveys. Given a combination of space limitations and incompatible dates, the only comparison here will be between BEEPS Trend B and the CPI 2002-2006. The results are similar in the case of the EU actual and prospective states; Poland was the only country in which the results went in opposite directions. Turning to the non-EU states, the picture is more mixed. In the case of six of the sixteen countries in Table 7, no comparison could be made because there was no 2002 score in the CPI. Of the remaining ten, the trend results were the same for Georgia, Kazakhstan, Moldova, Russia, Ukraine and Uzbekistan, but went in opposite directions in the cases of Albania, Belarus and Croatia (the difference was less marked in the case of Azerbaijan). Discrepancies between the CPI and ICVS results on the one hand (the GCB is excluded from consideration here, for obvious reasons), and the 2nd and 3rd BEEPS data 39 on the other, could be primarily a function of somewhat different target groups. This is clearly true of the ICVS and BEEPS. However, in that the CPI is a poll of polls mostly of business surveys, the differences between it and the 2nd and 3rd BEEPS are of concern. It is not possible to determine which is the more accurate – though TI certainly claims that there has been a high correlation between the aggregate perceptions in the CPI and the aggregate experiences of petty corruption reported in its Global Corruption Barometer since the correlation began to be tested in the mid- 2000s (Hodess and Lavers 2006: 9-10; Transparency International 2007: 10-11; Riaño 2009: 14). Despite these differences, there is broad agreement even between the CPI and BEEPS – at least until about the time the first group of CEE states was admitted to the European Union (the unavailability and changed format of the 2008 BEEPS renders later comparison impossible) – that EU applicant and member states have performed better than other post-communist states. Before finishing this brief consideration of the corruption situation in CEE and the CIS, one other paradox needs to be addressed. This is the significant mismatch between the EU’s assessments of Bulgaria and Romania and some of those produced by other IOs and international NGOs. There is a simple explanation for this; it relates to the point touched upon at various points above concerning target groups. One of the problems with corruption-oriented experiential surveys of either citizens or businesspeople is that they typically measure petty (low-level or administrative) corruption, rather than grand (high-level or political) corruption. This essentially narrow focus is understandable, since most citizens and businesspeople do not come into direct contact with high-level politicians, senior ministerial officials, etc. But it also helps to explain why such surveys are less useful than experiential surveys of crime generally, and why they should not be considered superior to perceptual surveys; both methods have advantages and disadvantages. Having made this point, EU concern about Bulgaria and Romania is not about everyday corruption levels and experiences; it is about high-level corruption by senior officials who have allegedly been siphoning off EU funds. There is no inherent reason why high-level corruption rates in a given state should correspond with low-level rates. This point is often overlooked in debates about corruption levels. 4. Conclusions As already emphasised, demonstrating correlations does not prove causality or indicate directionality. The previous section did demonstrate some reasonably strong correlations. But it also suggested that the impact of most IOs on corruption levels is typically low; the EU appears to be an exception to this. Yet EU influence cannot explain apparent improvements in the corruption situation in most CIS states. It is, of course, possible that targeted IO programmes are partly responsible for these improvements. However, since such programmes have often been directed at states that do not appear to have improved – and may even have deteriorated – their positive impact is, at most, patchy and limited. Moreover, even in cases in which IOs do ap- 40 pear to be exerting a positive influence, there is no question that numerous other factors are also at play. Among the variables that can help to explain patterns in individual states are the role of leadership (how genuine its commitment to fighting corruption is); GDP per capita; the end of war and sanctions; proximity to the West; religion and culture; and, importantly, the commitment to democratisation and the related issue of media freedom. A comparative exploration of any one of these – let alone all of them – would require at least a separate paper, and cannot be undertaken here. But it must be noted that factors other than the role of IOs are playing a significant role in explaining any patterns tentatively identified here. So is the role of IOs negligible or even irrelevant? Most IOs have been doing much in recent years to address the so-called ‘corruption eruption’. There can be no doubt that they have succeeded in raising corruption awareness, and in securing formal support from many governments for governance improvements designed to reduce corruption. Yet when we move from declarations to results, Kaufmann’s pessimistic assessment appears to be validated in the case of most CIS states – though it is acknowledged that the corruption situation in many countries might have been even worse were it not for the IOs’ various anti-corruption measures. On the other hand, his conclusion is not clearly supported in the case of most CEE countries. It has been argued here that part of the explanation for the apparent overall improvement in these is the influence of the EU on potential and actual member states. This influence manifests itself in conditionalities prior to admission and, very recently, in the withdrawal of substantial funding once a country has been admitted. Although it is too early to draw conclusions about the effectiveness of the latter approach, it does a priori have more bite than most other methods used by IOs. While targeted programmes such as the CoE’s Octopus may have had limited effect in some cases in the short term, sustained improvement is not generally a feature of countries that have not expressed a commitment to joining the EU. Given the EU’s recent more aggressive approach, Kaufmann’s comment about the currently ‘tepid’ approach of most of the international community has widespread but not universal applicability. Ultimately, most IOs have only limited leverage over sovereign states, and at a time when the term ‘cultural imperialism’ is frequently invoked, this may not be a bad thing. But it is perfectly reasonable for the EU to withhold funds or introduce conditionalities for countries that apply for major funding and/or membership from the EU. Indeed, the stick of withholding funds is a potentially powerful method for dealing with recalcitrant corrupt elites, and may prove far more effective than guidelines or recommendations. Unfortunately, as noted, this approach is at present too new to permit assessment. Nevertheless, we agree in principle with Klitgaard (1998: 6) that a shock is sometimes required to ‘disturb a corrupt equilibrium’. There are also other ways in which IOs could enhance their role, without risking being accused of bullying or double standards. If IOs are to use the stick, it should be stronger – and it should be used not merely against states, but also more than it typically is against transnational companies (TNCs) that offer bribes to officials in CEE and the CIS. This could be directly, such 41 as blacklisting companies in the way the WB has been doing since 1999 (the WB names companies and individuals it is ‘debarring’ – its preferred term – on its website; there were more than 130 firms and persons on the list in August 2009). It could also be done vicariously, by urging Western states to exert more pressure on companies suspected of being or known to be involved in corrupt practices. The scandalous case of the UK government, which blocked investigations into alleged massive bribery by arms manufacturer BAE in Saudi Arabia and Czechia (see Holmes 2009b), demonstrates that IOs could and should do more to combat the supply side of highlevel corruption, not just the demand side. Unfortunately, the troubled relationship between the USA and the UN over many years demonstrates clearly the limits to IO power; most simply do not have the leverage over developed states – typically, their principal source of funding – that they have over transition and developing states. Another example of this problem is the OECD. This organisation was highly critical of the UK’s handling of the BAE allegations in an October 2008 report, but was able to do little other than urge the British government to amend its legislation on corporate bribery to align it better with the OECD Convention London had signed a decade earlier (Daily Telegraph, 17 October 2008). When it is borne in mind that, according to 2009 TI assessments, only four out of 36 signatories to the Convention assessed (Germany, Norway, Switzerland and the USA) are actively enforcing it, while there is ‘little or no enforcement’ in 21 (seven of them CEE states), the limited impact of what have been hailed as significant IO initiatives comes sharply into focus (Heimann/Dell 2009: 6).8 Indeed, Managing Director of TI Cobus de Swardt declared in mid-2009 that ‘The failure of the OECD Convention would arguably be the single most serious setback to the fight against international corruption’ (Reuters 2009). While they often exaggerate this, Realists’ emphasis on the role of the state in international politics appears to be justified in these types of situation. Nevertheless, IOs could do more to publicise and criticise misbehaviour by developed states and TNCs. They could also make greater use of the carrot in their publicity role. For example, when TNCs resist meat-eating demands for bribes, as IKEA often does in Russia, their good citizen role should be highlighted by IOs. IOs should themselves set better examples than they often do; instructing transition states to lift their game should not be a case of ‘don’t do what I do, do what I say’. In assessing the success of IO attempts to reduce corruption in CEE, CIS and elsewhere, it should be noted that there are often tensions – sometimes blatant contradictions and hypocrisy – in the words and deeds of agencies such as the EU, the UN and the OECD. Virtually all IOs have experienced at least one major corruption scandal in recent years (e.g. the UN’s oil-for-food scandal; the EU’s Santer Commission and Eurostat scandals).9 Many IOs are at last taking this example-setting role 8 The OECD Guidelines for Multinational Enterprises have also been less than a resounding success – see Transparency International (2008: 4-5). 9 The EU’s anti-fraud organisation OLAF was created in 1999 as a direct response to the Santer Commission scandal. This reflects well on the EU, in that it quickly adopted concrete meas- 42 more seriously, but have been slow to do so. Thus the UNDP’s own Accountability System (Internal Audit and Oversight) dates only from December 2007, while the first Ethics Advisor within the organisation dates from 2008 (UNDP 2008: 29). The EU established systems much earlier. For instance, the Convention on the Fight Against Corruption involving Officials of the European Communities or Officials of Member States of the European Union was introduced in May 1997; but the subsequent corruption scandals demonstrate that the implementation of such schemes could be improved. IOs could also cooperate more than they do with each other, to minimise overlap and duplication. It is encouraging that many IOs have now secured observer status at meetings of other IOs. Nevertheless, it sometimes looks to an outsider as if there is competition between IOs in their anti-corruption endeavours. While some competition is desirable, too much can result in agencies essentially leaving it to others, with no-one accepting ultimate responsibility. Some IOs need to become more astute when funding and encouraging domestic anti-corruption NGOs; greater IO awareness of the fact that NGOs can have a vested interest in exaggerating the seriousness of problems in their country should help to reduce the problem of the ‘corruption industry’. Finally, IOs must be wary of becoming immune to corruption and/or succumbing to anti-corruption fatigue. Corruption will never be eradicated; but the fact that levels are so relatively low in many countries proves that it can be kept within manageable bounds. The business sector in many countries has since 2008 been indicating that the Global Economic Crisis is likely to increase the propensity of firms to attempt to secure contracts through questionable means, so that now is not an appropriate time to take the foot off the pedal. Since ‘it takes two to tango’, it would be a serious oversight to consider the impact of IOs only from their perspective. Post-communist states have their role to play too, and this can sometimes be in less than obvious ways. For instance, the desire by post-communist states to display commitment to fighting corruption and abiding by agreements such as the acquis communautaire – in short, to display their commitment to the rule of law, transparency and international agreements – can be exaggerated and, ironically, can foster corruption. During a June-July 2009 visit to Poland, a senior administrator informed me that her office now had to tender every time they wanted new pencils, rulers or staplers, since, they had been informed, simply renewing the contract with the existing (and apparently perfectly satisfactory) supplier would be seen as too ‘cosy’ and hence corrupt by the EU. This could be the result of a simple misunderstanding. But a petty-bureaucratic approach on inexpensive everyday products, by states claiming to be following EU regulations, can result in a loss of faith in a rules-based system and promote the development of a mentality that will circumvent the rules if these are seen as illegitimate. The rule of law must not manifest itself in laws on rulers! ures to reduce the likelihood of such a situation arising again. This said, OLAF has been criticised by, among others, Roman Prodi, for being slow and ineffectual. 43 IOs must be prepared for criticism and failure from countries they are trying to assist. Unfortunately, IOs can sometimes come off worse than national governments in any conflict between the two. Domestic leaders can use IOs as scapegoats – blame them – for the introduction of tough anti-corruption measures. And if these unpopular measures do not work – even if this is primarily because of the lack of political will on the part of national governments – IOs can again be blamed. Ultimately, while the role of IOs in reducing corruption is potentially significant, governments and citizenries must play the lead role. The fight against corruption depends more on political will and capacity within countries than on any other single factor. The will referred to here is primarily of domestic leaders (elites) and ‘staffs’ (in Weberian terms), though civil society has an important watchdog function to perform. In arguing that the primary responsibility for combating corruption rests with national governments rather than the international community, my position remains as it was at the turn of the millennium (Holmes 2001). But the national governments referred to here include those of developed states. Unless governments take corruption more seriously, most IOs will be seriously hampered in their endeavours to reduce corruption globally. Bibliography ACNielsen (1999) Key Findings from the Business Environment Survey of Private Sector Firms in the CEE and CIS (ACNielsen), http://siteresources.worldbank.org/INTECAREGTOPANTCOR/ Resources/beepsr99.pdf. Alvazzi del Frate, A./van Kesteren, J. (2004) Criminal Victimization in Urban Europe: Key Findings of the 2000 International Crime Victim Surveys, Turin. Alvazzi del Frate, A./Zvekic, U./van Dijk, J. (eds.) (1993), Understanding Crime Experiences of Crime and Crime Control, Rome. Center for Research on Population and Security (1986) Public Report of the Vice President’s Task Force on Combating Terrorism, http://www.population-security.org/bush_report_on_terrorism/ bush_report_on_terrorism.htm. Commission of the European Communities (2009a) Interim Report from the Commission to the European Parliament and the Council: On Progress in Romania under the Co-operation and Verification Mechanism, Brussels. Commission of the European Communities (2009b) Interim Report from the Commission to the European Parliament and the Council: On Progress in Bulgaria under the Co-operation and Verification Mechanism, Brussels. Deutsche Welle (2008) Scathing EU Report Exposes Bulgaria's High-Level Corruption, http:// www.dw-world.de/dw/article/0,,3493169,00.html. Dunga, E. (2008) Implementation of GRECO Recommendations and GRECO Experiences, paper presented at the Conference in the Occasion of the International Anti-Corruption Day, Sarajevo, 11 December, http://www.rai-see.org/doc/Implementation_of_GRECO_recommendations_and_ GRECO_experiences_Mr._Edmond_Dunga.pdf. Eigen, P. (2002) Time for a private sector clean-up, TI Q, December, 1. Eigen, P. (2003) WTO must address corruption in procurement, TI Q, September, 1. 44 Frankfurter Allgemeine Zeitung (FAZ.Net) (2008) EU stoppt Finanzhilfen für Bulgarien, http:// www.faz.net/s/Rub99C3EECA60D84C08AD6B3E60C4EA807F/Doc~EC7FCC987DB924065 A113ABA03461B829~ATpl~Ecommon~Scontent.html. Galtung, F. (2006) Measuring the immeasurable: boundaries and functions of (macro) corruption indices, in Sampford, C. et al. (eds.) Measuring Corruption, Aldershot, 101-130. Heimann, F./Dell, G. (2009) Progress Report 2009: Enforcement of the OECD Convention on Combating Bribery of Foreign Public Officials in International Business Transactions, Berlin. Hodess, R./Lavers, T. (2006) Report on the Transparency International Global Corruption Barometer 2006, Berlin. Hodess, R./Wolkers, M. (2004) Report on the Transparency International Global Corruption Barometer 2004, Berlin. Holmes, L. (2001) Crime, Corruption and Politics: International and Transnational Factors, in Zielonka, J./Pravda, A. (eds.) Democratic Consolidation in Eastern Europe: International and Transnational Factors, Vol. 2, Oxford, 192-230. Holmes, L. (2009a) Crime, Organised Crime and Corruption in Post-Communist Europe and the CIS, Communist and Post-Communist Studies 42, 265-287. Holmes, L. (2009b) Good Guys, Bad Guys: Transnational Corporations, Rational Choice Theory and Power Crime, Crime, Law and Social Change 51, 383-397. June, R. et al. (2008) A User’s Guide to Measuring Corruption, Oslo. Kaufmann, D. (2009) Aid Effectiveness and Governance: The Good, the Bad and the Ugly, Development Outreach, February, 26-29. Klitgaard, R. (1998) International Cooperation Against Corruption, Finance and Development 35, 3-6. Krastev, I. (2004) Shifting Obsessions: Three Essays on the Politics of Anticorruption, Budapest Miller, W./Grødeland, Å./Koshechkina, T. (2001) A Culture of Corruption? Coping with Government in Post-Communist Europe, Budapest. Mungiu-Pippidi, A. (2007) EU Accession is No ‘End of History’, Journal of Democracy 18, 8-16. Naim, M. (1995) The Corruption Eruption, Brown Journal of World Affairs 2, 245-261. Naim, M. (2005) Bad Medicine, Foreign Policy 147, 96. Pradhan, S./Anderson, J./Hellman, J./Jones, G./Moore, B./Muller, H./Ryterman, R./Sutch, H. (2000) Anticorruption in Transition: A Contribution to the Policy Debate, Washington. Pridham, G. (2007) The Effects of the European Union’s Democratic Conditionality: The Case of Romania during Accession, Journal of Communist Studies and Transition Politics 23, 233-258. Reuters (2009) Anti-corruption body raps UK, other OECD countries, http://www.assetrecovery. org/kc/node/1335c076-647e-11de-bacd-a7d8a60b2a36.0;jsessionid=00447E75FF2BA9289E1F 6BA0DAF6EBB1. Riaño, J. et al. (2009) Global Corruption Barometer, Berlin Sajó, A. (1998) Corruption, Clientelism and the Future of the Constitutional State in Eastern Europe, East European Constitutional Review 7, 37-46. Sajó, A. (2003) From Corruption to Extortion: Conceptualization of post-communist corruption, Crime, Law and Social Change 40, 171-194. Schmidt-Pfister, D. (2010) Transnational Advocacy on the Ground: Against Corruption in Russia?, Manchester. Siemaszko, A. (1993) Central and Eastern European Victimisation Rates: To Compare or Not to Compare?, in Alvazzi del Frate, A./Zvekic, U./van Dijk, J. (eds.) Understanding Crime Experiences of Crime and Crime Control, Rome, 87-92. 45 SpiegelOnline (2008) EU-Kommission streicht Bulgarien Hilfen in Millionenhöhe, http://www. spiegel.de/politik/ausland/0,1518,592707,00.html. Transparency International (1996) Sharpening the Responses Against Global Corruption, Berlin. Transparency International (2007) Report on the Transparency International Global Corruption Barometer 2007, Berlin. Transparency International (2008) Using the OECD Guidelines to Tackle Corporate Corruption, Transparency International Working Paper, No.3. UNDP (2008), Mainstreaming Anti-Corruption in Development: Anti-Corruption Practice Note, http://www.undp.org/governance/docs/Mainstreaming_Anti-Corruption_in_Development.pdf. Valkova, J. (1993) Czechoslovakia, in Alvazzi del Frate, A./Zvekic, U./van Dijk, J. (eds.) Understanding Crime Experiences of Crime and Crime Control, Rome, 493-502. Van Dijk, J./Van Kesteren, J./Smit, P. (2007) Criminal Victimisation in International Perspective: Key Findings from the 2004-2005 ICVS and EU ICS, The Hague. World Bank (2006a) BEEPS–at-a-glance – Europe and Central Asia Final, http://web.worldbank. org/WBSITE/EXTERNAL/COUNTRIES/ECAEXT/EXTECAREGTOPANTCOR/0,,contentM DK:20720934~pagePK:34004173~piPK:34003707~theSitePK:704666,00.html. World Bank (2006b) Europe and Central Asia – BEEPS at-a-Glance, http://siteresources.worldbank.org/INTECAREGTOPANTCOR/Resources/BAAGREV20060208ECA.pdf. Zvekic, U. (1998) Criminal Victimization in Countries in Transition, Rome. 47 Fighting Corruption Abroad. The EU’s Good Governance Export Tanja A. Börzel, Yasemin Pamuk and Andreas Stahn 1. Introduction With the End of the Cold War, good governance has been mainstreamed into the development strategies of major international organisations, such as the United Nations or the World Bank (Weiss 2000; Adam/König 2001; Hill 2006). The European Union (EU) is no exception. In fact, the EU has been among the first organisations to write good governance, together with human rights, democracy, and the rule of war, into its cooperation agreements with third countries. In its Communication on a Comprehensive EU Policy Against Corruption the EU has acknowledged that the promotion of good governance constitutes an essential element of a comprehensive anti-corruption policy (European Commission 2003f). Consequently, good governance has developed into an integral part of the EU’s foreign policy and is backed up by considerable financial and human resources. Following the comprehensive approach of the United Nations Development Programme the EU defines good governance as the ‘transparent and accountable management of human, natural, economic and financial resources for the purpose of equitable and sustainable development’ (European Commission 2003f: FN 46). Good governance is closely linked to clear decision-making procedures, transparent and accountable institutions, the rule of law and, in particular, the fight against corruption (European Commission 2003f: 21). Accordingly promoting good governance in accession, candidate and other third countries, in particular in the neighbourhood regions, constitutes the external dimension of the EU’s framework for the fight against corruption. This chapter aims at shedding light on the quality of the EU’s efforts to promote good in its various relations with third countries. Focusing on the EU’s development policy, the European Neighbourhood Policy (ENP), its relations with the countries of the Western Balkans, and enlargement policy, we attempt to trace the EU’s good governance configuration within these different contexts. In particular we ask with whom (channel of influence) and what exactly (content) the EU is seeking to promote. Lastly, we cast a look on the mix of instruments the EU applies in order to promote good governance in its external relations. Simultaneously, we cast a look on the mix of instruments the EU applies in order to promote good governance in its external relations. In the first part of our contribution we develop an analytical framework that conceptualises good governance along two analytical dimensions: 1) The specific good governance content that is promoted (input vs. output-related objectives), and 2) the

Chapter Preview

References

Zusammenfassung

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.