33255 Bosnia and Herzegovina Diagnostic Surveys of Corruption Prepared at the request of the Governments of Bosnia and Herzegovina by the World Bank ACKNOWLEDGEMENTS The report is based on the surveys conducted in BiH in June-July and September 2000 with 700 public officials, 350 enterprise managers, and 1,200 individuals (general public), whose agreement to be interviewed for the Diagnostic have made this report possible. The report would not have been possible without the collaborative work of the representatives of two survey agencies - Paul Barrett, and Rajna Cuculic (ACNielsen) Mirsada Muzur, Dino Djipa (PRISM Research). The report was prepared by Maria Shkaratan (World Bank). The report draws on the Anticorruption Diagnostic program designed and implemented in several countries of the region by ECA Anticorruption team. The conceptual basis for the report was the World Bank report Anticorruption in Transition - A Contribution to the Policy Debate. The opinions expressed in this report do not necessarily represent the views of the World Bank or its member governments. The World Bank does not guarantee the accuracy of the data included in this publication and accepts no responsibility whatsoever for any consequence of their use. The author is very grateful for leadership, advice and support provided by Shekhar Shah, for comments based on invaluable experience gained during anti-corruption work in several countries of the region received from James Anderson and Randi Ryterman, for guidance and comments provided at all stages of the project by Sebnem Akkaya, for a great opportunity to learn from the publications prepared by the ECA anti-corruption team under the leadership of Sanjay Pradhan, for detailed comments received from Robert Beschel, Joel Turkewitz, and Pablo Zoido-Lobaton (peer reviewers), and for support received from Christiaan Poortman, Elaine Patterson, and Simon Gray. ii TABLE OF CONTENTS TABLE OF CONTENTS .............................................................................................................................................i EXECUTIVE SUMMARY.........................................................................................................................................1 The Pattern Of Corruption In Bosnia And Herzegovina.....................................................................................2 High Level Of Public Concern With Corruption .............................................................................................2 Low Level Of Public Trust In The Governments.............................................................................................2 State Capture and Conflict of Interest................................................................................................................3 Public Administration Inefficiencies Reflected In Widespread Bribery In Public Offices ......................4 Distorted Business Environment and Impact on New Firms .........................................................................5 Unpredictability of the Legal and Policy Environment...................................................................................6 Effect on Poverty And Inequality .......................................................................................................................6 Support for Reforms ..............................................................................................................................................7 Anti-Corruption Reform Design ..............................................................................................................................8 1. INTRODUCTION..............................................................................................................................................9 2. THE PATTERN OF CORRUPTION IN BOSNIA AND HERZEGOVINA ...................................12 High Level Of Public Concern With Corruption................................................................................................12 Low level of Public Trust in the Governments....................................................................................................15 State Capture and Conflict of Interest..................................................................................................................18 Distortions In Public Institutions..........................................................................................................................19 Distorted Business Environment...........................................................................................................................25 Corruption's Impact on Poverty and Inequality.................................................................................................28 3. APPROACHES TO AN ANTI-CORRUPTION STRATEGY IN BOSNIA AND HERZEGOVINA ........................................................................................................................................................31 The Pattern of corruption and an Approach to Reform ....................................................................................31 Competitive Business Environment.......................................................................................................................32 Increasing Transparency And Accountability In The Public Sector...............................................................32 Rebuilding Public Trust By Empowering Citizens And Realizing The Potential Of Civil Society.............33 The Need For Action................................................................................................................................................34 ANNEX 1. ADDITIONAL CHARTS AND TABLES ......................................................................................36 Additional Charts On The High Level Of Public Concern With Corruption.................................................36 Additional Charts on the Low Level of Public Trust in the Governments and Negative Attitude to Corruption.................................................................................................................................................................38 Additional Charts on State Capture and Conflict of Interest..........................................................................39 Additional Charts on the Distortions In Public Institutions.............................................................................41 Additional Charts On The Cost Of Corruption To Households.......................................................................44 ANNEX 2. CORRUPTION IN THE BUSINESS ENVIRONMENT...........................................................45 Acquiring Financial Assistance from the Government .................................................................................45 Receiving Government Contracts .....................................................................................................................46 Participating In Public Tenders .........................................................................................................................47 Dealing With Courts............................................................................................................................................49 Dealing With Inspections ...................................................................................................................................51 Registering The Company..................................................................................................................................52 Obtaining Export/Import Permits......................................................................................................................53 ANNEX 3. PROCESS OF CORRUPTION.........................................................................................................57 ANNEX 4. HOW RESPONDENTS DEFINE CORRUPTION.....................................................................61 ANNEX 5. SURVEY METHODOLOGY: SAMPLING AND INTERVIEWING PROCEDURE.....64 General Public (1,200 Interviews)........................................................................................................................64 Public Official Survey (700 Interviews)...............................................................................................................65 Enterprise Survey (350 Interviews).......................................................................................................................66 ENDNOTES ..................................................................................................................................................................68 ii EXECUTIVE SUMMARY Recognizing that corruption is an impediment for the nation's economic and social development, the Government of Bosnia and Herzegovina (BiH), requested that the World Bank prepare a diagnostic study of corruption to inform the development of an anti-corruption program. The objective of this Diagnostic analysis is to identify the pattern of corruption and to make the underlying problems transparent, help distinguish priorities, help raise public awareness through the dissemination of the results, and to provide a partial foundation for the development of an anticorruption strategy and action plan. The report is based on surveys with 2,250 individuals, public officials, and enterprise managers conducted in June-July and September 2000. The urgency of anticorruption measures in BiH, where corruption may delay the reform process, cannot be overemphasized. Government "ownership" of the anti-corruption program, turning intentions into concrete, monitorable actions, are crucial for achieving progress in the transition process. In this context, the Anti-corruption Diagnostic report is intended to provide information and direction for the process of anti-corruption reform implementation. Box 1. State Capture and Administrative Corruption This report distinguishes between two main forms of corruption - state capture and administrative corruption. State capture refers to the actions of individuals, groups, or firms both in the public and private sectors to influence the formation of laws, regulations, decrees, and other government policies to their advantage as a result of the illicit and non-transparent provision of private benefits to public officials. Administrative corruption refers to the intentional imposition of distortions in the prescribed implementation of existing laws, rules, and regulations to provide advantages to either government or non-government actors as a result of the illicit and non- transparent provision of private gains to public officials. Source:Anticorruption in Transition: A Contribution to the Policy Debate. The World Bank, 2000. Executive Summary THE PATTERN OF CORRUPTION IN BOSNIA AND HERZEGOVINA The corruption pattern in BiH is characterized by (a) high level of public concern with corruption, (b) low level of public trust in the governments, (c) state capture and conflict of interest, (d) public administration inefficiencies reflected in widespread bribery in public offices, (e) distorted business environment and (f) a significant burden on poor households, exacerbating poverty and inequality. High Level Of Public Concern With Corruption Respondents believe that corruption is endangering the security of the BiH government, limiting foreign investment, causing rising inequality, and leading to increases in crime. Almost 100 percent of respondents in all three surveys said that corruption in BiH exists and more than half of them (60% of general public, 54% Chart A of public officials, and 52% Perceived Level of Corruption in Bosnia and Herzegovina of entrepreneurs) believe that corruption is very widespread 100% (Chart A). As many as 27 percent of public officials 80% admitted that in the past two 60% years they had been offered a respondents of 40% small present by a client for services provided, and 36 20% percent of public officials percentage 0% said that in the past two years public officials general public enterprise managers they had been offered either corruption exists corruption is very widespread money, or an expensive present, or a counter-service. Chart B Low Level Of Public Trust Perceptions of Dishonesty in Government In The Governments who 6.0 As a result, the level of public the trust in the governments is that 5.0 low. Only a very small think 4.0 respondents honest fraction of the respondents who to is it 3.0 believes in the honesty of the that 2.0 governments. The number of dishonest respondents is think 1.0 respondents who believe that of the governments are dishonest - ratio Council of Government of Cantonal Municipal is several times higher than government Ministers of BiH this Entity Authorities Authorities the number of those who public officials general public enterprise managers think that they are honest 2 Executive Summary (Chart B). The BiH governments will need to make serious efforts in order to rebuild public trust in authorities. The low level of trust also suggests that the issue that is essential for BiH is participation. The key will be to involve stakeholders ­ different interest groups within the government and representatives of different civil society groups ­ in the reform process. State Capture and Conflict of Interest The following outcomes of the BiH Anti-corruption Diagnostic indicate that state capture and conflict of interest are major problems in BiH: Enterprise managers interviewed consider policy and regulatory aspects of state capture the biggest problem for businesses. Out of 27 problems that the business environment can create for an enterprise, most of the respondents chose "monopolies" as the most serious obstacle. "Sale of Parliamentary votes and governmental decrees to private interests," which describes policy aspect of state capture, was also ranked among the most serious obstacles for business. Similarly, according to public officials' perceptions, policy and regulatory aspects of state capture constitute a serious problem at all government levels. Half of the respondents in the public officials survey said that accepting bribes for influencing the content of laws and other obligatory decisions - the most direct indication of state capture - is a widespread practice at the State, Entity and Local government levels. Economic conflict of Chart C interest was reported by the Conflict of Interest: Officials Running Businesses on the Side majority of respondents. Seventy-five percent of run 100% respondents think that at 80% least some of their colleagues colleagues run their own side 60% that the businesses on the side said on 40% (Chart C). Customs are at who 20% the top of the list ­ almost businesses 90 percent of respondents officials 0% of who work at customs think Health Police Taxation that these practices exist Customs Education Judiciary enterprises there. Using one's position percentage administration in the public office to Municipal State-owned protect interests of entrepreneurs with whom officials have political or other connections - another indication of economic conflict of interest - is perceived as a widespread problem by more than 50 percent of the public officials interviewed. 3 Executive Summary Other manifestations of Chart D state capture ­ political New and Small Firms Have to Pay Bribes, While Politically- conflict of interest. An connected Firms Do Not. evidence of the problems 40% related to political aspects of pay 35% to state capture can be found in 30% have responses of both enterprise 25% will managers and public respondents 20% officials to our survey of 15% companya questions. Fifty percent of 10% that enterprise managers said that 5% percentage- "competitors have access to 0% newly- small construction retailers, international with a with likelihood bribes started industry bars, cafes political pressure or politician on connections the board of to political influence that they use for relative much more likely to have to pay bribes than the average firm directors figures their own benefit" and that much less likely to have to pay bribes than the average firm this creates a serious problem for their own businesses. Half of public officials surveyed believe that receiving political payoffs for acting in favor of a political party or a political campaign is a widespread problem at all levels of the government. Unlawful political party financing is another strong indicator of the political aspect of state capture. Enterprise managers surveyed reported that up to 4 percent of their company turnover is spent on unofficial payments to political parties (Chart D). Enterprise survey data also demonstrate which types of firms are more likely to have to pay bribes according to the perceptions of the respondents. Newly-started firms appear to be at the top of the list. This implies that new firm entry, a badly needed source of growth in most transition countries including BiH, is severely hampered by corruption. Not surprisingly, the least likely to have to pay bribes are firms with political connections--it has been shown in other countries that political connections substitute for bribes. Public Administration Inefficiencies Reflected In Widespread Bribery In Public Offices The incidence of bribery among public officials within public offices is perceived by the employees of these offices to be strikingly high. The percentage of public officials who reported that certain illegal activities are present at their institutions is very high ­ 80 percent of respondents believe that bribing for positions, grading, and easy assignments is present at their organizations. Especially striking is the survey outcome demonstrating that one-third of the respondents believe that colleagues who refuse to ask for bribes or to take bribes are sometimes ostracized, victimized, or forced to leave their institution. Easy posts or assignments are attractive, and the managers are reported to distribute them in exchange for bribes. Seventy percent of respondents think that this practice exists in their institutions. Promotion is also subject to corruption. Seventy-two percent of public officials reported that in their organizations bribing is used to influence decisions on promotions or on transfers to better grading. Hiring decisions are also influenced by those 4 Executive Summary who are willing to pay for getting a job in a public office. Seventy-three percent of officials reported that bribery for being hired exists in their offices. There is a high positive Chart E relationship between bribery Level of Bribery and Transparency of Personnel Policies* and public administration inefficiencies. The link that 50% between the quality of institution 40% public administration and believing an corruption is clear - in a in theirs 30% professional, meritocratic, as officials 20% efficient public office, of such widespread 10% where public servants are is 0% well-paid, incentives to percentage low medium high bribery engage in bribery are low. level of meritocracy and transparency in personnel policy Offices that are * measured as the mean of three characteristics: personnel decisions are clear to the employees, personnel decisions are based on written personnel policy guidelines, decisions are based on professional experience, merit, characterized by the performance employees as having low level of meritocracy and insufficient clarity and transparency in personnel policies, are also described as more corrupt. Out of all indicators, clarity of personnel policies is the strongest indicator of inefficiencies (Chart E). In the public offices where personnel decisions are based on the criteria specified in written personnel policy guidelines, level of perceived bribery is low. There also is a positive relationship between nepotism and bribery. Analysis also shows that when the personnel policy in a public office is influenced by political connections, political affiliations, or political pressure, bribery is likely to be high. Distorted Business Environment and Impact on New Firms Corruption imposes high transaction costs for businesses. According to the information from the enterprise survey, every month businesses pay bribes in the form of gifts, tips, and money to representatives of public offices in amounts equal to18 Chart F percent of expenses of the average The Proportion of Bribes in Firm Expenditures * firm paying bribes or eight percent of the of 35% expenses of all firms (Chart F). The 30% average "standard rate" for bribes that 25% businesses report paying for expenses 20% connection to utilities, obtaining 15% licenses and permits, dealing with monthly 10% firm taxes and tax collection, gaining of 5% government contracts, dealing with 0% customs/import, dealing with courts, all percentage small state FDI and dealing with health/sanitation/fire /large private with noFDI as 1998) in1996-98 in1992-95 1992 with medium inspectors ranges from 120 to 609 after before founded founded founded KM. bribes (founded new * data refers to firms that paid bribes 5 Executive Summary Managers of the average business firm spend 10.5 percent of their time on communicating with public officials from various agencies trying to resolve bureaucratic issues. Not surprisingly, it is the new firms1 that must spend more time dealing with government bureaucracy. Managers of new firms said they must spend more than 20 percent of their working time dealing with government officials. Entry of new firms, the engine of growth, employment and income generation, is hindered by the interconnected obstacles of corruption and bureaucracy. Unpredictability of the Legal and Policy Environment Firms in all countries prefer stable, predictable environments in which to do business. Unpredictability of regulatory and legal decisions discourages investment, both domestic and foreign, and stymies the development of competitive business environment. In BiH, 65 percent of enterprise managers interviewed thought that changes in the regulatory environment are unpredictable. Other survey data show the link between this unpredictability and corruption: 67 percent of public officials interviewed said that corruption helps bias regulatory/legal decisions. Public enterprises and firms with foreign direct investment reported somewhat lower levels of unpredictability and new firms, small and private firms are at the other end of the scale. But the more striking result is that firms of all sorts reported high levels of unpredictability ­ changes in government economic policies are perceived as unpredictable by more than two-thirds of the interviewed. Effect on Poverty And Inequality When asked about the main Chart G consequences of corruption, 60 percent Poor Households are Hit Hardest by Bribery of respondents in the general public survey indicated "the poor get poorer and the rich get richer." This 7 scale) perception is supported by the actual annual 6 experiences with corruption reported in 5 in the surveys. A large number of (linear-log 4 households encounter corruption when bribes of 3 dealing with public officials: twenty income 2 percent of respondents who received share 1 services in health care, education, mean 0 police, judiciary, and some other household 6 7 8 9 10 11 institutions in the past year, paid bribes annual household income (linear-log scale) (money, gifts, or counter-services). These informal payments shrink household budgets,2 distort the structure of household budgets and make budget spending less predictable. The survey data suggest that unofficial payments in BiH affect poor families more than non-poor households. Chart G demonstrates that the lower the household income, the higher the share of household income paid in bribes. 6 Executive Summary Corruption has an adverse impact on the poor for several reasons. In many cases bribes are flat lump-sum payments and thus constitute a larger portion of poor families' budgets than of the budgets of more affluent households. Moreover, corruption makes public services less affordable for the poor and thus reduces accessibility of the services for poor families. Many poor families may even have to stop using public services, which leads to further reduction in their living standards. Although informal payments may seem similar to co-payments or implicit taxes, there are several key differences. Whereas co-payments or taxes may flow into institutional budgets and contribute to improved quality, unofficial payments simply flow into the pockets of the officials that demand them. Indeed, the quality of services may be deliberately worsened in an effort by some officials to increase the incentives of the public to offer bribes. Support for Reforms It is crucial for the anticorruption program success that the reforms are supported by civil servant who will be implementing them. Survey outcomes indicate that there is a strong support among public officials for various public sector reforms that are at the core of the BiH's anticorruption program, including civil service reform, transparency in political party financing, and civil society participation in government decision making ­ 60-80 percent of respondents said that they would support reforms in these areas. The most striking survey outcome is that reforms aimed at creating meritocratic civil service career gained more support than the rest of reforms ­ more than 80 percent of respondents are in favor of them as a matter of priority (Chart H). Chart H Support for Reforms Meritocracy: promotions are based on performance indicators A law ensuring transparency in political party financing Civil society monitoring of public sector activities Simplification and clarity of public administration rules A law on free public access to government information Privatization of more public services Decrease of the number of public officials, increase salaries 0% 20% 40% 60% 80% 100% percentage of officials 7 Executive Summary ANTI-CORRUPTION REFORMDESIGN The pattern of corruption that BiH Anticorruption Diagnostic reflects ­ high level of public concern with corruption combined with low trust in the governments, state capture and conflict of interest, public administration inefficiencies reflected in widespread bribery in public offices, distorted business environment ­ is difficult to tackle. This complex pattern of corruption needs to be addressed because without doing so the success of other structural reforms is likely to be low. This will also take time, a good deal of political commitment to tackling the problem, and the choice of sound entry points. Fortunately, the Government has already recognized the importance of this problem and the strong level of public concern about corruption suggests that a credible policy of fighting corruption will have the support of the public. To be successful, the strategy for reducing corruption in BiH should emphasize policy and institutional reforms that are aimed at: · Increasing transparency and accountability in the public sector. A start could be made in a number of areas - transparency in financial management, inclusive budgets, credible downstream controls over public sector operations, conflict of interest prohibitions, de-politicization of public administration, increasing accessibility of government decision making to the public. Transparent mechanism of government decision making and execution, which is supported by a professional and efficient civil service and open to public monitoring, is an essential goal to strive towards in order to reduce opportunities for both state capture and administrative corruption. · Introducing greater competition in the private sector, reducing regulatory burden on firms. De-concentration, competitive restructuring, and reducing barriers to entry are essential measures against state capture when it is supported by powerful monopolies and when anti-competitive barriers are created resulting in rents to a small number of privileged firms. · Rebuilding public trust by empowering citizens and increasing potential of civil society to form organizations that can participate in policymaking and institutional reforms. The key challenge is to involve stakeholders within the governments and outside of the governments in the reform process. Partnerships need to be created between the governments and civil society constituencies. Citizens can participate in the policy process by voting, joining a political party, or influencing the government through civil society groups that support certain interests and are organized to lobby the government. These groups play an important role ­ their aim is to represent civil society to influence policymaking collectively and transparently. 8 1. INTRODUCTION 1. The importance of controlling corruption is increasingly being recognized in countries around the world for the harmful impact corruption poses for economic, social, and political processes. Corruption deters foreign and domestic investment and inhibits the development of the private sector. Government budgets can become distorted through insufficient tax collection, diversion of budgetary funds, and overspending due to corrupt procurement. In addition to economic costs, corruption is associated with inequitable growth, poverty, social exclusion, lack of trust in authorities, and other political and social costs that endanger development further. 2. Recognizing that corruption is an impediment for the nation's economic and social development, the Government of Bosnia and Herzegovina (BiH), requested that the World Bank prepare a diagnostic study of corruption to inform the development of an anti-corruption program. The objective of this diagnostic analysis is to identify the pattern of corruption and to make the underlying problems transparent, help distinguish priorities, help raise public awareness through the dissemination of the results, and to provide a partial foundation for the development of an anticorruption strategy and action plan. The report is based on surveys conducted in BiH by the international survey research firm ACNielsen and their local counterpart, PRISM Research, in June-July and September of 2000. Three surveys were implemented, each provide information on different aspects of corruption and governance: 700 public officials, 350 enterprise managers, and 1,200 individuals (general public) were interviewed. The quantitative approach of the surveys was supplemented by the qualitative information of nine focus group discussions. 3. Transition countries are especially vulnerable to corruption due to the fact that they are undergoing institutional restructuring, when substantial changes in the rules that govern behavior of economic actors take place. This situation creates both incentives and opportunities for corruption. When public institutions are not functioning under clear and transparent rules and mechanisms that guarantee accountability to the public they can easily become corrupt and be used for private gain. 4. Two recent cross-country studies highlight the need to address corruption in transition economies and in the Balkans.3 The 1999-2000 Business Environment and Enterprise Performance Survey (BEEPS) conducted by the World Bank and the European Bank for Reconstruction and Development demonstrates that BiH is among the third of transition countries with the highest levels of state capture. The level of administrative corruption, approximated by the percentage of firm revenues paid in bribes, is not as high as in most of the Former Soviet Union countries, yet one of the highest in Central and Eastern Europe. The second study published in March 2001 by the International 1. Introduction Development Law Institute in Rome and Bulgaria's Center for Study of Democracy (CSD) covered Albania, Bosnia and Herzegovina, Bulgaria, Croatia, Macedonia, Romania and Yugoslavia. The graft index it employs, based on respondents' assessment of the extent of corruption, places BiH toward the middle of the group of countries studied - after Romania, Albania, Serbia, and Bulgaria, where corruption levels are higher. Box 1. State Capture and Administrative Corruption This report defines corruption as the use of public office for private gain or for the gain of groups closely affiliated with public officials. It further distinguishes between two main forms of corruption - state capture and administrative corruption. State capture refers to the actions of individuals, groups, or firms both in the public and private sectors to influence the formation of laws, regulations, decrees, and other government policies to their advantage as a result of the illicit and non-transparent provision of private benefits to public officials. Administrative corruption refers to the intentional imposition of distortions in the prescribed implementation of existing laws, rules, and regulations to provide advantages to government or non-government actors as a result of the illicit and non-transparent provision of private gains to public officials. Source: Anticorruption in Transition: A Contribution to the Policy Debate. The World Bank, 2000. 5. While cross-country studies give information on the relative levels of corruption in different countries, dwelling too much on the relative rankings can sometimes take attention away from the fact that corruption is a serious domestic problem that must be addressed. It can also distract attention away from a more detailed analysis of the unique corruption pattern in each country that is required to form the basis for a reform strategy. The diagnostic surveys on which this report is based help provide the country-specific information needed to understand this pattern. Nevertheless, while the surveys provide valuable information, they can only tell a part of the story. The development of an anticorruption strategy should be informed by other views and analyses from within the government, from civil society, and from the international community. 6. The BiH authorities have recognized that weaknesses in the institutional and policy environment create opportunities for corruption, and the newly created governments have stated that anti-corruption programs will be essential for their reform agenda.4 The Republika Srpska anti-corruption group is considering systemic measures against corruption based on internationally accepted corruption prevention instruments such as strengthening of legislative and legal state, promotion of transparency and integrity of business operations, policy reform, institution building, and participatory approach to anti-corruption program implementation and monitoring. It envisages undertaking 20 concrete immediate actions that include adoption of the draft anticorruption law and of the law on prevention of money laundering. One of the first actions in this direction was taken on March 14 when the Republika Srpska Government determined a draft law on free access to information in the Republika Srpska. 7. A similar approach is expected to be announced by the recently formed (on March 12, 2001) Federation Government. The anti-corruption programs of the Entities will be 10 1. Introduction coordinated at the State level. The Chairman of the BiH Council of Ministers announced that fighting against corruption (with the emphasis on institutional reforms such as judicial reform, development of Entity Treasury systems, transparency in public procurement) is the second most important item of the reform program. 8. The urgency of anticorruption measures in BiH, where state capture as well as distortions in the functioning of public offices leading to administrative corruption delay the reform process, cannot be overemphasized. Real commitment of the governments to implement anti-corruption reforms and government ownership of the anti-corruption program, turning intentions into concrete, monitorable actions, are crucial for achieving progress in the transition process. In this context, the Anti-corruption Diagnostic report is intended to provide information and direction for the process of anti-corruption reform implementation. 9. Section 2 describes the pattern of corruption in BiH as revealed by the surveys. The pattern of corruption highlights the costs that corruption is imposing on BiH and suggests key priorities for reform. The pattern of corruption is characterized by (i) a high level of public concern with corruption, (ii) a low level of public trust in the governments; (iii) state capture and conflict of interest; (iv) public administration inefficiencies and widespread bribery in public offices; (v) a distorted business environment; (vi) unpredictability in the legal and policy environment; and (vii) a link between corruption and poverty and inequality. Section 3 describes an approach to reducing corruption. 11 2. THE PATTERN OF CORRUPTION IN BOSNIA AND HERZEGOVINA 10. The corruption pattern in BiH is characterized by (a) high level of public concern with corruption, (b) low level of public trust in the governments, (c) state capture and conflict of interest, (d) public administration inefficiencies reflected in widespread bribery in public offices, (e) distorted business environment, (f) a significant burden on poor households, exacerbating poverty and inequality. HIGH LEVEL OF PUBLIC CONCERN WITH CORRUPTION 11. The success of an anti-corruption program depends significantly on how well the understanding of what corruption means for the average citizen is embedded in the reform design. This understanding should be based on information combining citizens' experience with corruption, their perceptions of how widespread and dangerous corruption is for the society, and their willingness to fight corruption.5 The survey results make it clear that citizens view corruption as a process that is seriously damaging society. Moreover, there is a consensus in society that corruption has severe consequences. The respondents believe that corruption is endangering the security of the country, limiting foreign investment, causing rising inequality, and leading to increases in crime. 12. The majority of respondents perceive the Chart 1 corruption level 6in BiH as Perceived Level of Corruption in Bosnia and Herzegovina high, and different types of experience reported in the 100% surveys by public officials, business managers, and 80% general public confirm their 60% perceptions. Almost 100 respondents percent of respondents said of 40% that corruption in BiH exists 20% and more than half of those percentage interviewed (60% of general 0% public officials general public enterprise managers public, 54% of public officials, and 52% of corruption exists corruption is very widespread enterprise managers) said that corruption in BiH is very widespread. (Chart 1).7 2. Pattern of Corruption 13. Survey data on respondents' actual experience with corruption support survey outcome on perceived high level of overall corruption. Both recipients and providers of public services were asked about their experience with bribes in service delivery. Data from the general public survey show that while during the last three years, 11 percent of respondents in the general public survey were asked by public officials to make unofficial payments (extra money, gifts or services) for public, educational, or medical services, during only two months before the survey was conducted, 7.4 percent of all respondents had this experience. Regarding how many people actually paid bribes notwithstanding whether the officials indicated that the bribe would be accepted, survey data show that in the two months prior to the survey, 8 percent of the respondents paid bribes, and during the last three years 17 percent of all respondents paid bribes. Similar answers are given by enterprise managers (Annex 1, Chart A1.1). We also asked only those who received services during the last year if they paid bribes. Responses to this question are consistent with the previous result ­ 20 percent of those who received health care, education, police, and some other services in the last year said that they had paid bribes for them. 14. To draw a more complete picture of the level of corruption in BiH, we also asked providers of public services about their experience with corruption. In the past two years, as many as 27 percent of public officials were offered a small present by a client for services provided, and 36 percent of public officials were offered either money, or an expensive present, or a counter-service. 15. Assessing trends in corruption is easier when the data for previous years are available. Unfortunately, we do not have the data for previous years for BiH. To substitute for the past data, we included in our surveys the following question: "When was the highest level of corruption in BiH: before the war, during the war 1992-95, in 1995-98, after 1998, or was it the same in all the periods?" The largest group of respondents - more than one-third of those interviewed - reported that corruption level in the country is the same today as it was in the past (Annex 1, Chart A1.2). Those who think that corruption is getting worse with time and reached the highest level in the last two years, more often than others say that corruption is "very widespread" in BiH today.8 16. Respondents' perception of how the situation will change in the next three years suggests that none of the three groups interviewed anticipates that things will be getting worse. At the same time, people surveyed had mixed expectations about the future. The largest group, with more than one-third of the respondents, believed that the level of corruption in a three-year time frame would be the same as today, though more than 20 percent also believed that corruption would be lower in two years (Annex 1, Chart A1.3). 17. Almost all respondents (97-99 percent in different surveys) agreed that corruption leads to very serious consequences. Respondents were asked to choose three main options from the list of eleven possible consequences or to suggest another answer to the question "What are the three main consequences of corruption in BiH?". The dangers of corruption that were marked by most of the respondents were: rising inequality, increase in crime, endangering security of the governments, moral decline in society (Chart 2). 13 2. Pattern of Corruption 18. The way respondents view consequences of corruption is broadly similar among income groups. There are also some differences: people in the highest-income group stress the danger of decline in foreign direct investment while no other income group names it among the most dangerous consequences of corruption. The respondents from the highest-income group, unlike others, do not put rising inequality on the list of most negative outcomes of corruption. At the same time, respondents who assess their financial status as "on the edge of survival" put economic inequality and human rights at the top of the list of dangers that corruption entails. Employees of the government sector put more weight on "decline in moral standards" than two other employment Chart 2 groups ­ enterprise owners and private Perceptions of Consequences of Corruption sector employees. The poor will get poorer and the rich 19. When answering the survey even richer question about efficacy of different anti- Crime increase in Bosnia and corruption measures, most of respondents Herzegovina preferred enforcement measures over It endangers the security of the state long-term institutional measures. The top three measures chosen by most of the Decline in moral standards respondents were severe punishment of those taking bribes, publishing corruption Human rights get infringed cases, and controlling public officials (Annex 1, Chart A1.4). Although Foreign investors will lose their interest to invest in Bosnia and Herzegovina enforcing punishment for corrupt activities is necessary, enforcement Moral decline of the society measures are aimed against individual corrupt actions rather than at the causes The number of those citizens believing that honesty works ­declines of corruption, and therefore are not usually sufficient. The fact that measures The loss of confidence in one's own abilities ­ the rule of money such as increasing transparency, setting up administrative rules, reducing The development of the private sector gets retarded government interventions in the economy, and promoting citizen The transition process gets retarded education are not cited prominently by respondents suggests a strong role for 0 20 40 60 80 public education, as well as the need for percentage of respondents in the general public the government to play a leadership role. survey Anti-corruption programs should address the root causes of corruption, which are embedded in the norms of behavior of economic actors and which can be addressed only if these norms are changed in the process of the institutional reforms. 14 2. Pattern of Corruption LOWLEVELOF PUBLIC TRUST IN THE GOVERNMENTS 20. The surveys clearly demonstrate that the level of public trust in the governments is low.9 At the same time, the analysis of the survey data shows that there is willingness in the BiH society to improve the situation and to achieve reduction in the level of corruption. The BiH governments will need to make serious efforts in order to rebuild public trust in authorities. The low level of trust also suggests that the issue that is essential for BiH is participation. The key will be to involve stakeholders ­ different interest groups within the government and representatives of different civil society groups ­ in the reform process. Only when the reforms are "owned" by the government and when partnerships are created between the governments and different civil society constituencies that progress in the anti-corruption efforts can be achieved. 21. According to survey outcomes, only a very small fraction of the population believes in the honesty of the governments. About 40 percent of respondents in both general public and public officials surveys perceive governments as dishonest and have little faith in it. Cantonal governments received lowest assessment by the respondents. Municipal authorities are considered more trustworthy than other levels of the government. The number of respondents who think that the governments are dishonest is several times higher than the number of those who trust the government (Chart 3). 22. At the same time, almost Chart 3 half of the respondents in both Perceptions of Dishonesty in Government surveys trust that all three levels of government are willing to solve corruption who 6.0 the problems. A possible that 5.0 explanation of the difference in think 4.0 respondents honest answers to these two questions who to is it 3.0 is that the public believes that that 2.0 there are individuals at all dishonest is think levels of government who are respondents 1.0 of honest and willing to - ratio implement anti-corruption Council of Government of Cantonal Municipal government Ministers of BiH this Entity Authorities Authorities measures, although corruption public officials general public enterprise managers is widespread in those institutions. It is a good sign that the public trusts in the willingness of government officials to reform institutions, and it is important for the government not to disappoint the public once the reforms are started. It is useful to note that government willingness to implement anti-corruption measures is perceived to be the same at the local level, at the Entity level, and at the State level (Chart 4). 23. Public support for the implementation of anti-corruption measures often depends on how acceptable or unacceptable the public finds corruption. A common view is that in many countries in the region, including BiH, the public is resigned to corruption as a stable cultural phenomenon that can only be reduced when a range of policy and institutional reforms is successfully implemented over a long period of time. 15 2. Pattern of Corruption 24. The survey data do show that 73 percent of the general Chart 4 public respondents and 71 Perceptions of the Willingness of the Governments to Solve Corruption Problem percent of public officials agree with the statement that in BiH "corruption is a part of 100% culture and mentality". 90% 80% Answering the question 70% "Would you take bribes if 60% respondents 50% offered?", 26 percent of of 40% respondents in public officials 30% 20% survey and 28 percent of percentage 10% those interviewed for the 0% Council of Ministers of BiH Government of this Entity Local municipal authorities general public survey said that they "would take bribes public officials general public entreprise managers on specific occasions." At the Chart 5 same time, there is a fairly Willingness to Report Corruption Cases strong belief among all If somebody asks you for a bribe, would you report it, publish, or groups surveyed, particularly complain somewhere? enterprises, that the governments are willing to solve corruption problems 100% 90% (Chart 4). 80% 70% 60% 25. Measures against respondents 50% of corruption include as an 40% 30% essential component 20% willingness of the public to percentage 10% 0% report corruption cases. If yes it depends on the circumstances no the public refuses to report public officials general public enterprise managers corruption cases to the responsible authorities, low- level corruption is likely to Chart 6 persist. Corruption usually Percentage of Respondents Who Reported to Authorities Cases occurs behind closed doors, When They Were Asked for Bribes (extra money, gifts, services) and nobody else but for Government, Public, Educational, or Medical Services participating parties are 100% present. Fortunately, the 90% outcomes of the surveys 80% 70% indicate that more than half of 60% the respondents are willing to respondents 50% of 40% report incidents of corruption. 30% The percentage of respondents 20% 10% who said that if somebody asks percentage 0% them for a bribe they would in the past two months in the past three years report it regardless of any general public enterprise managers circumstances is the highest 16 2. Pattern of Corruption among the enterprise managers ­ it almost reaches 50 percent of those surveyed in this group (Chart 5).10 Results also show that 16 percent of those who were asked for bribes actually reported them (Chart 6). The message that comes through is that there is room for public support of measures against corruption in public services and that widening opportunities of customers of public services to report bribing in service provision can be helpful as a mean of reducing the level of bribery in public offices.11 26. It is also crucial for the anticorruption program success that the reforms are supported by civil servant who will be implementing them. Survey outcomes indicate that there is a strong support among public officials for various public sector reforms that are at the core of the BiH's anticorruption program, including civil service reform, transparency in political party financing, and civil society participation in government decision making ­ 60-80 percent of respondents said that they would support reforms in these areas. The most striking survey outcome is that reforms aimed at creating meritocratic civil service career gained more support than the rest of reforms ­ more than 80 percent of respondents are in favor of them as a matter of priority (Chart 7). Chart 7 Support for Reforms Meritocracy: promotions are based on performance indicators A law ensuring transparency in political party financing Civil society monitoring of public sector activities Simplification and clarity of public administration rules A law on free public access to government information Privatization of more public services Decrease of the number of public officials, increase salaries 0% 20% 40% 60% 80% 100% percentage of officials 27. Poor service delivery creates opportunities for corruption because consumers have incentives to pay bribes in order to receive better services - whether it is faster service delivery, simplification of the procedure, good-quality service, or a decision of the official in their favor. Willingness to report poor service delivery as opposed to bribing for preferred service would support government efforts to introduce institutional measures leading to higher quality of services and reduced incentives for corruption. Almost half of the respondents indicated that they would complain about public services if they have reasons to do so (Annex 1, Chart A1.5). 17 2. Pattern of Corruption STATE CAPTURE AND CONFLICTOF INTEREST 28. The following outcomes of the BiH Anti-corruption Diagnostic indicate that state capture and conflict of interest are major problems in BiH: 29. Enterprise managers interviewed consider policy and regulatory aspects of state capture ("sale of votes to private interests", "monopolies") the biggest problem for businesses. Out of 27 problems that the business environment can create for an enterprise, most of the respondents chose "monopolies" as the most serious obstacle. "Sale of Parliamentary votes and governmental decrees to private interests," which describes policy aspect of state capture, was also ranked among the most serious obstacles for business. Chart 8 Surprisingly, the last Conflict of Interest: Officials Running Businesses on the Side positions in the ranking belongs to "corruption of run 100% regulatory and administrative authorities", and "rules to 80% colleagues obtain permits and to that side 60% establish an enterprise" the said on 40% (Annex 1, Chart A1.6). who 20% 30. Similarly, according to officials businesses of 0% public officials' perceptions, Health Police policy and regulatory aspects Customs Education Taxation Judiciary enterprises of state capture constitute a percentage administration serious problem at all State-owned Municipal government levels. Half of the respondents in the public officials survey said that accepting bribes for influencing the content of laws and other obligatory decisions - the most direct indication of state capture - is a widespread practice at the State, Entity and Local government levels. 31. Economic Conflict of interest was reported by the majority of respondents. One of the basic rules that are Chart 9 accepted by many countries New and Small Firms Have to Pay Bribes, While Politically- in order to protect connected Firms Do Not. government service from 40% corruption is the separation of pay 35% to business activity and public 30% service or, in other words, the have 25% will control of economic conflicts respondents 20% of interest. In our survey, we of 15% asked public officials if their companya 10% that colleagues own businesses. 5% percentage- 0% Seventy-five percent of newly- small construction retailers, international with a with likelihood bribes started industry bars, cafes respondents think that at least politician on connections the board of to political some of their colleagues run relative much more likely to have to pay bribes than the average firm directors figures their own businesses on the much less likely to have to pay bribes than the average firm 18 2. Pattern of Corruption side (Chart 8). Customs are at the top of the list ­ almost 90 percent of respondents who work at customs think that these practices exist there.12 Using one's position in the public office to protect interests of entrepreneurs with whom officials have political or other connections - another indication of economic conflict of interest - is perceived as a widespread problem by more than 50 percent of the public officials interviewed. 32. Other manifestations of state capture ­ political conflict of interest. Unlawful political party financing is another strong indicator of state capture. Enterprise managers surveyed reported that up to 4 percent of their company turnover is spent on unofficial payments to political parties (Annex 1, Chart A1.7).13 Enterprise survey data also demonstrate which types of firms are more likely to have to pay bribes according to the perceptions of the respondents. Newly-started firms appear to be at the top of the list. This reflects the fact that the new firms are at the stage of business registration, obtaining licenses and permits, moving into new offices, and so forth, when they have to deal with the authorities a lot. It also implies that new firm entry, a badly needed source of growth in most transition countries including BiH, is severely hampered by corruption. Not surprisingly, the least likely to have to pay bribes are firms with political connections--it has been shown in other countries that political connections substitute for bribes (Chart 9)14. 33. An evidence of the problems related to political aspects of state capture can be found in responses of both enterprise managers and public officials to our survey questions. Fifty percent of enterprise managers said that "competitors have access to political pressure or influence that they use for their own benefit" and that this creates a serious problem for their own businesses. Half of public officials surveyed believe that receiving political payoffs for acting in favor of a political party or a political campaign is a widespread problem at all levels of the government. All government levels received similar assessments of their involvement in state capture from the public officials interviewed, with a slightly less critical view about local level governments. On the average, one-quarter of public officials surveyed believe that different types of state capture are considerable in their institutions, with most prominent examples including accepting bribes ­ particularly from banks and public enterprises - in return for influencing the content of laws and other obligatory decisions, for privileges in public tenders, and for protecting interests of entrepreneurs with whom they have political connections.15 DISTORTIONS IN PUBLIC INSTITUTIONS 34. Corruption characterizes not only client-service provider relationships, but also interactions among public officials. Incidence of corrupt practices within public office points out distortion of administrative systems in public institutions. 35. The incidence of bribery among officials working in public offices is perceived by the employees of these offices to be strikingly high. Most public officials reported that certain illegal activities are present at their institutions ­ 80 percent of respondents 19 2. Pattern of Corruption believe that bribing for positions, grading, and easy assignments is present in their organizations. Especially striking is the survey result showing that one-third of the respondents believe that colleagues who refuse to ask for bribes or to take bribes are sometimes ostracized, victimized, or forced to leave their institution. In public enterprises, 37 percent of respondents confirmed that this practice exists there16 (Chart 10).17 36. Easy posts or assignments are attractive, and managers are reported to distribute them in exchange for bribes - 70 percent of respondents think that this practice exists in their institutions (Annex 1, Chart A1.10).18 Promotion is also subject to corruption--72 percent of public officials surveyed reported that in their organizations bribing is used to influence decisions on promotions or on transfers to better grading (Annex 1, Chart A1.11).19 Hiring decisions are also influenced by those who are willing to pay for getting a job in a public office --73 percent of officials reported that bribes are offered for being hired in their offices (Annex 1, Chart A1.12).20 Health, public enterprises, customs, and municipal administrations were the most vulnerable to this form of corruption. 37. While an assessment of Chart 10 the level of corruption in the Fear of Retributions country is important to Percent of officials who say colleagues who refuse to take bribes face understand the degree of the retributions (are ostracized, victimized or forced to leave the unit) problem, it does not direct 100% us toward solutions. In order 80% to be successful, an anti- corruption reform program 60% respondents should be focused, and of 40% actions should be carefully 20% sequenced. One approach to percentage 0% sequencing anti-corruption actions is to start Health Police Customs Education Taxation Judiciary institutional restructuring enterprises administration with the sectors or owned State Municipal government bodies where corruption-related problems Chart 11 are more worrisome than in Offers of Gifts, Services and Money are Common other parts of the government. For this, two 70% information about the level counter- past 60% of institutional problems in the 50% money, in different parts of the 40% 30% government would be offered present years20% useful. 10% officials of expensive 0% an 38. In the public officials or Police Health Bank Taxation Customs Judiciary survey, several questions Education percentage administration enterprises were used to evaluate service owned Municipal State corruption-related 20 2. Pattern of Corruption institutional problems in different sectors. The number of officials who had been offered bribes during the last two years, gives us a ranking of agencies (Chart 11). At the top of the list are tax authorities and Customs Administration. Almost 60 percent of respondents from these agencies reported that they had been offered bribes in the last two years. It is important to note that this indicator does not measure the level of corruption per se, but helps to see the degree of public sector distortions, which might be the responsibility of particular public sector institutions, but also might reveal more general problems with governance. 39. To look at the same problem from Chart 12 another perspective, we asked enterprise Firms' Experiences With Bribery managers about their experience with corruption in services that they obtained Environmental protection from different institutions. Information Construction permissions presented in Chart 12 reflects their answers and shows that enterprise Traffic police managers are more often asked to pay Assigning licenses, export/import permissions bribes by officials in environmental Business authority protection offices, by officials issuing Notary offices construction permits and by traffic police. Certification authority Courts 40. Ranking of government bodies Labor office based on perceptions of all three groups of respondents of how corrupt different Auditor services agencies are gives us the following Tax supervision, tax authority picture21: in both public officials and Fire supervision general public surveys most of Business registry respondents marked the same agencies as very corrupt - customs, Entity Customs offices government, local authorities, tax Hygiene supervision authorities, public enterprises. The next Labor security authority corrupt are perceived Cantonal authorities, presidency of BiH, and Assigning licenses Vehicle registration and official technical checks parliament. The least corrupt are of vehicles considered the army, education, and Health insurance/social insurance companies international organizations. Answers State business supervision received from enterprise managers suggest some differences in their Banking services perception of which agencies are more Telecommunications corrupt as compared to the opinions of Energy (water, gas, electricity) the officials and the general public. 0% 5% 10% 15% 20% Enterprise managers named as more percentage of enterprises asked for bribes (among those that contacted corrupt tax authorities, real estate public officials) cadastre and land commission, government, ministries, and justice 21 2. Pattern of Corruption system. The army, education, and health facilities were considered the least corrupt (Annex 1, Charts A1.13). 41. How do public officials perceive the level of corruption in the institutions where they do not work? Chart 13 shows how those who are not employed by a particular agency/sector assess the level of corruption there. When describing their own institution, respondents are likely to be less critical, and their assessment of how corrupt the institution is increases approximately from 20 to 100 percent when they are not employed by the institution that they are evaluating. At the same time, ranking of government institutions based on opinions of all respondents and ranking based on opinions of those not employed by the institution being assessed, is very much the same. This outcome shows the stability of public perceptions of how corrupt different institutions are and supports the outcome described above that lists customs, government, local authorities, tax authorities, and publicly-owned enterprises as institutions that are perceived as the most problematic by the respondents (Chart 13). 42. Is corruption more Chart 13 widespread among higher-level Public Officials' Perceptions of Corruption* staff? Most of the respondents think that senior staff are much more corrupt than lower level Customs authorities employees. However, due to the Local authorities (Opcina ­ municipality) sample structure, this might not Tax authorities reflect the real situation. The Government of this Entity proportion of senior level officials publicly owned enterprises in public institutions is low. As a Council of Ministers of B&H result, the representative sample Judiciary includes only a few people that Banks have senior level positions. Police Considering the sample Health care system composition, the message that Education comes across from this outcome is 0% 10% 20% 30% 40% 50% 60% that mid-level and low-level percentage of officials who believe that corruption is very widespread employees of public institutions consider senior managers much more corrupt than themselves (Annex 1, Chart A1.14). 43. Bribery and public administration inefficiencies. Anti-corruption reforms include as an important component civil service reforms. The link between the quality of civil service and corruption is clear - in a professional, meritocratic, efficient public office, where public servants are well-paid, incentives to engage in bribery are low. However, the cost of the full range of public administration reforms is often high and outcomes are not immediate. The data below demonstrate high positive relationship between bribery and public administration inefficiencies. This result could be helpful in locating entry points for initial actions in the area of public administration reforms.22 22 2. Pattern of Corruption 44. While most public officials do not agree with the statement that the institutions where they work cannot offer quality services or are characterized by inefficient work performance, every fourth respondent thinks that these problems are widespread in their workplace. Banks and publicly-owned enterprises are in the worst situation in this respect (Table 1). Table 1 Inefficiency and Poor Quality, as Assessed by Public Officials Themselves Percent who believe problems are widespread State owned Municipal Bank enterprises Health Judiciary Customs administra Taxation Education Police tion inefficient work performance 37% 32% 24% 20% 19% 24% 20% 23% 16% does not provide quality services 30% 21% 15% 18% 21% 22% 13% 18% 11% 45. Chart 14 shows a negative Chart 14 relationship between the level of Bribery and Transparency of Personnel Policies* bribery and the degree of meritocracy, clarity, and that 50% transparency of the personnel institution 40% believing an policy. Offices that are in theirs 30% characterized by the employees as officials 20% as having low level of of such widespread 10% meritocracy and insufficient is 0% clarity and transparency in percentage bribery low medium high personnel policies, are also level of meritocracy and transparency in personnel policy described as more corrupt. * measured as the mean of three characteristics: personnel decisions are clear to the employees, personnel decisions are based on written personnel policy guidelines, decisions are based on professional experience, merit, performance 46. Out of three indicators used Chart 15 for the index of meritocracy, Bribery and Nepotism transparency, and clarity of personnel policies, one appears 50% that to be more significant that 40% others.23 It is clarity of personnel institution believing an policies. In the public offices in 30% theirs where personnel decisions are as officials 20% based on the criteria specified in of such widespread written personnel policy is 10% guidelines, level of perceived percentage bribery 0% bribery is low. This outcome high medium-high medium-low low leads to relatively level of nepotism measured as personal and family connections-based personnel decisions straightforward reform recommendations: introduction of clear personnel guidelines in public offices. Unfortunately, in a corrupt environment the implementation of meritocratic personnel policy based on written guidelines might be blocked by the groups that are interested in maintaining status quo. However, an introduction of the guidelines could be considered 23 2. Pattern of Corruption the first reform step that would Chart 16 create opportunities for other Bribery and Politicization groups to change the 50% environment. that 40% institution 47. The relationship between believing an in 30% the level of bribery and the theirs as degree of nepotism is reflected officials 20% of such in Chart 15. It demonstrates a widespread is 10% positive relationship between percentage nepotism and bribery. It is worth bribery 0% low medium-low medium-high high mentioning that other types of political connections, party affiliation, political pressure as a basis for personnel decisions personal preferences in personnel actions - connections Chart 17 with people from the same Bribery and the Use of Unofficial Payments for Personnel region or connections within the Decisions organization - while being positively correlated with 50% that bribery, have lower level of 40% correlation with it than family institution believing an 30% connections. in theirs as officials 20% 48. When personnel policy in a of such widespread public office is influenced by is 10% political connections, political percentage bribery 0% affiliations, or political pressure, low medium-low medium-high high bribery is likely to be high bribing for personnel decisions (Chart 16). The issue of political conflict of interest was discussed Chart 18. Proportion of Bribes in Firm Expenditures * earlier as one of the major the indications of state capture. of 35% Correlation between bribery and 30% the level of politicization of 25% expenses personnel decisions 20% demonstrates that there is a 15% monthly positive statistical relationship 10% of firm between state capture and 5% 0% administrative corruption, and it all is likely that they reinforce each percentage small state FDI /large private 1992 with noFDI other. as 1998) in1996-98 in1992-95 with after before medium founded founded bribes (founded founded 49. Is there a positive new relationship between the level of * data refers to firms that paid bribes bribery for service delivery, for speeding up the delivery of the service, for deciding in favor of a specific client, or for a better service delivery, on the one hand, and the degree to which personnel decisions (hiring, promotions) are based on 24 2. Pattern of Corruption bribes, on the other hand? Chart 17 demonstrates that higher bribery in service delivery is correlated with higher bribery for personnel decisions. This means that employees in public offices with highly corrupt personnel decisions take more bribes from clients. Even if public officials in such offices don't share bribes (as our respondents indicate answering another survey question) with the managers, however, bribes for hiring and promotion can be considered a form of implicit payment for the opportunity to be able to extort bribes from clients. DISTORTED BUSINESS ENVIRONMENT 50. Although a complete assessment of business losses due to corruption goes beyond the scope of this paper, the enterprise survey is designed to estimate two types of costs that corruption imposes on businesses ­ financial burden of bribes and losses of management's time on resolving bureaucratic issues with public officials. 51. Corruption imposes high transaction costs for businesses, particularly new entrants. According to the information from the enterprise survey, every month businesses pay bribes in the form of gifts, tips, money etc. to representatives of public offices in the amount that constitutes 18 percent of expenses of the average firm paying bribes or eight percent of expenses of all firms.24 The share of bribes in expenses is the highest for public firms and firms with foreign direct investment25 and the lowest for small firms, which probably cannot afford to spend more on bribes (Chart 18). 52. The average "standard rate" for bribes that businesses report paying for several services including connection to utilities, obtaining licenses and permits, dealing with courts and customs, etc. ranges from 120 to 609 KM.26 Table 2 lists mean amounts of bribes that respondents paid for each type of service. It is worth mentioning that amounts of bribes for the same services reported by public officials are much smaller. It is difficult to explain this survey outcome without additional information, but the stark disconnect between perceptions of public officials and the actual experience of enterprise managers is worth mentioning. One explanation may be that public officials are averaging bigger- size bribes for business services from enterprise managers and more numerous smaller unofficial payments for public services from general public. Table 2 Average Size of Bribes Type of service Mean amont of bribe (KM) as reported by as reported by public officials entrepreneurs To get connected to and maintain public services (electricity, telephone) 5 288 To get licenses and permits 9 463 To deal with taxes and tax collection 7 609 To gain government contracts 11 504 To deal with customs/imports 9 601 To deal with courts 8 216 To deal with health/sanitation/fire inspectors 7 120 25 2. Pattern of Corruption 53. Managers of the average business firms spend 10.5 percent of their time on communication with public officials from various agencies trying to resolve bureaucratic issues. Different types of firms ­ small and medium/large, with foreign direct investment and without it, public and private ­ spend from 8-14 percent of their time on resolving bureaucratic matters with public officials (Chart 19). Not surprisingly, it is the new firms27 that must spend more time dealing with government bureaucracy. Managers of new firms said they must spend more than 20 percent of their working time dealing with government officials. Entry of new firms, the engine of growth, employment and income generation, is hindered by the interconnected obstacles of corruption and bureaucracy. 54. Unpredictability of legal and policy environments and corrupt regulatory and legal processes negatively affect businesses. Unpredictability of regulatory and legal decisions increases business risk and thus creates impediment to Chart 19 investment, including foreign Percentage of Working Time That Has to Be Spent With investment, and to the Officials On Bureaucratic Procedures development of competitive business environment. In BiH, 25% 20% 65 percent of enterprise time of 15% managers interviewed thought 10% that changes in the regulatory 5% environment are unpredictable percentage 0% and 70 percent considered all small state FDI changes in policy environment 1992 large private with noFDI 1998) in1996-98 in1992-95 and with unpredictable (Chart 20). Other after before medium founded founded (founded founded survey data show the link new between this unpredictability and corruption: 67 percent of public Chart 20 officials interviewed said that Unpredictability of Business Environment corruption helps bias 100% regulatory/legal decisions (Annex 1, Chart1.9). Public 80% managers enterprises and firms with 60% foreign direct investment reported somewhat lower levels enterprise 40% of of unpredictability and new 20% firms, small and private firms are 0% at the other end of the scale.28 percentage "changes in government economic and financial "changes in rules, laws, and regulations that policies that materially affect my business materially affect my business interests are interests are unpredictable" unpredictable" 55. Public enterprises and firms with foreign direct investment reported lower level of unpredictability of changes in the regulatory environment ­ more often than in other firms, the management there believes that changes in rules, laws and regulations are more or less foreseeable. New firms, small and private firms are at the other end of the scale ­ according to their answers to the same question, they perceive regulatory environment as more predictable. (Chart 21).29 26 2. Pattern of Corruption Chart 21 56. Changes in government Unpredictability of Laws, Rules, and Regulations economic policies are perceived as unpredictable by 70 percent of the 100% enterprise managers interviewed. 80% believing These changes negatively affect 60% firms founded before 1992 (old managers 40% firms) more than any other group of unpredictable are 20% companies. The least vulnerable are enterprise 0% again firms with foreign direct of changes all investment and public companies small state FDI /large private 1992 with noFDI with (Chart 22). The explanation could 1998) in1996-98 in1992-95 after before medium percentage be the same as with vulnerability to founded founded (founded founded new changes in legislation - firms with foreign direct investment and public enterprises are better- Chart 22 Unpredictability of Economic Policies connected and thus better- informed. In turn, connections as 100% well as access to information that 80% is vital for businesses open doors believing 60% for them to influencing policies. managers 40% unpredictable 57. The level of confidence in are 20% the firm's ability to "voice" their enterprise 0% of concerns related to changes in changes all small /large state private FDI laws and policies is relatively low with noFDI 1998) in1996-98 in1992-95 1992 with medium - only 23 percent of managers percentage after before founded founded (founded founded interviewed believed that their new Chart 23 opinions would be taken into Enterprise Voice in Matters of Law and Policy account by the government. Out of all firms, newly-created 50% the enterprises (those founded in 40% account 1999-2000) express much more 30% believing into confidence in that "in case of 20% important changes in laws and managers concerns 10% policies affecting my business, the their 0% government takes into account all concerns voiced by me or my enterprise takes small /large state FDI of private 1992 with noFDI with business association". The least 1998) in1996-98 in1992-95 after before medium confident in being able to voice founded founded government (founded founded percentage new their concerns to the government are firms with foreign direct investment (Chart 23). This outcome could be explained by new firms' lack of experience with these issues and consequent overestimation of their ability to participate in government decision-making processes.30 27 2. Pattern of Corruption CORRUPTION'S IMPACT ONPOVERTY AND INEQUALITY 58. Corruption has an adverse impact on the poor for several reasons. In many cases bribes are flat lump-sum payments and thus constitute a larger portion of poor families' budgets than of the budgets of more affluent households. Corruption makes budget spending less predictable, and distorts the structure of household budgets. Corruption makes public services less affordable for the poor and thus reduces accessibility of the services for poor families. Many poor families have to stop using public services, which leads to further reduction in their living standards. Corruption affects the quality of public services since bribes are often paid not only for an access to the service, but also for the services of higher quality. The survey data suggests that in BiH poor families are more affected by corruption than non-poor households. When asked about the main consequences of corruption, 60 percent of respondents in the general public survey indicated "the poor get poorer and the rich get richer." This perception is supported by the actual experiences with corruption reported in the surveys. A large number of respondents encounter corruption when dealing with public officials - 20 percent of respondents who received services in health care, education, police, judiciary, and some other institutions in the past year, paid bribes (money, gifts, or counter-services). These informal payments shrink household budgets,31 distort the structure of household budgets and make family budget spending less predictable. The survey data suggest that unofficial payments in BiH affect poor families more than non-poor families. 59. On average, 19.6 percent of respondents who received services in health care, education, police, judiciary, and some other institutions in the past year, paid bribes (money, gifts, or counter-service) (Table 3). However, those of them who need to use services frequently, have to pay a bribe each time when they receive the service, and as a result bribes put a serious burden on those families that need services often. Table 3 32 Bribery in Everyday Life number of percent of institutions respondents who respondents who received services gave bribes* Health 610 26% Education 600 7% Police 147 17% Judiciary 157 5% Other 571 5% weighted average 20% gifts, counter-services, or unofficial money for services 28 2. Pattern of Corruption 60. Although the average annual amount of bribes paid by ordinary citizens is not very high as compared to average income, the effect on the poor is notable.33 Poor families suffer from bribes more than non-poor households, especially when they have a need to obtain health or other necessary services on a regular basis. The group with the lowest income suffers the most ­ bribes constitute more than 7 percent of their annual income. Our analysis demonstrates a negative relation between household income and the share of bribes in household income.34 The lower the household income, the larger the share of bribes (Chart 24 ). 35 61. Low quality of public sector Chart 24 Poor Households are Hit Hardest by Bribery services may also be a source of increased inequality in terms of either material wealth or living standards 7 understood in more general terms. If 6 dissatisfaction with public services is annual scale) in 5 high, low-income customers would be (log 4 either forced to pay for alternative bribes private sector services, which could of income 3 seriously affect their household budget, 2 share or to accept low-quality public services, 1 which would increase the socio- mean household 0 6 7 8 9 10 11 economic distance between them and annual household income (log scale) more affluent groups. 62. The level of satisfaction with services as reported by the respondents is different depending on the type of service (some results are reflected in Chart 25). The analysis shows that respondents are highly satisfied with elementary school services and gave good assessment of secondary school services. Universities received lower evaluation. Police service was assessed as having low quality. Levels of satisfaction with private and public health services differ substantially. Private health service customers demonstrate much higher level of satisfaction with the quality of services than the customers of the public health services (Annex 1, Chart A1.15).36 It is important to emphasize that this outcome does not suggest that privatization of health services could be a solution. While solving the problem of service quality, privatization of health services would create a bigger problem of restricting access to health care for the large groups of less affluent population, and therefore does not appear to be a reasonable solution. 29 2. Pattern of Corruption Chart 25 Satisfaction With the Quality of Services General public education health care services police very very very dissatisfied very dissatisfied satisfied very very fairly 5% satisfied 7% 16% fairly satisfied dissatisfied dissatisfied 22% fairly satisfied 26% 37% 10% dissatisfied 14% 11% neutral fairly neutral 20% fairly fairly dissatisfied 10% satisfied neutral satisfied 23% 43% 21% 35% 63. These results demonstrate that there is room for improvement of public sector services that are used by general public: among institutions that we looked at, only elementary and secondary schools received high evaluation, while the quality of services that the public receives from the police, the universities, and the health sector, was considered low by the respondents. 64. In some countries of the Chart 26 region, quality of public Satisfaction With the Quality of Educators services used by the general population and amount of 12 to bribes paid for these services educators 10 are negatively correlated. The satisfied with 8 analysis suggests similar 6 negative correlation between 4 the size of bribes and dissatisfied respondents satisfaction with the public of 2 services in BiH. Respondents ratio 0 elementary school secondary school university report higher bribes paid in respondents institutions that, in their view, provide low quality services for the average customer (Chart 26). 30 3. APPROACHES TO AN ANTI-CORRUPTION STRATEGY IN BOSNIA AND HERZEGOVINA THE PATTERN OF CORRUPTION AND ANAPPROACH TO REFORM 65. The BiH Anticorruption Diagnostic reflects a pattern of corruption that is difficult to tackle. There is a high level of public concern with corruption and a low level of trust in the governments. Survey responses highlight that the general perception is that state capture is a common phenomenon. Public administration inefficiencies coexist with widespread bribery in public offices. And firms suffer from the high frequency of bribe payments and unpredictability in the business environment. This complex pattern of corruption needs to be addressed to make the success of other structural reforms possible. The task will take time and political commitment will be critical. The choice of sound entry points will be important for building credibility of the program. 66. Despite its daunting aspect, the challenge of addressing corruption and state capture becomes manageable when the problem is unbundled into its constituent parts. In BiH, many types of institutional distortions should be tackled by the anti-corruption strategy. Box 1. Designing effective anti-corruption strategies In countries with high level of state capture and relatively high, but less worrisome, level of administrative corruption, the major challenge is to break the political stranglehold on further progress in reform. Combating state capture should become the key focus of the anticorruption strategy. Technocratic reforms may serve as entry points into anticorruption work, but a reform program should be designed to broaden formal channels of political access and to increase the accountability of political leaders to a wider range of constituencies. Efforts to promote collective action among anti-corruption constituencies and competing interest groups (SME, trade associations, professional societies) should play an important role in anti-corruption strategy. In such countries, the early promotion of political reforms and partnerships with civil society has a potential to achieve a considerable impact. Targeting state capture requires measures on three fronts: (a) decreasing the gains to firms that engage in state capture (b) increasing the cost to politicians of state capture, and (c) building credible constituencies in and outside the government. Decreasing the gains to captors may entail de-concentrating vested economic interests through competitive restructuring of monopolies, reducing barriers to entry, and increasing transparency in corporate governance. Increasing the costs to politicians requires efforts to foster collective action among potential countervailing interests, such as small and medium-size enterprises, to obtain political access. The goal is to increase the number and diversity of economic actors competing through more transparent and open conduits of political access for a limited pool of rents. Source: Anticorruption in Transition: A Contribution to Policy Debate. World Bank, Washington, 2000. 3. Approaches to Anti-Corruption Strategy The formal laws and regulations that define norms of behavior of economic agents (businesses, public agencies, individuals) need to be examined and improved where appropriate, and the inefficiency of law enforcement must be addressed. Distortions in the power structure caused by the capture of the state by special interests must play a central role in the anti-corruption strategy. The inefficiency of public organizations­ unpredictability of service delivery, lack of public information about functions and rules, arbitrary and non-transparent decision making practices within public organizations­must give way to more efficient and transparent institutions. 67. A central theme of an anticorruption strategy must be to empower civil society to participate in the policy process and to monitor the design and implementation of anticorruption reforms. Increasing public awareness and knowledge of the consequences and causes of corruption, as well as focusing on activities that help build transparency and consensus for change in the institutions of governance, are the main measures of the bottom-up approach to anti-corruption reforms. Building coalitions of civil society, government, private sector, and the media will lead to increased responsibility of the government for the reform process. Reforms that increase transparency in governance institutions and empower civil society are key for weakening the vested interests that engage in state capture. COMPETITIVE BUSINESS ENVIRONMENT 68. Reducing the regulatory burden on firms and creating a competitive business environment should be key components of an anti-corruption strategy. When state capture in concentrated sectors generates rents for a small number of privileged firms, de- concentration, competitive restructuring, and reducing barriers to entry are essential measures against state capture. The experience of other transition countries demonstrates that improved regulatory performance means more than just reducing the number of regulations in force. It also requires the definition of clear rules regarding when government intervention is appropriate and improvements in the quality of regulatory instruments. A rule-based regulatory system that relies on rules and procedures that encourage both government and private firms to take regulations seriously can foster the needed transparency in regulatory processes. INCREASING TRANSPARENCY AND ACCOUNTABILITY IN THE PUBLIC SECTOR 69. The pattern of corruption described in this report shows the importance of transparency and accountability in the public sector. There are a number of improvements in public administration that can be made. Building transparent and efficient systems of financial management can help reduce the scope for diversion of funds and other abuses. A system that generates comprehensive budgets as well as credible downstream controls (e.g. through the functioning of an effective Treasury, an independent and capable audit institutions, effective internal audit procedures and sound procurement system) removes some of the uncertainty faced by public officials and increases the potential costs of corrupt behavior. A meritocratic, de-politicized public administration similarly increases the career benefits of honesty and provides incentives 32 3. Approaches to Anti-Corruption Strategy for professional development. Whistle-blower protections can build confidence that public officials revealing abuses will not face repercussions. Obligating financial institutions to report large or suspicious transactions to government authorities can help reduce money laundering, making it difficult to conceal large-scale corruption. 70. Lack of clarity in the division of responsibilities between government institutions and different levels of government and lack of control over decision making processes within the government can lead to rent seeking and state capture, potentially hindering the reform process. Building more transparent mechanisms of government decision making and execution, supported by a professional and efficient civil service and open to public monitoring, is an essential goal to strive towards in order to reduce opportunities for both state capture and administrative corruption. 71. Clearly defining responsibilities for corruption in public office and in political processes are essential steps in the anti-corruption program. Conflict of interest provisions, as often covered under a code of ethics, clarify situations of conflict of interest, the public official's responsibility to report situations of conflict of interest, methods of reporting conflicts of interest, methods of how conflicts can be resolved, and penalties for failure to report conflicts. Codes of ethics, or related legislation, often call for disclosure of assets and income by high level public officials. Such disclosures by high-level officials are most effective when they are made public, since every interested person becomes a potential ally in discovering inconsistencies and hidden conflicts of interest.37 72. Another area in which transparency is vital is in the process of party financing. Laws in this area typically regulate the circumstances under which a politician or political party can accept a gift, favor, or donation. At minimum, they should require the politician or party to disclose to the public the sources and value of any gifts, favors, or donations they receive.38 Such laws are most effective when they also set limits on the value of gifts, favors, or donations that can be legally accepted and, in some cases, prohibit their acceptance altogether. Clear guidelines in this area provide needed guidance to public officials and citizens alike on what behavior is and is not acceptable. 73. Transparency in the relationship between politicians and economic interests is also strengthened when the rules of lobbying are formalized, clear, and err on the side of openness. Laws on lobbying establish the procedures which individuals or interest groups must use to meet with politicians. Such a law may require politicians to register the names of interest groups (including individuals representing special interests) with whom they meet, and make this registry available for public scrutiny. REBUILDING PUBLIC TRUST BYEMPOWERING CITIZENS AND REALIZING THE POTENTIAL OF CIVILSOCIETY 74. Increasing the potential of civil society to participate in policymaking and institutional reforms is critical for the reform success. The key challenge is to involve stakeholders within the governments and outside of the governments in the reform 33 3. Approaches to Anti-Corruption Strategy process--when government and civil society constituencies work in partnership they can be particularly effective, influencing policymaking collectively and transparently. 75. Citizens can participate in the policy process by voting, joining a political party, or influencing the government through civil society groups (business associations, NGOs, community groups, professional associations, labor unions, advocacy groups, etc.) that support certain interests and are organized to influence the government. These groups play an important role as intermediaries between competing interests and the government. 76. Reforms aimed at transparency in the public sector can work to empower civil society, and increase the benefits of collective action. One important way to introduce transparency in the public sector is to ensure public access to public-sector information. Such a system of free access to information typically consists of: (i) a law guaranteeing public access, (ii) regulations establishing responsible authorities, procedures for classifying information as public or private, procedures for releasing information, fees (including specifying which information will be provided free of charge), procedures for appealing denials of information, and sanctions for violating regulations; (iii) special units, in all key public-sector bodies, responsible for responding to public requests; and (iv) a low-cost system for appealing denials of information. 77. A free and independent media is closely associated with control of corruption.39 The media can be most effective in reducing corruption when there is free access to public information (see above). Moreover, at the implementation stage of an anti- corruption strategy, the media and civil society are both important allies, especially when fighting high-level corruption requires national attention and organizations willing to push for change. 78. Transparency can be further enhanced through the implementation of a sunshine act. The principle provision of such a regime is to require that meetings of key public- sector bodies are open to public observation, except when government secrets are discussed. Similarly, openness would be strengthened by a notice and comment procedure, which would require publicizing draft laws, decrees, and regulations and giving the public the opportunity to comment on the acts before they are adopted. For even greater effect, public hearings could be extended to large procurements. Consultative mechanisms such as these would allow for an open and transparent dialogue between constituencies and policies adapted to realities on the ground. THENEED FOR ACTION 79. The BiH governments have recognized the importance of reducing corruption. While, to date, considerable progress has been achieved in strengthening the fiscal management and introducing consistent and harmonized policies at the State and Entity levels, the remaining challenges BiH face are formidable. Interests that benefit from the status quo have every incentive to stall reforms. But there are also many who would benefit from the reforms. Increasing their awareness of the potential gains from addressing corruption and state capture squarely can help foment the ability for civil society to collectively push for change. Building on the heightened public awareness of 34 3. Approaches to Anti-Corruption Strategy the costs of corruption, reformers will be better able to advance an anticorruption agenda that tackles the systemic institutional weaknesses that facilitate corruption. For it is only by undertaking concrete actions, rigorously monitored by stakeholders inside and outside of government, that progress can be made. 35 ANNEX 1. ADDITIONAL CHARTS AND TABLES ADDITIONAL CHARTS ON THE HIGH LEVEL OF PUBLIC CONCERN WITH CORRUPTION The following charts provide further illustrations on the level of corruption as perceived and experienced by respondents and the trends as perceived by respondents. Chart A1.1 Level of Corruption in Bosnia and Herzegovina. Experience with Bribes. Percentage of respondents who were asked for Percentage of respondents who gave unofficial unofficial payments (extra money, gifts, or services) by payments (extra money, gifts, or services) to public public officials. officials. 20% 20% 15% 15% respondents of 10% respondents 10% of 5% 5% percentage percentage 0% 0% general public enterprise managers general public enterprise managers in the past two months in the past three years in the past two months in the past three years Chart A1.2 When Did Corruption Reach the Highest Level in Bosnia and Herzegovina? Public officials General public Entreprise managers Before During the Before Before the war, war 1992 the war, During the the war, prior to ­1995 prior to During the war 1992 prior to 1992 9% 1992 ­1995 1992 war 1992 1% 3% ­1995 13% 1% 10% Inall those After the In all periods it war 1996 ­ those Inall After the was the 1998 periods it After the those war 1996 ­ same 22% was the war 1996 ­ periods it 1998 32% same From 1998 was the 28% From1998 28% 1998 From 1998 same till now till now 22% till now 34% 35% 36% 26% Annex 1. AdditionalCharts and Tables Chart A1.3 Perceived Level of Corruption in Bosnia and Herzegovina in Three Years 40% 35% 30% 25% respondents public) 20% of 15% (general 10% percentage 5% 0% Much lower A little lower Approximately A little higher Much higher the same 37 Annex 1. AdditionalCharts and Tables ADDITIONAL CHARTS ON THE LOW LEVEL OF PUBLIC TRUST IN THE GOVERNMENTS AND NEGATIVE ATTITUDE TO CORRUPTION These charts provide further information on the opinions of respondents about the most efficient means of fighting corruption, and the willingness to complain about public services. Chart A1.4 Perceived Efficiency of Different Means to Fight Corruption To publish the corruption cases ­ giving always the name and the institution Severe punishment of those taking bribes To increase control of public and state officials To pay a better salary to public (state) employees Severe punishment of those giving bribes To professionalise the work of state officials (those should stay in their positions regardless of the election results) To improve the work of the courts Transparency, setting up rules To decrease bureaucracy Citizens enlightenment and education To reduce the public administration interventions into economy 0% 10% 20% 30% 40% 50% percentage of public officials who selected the measure as efficient 38 Annex 1. AdditionalCharts and Tables Chart A1.5 Would you Complain About Poor Public Services? 60% 50% 40% respondents 30% of 20% percentage 10% 0% yes it depends on the circumstances no public officials general public ADDITIONAL CHARTS ON STATE CAPTURE AND CONFLICT OF INTEREST These tables and charts provide further information on the perceived types of corruption in various sectors and institutions, the problems enterprises face doing business, and unofficial contributions to political parties. Table A1.1 Level of Various Forms of Corruption Including State Capture. Percentage of public officials believing that the following problems are widespread in their institution. 40 Municipal Public Banks enterprises Health admini- Taxation Judiciary Customs Education Police stration embezzlement or misuse of government funds 27% 28% 23% 14% 24% 10% 27% 14% 18% accepting bribes in return for influencing the content of laws and other obligatory decisions 30% 30% 24% 21% 20% 24% 25% 17% 19% accepting bribes in return for influencing specific government regulation decisions accepting bribes in return for speeding up 23% 30% 24% 19% 16% 16% 21% 13% 11% the process of getting licenses and permits 30% 34% 27% 25% 31% 27% 33% 20% 16% accepting bribes in return for reducing fees or taxes paid by enterprises 37% 32% 25% 27% 31% 29% 19% 18% 13% receiving political payoffs to benefit political party of campaign 30% 28% 28% 30% 18% 22% 17% 17% 16% accepting bribes in return for providing privileges in state contracting, granting concessions, etc. to private firms 43% 32% 29% 27% 22% 27% 21% 18% 16% using position to protect the interests of entrepreneurs or others with whom they have political or other connections 37% 33% 32% 27% 27% 27% 21% 17% 18% mean 32% 31% 26% 24% 24% 23% 23% 16% 16% 39 Annex 1. AdditionalCharts and Tables Chart A1.6 41 Serious Obstacles for Businesses Monopolies Sale of Parliamentary votes and Presidential decrees to private interests Shortage of loan resources Shortage of qualified workers Tax Rates Bureaucracy Slow courts Inflation Public tender conditions The non-transparent tax regulations Labour costs Rules to obtain licenses in general Rules to establish an enterprise Rules to obtain an import/export license Corruption of regulatory and administrative authorities 0% 5% 10% 15% 20% 25% 30% 35% percentage of enterprise managers who selected the characteristic as a serious obstacle for their business Chart A1.7 Unofficial Payments to Political Parties in a Company "Such as Yours" 5% 4% percentage as 3% parties 2% turnover 1% political to annual 0% of all small /large 1992 allstate state FDI private with noFDI payments 1998) in1996-98 in1992-95 large with after before medium founded founded (founded founded unofficial new 40 Annex 1. AdditionalCharts and Tables ADDITIONAL CHARTS ON THE DISTORTIONS IN PUBLIC INSTITUTIONS These charts provide further information on the frequency of various types of corruption, and the sectors or bodies that are perceived to have widespread corruption. Chart A1.8 How Common is Corruption in the Following Areas? 100% 80% officials 60% public of 40% 20% percentage 0% to obtain public to speed up to bias a to buy public to assign services delivery of public regulatory/legal contracts licenses, permits services decision corruption exists corruption is widespread Chart A1.9 Level of Bribery in an Institution "Such as Yours" 35% 30% officials 25% 20% public of 15% 10% 5% percentage 0% It does not exist It is not very It is fairly It is very It exists ­ but I widespread widespread widespread cannot judge the extent now two years ago Chart A1.10 Bribing for Posts or Assignments In an institution such as yours, are managers offered bribes for posts or assignments that are easy? 100% 80% 60% respondents of 40% 20% percentage 0% Bank Police Health Customs Judiciary Taxation Education administration enterprises Municipal State-owned 41 Annex 1. AdditionalCharts and Tables Chart A1.11 Bribing For Promotion In an institution such as yours, are managers offered bribes for promotion or transfer to better grading? 100% 80% 60% respondents of 40% 20% persentage 0% Health Police Bank Customs Education Judiciary Taxation enterprises administration State-owned Municipal Chart A1.12 Bribing For Positions In an institutions such as yours, are bribes offered to receive positions? 100% 80% 60% respondents of 40% 20% percentage 0% Bank Police Health Customs Judiciary Taxation Education administration enterprises Municipal State-owned 42 Annex 1. AdditionalCharts and Tables Chart A1.13 Perceptions of the Level of Corruption in Different Institutions Percentage believing that corruption is very widespread in this field public officials and general public enterprise managers Customs authorities Government of this entity Tax authorities Real estate cadastre (land titles office) and land Local authorities (Opcina ­ municipality) commission Tax authorities ministries State/socially owned enterprises federal parl Cantonal authorities Justice ­ courts and prosecution Presidency of B&H administration Parliament of this entity Local authorities (Opcina ­ municipality) Council of Ministers of B&H International organisations Justice ­ courts and prosecution Federal governnment Agency for privatization Presidency of B&H Private companies Private companies Public companies (Elektroprivreda, Vodoprivreda etc.) Cantonal authorities Police Police Health care system Public companies (Elektroprivreda, Vodoprivreda etc.) Banks Banks Real estate cadastre (land titles office) and land commission Customs authorities International Organizations Health care system Education Education Army Army general public public officials 0% 10% 20% 30% 40% 50% 60% 70% 0% 10% 20% 30% 40% 50% 60% Chart A1.14 Bribing Among Public Officials of Different Ranks among senior management among middle management among rank and operational staff It exists ­ It is not It does but I very It exists It exists ­ It does not exist It does It is not cannot wide- ­ but I but I not exist It is not 5% judge the spread cannot not exist very cannot 4% very extent 29% 14% wide- judge the judge wide- 19% spread extent the spread 39% 18% extent 14% 22% It is fairly It is very It is fairly It is very It is fairly wide- It is very wide- wide- wide- wide- spread wide- spread spread spread 33% spread spread 31% 33% 14% 8% 17% 43 Annex 1. AdditionalCharts and Tables ADDITIONAL CHARTS ON THE COST OFCORRUPTION TO HOUSEHOLDS This chart provide further information on the ways in which households are affected by corruption, and their assessments of the quality of services. Chart A1.15 Level of Satisfaction With the Quality of Private and State Medical Facilities Ratio of the number of respondents who were satisfied to the number of those who were dissatisfied to 100 80 satisfied 60 dissatisfied 40 respondents 20 of respondents - ratio with the care of the with the length of waiting with the facilities medical staff for medical care state medical facility private medical facility 44 ANNEX 2. CORRUPTION IN THE BUSINESS ENVIRONMENT Among the most often cited consequences of corruption in analyses of other countries is its damaging effect on the business environment. Corruption creates barriers to entry, adds to the uncertainty of the business climate and hence increases business risks, reduces investment including foreign direct investment, encourages excessive public spending and infrastructure investment, and on the whole leads to lower levels of investment and growth. In our survey, in order to determine the level and the sources of distortions in the business environment, enterprise managers were asked questions about obstacles to doing business that they face in dealings with different public institutions, about the incidence and the size of bribes and other costs that corruption imposes on firms, about certain characteristics of the business environment that increase the risks of doing business, and about corruption and other distortions that they face in communication with different public offices. Acquiring Financial Assistance from the Government Traditionally, big public enterprises in BiH relied on government transfers and/or subsidies. Efficiency was not the primary goal. Although market-based incentives were introduced during the post-war years and many managers were able to restructure old enterprises or to create new competitive businesses, support from the government is still an important issue in BiH. The analysis of the enterprise survey shows how government transfers and/or financial assistance is distributed among different types of companies and helps us understand what type of companies do and do not use them. The system of government transfers and development grants makes 58 percent of surveyed enterprises eligible for them, according to the enterprises. Small and medium public enterprises constitute the groups with the highest eligibility rate (76%) and newly-created enterprises--those created in 1999-2000--have the lowest eligibility rate (39%). One-third of the companies that are eligible for such government assistance never applied for them during the last two years. While 42 percent of eligible private firms did not apply for government support, only 10 percent of eligible large public companies and 23 percent of all eligible public companies follow this pattern of behavior. Annex 2. Corruption in the Business Environment Table A2.1 Eligibility for Subsidies and Development Grants and Support Received types of companies new founded all (founded founded in founded in large after 1996-98 1992-95 before all state private FDI no FDI 1992 state 1998) Eligible for subsidies or development grants 58% 39% 46% 56% 71% 76% 58% 50% 68% 55% Percentage of eligible firms that DID NOT apply for state subsidies in the last two years 34% 33% 41% 35% 25% 23% 10% 42% 39% 34% Percentage of eligible firms that received subsidies or development grants in the past two years 51% 67% 53% 54% 51% 43% 51% 59% 43% 53% Receiving Government Contracts Government contracts can be a substantial source of corruption. With the objective to understand what types of enterprises rely on the government budgets and to evaluate the level of bribes for government contracts, we asked our respondents if they had any contracts with the government and publicly-owned companies, what share of annual income comes from the sales under these contracts, and what percentage of the contractual value is typically given as a bribe. Government contracts are more likely to be signed with public enterprises and with firms with foreign direct investment. Approximately 50 percent of public firms' revenues come from the sales under government and government contracts. However, all other groups of companies also receive government contracts, including 29 percent of private enterprises, which have signed at least one such contract in the last two years. The analysis gives no evidence that companies of any type are discriminated in this respect (Chart A2.1). Chart A2.1 Companies That Had Government Contracts in the Last Two Years 60% 50% 40% companies of 30% 20% 10% percentage 0% all small /large FDI 1992 allstate state private with noFDI 1998) in1996-98 in1992-95 large with after before medium founded founded (founded founded new Another indicator of the enterprises' reliance on the government budget is the proportion of sales to budgetary institutions: 21 percent of respondents said that over the past two years their company sold its products / services either to a budgetary institution/company, or to an institution/company with government contribution, or to a company whose major shareholder is the government (Chart A2.2). 46 Annex 2. Corruption in the Business Environment Chart A2.2 Share of Company Income that Comes From Sales to a Budgetary Institution or a Company With a Major Shareholder From the Government 30% 25% income 20% of 15% 10% 5% percentage 0% all small /large 1992 allstate state private FDI with noFDI 1998) in1996-98 in1992-95 large with after before medium founded founded (founded founded new Similarly to the case with export and import permits (see section on permits below),42 the data suggest that there might be a standard rate for government contracts, which equals 4 percent of the contract value. The variance of the answers to the question "what percentage of the contract value is typically given as a bribe?" is low and 47 percent of respondents said that 4 percent of the contract value is paid as bribes. If there is a standard "bribe" rate for contracts, it might be tempting to view this bribe as a tax and view it as somehow less damaging for development than other types of corruption. But unlike taxes, bribes are marked by lack of transparency and therefore uncertainty. They further create unequal opportunities for businesses depending on their experiences with corruption, connections in public offices and access to those who take bribes. And unlike a tax, the proceed of bribes flow into the pockets of corrupt officials rather than the state's coffers where could be put to use in fighting poverty or achieving other social goals. As Chart A2.3 shows, firms of all types reported roughly similar levels of the "contract bribe." Chart A2.3 Bribes for Contracts as Percentage of Contractual Value 5.0% value 4.0% 3.0% contractual 2.0% of 1.0% 0.0% percentage all small /large 1992 allstate state FDI private with noFDI 1998) in1996-98 in1992-95 large with after before medium founded founded (founded founded new Participating In Public Tenders Public tenders can create opportunities for corruption if they are marred by a lack of transparency, if the information about them is not accessible or if the process of tendering is too complicated. Comparing different types of firms in how actively they participate in tenders and in how they evaluate the level of bribery in tenders, and also by analyzing the reasons for non-participation in tenders, can shed light on weaknesses surrounding tenders. 47 Annex 2. Corruption in the Business Environment The level of participation in tenders ­ percentage of firms that "tender for every contract for which they are eligible" ­ is high for all types of companies, ranging from 20 percent to 45 percent. Percentage of firms that actively participate in tenders is the highest among public enterprises and enterprises with foreign direct investment, where it equals 45 percent. All other types of companies are more or less equal in this respect (Chart A2.4). Chart A2.4 Firms That Tender For Every Contract For Which They Are Eligible 50% 40% firms of 30% 20% percentage 10% 0% Do firms pay bribes for winning tenders? 73 percent of respondents believe that firms at least sometimes have make unofficial payments in order to win a contract tender organized by the public sector and 25 percent think that bribing for tenders is the usual practice (Chart A2.5). Chart A2.5 Bribes for Contract Tenders in Public Sector 100% 80% managers 60% percentage enterprise 40% believing that of bribes are paid at 20% least sometimes 0% percentage percentage believing that bribes are usually paid Why do companies choose not to participate in public tenders? Chart A2.6 shows that unfair competition is mentioned more often than other reasons. Other reasons that respondents found important for not participating in tenders organized by public institutions are that "public procurement is not relevant for our business", "excessive prepayment is required" and "did not have information in time". For tenders organized by private institutions, important reasons also were that the firms "did not have information on time" that "sole source contracts are easier", that they "had to make too many bribes", and that "conditions for tender were not transparent". Overall, the analysis demonstrates that although the level of participation in tenders is high for all types of companies, public enterprises and firms with foreign direct investment participate almost twice as often as the average firm in tenders. The median number of contracts won is twice as big 48 Annex 2. Corruption in the Business Environment for public enterprises as it is for private companies. The level of bribery for winning a tender is considered high by the respondents. Chart A2.6 Reasons For Not Participating in Tenders organized by public institutions organized by private institutions competitionwas competitionwas unfair unfair public procurements did not have the not relevant to our information in time business expensive:prepayment sole source contracts required are easier did not have the had to make too information in time manybribes conditions of the had to make too tender were not manybribes transparent conditions of the public procurements tender were not not relevant to our transparent business no personal complex process: too connections with the many documents to organizers submit sole source contracts expensive:prepayment are easier required complex process: too no personal many documents to connections with the submit organizers 0 0.4 0.8 1.2 0 0.4 0.8 1.2 1.6 ratio of those believing that the reason is ratio of those believing that the reason is important to those considering it unimportant important to those considering it unimportant Dealing With Courts According to opinions expressed by business managers who were interviewed for the enterprise survey, slowness of the courts is the most problematic obstacle to doing business in BiH43. It was selected by 87 percent of respondents and ranks above all other 27 problems listed as negative characteristics of the business environment. Do people use courts to resolve business disputes or do they choose to use informal channels? The survey data show that businessmen use courts much more often than informal channels ­ courts were used by 41 percent of those interviewed and informal channels only by 10 percent (Chart A2.7). 49 Annex 2. Corruption in the Business Environment Chart A2.7 Court System How did you resolve problems that could be resolved through courts? 45% 40% 35% managers 30% 25% 20% enterprise of 15% 10% 5% percentage 0% through the courts through informal channels didn't pursue the matter Most of those who did not use the court system to resolve these problems cite as the main reason complicated court procedures. Another reason mentioned by several respondents is when the issue concerned "was not substantial". Those who used courts for resolving relevant business matters were asked about their experience. As many as 66 percent of respondents in the enterprise survey who dealt with courts considered them fair, unbiased, and objective. At the same time, 46 percent of the same group of people (those who dealt with courts) said that the courts are corrupt. Courts are regarded as not fast, with unnecessary delays by 77 percent of respondents who dealt with them (Chart A2.8). Moreover, firms that felt the process was unfair were also very likely to declare it corrupt (Chart A2.9). Chart A2.8 In Solving Your Company's Business Matters, The Courts Are... fair, unbiased, objective not corrupt fast strongly strongly tend to disagree agree strongly strongly agree 19% 21% agree disagree 20% 23% 20% strongly disagree tend to 49% disagree 15% tend to agree tend to tend to tend to 45% disagree agree disagree 26% 34% 28% 50 Annex 2. Corruption in the Business Environment Chart A2.9 Fairness of Courts and Perceptions of Corruption By Enterprise Managers Subsample of those who found Subsample of those who found courts fair, unbiased, objective courts unfair, biased, not objective process was not process process corrupt was not was 13% corrupt process corrupt 87% was 74% corrupt 26% Slowness and inefficiency of courts are problems in themselves, but they also are correlated with corruption. Among enterprises that were involved in court proceedings, those that found proceedings to be slow report them as corrupt more often than those who found them fast. Out of the respondents who said that courts were slow, 50 percent also report them as corrupt, while among enterprise managers who said that courts were fast only 35 percent consider them corrupt (Chart A2.10). Chart A2.10 Enterprise Managers Who Found The Court Proceedings Fast , Also Found Them... fast slow process was not corrupt process 64% was process process corrupt was was not 36% corrupt corrupt 50% 50% Dealing With Inspections Thirty-four percent of enterprises surveyed were visited by financial police over the last two years in connection with an alleged breach of finance laws. After being investigated, 36 percent of those visited were considered breaking the laws. Every fifth respondent out of those who were regarded as breaking the law paid a bribe either to reduce the size of the amount due, or to close the case. Chart A2.11 demonstrates the respondents' perceptions of tax avoidance practices of other firms in their market segment. The enterprise managers were asked what proportion of sales competitors keep off the books. According to the answers, firms hide from inspection 28 percent to 41 percent of sales. Only 4 respondents out of 151 who answered the question said that the competitors do not keep any percentage of sales off the books. 51 Annex 2. Corruption in the Business Environment Chart A2.11 Hiding Profits off 50% keep 40% 30% competitors 20% that books 10% the sales 0% of all small /large allstate state FDI private with noFDI 1998) in1996-98 in1992-95 1992 large with before medium percentage after founded founded (founded founded mean new Among the surveyed enterprises that are liable to health and safety inspections, 80 percent were visited over the last two years. Only 7 percent said that they had paid bribes and only in one case did the inspector ask for a bribe. In all other cases the entrepreneur either "wanted to give it" or "knew that this is the way it goes." There is some evidence (see Chart A2.12) of a positive relationship between the number of visits by health and safety inspections in the last two years and the incidence of bribes given to the health and safety inspectors in the last two weeks. The enterprises in our sample that reported that they had given a bribe experienced more visits. Chart A2.12 Number of Visits by Health and Safety Inspections and Bribing 10 9 8 7 visits of 6 5 number 4 3 mean 2 1 0 did not give a bribe gave a bribe Registering The Company Company registration in BiH is usually done either by the owner of the company or by a company's employee. Only 18 percent of respondents reported that they had used a third party to handle the process of business registration. It is worth mentioning that none of the newly-created enterprises (those founded in 1999-2000) used a third party. At the same time, public enterprises used a third party very often - 40 percent of public firms who were involved in our survey reported so. Another group of firms that used this method of company registration are enterprises founded before 1992. Since 72 percent of public enterprises were founded before 1992, the prevalent use of third-party facilitators among older firms is not surprising (Chart A2.13). 52 Annex 2. Corruption in the Business Environment Chart A2.13 Using Third Party For Company Registration 50% party 40% third the 30% used 20% that 10% firms 0% of all small /large state FDI private 1992 with noFDI 1998) in1996-98 in1992-95 with medium percentage after before founded founded (founded founded new Why do companies use third parties for company registration? Mostly because the managers want to make sure that no procedural mistakes are made (59 percent of respondents), sometimes because they want to save time (23 percent), sometimes the reason is that the third party has connections ­ "knows the right people" (18 percent) (Chart A2.14). Use of third parties is also known to help hide corruption ­ the bribe in this case might be expensed as a fee for a business service provided by the third party. Chart A2.14 Reasons For Using the Third Party to save time 23% to avoid procedural mistakes 59% to use someone who knows the right people 18% Obtaining Export/Import Permits Interruption of trade links during the war and the high level of domestic demand created by foreign aid and remittances reduced the incentives for firms to look to international markets for business. Re-establishing a strong regime of foreign trade is especially important for BiH due to destruction of the industry during the war. Success in creating incentives for businesses to access the global markets depends, among other matters, on the efficiency of the processes by which export and import permits are issued. The average enterprise in BiH needs to obtain about eight export/import permits annually in order to function legally. Firms with foreign direct investment on the average need more permits than other firms. Large public companies had, on average, five permits (Chart A2.15). 53 Annex 2. Corruption in the Business Environment Chart A2.15 Number of Export/Import Permits the Average Firm is Required to Have 12 10 8 permits of 6 4 number 2 0 all small /large state state FDI private with noFDI 1992 1998) in1996-98 in1992-95 large with after before medium founded founded (founded founded new Nearly 32 percent of the firms in the sample exported products in the past two years. Large public enterprises are more export-oriented than any other group of firms, 60 percent of them did export products in the last two years (Chart A2.16). Chart A2.16 Firms That Exported Products in the Past Two Years 70% 60% 50% firms of 40% 30% 20% percentage 10% 0% Considering that many companies need to obtain several export permits annually, the licensing procedures could be a source of corruption if they are not clear and transparent and hence create obstacles for doing business. We asked respondents identical questions about clarity and transparency of processes of receiving import and export permits. Chart A2.17 shows how the responses to the first question were distributed among different types of firms. Almost all respondents ­ 91 percent of firms that tried to receive export permits and 93 percent of those who had experience with obtaining import permits - found the process clear and transparent. 60 percent of the firms got all the export permits they tried to obtain, and another 29 percent received almost all export permits. Out of those firms that tried to obtain import permits, 69 percent received all the permits, another 22 percent received most of them. Overall, analysis demonstrates that the process of obtaining permits as viewed by the respondents is clear, transparent and efficient. 54 Annex 2. Corruption in the Business Environment Chart A2.17 Process of Obtaining Export Permits 100% process 80% the 60% clear found and 40% that 20% firms 0% of transparent all small state FDI private 1992 /large allstate with noFDI 1998) in1996-98 in1992-95 large with before medium percentage after founded founded (founded founded new The time spent by firms on obtaining permits depends, among other things, on how many signatures are required for each permit. According to respondents' experience, the number of required signatures is low ­ three on the average for either export or import permits (Chart A2.18- A2.19). Information about the process of obtaining permits and the number of signatures required shows whether there are distortions in this process that create opportunities for corruption. Although the process was reported to be mostly transparent, clear, and efficient, and the number of required signatures not excessive, the enterprises nevertheless reported paying a certain share of the permit value as bribes ­ on the average, the bribes constitute 2.6 percent of the export permit value and 2.8 percent of the import permit value (ChartsA2.20-A2.21). A more important observation to be made is that the variance is low, and 85 percent of respondents who answered this question said that the size of the bribes equals 2 percent of the permit value, suggesting that the sizes of the needed bribes are fairly well known. Chart A2.18 Number of Public Officials' Signatures Required for Export Permits 4.0 3.5 3.0 signatures 2.5 of 2.0 1.5 1.0 number 0.5 mean 0.0 all small /large state private FDI 1992 allstate with noFDI 1998) large with after in1996-98 in1992-95 before medium founded founded (founded founded new 55 Annex 2. Corruption in the Business Environment Chart A2.19 Number of Public Officials' Signatures Required for Import Permits 4 3.5 3 signatures 2.5 of 2 1.5 number 1 0.5 mean 0 all small /large allstate state FDI private noFDI 1998) 1992 with large with after in1996-98 in1992-95 before medium founded founded (founded founded new Chart A2.20 Bribing for Export Permits 4% value 3% permit of 2% 1% percentage 0% all small /large allstate state FDI private 1992 with noFDI 1998) in1996-98 in1992-95 large with after before medium founded founded (founded founded new Chart A2.21 Bribing for Import Permits 4.0% value 3.0% permit 2.0% of 1.0% percentage 0.0% all small /large allstate state FDI private with noFDI 1998) in1996-98 in1992-95 1992 large with medium after before founded founded (founded founded new 56 ANNEX 3. PROCESS OF CORRUPTION An understanding of the mechanics of bribery is useful for two main reasons. First, by describing a process, as opposed to an outcome, it focuses attention on the behavior of actors involved at different stages. Second, by looking at the mechanics of bribery one might learn at which stages of the process institutional norms are being broken. This can provide insight into possible means to fight corruption. Step 1. Offering a bribe ­ asking for a bribe. What will happen if a citizen, a local entrepreneur, or a foreign entrepreneur comes to an institution such as yours? Most respondents report that the situation would be solved in accordance with the law. In 15-22 percent of cases the customer will offer a bribe. Foreign enterprise managers offer bribes more often than local enterprise managers or citizens. In 7-10 percent of the cases, the official will ask for a bribe. It is less likely that he/she will ask a foreign businessman, but more likely if the visitor is a local (Chart A3.1). At this stage of the process, the customer usually takes the initiative. Chart A3.1 What Will Happen If a Citizen, a Local Entrepreneur, or a Foreign Entrepreneur Requests Services From an Institution Such as Yours? 70% 60% officials 50% 40% public of 30% 20% 10% percentage 0% the worker of the institution the person offers the bribe the application would be would indicate that a bribe would himself/herself processed in accordance with be very suitable the law citizen local entrepreneur foreign entrepreneur Chart A3.2 Who Initiates Bribes entrepreneur offers a bribe spontaneously official 32% indicates that a bribe is expected 68% Annex 3. Process of Corruption However, according to the data from the enterprise survey, in cases when bribes are paid the situation is the opposite: it happens less often that an entrepreneur offers a bribe spontaneously (32 percent of cases) than when he/she gives a bribe after the official indicates that it is expected (68 percent of cases) (Chart A3.2). Step 2. Providing the service. Even when the bribe is paid, it does not necessarily mean that the service is provided. Only one-fourth of the respondents think that "once the bribe is accepted, the service is always provided". About 40 percent of the respondents believe that "once the bribe is paid, the service is usually provided". 18 percent said that if the bribe is paid, the service "sometimes is provided" and the rest of the interviewed even said that the service is "rarely provided" (Chart A3.3). Chart A3.3 Once The Bribe is Accepted, Is the Service Provided? 50% 40% 30% respondents of 20% 10% percentage 0% always provided usually provided sometimes provided never provided public officials enterprise managers Step 3. Sharing the bribe . How acceptable are corrupt practices in the public office? If someone accepts a bribe, do they usually hide it from colleagues or do they share them? The survey outcomes demonstrate that bribes usually are not shared, although in about 10 percent of the cases there is a habit of sharing illegal gains. In public administration, one of the ways to reduce opportunities for corruption is to rotate employees from one type of work to another. It is assumed that in this case officials can not obtain a stable group of clients who are aware of the fact that the official takes bribes and who know the rates to be paid. If bribe-sharing is widespread, rotation would not reduce corruption ­ bribetakers will share bribe-giving clients (Chart A3.4). Chart A3.4 If Someone in an Institution Such As Yours Accepts a Bribe, They Usually... 50% 40% officials 30% public of 20% 10% percentage 0% share payoffs with shares with seeps all for there are no bribes superiors colleagues him/herself taken at all here at an institution such as yours at your institution 58 Annex 3. Process of Corruption Step 4. Reporting bribes. Taking action against bribe-takers. How often are enterprise managers who offer or pay bribes reported to authorities? More than half of the enterprise managers surveyed believe that it happens "sometimes", 15 percent think that it occurs "usually", and only 22 percent stated that it is never the case (Chart A3.5). Unlike ordinary citizens who pay bribes for public goods that they are entitled to receive free of charge or for a reasonable price and who are in most cases loosing from bribing, enterprise managers gain by obtaining access to resources that bring a superior market position for the company. Therefore, they are more likely to actively support the bribes market. As chart A3.7 demonstrates, the enterprise managers interviewed are aware of the fact that not only can public officials be prosecuted for taking bribes, but businessmen giving bribes can also be penalized. This awareness could be a factor reducing the incidence of bribes initiated by enterprise managers. Chart A3.5 Perception of the Likelihood of Reporting Bribes and Penalizing Offenders 60 50 managers 40 30 enterprise of 20 10 0 percentage always usually sometimes never would official/enterprise manager who takes/offers bribes be reported to authorities? would action be taken against an official taking bribes? According to respondents' perceptions, actions that can be taken against both public officials and enterprise managers for bribing and for taking bribes, are quite severe. More than 15 percent said that dismissal is a probable outcome, and 35 percent believe that prosecution is likely. Chart A3.6 Perception of Penalties for Bribery 40 35 managers 30 25 20 15 enterprise of 10 5 0 dismissal transfer, demotion, loss prosecution fine percentage of privileges what action would be taken if a public official is reported for requesting or accepting a bribe? what is most likely the consequence for entrepreneur if he/she is reported giving bribes? Punishment for bribes. Public officials' experience with how corruption in public services is penalized is important to understand in order to use the means other than punishment for reducing corruption in cases where fear of punishment does not work. An example of the case when fear of punishment does not stop bribing is when the gain is too big and expected value of bribing is too high to be devalued by the risk of punishment. Another example is when punishment is perceived 59 Annex 3. Process of Corruption as a rare occasion and the risk to be punished is so low that even with low expected return the rationale will be to bribe. Chart A3.7 shows how different the institutions are in regard to punishing for bribery. The percentage of respondents saying that in their institution people were not penalized for bribery is the highest in health and at public enterprises. On the other hand, institutions with the most severe punishment (actually used) are banks, customs, and police. Chart A3.7 Punishment That Was ACTUALLY Used in Different Institutions Health Education Police would receive would not would not would not would be criminal be punished be punished be punished would be prosecution 15% 5% punished 42% punished modestly 9% no such modestly no such would be 12% case 22% case would be dismissed 45% 38% punished 26% would be more would be punished severely punished would be would more would 17% more punished receive severely receive severely modestly criminal would be 8% criminal would be 10% 13% prosecution dismissed prosecution dismissed 3% 7% 12% 16% would receive Bank Customs Judiciary criminal would not be would not be would be would be prosecution would not be punished punished would be punished punished 6% punished 5% 10% punished modestly modestly more would be 12% 15% 8% severely dismissed no such would be 10% 18% case punished 33% more severely would be would be dismissed would 13% would punished would be 15% receive no such receive more punished would be criminal case criminal severely modestly dismissed prosecution 49% prosecution 29% 35% 21% 13% 8% Municipal administration Taxation State owned enterprises would not be would be would be punished would not be would be no such 12% would be punished would not punished punished punished punished more case be punished 14% modestly more modestly severely 33% 7% severely no such 27% 13% 10% would would be 9% case receive 52% punished would be criminal would be dismissed modestly prosecution dismissed 8% 13% 6% would 17% would would be no such receive receive punished case criminal criminal would be more 44% prosecution prosecution dismissed severely 9% 5% 15% 6% 60 ANNEX 4. HOW RESPONDENTS DEFINE CORRUPTION What meaning does the term corruption have in everyday language in BiH? We need to know it in order to form the language for the anti-corruption campaign and of any publicly disseminated anti-corruption information. The definition of corruption used by anti-corruption specialists is complex, and needs to be conveyed to the public in order to avoid misunderstanding. If the definition of corruption used in the anti-corruption campaign is different from the perception of this term that is prevalent in society, the messages that are sent to the public during the campaign might be misinterpreted. In particular, the general public might have a narrow interpretation of the notion "corruption" that implies only bribes, and in this case stressing that corruption includes other forms of abuse of public office for private gain is important. The definitions of corruption that are used by many authors who write on anti-corruption issues can be summarized in the following way: Corruption is the use of public office for private gain or for the gain of groups closely affiliated with public officials. Different types of corruption include bribery for government contracts and services; theft or misappropriation of budget funds and public assets; patronage in civil service recruitment or dispensation of government benefits; and influence peddling. In the countries of Central and Eastern Europe, the Baltics, and the Commonwealth of Independent States, the simultaneous processes of developing a market economy, designing new political and social institutions and redistributing social assets have created fertile ground for corruption. In many countries, corruption has had a significant impact on this process, encoding advantages in these new rules and institutions for narrowly vested interests and distorting the path of economic and political development. In order to tailor anticorruption strategies to address specific problems and to target reforms more effectively, it is important to distinguish between two main forms of corruption - state capture and administrative corruption. State capture refers to the actions of individuals, groups, or firms both in the public and private sectors to influence the formation of laws, regulations, decrees, and other government policies to their advantage as a result of the illicit and non-transparent provision of private benefits to public officials. Administrative corruption refers to the intentional imposition of distortions in the prescribed implementation of existing laws, rules, and regulations to provide advantages to either government or non-government actors as a result of the illicit and non-transparent provision of private gains to public officials.44 We asked both a closed-ended and an open-ended question on the definition of corruption. The closed-ended question was asked in the way that limited the respondents to choosing one from ten optional answers (Table A4.1 summarizes the outcomes), and the open-ended question gave the respondents an opportunity to give a definition that combines as many forms of corruption as they wanted (see Table A4.2). The outcomes are similar, but not the same. When limited to one reply (closed-ended question), most of the respondents (68% of public officials, 61% of general public, and 56% of enterprise managers) define corruption as giving (receiving) bribes or as power abuse Annex 4. How Respondents Define Corruption to get material or other advantages. Both types of corruption are mentioned substantially more often than any other by all three groups of respondents. Enterprise managers put power abuse above bribes (mention it more often), while in public officials and general public surveys giving (receiving) bribes has been chosen more often than power abuse, which was the second most commonly-mentioned option. In the answers to the open-ended question on definition of corruption, "giving bribes" and "abuse of position to gain material wealth or other advantage" are again at the top of the list. In addition, another type of corruption - nepotism ("giving preferences to friends and relatives") - is mentioned very often by all three groups of respondents. "Collaboration between public servants and criminals" has also been found as one form of corruption by many respondents. The definitions used by the respondents are consistent with the meaning of the term "corruption" adopted by specialists. Therefore, it is unlikely that the public misinterprets the term "corruption" in mass media messages even if a definition of the term is not presented. Table A4.1. Definition of corruption. Closed-ended questions. percentage of percentage of public officials individuals/ Giving bribes, bribery 38.4% general public Power (function) abuse to get material Giving bribes, bribery 37.6% advantages or other advantages 29.5% Power (function) abuse to get material Giving or receiving gifts, donations ,,attentions" 5.3% advantages or other advantages 23.7% Deceitful money making 5.3% Giving or receiving gifts, donations ,,attentions" 7.0% Collaboration of public officials with the crime Favoritism, preferring of relatives and other scene 5.2% close people 5.7% Immoral officials 3.8% Collaboration of public officials with the crime Illegal ,,cooperation" of public officials and scene 5.4% entrepreneurs 3.3% Deceitful money making 4.7% Favoritism, preferring of relatives and other Immoral officials 4.2% close people 2.9% Illegal ,,cooperation" of public officials and Political protectionism ­ favoritism based on entrepreneurs 3.3% loyalty to a political party 2.9% Willful collapse of a firm or company 3.2% Political protectionism ­ favoritism based on Willful collapse of a firm or company 2.5% loyalty to a political party 3.2% percentage of enterprise managers Power (function) abuse to get material advantages or other advantages 31.2% Giving bribes, bribery 24.7% Giving or receiving gifts, donations ,,attentions" 9.1% Collaboration of public officials with the crime scene 8.8% Intimidation, terror, squeezing, racketeering 6.5% Illegal ,,cooperation" of public officials and entrepreneurs 5.0% Favoritism, preferring of relatives and other close people 5.0% Immoral officials 3.2% Political protectionism ­ favoritism based on loyalty to a political party 2.9% Willful collapse of a firm or company 1.8% Deceitful money making 1.8% 62 Annex 4. How Respondents Define Corruption Table A4.2. 45 What does the word corruption mean, in your opinion? Open-ended question, coded answers. General public Public officials Enterprise managers percentage of percentage of percentage of definition of corruption respondents definition of corruption respondents definition of corruption respondents Giving bribes 85.5% Giving bribes 82.2% Giving bribes 78.5% Abuse of position to gain Abuse of position to gain Abuse of position to gain material material wealth or other material wealth or other wealth or other advantage 56.1% advantage 60.5% advantage 57.9% Giving preference to friends and Giving preference to friends and relatives 50.4% Acquiring money through deceit 53.4% relatives 39.1% Giving and receiving gifts, Giving preference to friends and Collaboration between public donations, and "special attention" 43.4% relatives 43.0% servants and criminals 38.5% Giving and receiving gifts, Collaboration between public Collaboration between public donations, and "special servants and criminals 43.2% servants and criminals 41.3% attention" 37.6% Illegal collaboration of public servants and owners/directors of Purposepul bringing of Acquiring money through deceit 41.3% enterprises 37.7% enterprises to bankruptcy 36.8% Illegal collaboration of public Giving and receiving gifts, servants and owners/directors of donations, and "special enterprises 40.8% attention" 37.3% Acquiring money through deceit 36.8% Illegal collaboration of public servants and owners/directors of Immoral officials 39.0% Immoral officials 36.4% enterprises 35.3% Purposepul bringing of enterprises Purposepul bringing of to bankruptcy 38.0% enterprises to bankruptcy 36.4% Immoral officials 33.2% Giving advantage on the basis of Giving advantage on the basis of Giving advantage on the basis of belonging to or supporting belonging to or supporting belonging to or supporting political party 30.6% political party 28.1% political party 31.2% 63 ANNEX 5. SURVEY METHODOLOGY: SAMPLING AND INTERVIEWING PROCEDURE The sampling was done according to sociological techniques that are relevant for this types of surveys. Quotas were set on occupation/business types, etc., to ensure a cross-section of opinions and robust analyses; these are detailed below. All three surveys ­ general public survey, public officials survey and enterprise survey - were conducted following face-to-face interview methodology. Before the beginning of each interview the respondents were presented with a special letter in which all necessary details were listed regarding the purpose of the survey, conductor of the survey1 and the ways in which their identity would be protected. GENERAL PUBLIC (1,200 INTERVIEWS) Random sampling was used within each Entity. Population Statistics In order to get the most accurate estimates of current population statistics, the OSCE Voter Registration database was utilized as the basis for sampling. The OSCE data tells us the number of registered voters for each municipality, their age and gender. This data was utilized as it is currently the most accurate estimate of population parameters (in view of the fact that over 85 percent of the 18 years and over population registered to vote). Data from Statistical Institute are based only on population estimates received through municipal governments. Target Population The target population for this survey included residents age 18 and over. Selection of Primary Sampling Points To obtain 1200 valid interviews, an initial sample of 1350 was used. It was divided into two units, 750 interviews were conducted in the Federation , and 450 - in Republika Srpska. The sample is representative of the regional units within the Entities. Selection of Secondary Sampling Points Once the primary sampling points were selected, within each municipality, from the list of settlements/neighborhood councils/census areas and streets, a proportional number of sampling 1An international survey firm ACNielsen Annex 5. Survey Methodology: Sampling and Interviewing Procedure points was selected. The number of blocks within each primary unit was divided evenly among the secondary units. Selection of Households This sample used a "Random Walk" technique for selecting households. Interviewers were given an address of the sampling point. Starting at this address, the interviewer proceeded in that direction and chose the second house as the first household for interviewing. If the second house was an apartment building (or has multiple dwelling units), she chose the second dwelling unit in this building to start the interviewing. After the interview, she passed the next apartment or house and chose the fourth dwelling unit for the second interview. She proceeded in this pattern until the end of the street. At the end of the street, she moved to the next street on her right and continued the process until she completes the 5 interviews for that sampling point. Selection of Respondents In this survey a technique of random selection "by Closest Birthday" was used. At the door, the Interviewer asked to speak to a household member whose birthday is closest to the date she is interviewing. If that person was not at home, or that person did not meet the quota for age/gender group, she arranged to revisit that house if possible. Interviewers were directed to go on to the next unit if they cannot get the appropriate respondent after two attempts. This method ensures random selection of respondents. Respondent refusal The interviewer recorded all incidence of refusal of participation in the survey (including data of gender, estimated age, and reason given for refusal). The calculated response rate was 75 percent. Control The work of interviewers was controlled during and after the fieldwork. Approximately 15 percent of randomly chosen surveyed respondents were controlled. A team of independent controllers conducted the control checking the manner in which interviewers were applying the Random Walk technique in selecting the households, as well as if the interviewed members of households were selected according to their birth date. In some cases the controllers were also double-checking some of the answers given by the interviewed persons. PUBLIC OFFICIAL SURVEY (700 INTERVIEWS) Presently there is no institution in Bosnia and Herzegovina that would have information about the officials employed in different sectors or the addresses of all institutions and organizations in the public sector. Because of that it was impossible to prepare a stratified sample of public officials. As an alternative, random selection of officials was possible in randomly selected institutions. However, it was inappropriate to conduct the interviews in the institutions because of the subject of the research. Therefore, respondents for the public officials survey were selected during the general public survey interviewing and were interviewed at home. The above-described sample of the households was used as a frame for the sample of the officials. One of the survey objectives was comparison of different public institutions, and for this reason it was necessary to have groups of respondents employed by various institutions large enough for 65 Annex 5. Survey Methodology: Sampling and Interviewing Procedure the analysis of the differences among institutions. The sample (N=700 officials) was divided in two parts. For the first part (n=400), quota sampling was used - 50 respondents from eight main public institutions/sectors/agencies (Customs, Taxation, Judiciary, Police, Health sector, Education, Municipal Administration, publicly-owned enterprises) were chosen for interviews. The second part (n=200) was based on a random selection form different government institutions including the eight institutions used for quota sampling. At the time when interviews for the general public surveys were conducted, interviewers asked the respondent if any other member of the household worked as a public official. Whenever they received a positive answer, the interviewers checked if those members were fitting into the sampling quotas for public officials survey. If several members of a household were employed in the public sector, than the second person to be interviewed (public official) was selected based on their birth date, that is the persons whose birthday was the closest after the date of the interview was selected. The sample of public officials selected in the above manner had the same structure as the general public sample being composed of three representative units (N=200), each for one of the regions with majority population from one of the major B&H nations. Every unit was representative in relation to regions, type and the size of the populated areas. ENTERPRISE SURVEY (350 INTERVIEWS) The key problem for designing a representative sample for companies in BiH arises from the fact that a single register of companies does not exist. Statistical Institutes of the Federation of B&H and the Republic of Srpska have their own databases for the two entities: Federation B&H and the Republic of Srpska. However, in both cases the databases are updated only occasionally and not on a strict schedule. In Republica Srpska, the register did not include the information on company size (number of employees) and could not be used for this reason. The sample was designed to be representative for the three parts of the country, each with predominant population being of one of the three different main nations of Bosnia and Herzegovina, for specific geographic regions, each of the 6 key economic sectors (agriculture and forestry; manufacturing; construction; trade, wholesale, retail; professional services; personal services), different ownership types and the size of companies. The sampling was based on quotas formed respecting the ownership structure and the size of companies (number of employees), constructed separately for each geographic region and economic sector. The total size of the sample for companies in Bosnia and Herzegovina was N=350. The sample was constructed in the following way: 1. From the database of 40000 companies registered in the Federation of B&H, companies operating in one of the 6 selected economic sectors were chosen. In the Republic of Srpska, for which a register with the necessary information was not available, interviewers had a task to visit authorities in selected municipalities to compile the lists of companies fitting in the above-described quotation cells. 2. The total sample (N=350) was divided in three units: one for the part of Federation with predominantly Bosnian population, one for the part of Federation with predominantly Croat population and one for the Republic of Srpska as part of Bosnia and Herzegovina with 66 Annex 5. Survey Methodology: Sampling and Interviewing Procedure predominantly Serb population. The total number of companies in each of the three units was more or less proportionate to the number of companies registered in each of the three regions. 3. The quotas were than formed for each of the 5-6 geographic regions in the relevant unit, proportionate to the size of those regions. Apart from the regional center, at least one more municipality was selected for each of the regions. Only those municipalities in which interviews were previously held with general publics/public officials were selected. 4. Following that, quotas of companies were formed based on the ownership structure (most of the company privately owned vs. most of the company under government ownership) and the size of companies (number of employees), within each region and for each of the 6 economic sectors. 5. Interviewers visited all selected companies to set up/conduct interviewers with owners, directors or assistant directors. 67 ENDNOTES 1 The other subset of firms that spend significantly larger fractions of time dealing with bureaucracy are firms with foreign direct investment. Managers of such firms report spending nearly 20 percent of time dealing with government officials. 2 Out of the respondents who received health care, education, police, and some other services in the last year, 20 percent said that they had paid bribes for it. Only during the two months before the survey was conducted, 7.4 percent of all respondents were asked by public officials to make unofficial payment (extra money, gifts or services) for government, public, educational, or medical services. During the last three years, 17 percent of all respondents paid bribes. 3International comparisons of such a comple x phenomenon as corruption are difficult to make and a certain margin of error should be assumed in analyzing the data. 4 The Chairman of the BiH Council of Ministers announced that fighting against corruption is the second most important item of the reform program. The Republika Srpska (RS) government formed an anti- corruption group comprising of high-level government officials and headed by the RS Vice President and published an anti-corruption implementation plan. Similar actions are expected from the newly formed Federation BiH government. 5 The analysis is based on both experiences of the respondents with corruption and their perceptions of different aspects of corruption-related problems. Both public experiences and perceptions are important to understand when developing policies since the success of policy implementation depends on people's reactions to the rules of behavior that policies define. These reactions are based on a combination of personal experience and perceptions of reality. 6 A precise assessment of the level of corruption is problematic. However, it is possible to give a reliable although approximate estimate of how widespread corruption is, and our surveys are designed to enable us to do it. We estimate the level of corruption using data about perceptions as well as experiences of both recipients and providers of public services. This approach allows examination of the issue from different sides and helps reduce errors by comparing answers to similar questions. 7 As these outcomes show, the share of those who think that "corruption is very widespread" is lower among businessmen and public officials than among the general public. Both businessmen and public officials more often than general public survey respondents answer this ques tion with caution ­ they accept the existence of corruption, but say that "it is hard to tell how widespread it is". In all three groups, more than 90 percent of those interviewed picked one of these two answers, and less than nine percent of the respondents said that "corruption is not widespread" or it "does not exist" in BiH. The remaining 1 percent said that they "do not know if there is corruption in BiH". 8 One needs to be cautious when interpreting answers to questions asked today about level of corruption in the past. These answers reflect not only experience of the respondents, but also dissatisfaction with the current situation. The more problems people encounter in everyday life in the present, the more they tend to underestimate difficulties that they came across in the past. Also, corruption-related problems are widely discussed in mass media today, while it was not the case prior to transition. As a result, people know more about corruption in contemporary life than about the same facts in the past, and many of them substitute knowledge for personal experience. For these reasons, answers that respondents gave to the question Endnotes "When did corruption reach the highest extent in BiH?" reflect both how people perceive changes in the level of corruption over time and how they perceive the level of corruption in contemporary life. 9 Before the anti-corruption campaign is launched, it is important to estimate the level of public trust in authorities. Although this indicator is sensitive to political events and hence might be volatile during short periods of time around the events, it is relatively stable over the longer time periods. However, it is important to note that the surveys took place before the elections and reflect the opinions that the respondents had when the previous governments were in the office. 10 Verbal behavior is not the same as actions, and it is likely that less people will actually report bribes when faced with corruption in real life. However, the data on verbal behavior is indicative of real actions the same way as perceptions are indicative of behavior. 11 In order to assess willingness of the public to fight corruption, it is also important to understand the motivation of people. For example, if reporting bribery in public office by another public servant is based on envy to a more affluent colleague who benefits from corruption, this might not be a step toward reduced corruption, but instead might lead to a situation when officials who are involved in bribery are replaced by others who will be taking bribes as well. In this case, the outcome may be quite the opposite to desirable ­ instead of reduced corruption, we might observe increased amount of bribes requested (if the risk of losing a position that brings illegal income increases, the payoff should rise as well). Answering an open-ended question "If you decide to report a bribe, for what reasons would you do so?", most of respondents mention "to fight corruption", "moral reasons", "moral duty", "civic duty", "in accordance with the law", "injustice", "respect, self-respect, principle", "better tomorrow". Answering the question "For what reasons would you refrain from reporting corruption?", the respondents mentioned "security, safety", "nobody to report to". Although answers showing negative motivation (for example, "revenge to a more successful colleague") are not likely to be found in responses to survey questions, the fact that the vast majority of replies are aligned with the response "to fight corruption" is a promising outcome. This type of response demonstrates that people have similar attitude to corruption and that based on this common attitude, civil society groups can be formed with the objective of supporting the government in implementing anti-corruption measures. 12 When those who refused to answer this question are included in the calculations, the percentage of those who give a positive answer to this question is still high - 51-72 percent. 13 In cross-country analysis, the BEEPS survey demonstrates that political contributions as means of influence in BiH are especially popular among firms that report actively capturing the state. 14The index used in this chart was calculated in the following way: Answering the question "Are the following types of companies more likely or less likely to pay bribes than the average?", the respondents were choosing one out of five optional answers. The index is calculated in the following way: each optional answer was assigned a weight, and the weighted average of the percentage of respondents who picked different answers was calculated for each group of enterprises listed in the question. The formula used was as follows: Index = (a1*3+a2*1+a3*(-1)+a4*(-3))/4, where: a1 is the percentage of respondents who believe that the corresponding group of firms is "much more likely" to have to pay bribes than others, a2 is the percentage believing that these firms are "a bit more likely" to have to pay bribes than others, a3 is the percentage believing that these firms are "a bit less likely" to have to pay bribes than others, a4 is the percentage believing that these firms are "much less likely" to have to pay bribes than others. 15 The institutions that are most likely to be perceived by their employees as substantially involved in policy capture (a type of state capture) are publicly -owned enterprises and banks. Regulatory capture (another type of state capture) is attributed to publicly -owned enterprises and the health sector. Capture by political parties is more typical for municipal administration and banks. 69 Endnotes 16 When those who refused to answer this question are included in the calculations, the percentage of those who give a positive answer to this question is still high - from 24-34 percent in different institutions. 17 The data from the public officials survey also show that 65 percent of respondents believe that corruption exists in different public sector areas and 22 percent think that corruption is widespread in "institutions such as theirs" (Annex 1, Charts A1.8 - A1.9). 18 53-70 percent if refusals to respond are included in the calculations. 19 44-70 percent if refusals to respond are included in the calculations. 20 40-69 percent if refusals to respond are included in the calculations. 21 Respondents were asked to grade each of the 20 different BiH government institutions on a scale from "corruption is very widespread in this field" to "corruption does not exist in this field". Ranking is based on the percentage of respondents who though that corruption is "very widespread" in a particular institution. 22 This section is based on the analysis of public officials' responses to two questions: the first one is used to estimate the perceived level of bribery at the institution where the respondent works, and the second one gives information that helps assess different types of imperfections in public administration policy in the institution where the respondent works. The first question is "How widespread are practices of paying tips or making payments "under the desk" for the delivery of public services, for speeding up the delivery of public services, for deciding in favor of a specific person or just for a better service quality in an institution such as yours?" The graphs in this section show the percentage of respondents (public officials) who found that these practices are widespread in the institutions such as theirs. The second question is complex and consists of several statements that we used in different combinations to construct estimates of various aspects of public administration: "To what degree the following statements dealing with decisions in the field of the personnel policy (promotions, transfers, salary increases) are correct: (1) decisions are clear to employees, (2) decisions are based on criteria specified in written personnel guidelines, (3) decisions are based on professional experience, merit, performance, (4) decisions are based on family and personal connections, (5) decisions are based on regional connections, (6) decisions are based on patronage within institution, (7) decisions are based on unofficial payments?". 23 This outcome is based on probit regressions - we regressed variables of public administration quality (in different combinations) on perceived level of bribery. 24 See endnote 1. 26 For comparison, the salaries in public sector in September 2000 ranged from 387KM to 854KM in the Federation and from 369 to 495 in the RS. 27 The other subset of firms that spend significantly larger fractions of time dealing with bureaucracy are firms with foreign direct investment. Managers of such firms report spending nearly 20 percent of time dealing with government officials. 28 By comparing predictability of government actions as perceived by different types of businesses, we are trying to estimate the level of vulnerability of these businesses. In particular, we look at three types of risks ­ unpredictability of the regulatory environment, unpredictability of the policy environment, and ability of the business community to monitor government policy and regulatory decisions. The analysis shows that vulnerability of businesses to changes in policy and regulatory environment is very high. New firms, small and private firms appear to be more vulnerable to these changes. 29 This outcome - lower perceived vulnerability of public enterprises and firms with foreign direct investment - might be explained by their connections within power structures, which enable them to be informed about planned changes in laws and regulations or even to influence legislative processes. 70 Endnotes 30 It could be that new firms are more often than other companies members of business associations, and that firms with foreign direct investment less often have business associations' membership. If this is the case, it would explain this outcome. However, this guess is not confirmed by the data ­ the share of business association members among new firms is much lower than among firms with foreign direct investment (16 percent and 52 percent respectively). 31See endnote 2. 32 Calculated as percentage of those who gave bribes out of all respondents who received services from these institutions. 33The mean annual size of a bribe is 21.6 KM. 34Correlation between these two variables is -0.91. 35 The graph shows the relationship between income and share of bribes in income in the linear-log form. Each dot on the graph represents the average share of bribes in annual income for each income group. 36Who are the customers of private sector health services? Is it a group characterized with higher education and higher income level? The analysis demonstrates that this is not the case - there is no difference in terms of income or education between respondents who used public health services and those who used private health services. Therefore, we can not conclude that low-income groups or groups with lower levels of education are excluded from this category and thus are more likely to receive lower-quality services. However, there is at least one social group that is partially excluded from using private health services. Those are pensioners. Pensioners are 43 percent less likely to be found among private health service customers than among consumers of public health services. 37 For example, the income and asset declarations for politicians in the United States are scanned and are available on the world wide web. Disclosures for candidates for President, the Senate, and the House of Representatives are available at the http://www.opensecrets.org . 38 In the United States political organizations are required by law to submit detailed information on contributors and recipients of funds. These filings are publicly available at http://www.irs.gov/bus_info/eo/8871.html . Indeed, every individual contribution of more than $200 must be identified and publicly disclosed: http://www.fec.gov , and http://www.opensecrets.org . States have similar requirements regarding state and local legislatures and political offices. 39Despite the fundamental importance of investigative journalism as a check against abuse of public office, freedom of the media is in many countries diminished by a number of factors, including concentrated ownership of the media, practical limits on freedom of information, poor management of press relations by public-sector bodies, liberal use by public officials of libel and defamation laws as a means for intimidating journalists, and inadequate measures to protect the safety of journalists. 40 This table is based on closed-ended questions with six optional responses: (1)"problems do not exist", (2)"problems are not very widespread, (3)"problems are fairly widespread", (4)"problems are very widespread", (5)"problems exist, but I can not judge the extent", (6)"do not know". The numbers in the table are calculated as the ratio of those who chose either of the answers (3), (4), (5) and those who gave any answer (1)-(6). 41 Chart 9 lists 27 negative characteristics of the business environment rated according to enterprise managers' opinions on how problematic they are for the operation and growth of their business. The respondents were asked to evaluate each of the 27 characteristics of the business environment on the scale from "no obstacle to "serious obstacle". The characteristics were then ranked according to the percentage of respondents who think of them as of "serious obstacles". 71 Endnotes 42 In the case with permits, 85 percent of respondents who answered this question said that the size of the bribes equals 2 percent of the permit value. 43 It is important to mention that court slowness appears at the very top of the ranking when we ask about any obstacles to doing business - from "minor" to "serious",- but when we ask about "serious obstacles", all institutional imperfections including court slowness receive lower positions in ranking compared to policy and regulatory capture (see section 3.1). 44Anticorruption in Transition: A Contribution to the Policy Debate, The World Bank, 2000. 45 Many of the respondents gave a complex answer to this question. In these cases, each answer received several codes (for example, an answer from one respondent could receive three codes: "giving bribes", "abuse of position to gain material wealth", and "acquiring money through deceit"). This explains why the numbers in this table are so high. 72