Conditions for Effective Large-Scale Anticorruption Efforts and the Role of External Actors: What Does the Slovak Experience Tell Us?

The paper looks at what policy-makers can do to decrease corruption in developing and transition countries, based on an in-depth examination of effectiveness of actual anticorruption measures in Slovakia. The research presents a synthesis of 12 case studies where measures in the sectors most associated with corruption as well as horizontal measures were analysed. The research shows that corruption can be decreased significantly within several years and external actors can play a substantial role in the process. An overall decrease in corruption can be based on aggregation of individual sectoral changes in areas most suffering from graft. In particular, the Slovak strategy was based on a sector-by-sector economic approach to resolving supply-demand imbalances based on either liberalization/privatization, limitations on discretion or managing supply and/or demand. Horizontal reforms complemented by sectoral reforms with their strong focus on increasing transparency. Concerning the role of external actors, we conclude that even when there is a domestically driven anticorruption effort, the external actors can still help significantly by serving as sources of inspiration, legitimacy, know-how and funding for reform design and implementation.


Introduction
Academic interest in the issue of corruption in developing countries has a long pedigree, (e.g. Huntington 1968, Merton 1968, Myrdal 1968, Rose-Ackermann 1978 but the research has grown massively in 1990s and 2000s when the corruption was identified as one of the key problems in development and its prominence as a political issue has also risen dramatically. There is now extensive theoretical and empirical In addition to these factors, the Slovak case study can also be interesting because it examines a country with high, but not an extreme level of corruption, which managed to reduce it significantly. While the next section deals with the issue in more detail, a brief comparative note suffices to explain why Slovakia might be of interest. In 2000, Slovakia shared the same level of the Transparency International Corruption Perception Index (3.5) with four other countries -Argentina, Bulgaria, Ghana and Senegal. Eight years later, a substantial gap has opened between the five countries with Slovakia improving CPI to 5.0. On the other hand, CPIs of Ghana, Bulgaria and Senegal rose only by 0.2, 0.1 and 0.1 points respectively, and the Argentina's score actually declined to 2.9.
In this respect, the Slovak case study fits within the definition of the case study by Gerring (2004) as 'an intensive study of a single unit with an aim to generalize across a larger set of units' (p. 341). As noted by Yin (2003), case studies are appropriate instruments for social science research when the phenomenon being studied is highly complex and 'when "how" or "why" questions are being posed, when the investigator has little control over events, and when the focus is on a contemporary phenomenon with some real-life context'. (p. 1) In such a context, Gerring (2004) and Van Evera (1997) emphasise the complementarity of single-unit and cross-unit research designs rather than superiority of one or another.
While the research focuses on a single country, it presents a synthesis of a rich material of 12 case studies where measures in the sectors most associated with corruption as well as horizontal measures were analysed. It examines through what 4 mechanisms they contributed to the reduction of corruption and what was the role of external actors and aid in their design, passage and implementation.
The paper is structured as follows: the following section deals with the anticorruption strategy in Slovakia and the environment in which it was undertaken, including the 12 cases. This sets the stage for the overall analysis of the mechanisms of anticorruption efforts and the role of the external actors. The conclusion summarises findings of the paper, including potential for their generalisation.

The Slovak anticorruption strategy and its results in an international perspective
Anticorruption efforts in Slovakia need to be understood in the overall context of the Slovakia's transition to democracy and market economy during 1990s and 2000s.
After initial period of common transition with the Czech Republic within Czechoslovakia, Slovakia became independent in 1993. The 1992The -1994The and 1994The -1998 governments of Vladimír Mečiar conducted policies, which led Slovakia to be excluded from the first round of EU enlargement negotiations in 1997, achieved a near-collapse of the economy in 1998 and were marked by widespread problems with rule of law as well as corruption, particularly in privatization. In 1998, the opposition overwhelmingly won the parliamentary elections (with approximately 60% of the vote) with the rule of law, the EU integration and the corruption being among the key topics where the electorate felt a need for change. (Sičáková-Beblavá and Beblavý 2006) Therefore, the new Government led by Mikuláš Dzurinda felt a strong commitment to deal with corruption issues. However the discourse dealt nearly exclusively with privatization. Wider debate about anti-corruption strategies was 5 missing from the policy mainstream. (ibid.) At the same time, a 1999 survey (World Bank 2000) found that corruption was perceived to be widespread in Slovakia and two thirds of the interviewed had already encountered the problem in person at least in one sector/institution. In other words, while at the time of 1998 elections, corruption was a politically salient topic only within the context of privatization, Slovakia was ready for a much more wide-ranging change both in attitudes and policies.
Over time, the strong and sustained public pressure for anticorruption policies became one of the key drivers of the anticorruption efforts, together with EU accession and fiscal pressures. Corruption has been consistently rated as the fourth most serious problem for the politicians to focus on, behind unemployment, standard of living and health care (FOCUS 2006). In the vigorous and dynamic Slovak democracy, this has been crucial in keeping the interest of the political elite in the issue and in emergence of senior political figures who base their credibility on their anti-corruption efforts. The anticorruption strategy combined specific sectoral reforms (e.g. bank privatization) with horizontal policies (e.g. freedom-of-information legislation). Both types of policy change took place within the overall environment of rapid reforms on all fronts in Slovakia since 1998. This environment was caused by many activities, but the EU accession and fiscal pressures were among the most relevant. The rapid policy change contributed to the overall feeling that when change is possible in so many areas, it should be also possible to do something about corruption. More importantly, deep structural reforms of key elements of the public sector made it 7 possible to remove or diminish, at low political cost, some of the structural underpinnings of corruption.
During the same period, Slovakia has seen significant absolute improvement in the corruption situation over the last decade (as measured by the Transparency International Corruption Perception Index). Compared to the three neighbouring countries with most similarities -i.e. the Czech Republic, Hungary and Poland, we can see that the improvement was much higher than for the others between 1999 and 2008. Slovakia's improvement in the CPI has also outstripped the average of the new member states significantly over the same period: 1.3 points vs the average 0.54 point improvement for all postcommunist new EU member states. On a more global scale, only Latvia, Serbia, South Korea and Colombia have seen improvements of a similar size in recent years. Therefore, while not unique, the Slovak improvement is exceptional and not due solely to the EU accession process.
[Insert Table 1 here] Let us now look in more detail at the specific anticorruption reforms. The paper synthesises results of 12 case studies researched in Slovakia in 2007 and 2008. Ten of these were published in Beblavá and Beblavý (2008), the rest is unpublished. For each of the case studies, we present a brief description of the policy changes, their relationship to anticorruption efforts and estimate of their impact on corruption as well as a brief look at the role of the external actors. As already mentioned, we divide them into two groups -sectoral and horizontal changes. Each case study contained sections describing the reforms, the mechanisms through which they contributed to corruption control including reference to the relevant academic literature, the process of their approval including analysis of key interest groups supporting or opposing the change, and an analysis of the role of the external actors in the approval. Authors were provided with a unified template for the case study.
The case studies were prepared by a team of 8 researchers from the Comenius University under the auspices of a research project focusing specifically on Slovak 9 anticorruption strategies. Drafts were reviewed by two referees and returned to authors for finalisation.

Limiting discretion in active labour market policy. In 2004, the Slovak
Government enacted a new Act on Employment Services. The act brought substantial changes in decision-making concerning job and public works subsidies. It eliminated the power of local and national stakeholder committees to make decisions on allocating the subsidies on a discretionary basis and gave all employers/self-employed right to the subsidies after fulfilling clearly stipulated conditions in the law and secondary legislation. The size of the subsidies was also determined based on a formula. The impact on corruption was through eliminating discretion and balancing demand and supply. The perception of widespread corruption in labour market agencies decreased from 20% to 14% between 1999 and 2006 and the number of people who denied existence of corruption increased from 6% to 16% in the same period (from 9% to 16% in the two years after introduction of the reform). (Beblavý 2008a) There was essentially no external influence in preparation and implementation of the new policies, even though ESF funds provided the required financing.
However, the overall amount spent was not significantly higher than in the previous years when it was domestically funded. Portugal, but otherwise they played essentially no role.

Company Register reform. The Company Register in
Tax reform. In this area, two reforms -of taxes and tax administration -took place.
Slovakia introduced a radical simplification of taxes in 2004, introducing a unified VAT and income tax rates at 19% and eliminating most deductions and special regimes. In the tax administration, there was simultaneously a reshuffling of 13 responsibilities between tax and customs administration, with tax authorities responsible for direct tax collection and the customs authorities for indirect taxes.
During the same period, taxes (particularly indirect ones) were heavily influenced by harmonization of legislation with EU requirements. As a result, there was a decline from 40% to 15% between 1999 and 2006 of those who see widespread corruption among tax authorities while the decrease for customs was only from 42% to 30%. The steep decline for tax authorities compared to customs authorities is particularly instructive given the fact that both institutions are managed by the ministry of finance, have similar structures and, in Slovakia, they both collect taxes (tax authorities primarily collect income taxes, while customs mainly collect indirect taxes). It indicates that the change in perceived corruption in administration of direct taxes was much higher than for indirect taxes, possibly related to introduction of the flat tax. In terms of influence of external actors, EU influence was important with regard to indirect taxes where there were extensive accession requirements related to tax administration, but also tax structure. In terms of the flat tax, external actors, but mainly private ones, were important in initiating policy transfer, but the actual design was funded and conducted domestically. other reforms, but ultimately delayed until 2005, brought a fundamental overhaul of the healthcare system. All hospitals were to be corporatized and restructured (and some even privatized), insurance companies were to become profit-seeking health management organizations, insurers and providers were to freely negotiate prices and quantities, the government was supposed to clearly determine the extent of the free healthcare package and the co-payments for diagnoses not in it and a strict hard budget constraint was to be introduced for all actors involved. (Beblavý 2008b) However, many of the reform steps were not completed due to political controversy internal audit in local and regional self-governments and allowed the Supreme Audit Office to audit, for the first time, all transaction of the local and regional selfgovernments (requiring a constitutional amendment). The latter steps followed extensive decentralisation that took place between 2002 and 2004 and which dramatically expanded the scope of the decentralised authority. (Láštic 2008a). There is no prima facie evidence for the effect of these steps on perceived corruption, as the evaluation of corruption as widespread stagnated for local governments at 27% vs 28% as well as for ministries at 43% vs 45% (comparison between 1999 and 2006).
Much of the legislation has been introduced due to EU requirements related to management and disbursement of Structural Funds and there is little evidence that its effect has spread beyond these. With regard to local and regional government, analysts ( with the public sector as well as the private sector and increasing transparency to make it easier to uncover when the rules are broken. The key problem for the effectiveness of the law was that its provisions were enforced by politically controlled institutions -a cross-party parliamentary committee or the ministry of finance. This became clear in 2008 when media or independent observers pointed to several clear violations of the law by both government and opposition parties, but there was no enforcement. However, due to media pressure, all parties took steps to avoid the same violations in the future. Therefore, the story of the new law has been one of gradual effect primarily through the sanction of the public opinion. There has been no measurable external influence in case of the changes in financing of political parties.

Effectiveness and mechanism of reducing corruption
This section summarises the data on effectiveness of the individual reforms and mechanisms they used to decrease corruption, based on the information provided in the previous section. Therefore, it is first necessary to explain the methodology used for the assessment and the summary With regard to effectiveness, the methodology is different for sectoral and horizontal reforms. For sectoral reforms, the assessment is primarily based on the change in perceived corruption in the given sector, if such data are available. When necessary, they are complemented by evaluations of other observers or by data on changes in inputs/outputs. In case of horizontal reforms, it is not possible to link individual reforms with specific outcome data on corruption. Therefore, we have to rely solely on analysis of data concerned with inputs, outputs and processes to measure the potential contribution of these reforms to corruption control.

20
To understand the anticorruption strategy in Slovakia and its implication, we also need to understand mechanisms through which individual reforms actually aimed to decrease it. To classify the instruments, we use a simple framework developed by Beblavý (2007), using Huther andShaw (2000), based on concepts developed by Becker (1993), Rosen (1993) and Becker and Stigler (1974). This framework recognises that to decrease incentives for corruption, one can either decrease the benefits or increase the costs of corruption.
Decreasing benefits of corruption can be through one or more of the following:  (1993), Rijckeghen and Weder (1997) or Becker and Stigler (1974). Increasing the probability of discovery of corrupt behaviour can also be included here. Steps such as increasing publicly available information and/or the number of agents with incentives to discover corruption (e.g. the US system of rewarding whistleblowers) or better work of law enforcement agencies and other agencies of self-restraint (see Schedler et al. (2000)).
[Insert Table 2 here] Table 2 summarises the findings for the 12 Slovak case studies. We see that the strategy was a combination of decreasing benefit of corruption through structural reform with several horizontal reforms aimed primarily at increasing transparency.
Relatively limited attention was paid to increasing costs of corruption. In other words, the Slovak strategy can be summarised as: "economic reform + transparency".
However, the Slovak case has important implications that go beyond this simple equation. First of all, the economic reform cannot be identified solely with privatization and liberalization. Depending on the sector, two other techniques were also used: limiting discretion and increasing supply/managing demand. Therefore, even when economic approach is adopted towards reduction of corruption, it can contain a far more sophisticated array of instruments. Secondly, the horizontal transparency efforts formed an integral part of the package.

Role of external actors
In this section, we look at the role of external actors (foreign governments and intergovernmental organisations -Accession conditionality when a certain action becomes a de jure or a de facto condition for entry into a "club". Usually, this type of conditionality is used by the European Union in its accession negotiations, but is not limited to the EU -Membership conditionality when a certain action is an obligation imposed on existing members of a "club". Again, this is mostly the case for the EU, but is not exclusively so. Table 3 presents a summary of our evaluation of the role of external actors in the 12 reforms according to this classification. The table shows that the role of external actors was much more important in policy transfer and technical assistance than in forcing the reform through conditionality. Within the group of conditionalities, accession conditionality was more important than financial conditionality while the membership conditionality has so far not been relevant for the Slovak case. Overall, however, it is surprising that even the strongest conditionality was present only in a minority of reforms. In other words, the political will to enact the reforms was mostly home-grown. Also, looking in more detail at the cases where financial conditionality was present, the size of the financial support was relatively small compared to the overall fiscal or sectoral expenditure.
[Insert Table 3 here] On the other hand, the role of external actors in inspiring and/or designing reforms was crucial and was present in ALL the reforms that were graded as having high or medium effect on corruption. Therefore, even when there is a domestically driven anticorruption effort, the external actors can still help significantly by serving as sources of inspiration, legitimacy, know-how and funding for reform design and implementation.

Conclusion
The The first lesson of the Slovak case is optimistic, even if not ground-breaking. The corruption can be decreased significantly within several years and external actors can play a substantial role in the process. It is relevant to emphasise this finding in a world where scepticism about anticorruption efforts is rife. The second, more intellectually interesting lesson is that an overall decrease in corruption can be based on aggregation of individual sectoral changes in areas most suffering from graft. Even though Slovakia had official anticorruption strategies during this period, some of the key reforms were either not included or mentioned only marginally either because they had been approved before the first strategy was approved or because they were 25 primarily sectoral strategies aimed at improving the overall situation, with the reduction in corruption only one of many objectives.
The paper classified all reforms according to their effectiveness as well as the mechanism through which they contributed to the anticorruption efforts. Its conclusion is that effective anticorruption reforms in Slovakia usually combined structural reforms that decreased scope for (and benefits of) corruption and increased transparency. The paper shows that the decrease in the potential benefits of corruption was achieved not only through the more traditional recipes of privatization and liberalization, but also through introduction of clear decision-making rules to limit discretion and through introduction of policies that increased supply or managed demand. In other words, even though the Slovak strategy was based on a hard-headed economic approach to resolving supply-demand imbalances, there was a variety of instruments that were utilized rather than just conventional ones belonging to the Washington consensus. On the other hand, there was only limited emphasis on making it more costly to engage in corruption through increased repression.
The third group of findings concerns the role of external actors. Our finding was that the Slovak anticorruption drive was primarily underwritten by the electoral pressure, with voters consistently rating corruption as one of the main problems facing the country. Therefore, despite the importance of the process of EU accession, the role of external actors was much more important in policy transfer and technical assistance than in forcing the reform through conditionality. On the other hand, the role of external actors in inspiring and/or designing reforms was crucial and was present in nearly all the reforms. Only a small minority of reforms was completely home-grown