Researchers Alliance for Development Written Submission on Consultations on Strengthening World Bank Engagement on Governance and Anti-Corruption The Researchers Alliance for Development (RAD) is an informal, action-oriented, multidisciplinary network of researchers and academics that provides a platform for dialogue and exchange of ideas on international development. The RAD currently comprises more than 600 representatives of academic institutions, research centers, as well as research units in NGOs, bilateral agencies, the private sector and trade unions from all over the world. Its main objective is twofold: 1. Facilitating interaction between the academic community and the World Bank; 2. Mobilizing the academic and student community on development issues and curricula, facilitating mutual flow of knowledge. The activities of the RAD are coordinated by the Steering Committee composed of 10 prominent academics. The RAD Secretariat is located at the World Bank Paris Office and is coordinated by the Development Policy Dialogue Team. www.worldbank.org/rad The comments were collected through electronic consultations with the RAD, throughout December 2006 and early January 2007. Researchers who participated in the consultations represent universities, research institutes, NGOs, private enterprises and government agencies in the following countries: Bangladesh, Cambodia, Ghana, Iran, Iraq, Italy, Japan, Malaysia, Nigeria, Norway, Portugal, Sri Lanka, Tanzania, and Vietnam. Compiled by: Anna Kuznicka, Development Policy Dialogue, External Affairs, The World Bank Edited by: Odd-Helge Fjeldstad, Chair, Researchers Alliance for Development; Research Director, Chr. Michelsen Institute, Norway. 1
Executive Summary Main lessons for the World Bank from RAD-affiliated researchers: In order to improve the chances of making a positive difference in countries with weak governance and severe corruption, the World Bank should acknowledge that (a) political will is always partial, qualified and temporary; (b) economic resources are usually seriously inadequate; and (c) governance institutions have significant weaknesses, and may require a long time to improve. Thus, the Bank should concentrate its attention on those factors which it can influence, and which have an impact on the effectiveness of governance and anti-corruption reforms. The Bank should support efforts to strengthening democratic institutions of accountability, such as the Parliament s public accounts committee, the auditor general s office, the auditor general s annual report, a free press, and contribute to building capacity of local leaders as a long-term approach to curbing corruption. In this respect, it is of key importance that the Bank engages more with youth, in order to promote good governance as a value among future leaders. The World Bank should look at the term civil society more broadly and include youth organizations, religious groups, trade unions or vernacular (grassroots) organizations among its partners. It is recommended that the Bank invest time and resources in identifying a pool of reliable persons representing the civil society at large, to be consulted on a regular basis. Such partnerships can serve the purpose of raising awareness and giving a voice to citizens on better governance and anti-corruption. In scaling up aid, it is tempting to resort to aid modalities that facilitate easy disbursement. Although budget support is suitable for developing countries with good institutions, it is a poor choice where institutions are weak. Corruption makes budget support ineffective, and in some countries even counter-productive. Giving budget support to highly corrupt countries or countries with a high degree of political corruption is unlikely to spur development. Adequate discussion of the role of aid in fuelling corruption is missing from the current debate. Transparency and accountability of donor agencies in this respect are also limited. The Bank should apply similar criteria for transparency and accountability on its own operations as it demands from governments in the developing world. There is no simple way to mitigate fiduciary risk, and corruption is not properly covered by existing fiduciary risk assessment tools. Recently, a harmonized framework for assessing fiduciary risk has been developed by the Public Expenditure and Financial Accountability (PEFA) working group. Comprehensiveness and proper classification of the budget and fiscal information implies that officials will find it harder to divert or misuse public funds. Transparent and good fiscal information, available to the public, makes misuse easier to detect, as does effective internal and external control. Nevertheless, the PEFA framework is inadequate for capturing corruption. Important determinants of corruption are left out, and the indicators are therefore insufficient to judge whether corruption is likely to be high or low in the country assessed. In particular, the indicators of central government accountability focus largely on the characteristics of 2
control mechanisms, rather than the ability of other agents to sanction government officials, where misuse of funds occurs. Hence, the PEFA framework provides more of a technical summary of a Public Finance Management system, omitting important political and cultural aspects relevant for corruption. This is a challenge the Bank is recommended to address. In order to minimize the risk of corruption in World Bank funded projects/programs, proper monitoring mechanisms are required, involving formal accounting and auditing mechanisms, public expenditure tracking surveys (PETS) as well as monitoring by local stakeholders. Projects should be monitored from the planning phase through to implementation, including the pre-qualification, tendering, procurement and construction processes. Impacts should be evaluated and communicated to the public. This requires a set of indicators of change induced by the project or program. Such indicators should be based on a set of data that is easily accessible, easily compiled and easily maintained over time. Efforts to fight the supply side of corruption need to be strengthened. Firms detected for paying bribes should be blacklisted from future contracts and named-andshamed. Anti-corruption efforts in natural resource extraction projects (oil, minerals, forestry, fisheries and wildlife) require urgent priority due to the damaging impacts of corruption in these sectors on state revenues and, thus, financial resources available for infrastructure and pro-poor service delivery. The EITI is a promising approach in this respect (though its implementation will require thorough following-up by the Bank and other stakeholders). Generally, it is important to report all cases of malpractice, as well as all success stories. Improved donor coordination is important, and the World Bank should aim to play a prominent role to achieving this. Promising lessons come from, for instance, Tanzania, where the Bank and bilateral donors have formed joint ventures to ensure efficiency and avoid waste by pooling resources together in civil service reforms. It is, however, important not to try to develop a one-size-fits-all approach and apply the same solutions to all countries receiving aid. Moreover, it should be acknowledged that aid is highly political, and the donor community itself is composed of many and sometimes conflicting interests and priorities. It must not be forgotten that corruption is a complex, multi-layered phenomenon that cannot be tackled but by a comprehensive, long-term, collective approach. Perhaps the most easily overlooked lesson about governance and anti-corruption reforms is that it takes a long time to achieve fundamental reform of institutions. The experience of Western countries illustrates this well. It took centuries for today s developed countries to build reasonably effective and accountable public institutions, while many poor countries are attempting to achieve this in a couple of decades. It is therefore not surprising that the record to date has been full of failures and disappointments. Sustainable change demands sustained effort, commitment and leadership over a long time. Mistakes and setbacks are a normal and inevitable part of the process. The big challenge is to use failures as learning opportunities, rather than as excuses for abolishing reforms. 3
1. Engagement Strategy: What should the Bank do in countries with weak governance and severe corruption, where government leadership is not consistent in tackling these problems? In such circumstances, how should the Bank remain engaged to help reduce poverty? Are there circumstances under which the Bank should restrict or stop lending? How can the Bank ensure consistent treatment across countries? World Bank-imposed conditionality is no longer (has it ever been?) an effective stick to curb corruption in borrowing countries, partly because of increasing unconditional aid from new donor countries like China. The Bank should, however, recognize that some conditions are profoundly difficult to change. In order to improve the chances of making a positive difference in countries with weak governance and severe corruption, the Bank and bilateral donors - need to acknowledge that: Political will is always partial, qualified and temporary. Economic resources are usually seriously inadequate. Governance institutions have significant weaknesses, and may require a long time to improve. The Bank needs to concentrate its attention on those factors that it can influence, and that have an impact on the effectiveness of governance and anti-corruption reforms. Occasionally, the Bank may use an emphasis on anti-corruption work to expedite a response from government and, if that is not forthcoming, Bank support may be curtailed. In relation to anti-corruption policies, Bank support may be terminated either because of a failure by the government to deliver as agreed or because of a failure by the government to account properly for the disbursement of funds. However, financial support to institutional reforms and anti-corruption initiatives is too rarely predicated on: Suitable pre-funding assessment. Consideration of which aspects of the governance and corruption problem the Bank-supported program/project is best equipped to tackle. Meaningful measurement criteria for assessing success or failure. Hence, the Bank and other involved donors are advised to collectively, on a countryspecific basis, undertake a review under three categories: Markets and Context: Identifying what is the threat. What types and levels of corruption exist and what threats do they pose to politics, administration, public perceptions and democratization? These need to be identified, in order to develop the strategy and an institutional shape. Management and Managed Work: The context should determine the institutional requirements and the organization of the work to be undertaken by the specific government institutions. 4
Measurement and Performance: Strategy and organizational design is needed to ensure that focus and funding is fully translated into the anticipated delivery of, or improvement in, performance. In order to minimize the risk of corruption in Bank funded projects/programs, proper monitoring mechanisms are required, involving formal accounting and auditing mechanisms as well as monitoring by local stakeholders. In this respect, it is relevant to work with reliable civil society organizations that are close to the local people and committed to good governance. This requires a mapping by local Bank staff of the NGOs that operate in the country. Projects should be monitored from the planning phase through to implementation, including the pre-qualification, tendering, procurement and construction processes. Impacts should be evaluated and progress reports communicated to the public, for instance through popular newspapers. This requires a set of indicators of change induced by the project or program. Such indicators should be based on a set of data that is easily accessible, easily compiled and easily maintained over time. Public expenditure tracking surveys (PETS) and service delivery surveys in key sectors supported by the Bank should become institutionalized. The Bank has applied these monitoring tools with success in some countries, like Uganda and Honduras, to locate and quantify political and bureaucratic capture, as well as to identifying incentives and corrupt practices in service delivery. However, there are indications that the Bank in recent years has not seriously applied these methodologies, or at least not exploited their potential as anti-corruption measures. The Bank should also put forward strong criteria for its assistance to be continued if progress is not sufficient. For this the Bank can invest in parallel projects - preferably for the same target group but through different institutions. If progress is not sufficient in one project, the Bank can stop lending (and make the reason public) and transfer the remaining funds to other programs/projects where progress is documented. Banks and financial intermediaries are potential allies to fighting corruption. A regional and transnational approach is required for the efficacy of governance policies in areas of high corruption. For political corruption the measures should be tied to structural reforms and deregulation. The World Bank should continue its support to strengthening the judicial system and to building democratic institutions of accountability, such as the Parliament s public accounts committee, the auditor general s office, the auditor general s annual report, a free press, and contribute to building capacity of local leaders as a long-term way to curbing corruption. In this respect, it is of key importance that the Bank engages more with youth, in order to promote good governance as a value among future leaders, and as part of a long-term strategy to fight corruption. 2. A Multistakeholder Approach: While government is the key counterpart of the Bank, how can the World Bank Group better engage non-governmental stakeholders, including civil society, media, and the private sector? 5
Building and strengthening existing strategic partnerships with multiple stakeholders is essential for the World Bank. The Bank already works with a range of NGOs, academic associations, youth organizations, groups of parliamentarians and enterprises. The Bank should engage more with networks such as the RAD and the PNoWB that know the Bank well and can provide useful inputs on many issues. The World Bank should look at the term civil society more broadly and include youth organizations, religious groups, trade unions or vernacular (grassroots) institutions among its partners. The Bank is recommended to invest time and resources in identifying a pool of reliable persons representing the civil society at large, to be consulted on a regular basis. Before engaging with any stakeholders, the Bank needs to ask itself: What organizations are active in the country? What is their level of participation in public affairs? What is their capacity? Are they in the position to challenge the government? Who are the owners of these organizations? (Do they have any vested interests?) Do they have national affiliation? Are they neutral? Do they represent the opinion of the majority at the grassroots level? Does the World Bank have the capacity to work with them? When these key questions are answered, the Bank needs to engage in real dialogue with the selected stakeholders, making sure that consultations are not a pure formality (cheap talk), but result in genuine input. Dialogue is useful to clarify ends and expectations, and to coordinate efforts. However, some NGOs in developing countries are merely rent-seeking instruments with no real development impact. The funds that are allocated to such NGOs do not reach the intended target groups, but are captured by local economic and political elites. Thus, NGOs should be reviewed to identify whether they do genuine work before allocating funds to them. An effective monitoring system of the NGO sector, combining formal audits and surveys among the target group, is indispensable. If mismanagement of funds is revealed immediate action is required implying prosecution and withdrawal of funding. Such measures will not provoke immediate change, but might bring about positive developments in the longer run. 3. Mitigating Fiduciary Risk: How can the Bank ensure that its grant and loan proceeds are used for their intended purposes, while helping countries build their own systems and capacity? There is no simple way to mitigate fiduciary risk, and corruption is not properly covered by existing fiduciary risk assessment tools. To assess the quality and characteristics of the 6
public financial management (PFM) systems of partner countries, the Bank and bilateral donors employ various types of fiduciary risk assessment tools. Recently, a harmonized framework for assessing fiduciary risk has been developed by a Public Expenditure and Financial Accountability (PEFA) working group. Several donors have signaled that they will substitute the PEFA framework for older methods of assessment. The PEFA framework consists of 31 high level indicators that assess the PFM system of a partner country. Some of the PFM characteristics captured by the framework will in part determine whether corruption will flourish or not. Comprehensiveness and proper classification of the budget and fiscal information, implies that officials will find it harder to divert or misuse public funds. Transparent and good fiscal information, available to the public, makes misuse easier to detect, as does effective internal and external control. Nevertheless, the PEFA framework is inadequate for capturing corruption. Important determinants of corruption are left out, and the indicators are therefore insufficient to judge whether corruption is likely to be high or low in the country assessed. In particular, the indicators of central government accountability focus largely on the characteristics of control mechanisms, rather than the ability of other agents to sanction government officials where misuse of funds occurs. In sum, the PEFA framework provides more of a technical summary of a PFM system, omitting important political and cultural aspects relevant for corruption. This is a challenge the Bank is recommended to address. In scaling up aid, it is tempting to resort to aid modalities that facilitate easy disbursement. Although there is as yet no empirical study of the specific impact of corruption on the effect of budget support versus project support, we know that corrupt countries have less productive and less pro-poor public spending. Several studies, including World Bank research, show that corrupt countries spend less on education and health, and more on the military. As human capital formation is conducive to growth, whereas military spending may be detrimental thereto, this may partly explain why corrupt countries have lower rates of growth. Similarly, spending on education and health is directly relevant for poverty reduction, and public resources therefore have less of an impact on poverty in corrupt countries. The implication for budget support is that if aid is less than perfectly fungible, budget support has a lower impact on growth and poverty reduction in corrupt countries. Though budget support is suitable for developing countries with good institutions, it is a poor choice where institutions are weak. Corruption makes budget support ineffective, and in some countries even counter-productive. Giving budget support to highly corrupt countries, such as Bangladesh, or countries with a high degree of political corruption, such as Malawi, is therefore unlikely to spur development. If budget support evaluations and fiduciary risk assessments do not pay sufficiently attention to the issue of corruption, allocations of budget support are bound to be flawed. A collateral program of institutional building appears indispensable. The measures against corruption should involve local institutions and political bodies, 7
with endorsement and open support of the citizens. Grants and loans should not be disbursed as a lump-sum payment, but must be based on progress in the field. Hence, there is a need to effectively monitor the progress of designated projects or activities. Civil society actors need to be aware and continuously informed about the Bank s lending to a country, and be facilitated to raise their voice, and when required put pressure on the government for change. 4. Global Collective Action: Should donors have a more common approach? How can mixed signals by donors be avoided? How can the role of multinationals from industrialized countries in corruption be addressed? Ideally, donors should have a common approach in a given country and thereby not overburden governments in recipient countries with many uncoordinated and sometimes conflicting initiatives. Lessons from, for instance, Tanzania, where the Bank and bilateral donors have formed joint ventures to ensure efficiency and avoid waste by pooling resources together in civil service reforms, are promising. It is, however, important not to try to develop a one-size-fits-all approach and apply the same solutions to all countries receiving aid. Moreover, it should be acknowledged that aid is highly political, and the donor community itself is composed of many and sometimes conflicting interests and priorities. It is important that donors, as a collective, aim to reduce harmful practices in development aid, such as insufficient control of funds and over-funding which may undermine the incentives for poor countries to build institutions for domestic revenue generation and, thus, to achieving fiscal self-reliance. The relationship between statebuilding and taxation in the developing world has received almost no attention by the aid community, including the Bank. There is a strong argument in the literature on statebuilding in Western countries that a substantial governance dividend can be gained from mobilizing domestic financial resources through the tax system. A virtuous circle may be generated whereby the generation of government tax revenues leads to improved service provision, which in turn increases citizens willingness to pay their taxes. However, high levels of foreign aid that are equal to or exceed government budgets, raise important questions about disconnections between governments and their citizens. One question is whether the attention paid by governments to the Bank s demands tends to displace attention from the much longer term and difficult task of improving domestic revenue collection. Increasing aid dependency means that revenue sustainability is an urgent matter for the Bank and developing countries alike. The way aid is disbursed in some aid-dependent countries may fuel corruption. Disbursement goals and a push to spend make it difficult for the Bank and other donors to monitor or evaluate adequately the quality of their assistance. Their tendency to move into new areas and activities at the same time adds to the problem of too much aid chasing too little absorptive capacity in poor countries. Thus, in situations characterized by poor control and monitoring, the strategy of recipient responsibility and local 8
ownership may prove completely irresponsible. A secure source of foreign aid can be like a diamond mine from which corrupt officials and politicians can extract rents. Neither the Bank nor bilateral donors are willing to admit that they have weak control over their spending as they do not want to be seen as supporting non-performing and corrupt activities. Adequate discussion of the role of aid in fuelling corruption is missing from the current debate. Transparency and accountability of donor agencies in this respect are also limited. This may partly be explained by the theory of political economist William Niskanen on the behavior of bureaucracies. He argues that to sustain and expand its budget, a bureaucracy needs a monopoly of knowledge. The better it can control information, the better the bureaucracy will persist. This may be why the process of appraisal, implementation and evaluation of aid is often carried out by a limited number of consultancy companies with close and long-term affiliations with the donors. As such, evaluation is not a tool for acquiring new insights to improve future performance, but a tool for legitimizing ongoing activities and policies. The prescription of good governance through improved accountability and transparency should be introduced in the Bank and donor agencies as well. To address the question of corruption, the Bank should focus on its own role in sustaining the problem. To establish credibility, more openness about weaknesses in the aid system is needed. Transparency International annually publishes a corruption perception index (CPI), ranking countries according to their levels of corruption. There seems to be a need for an index ranking donor agencies, including the Bank, according to similar criteria, to make the public more informed and the agencies more accountable about this important issue. More than three decades ago the Swedish Nobel Laureate in Economics, Gunnar Myrdal, wrote in Asian Drama (1968): One problem of considerable importance requiring specific attention is the role of Western business in feeding corruption in South Asia. He argued that Western businessmen undermined the integrity of politicians and administrators through bribes, and that this effect was strengthened by aid. This issue is essential to address if the current strategies to improving governance and fight corruption are to succeed. Many bribes paid in the course of international business originate in companies headquartered in the same industrial countries whose governments are now calling for anti-corruption campaigns in developing countries. Large-scale, capital-intensive infrastructure projects like big dams have been particularly attractive, for both donors and recipients. The donor country receives return on its investments through supplies of equipment and expertise and, in the case of export credits, through the repayment of loans. On the recipient side, senior bureaucrats and ministers negotiating such contracts have at their fingertips a capital-intensive project with options for overpricing and kickbacks. Kickbacks in this context are payments by the Western companies to agents of the recipient government. The agents pocket money (or deposit them in foreign bank accounts) in exchange for ensuring that a particular firm obtains the contract or receives a higher price. According to some observers, the standard kickback paid in infrastructure 9
projects in Kenya in 1990s, for instance, was never below 10 per cent of the value of the contract. Such contracts may leave the developing country with a mountain of debts and, thereby, undermine prospects for economic recovery. Multinationals must be made aware of the need to distance themselves from such practices and penalized severely if detected by their countries of origin. In recent years, pressurized by international NGOs, multinationals have declared readiness to contribute to fighting corruption, but so far they have not shown much action. The World Bank needs to join the efforts of the organized civil society in monitoring and reporting on corrupt practices in multinationals (blacklisting). All cases of corruption should be reported to the public, making it possible for buyers to make informed consumer choices. Multinationals often shrug off allegations of corruption by claiming two-facedly that the acts are committed by the local contractors upon whom they have no control. Pressure needs to be exerted on them to acknowledge responsibility for what happens along the entire production distribution chain. 4. Tracking Change: How should progress be monitored? Progress should not be monitored based on GDP, poverty rate or HDI, but by using local definitions, established together with local partners at the start of the program/project. The monitoring should be conducted through regular, measurable targets of the project in place. Although users of governance indicators easily get lost in the jungle of hundreds of existing indicators, it makes sense to measure progress with regard to corruption using existing comparative indices. In particular the governance indicators developed by the World Bank Institute (the KKZ -indicators), prepared annually from 2006 and covering more than 200 countries, may be useful. Comparative indices can be prepared both country-wide and regionally, offering a chance for less corrupt regions to derive profits from their success, for example by attracting more foreign investors. Progress, once measured, should be communicated broadly, by way of success stories setting positive examples. However, perhaps the most easily overlooked lesson about governance and anti-corruption reforms is that it takes a long time to achieve fundamental reform of institutions. The experience of Western countries illustrates this well. It took centuries for today s developed countries to build reasonably effective and accountable public institutions, while many poor countries are attempting to achieve this in a couple of decades. It is therefore not surprising that the record to date has been full of failures and disappointments. Sustainable change demands sustained effort, commitment and leadership over a long time. Mistakes and setbacks are a normal and inevitable part of the process. The big challenge is to use failures as learning opportunities, rather than as excuses for abolishing reforms. 10
Other comments: It is important to tackle corruption at its roots and fight it in every area of life in a continuous way. It needs to be remembered that corruption in many developing countries is broader than just two parties: the one who gives a bribe and the one who accepts it. It involves the entire society, where the majority not necessarily engages in corrupt acts, but silently approves or accepts instances of bad governance. It may take a generation or more to alter the mentality that allows for corruption in the public sphere. And the change in popular attitudes cannot take place without adequate education programs and good communication efforts. Sustained development cannot grow from an institutional framework that fosters corruption. 11