WB/IMF roles


Beyond Meltzer

14 June 2000 | Briefings

Where next on IMF and World Bank reform?

by Angela Wood

April 2000

  • 1. Role of the IMF
  • 2. Conditionality
  • 3. charges on loans
  • 4. Transparency and governance
  • Reforming The World Bank And Regional Development Banks
  • 1. Country and regional focus of lending
  • 2. Private financial flows and Official Development Assistance
  • 3. Structure of the World Bank Group
  • 4. Transparency
  • Conclusion
  • Acknowledgements
  • Summary Of Reform Proposals (Word 95 download)
  • Introduction

    In March, the US Congressional Committee set up to examine reform of the multilateral institutions, led by Allan Meltzer, presented its recommendations. The report proposes extensive reforms to these institutions which would completely change their nature and purpose, and it has renewed fears that the reform debate will be skewed towards right wing interests.

    This briefing is an initial response to the Meltzer report and debates about how to reform the IMF and World Bank. It aims to guide NGO policy people, journalists, officials and researchers as to what the report says and help them think through some of the implications. It critiques some of the report’s recommendations without trying to outline a comprehensive alternative reform agenda, or discuss the desirability and achievability of abolishing the organisations. For more on the latter, see the recent articles by Walden Bello and Patrick Bond in Focus on Trade, Number 48, April 2000 (at www.focusweb.org). See also: Oxfam International, Reforming the IMF, ( advocacy@oxfaminternational.org); and Jo-Marie Griesgraber, Statement Before the House Banking Committee, ( coc@coc.org).

    In general, whilst the report seems to advocate some of the right policies, though often for the wrong reasons, many of its proposals are often no more acceptable, or even worse, than the practices it objects to. The danger is that these measures would allow the US to retreat from involvement in global governance, whilst requiring greater compliance with non-negotiable conditionality and intrusion into developing countries.

    The report starts from the assumption that financial markets are largely efficient. Those problems that exist can largely be addressed by actions taken at the national level, such as providing more information and opening borders to free flowing capital. Hence, its recommendations would transform the IMF into an institution to serve the needs of private investors and facilitate the flow of private capital rather than an institution which has responsibility to ensure global stability and prosperity for all.

    Its second underlying assumption is that private finance can substitute for official development assistance. Thus, the World Bank is given a supporting role to facilitate private sector flows whilst focusing its lending on projects and countries which are undesirable from a private sector perspective.

    Both these assumptions are highly questionable. Firstly, greater transparency and more sophisticated financial sectors may be beneficial but these are not going to address serious market failures, such as herd behaviour, and they may not be priority considerations for countries mired in poverty. Secondly, governments should implement good macroeconomic policy but it should be focussed on the needs of domestic economy not foreign investors. Thirdly, private finance can be beneficial to some countries but for many it is too costly and is likely to lead to unsustainable debt burdens if too much reliance is placed on it. Fourthly, private finance is motivated by profit. Whilst many necessary development projects have a high social rate of return they often are unattractive to private investors, either because they are too risky or because they take too long to generate returns.

    Several areas are neglected in the report, the most glaring of which are: the failure to reconsider the IMF‘s macroeconomic policy advice; no proposals to involve private investors in crisis resolution mechanisms; and no examination of the governance structures of the Bank and Fund.

    Reforming The IMF

    1. Role of the IMF

    1a: the IMF should not engage in long-term lending for structural adjustment

    The IMF should be scaled back, specifically it should not be engaged over the long-term in countries through back-to-back lending programmes. The IMF‘s record with structural adjustment has been very poor, the IMF does not have the expertise to advise countries appropriately on structural or social issues and the terms of the PRGF finance are not sufficiently concessional. With regard to this latter point, the proposal to write off all debts owed by the Heavily Indebted Poor Countries to the IMF is long overdue and very welcome.

    There is still a role for the IMF to support stabilisation measures in the poorest countries, for example, in response to terms of trade shocks and natural disasters as well as financial crises. However, it policy advice and conditionality must be confined to the immediate macroeconomic concerns and should be consistent with poverty reduction objectives.

    1b: quasi Lender of Last Resort

    The report suggests that the IMF would perform this function by borrowing from the market or from bilateral sources. It is not clear what the implications of the former would be and this is not addressed. An alternative proposal which should be given serious consideration is for the IMF to issue Special Drawing Rights which could be “retired” as countries paid them back. Drawing on bilateral resources does not seem to be a good option because these will be limited. To be an effective Lender of Last Resort the IMF should not be constrained in its ability to lend.

    Also, rather contradictorily, the report recommends that borrowing from the IMF should be limited; an amount based on a year’s tax revenue is suggested. Although repayment considerations are important, which suggests looking at the terms of IMF lending, given the scale of recent financial crises this could mean that the IMF would be unable to provide enough resources to act as a Lender of Last Resort, even if lending rules were changed so that it could commit its resources very quickly.

    1c: the role of the private sector in crisis resolution

    Whilst the report acknowledges that the private sector should not be bailed out by the public sector there is no mention of what role the private sector should take in the bailout process and how it could be organised. It cannot be assumed that private investors would organise themselves to participate informally, nor, if they did so, that the outcome would be equitable to both investors and creditors. In the case of the Korean restructuring, private investors agreed to restructure debts only after the government had publicly guaranteed repayment.

    It is important to establish clear rules and processes for involving the private sector in working out rescheduling agreements and provision of new resources. The failure to do so leaves a gaping hole in the international architecture. Technically it is possible to establish clear, multilateral procedures for renegotiating debts, it is therefore unacceptable that progress has stalled on a debt standstill or stay of litigation procedure because there is a lack of political will amongst the G7 to address this issue.

    Moreover, the report says nothing about how the IMF might assist countries to stabilise and prevent an unsustainable build up of private flows through, for example, careful use of capital controls.

    2. Conditionality

    2a: Pre-qualification versus ownership

    The report misleads when it suggests the IMF should abandon conditionality. In fact, it proposes to front-load conditionality by insisting that countries pre-qualify for assistance. This would be even more constraining on governments, who at the moment at least have some leverage to negotiate timing and sequencing of conditions and to tailor them to the country context.

    The Meltzer report suggests some conditions for pre-qualifying the most insidious of which is free access to financial markets. This is effectively capital account liberalisation, which the IMF has not yet been given the mandate to impose. The report demonstrates the danger of pre-set conditions, that is, that it is likely to be the G7 and the financial and business interests that they represent who will set the conditions which borrowing countries will have to comply with.

    It would be a step backwards to move away from the ownership approach that the IMF has started to embrace through the Poverty Reduction Strategy Papers. The way forward must be towards greater ownership and to more coherence between the macroeconomic framework and development and poverty reduction goals. All countries, not just the poorest, should be allowed to define their reform priorities nationally which should form the basis for agreements with the IMF.

    2b: Pre-qualification causes more problems than is solves

    Pre-qualification also raises many of the problems which have made the Contingency Credit Line (CCL) a failure (after a year no country has signed up to it and the IMF is now reviewing how it can be made more attractive). For example, the IMF would not be able to deny finance to those countries that do not prequalify if they were systematically important because crises in these countries could have ramifications for others and the entire global financial system. It would be unfair, therefore, to deny smaller countries which do not pre-qualify assistance. Whilst the report does not discuss this issue, Allan Meltzer has suggested that loans would be provided to non-qualifiers at even higher penalty rates. However, would these loans be provided with conditions attached? The Meltzer Report argues that conditions do not work and if a country was required to implement them first, as it presumable would, this would delay the loans, which is a problem with current IMF arrangements.

    Moreover, what is regarded as the “correct” set of policies is constantly changing. In the pre-Asia crisis climate, Chile or Malaysia might have been rejected, only for it to be belatedly acknowledged that their rejection of the full neoliberal recipe had actually been a source of strength rather than weakness.

    Also, what happens if countries qualify but then their policy environments change? The act of disqualifying a country could precipitate the crisis qualification was trying to avoid. These problems were all identified with the Contingency Credit line and are not resolved in the Meltzer Report.

    2c: Limiting the number of conditions

    The Meltzer report is correct that there should be less conditionality and that conditionality to a large extent has failed. Part of the explanation for this is that governments have not witnessed good results from the reforms they have made; they have been technically difficult to implement; and they have been politically unacceptable. The IMF itself has acknowledged the failure of its traditional conditionality but rather than reassessing its policy prescriptions it has encroached into areas of structural reform and imposed more conditionality ie “second generation” reforms. However, the IMF does not have the expertise to advise governments on structural reforms. Certainly, the IMF should not impose structural conditions during periods of crisis. Whilst leverage may be greatest at such times so is the potential for harmful social impacts, as was witnessed in Indonesia when the IMF insisted on rapid closure of several banks and restructuring of others. Particularly, as it is often the less well off who are most hurt and less able to protect themselves.

    The IMF should be restricted to imposing only macroeconomic conditions and it should be required to assess, before implementation, the expected social impacts. Where reforms have negative impacts on poorer sections of society, the IMF should be required to suggest to governments alternative macroeconomic strategies.

    All too often it has been ordinary people in crisis hit countries who have borne the burden of the IMF‘s policy conditions. This was starkly seen in the East Asia crisis where, for example, very high rates of interest imposed by the IMF led to a credit crisis for small and medium enterprises, worsening unemployment and stalling recovery. It is essential that the content of IMF macroeconomic advice is seriously reconsidered in the light of mechanisms such as the Poverty Reduction Strategy Papers, the Comprehensive Development Framework and National Strategies for Sustainable Development which aim to achieve coherence between poverty reduction and sustainable development objectives and macroeconomic, structural and social policies. The IMF needs to demonstrate that it is rethinking its policy advice and that it can be flexible about what it will accept as a good macroeconomic strategy bearing in mind these goals. It should engage more openly with academics, NGOs and others outside the IMF in this debate.

    3. charges on loans

    The Meltzer report calls for penalty rates of interest to be charged on IMF loans and to accord them senior status to ensure that countries only come to the IMF as a last resort. As pointed out in the Dissenting report from Fred Bergsten et al, “the Fund already has de facto senior status and has already sharply increased its lending rates, however, and a crisis country in any event is motivated primarily by acquiring additional liquidity rather than by the terms thereof.”

    Meltzer proposes to hit countries twice, first with up-front conditionality and then with penalty rates of interest. As the IMF is now beginning to recognise as it reconsiders the terms of the CCL, countries need to be given some incentive for adhering to up-front conditions. Charging penalty rates is not conducive to encouraging countries to forgo policy autonomy.

    It is not appropriate to offer the poorest countries loans at rates near or above market rates, they simply can not afford them. Therefore it is not appropriate to eliminate all concessional financing.

    4. Transparency and governance

    4a: improving transparency and accountability

    The report makes some useful recommendations for improving transparency in particular it calls for: publication of the details of lending programmes and Article 4 reports; and that the Executive Board should be required to take votes which should be recorded and published, and summaries of Board meetings should be published after a time lag.

    In addition to (or instead of) these the IMF should:

    • publish all staff reports, including internal evaluations and reviews;
    • publish agendas of board meetings at least a week prior to meetings and minutes;
    • draft Letters of Intent should be made available to stakeholders in countries;
    • the IMF‘s new evaluation mechanism should systematically evaluate all programmes as well as conduct periodic reviews of policy and operations, the results of which should be disclosed publicly.

    A significant oversight is the report’s failure to address the IMF‘s governance structure. The current system for allocating IMF votes is unfair and obstructs developing countries from participating in decision making processes. Asian countries are particularly disgruntled about their restrictive voting (and borrowing) capacity given their importance in the global economy, whilst 43 Sub-Saharan African countries are represented by only two executive directors on the IMF Board and account for less than 5% of the votes. Contrast this with the US which has veto power with its share of 18% of the vote. This inequality is mirrored on the IMF‘s governing body, the International Monetary and Finance Committee.

    Given this scenario it is not surprising that IMF advice does not reflect developing countries’ interests and that programmes breakdown or that countries, such as Malaysia, prefer to opt out of the multilateral system and chance it on their own or develop regional structures which better suit them.

    At the very least voting rules should be changed so that US can no longer veto decisions. Also, there should be a more equitable allocation of votes, perhaps based on national income and population, which should lead to a better balance of developing countries, including the poorest, represented on the governing bodies. The formula used to allocate votes and quotas should be objective and transparent. The recent report on the review of the quota formula should be published immediately.

    Reforming The World Bank And Regional Development Banks

    1. Country and regional focus of lending

    1a: lending to middle income countries should be focussed on poverty reduction

    The report argues that lending should be phased out for countries with annual per capita incomes of $2500+ and should cease completely for countries with per capita incomes of $4000+. Whilst it is legitimate that these countries should seek private finance for profitable investments, there is still a case to be made for providing multilateral loans for projects which are not attractive to private investors but which offer a high social return, for example, providing rural water supplies.

    Many middle-income countries which are able to attract private finance have large numbers of people living in poverty, for example, China, India and Brazil. However, much multilateral lending to middle income countries is not focussed on poverty reduction or reducing inequality, nor is it framed within a coherent development strategy. Multilateral lending should be able to demonstrate clear impacts on these goals.

    1b: competition between the World Bank and regional banks is healthy

    The report argues that to reduce the overlap between the World Bank and the Regional Development Banks (RDBs) that the Bank should only lend to countries in Africa, Europe and the Middle East. This assumes wrongly that the RDBs perform well and that their lending is poverty focussed. In fact they are often weaker than the World Bank in the areas of poverty focus, transparency and participation.

    However, there is a case for improving the effectiveness of the RDBs, particularly because they are likely to be more aware of the issues and problems important to countries in the region and therefore more able to provide appropriate advice. Eliminating World Bank lending in some regions would remove a source of competition which could lead to both the World Bank and RDBs being less effective.

    2. Private financial flows and Official Development Assistance

    2a: private finance is often too unstable and too costly for development purposes

    Whilst there is a case for government’s to encourage long-term inflows of capital there is also a case for enabling governments to direct this finance to projects which will have a positive, sustainable development impact. Governments should not be prevented from doing so by conditionality or rules imposed at the multilateral level.

    However, much of the private capital is invested on a short term basis and can be pulled out of a country very quickly. Such actions may not be related to problems in the host economy but simply because investment options are more attractive elsewhere, especially if interest rates rise in developed countries. The uncertainty of short-term flows makes this source of finance impractical for development planning and unsuitable for development projects which often take several years to generate returns.

    Furthermore, whilst Foreign Direct Investment (FDI) may be preferable to short-term flows because investments are more long term, FDI often leads to a larger outflow of resources (via profit remittances etc) than it brings in. The cost of this finance very quickly can become unsustainable as governments get locked into a cycle of attracting increasingly large flows of capital to maintain a positive balance of payments. Also, since governments need to maintain a “favourable environment” for private flows they often subordinate domestic needs to the needs of foreign private investors. With an increasing number of countries being encouraged to attract foreign investment the likelihood is that the benefits of FDI to the host country will be eroded as developing countries compete amongst themselves to offer more attractive terms to investors.

    Therefore, whilst there is a case for graduating middle income countries from multilateral finance, overall, it cannot be assumed that private finance can substitute for development assistance in all countries. Moreover, the primary objective of aid should not simply be to facilitate the flow of foreign private investment, other uses may be much more effective at reducing poverty or inequality. If it is used to facilitate private investment then the focus should be to encourage domestic investment not foreign investment.

    2b: concessional lending and grant financing for social sectors and infrastructure

    Cutting Bank lending to middle-income countries would remove an important source of funds for concessional lending. The report simply proposes to replace this with more bilateral finance. Whilst more bilateral development assistance would be welcome, it is doubtful that the World Bank should control more bilateral aid money. Moreover, it is extremely unlikely given the current political climate that a substantial increase in aid money will be forthcoming. Thus, cutting lending to middle income countries may well result in fewer concessional resources available for the poorest countries.

    Although the Bank already provides highly concessional finance through the International Development Association, the report advocates the use of grants for social sector and infrastructure projects. The Bank is channelling an increasing amount of lending into these areas and into social safety nets. Given that these types of projects can take many years to generate returns, and often then only indirectly, or may not do so at all, especially where safety-nets are being applied in crisis situations, it would be sensible to reconsider the terms under which the Bank lends for them with a view to providing grants or more enhanced concessional lending.

    2c: cutting out corruption and improving project success

    The suggestion that the Bank and RDBs should bypass ineffective governments altogether and issue contracts directly to private service providers in areas such as health and education is at best a short-sighted solution and is effectively unworkable. Firstly, the report suggests that contractors would not be paid until after the project is completed and evaluated as satisfactory. Its hard to see how this could work given that infrastructure projects in particular can take several years to plan and then implement and that the resources available in developing countries are limited which is why lending is necessary. Secondly, the report does not make clear who would be responsible for repayment, the government should not be required to guarantee repayment if it is not involved in the project/lending process. Thirdly, given the size, distance from borrowers, and structure of the World Bank it is not able to effectively provide small scale project loans.

    Whilst corruption needs to be addressed, it is important to examine how the IMF and World Bank encourage corruption through their structural adjustment programmes. For example, poorly conceived plans to cut civil service costs and rushed privatisation schemes have both led to increased corruption.

    Also, more attention should be focused at the institutional level to improve how projects are designed and managed. Staff members should be held accountable for project failures and incentive structures should be linked to project outcomes. Changing the Bank’s name to the World Development Agency, as the Meltzer Report recommends, would be beneficial to identify that the Bank is a development institution not a profit driven finance institution.

    3. Structure of the World Bank Group

    The report recommends that the Multilateral Investment Guarantee Agency (MIGA) should be scrapped and the International Finance Corporation (IFC) should be merged into the World Bank. Whilst the former recommendation is not problematic the latter could be.

    MIGA provides a “win-win” opportunity for private investors. Not only are they are able to demand high rates of interest to cover themselves against the risks of investing in developing countries, such as policy changes, MIGA also provides them with guarantees against these same risks. Also, because the cost of the guarantees, if used, are borne by the developing country this acts to discourage governments from taking actions which would necessitate their use. The private sector is quite able to provide its own risk insurance and therefore it would be sensible to scrap MIGA.

    The authors of the Meltzer report seems to be unaware that some IFC departments have already merged with the World Bank. A further merger should be taken very cautiously, it could result in a conflict of interests, for example, if the Bank was to lend to governments and the private sector for the same project.

    4. Transparency

    The report calls for annual, independent evaluation of the World Bank’s effectiveness. Regular, independent evaluation would be welcome. In addition the World Bank’s Operations Evaluation Department should publish all its evaluation reports, particularly programme evaluations. Plus, draft CASs should be made available to stakeholders and final CAS documents should be published.

    Also, the World Bank’s governance structures are as weak as the IMF‘s (see above). Several steps should be taken to improve its governance and the transparency of its decision making processes:

    • formal votes should be taken and the results of how each country voted made public;
    • voting and quota allocations should be more equitably allocated based on a transparent and objective formula;
    • agendas of board meetings should be published at least a week prior to meetings and the minutes of meetings should be disclosed.


    A thorough review of the policies, functions, governance and structures of the multilateral financial institutions is essential. The Meltzer report has put this issue on the international agenda but it must not set the reform agenda.

    A review should be transparent and involve developing country governments and other stakeholders such as NGOs and civil society organisations, academics, and parliamentarians. It would be very inappropriate if it is orchestrated by the G7and therefore it must be led by an independent, representative body elected through a transparent process. The review should also examine what aspects of the global architecture are still missing and take forward the stalled reform efforts initiated in response to the financial crises of 1997 and 1998.

    Annex 1

    Summary Of Reform Proposals (Word 95 download)


    Thank you to Duncan Green, CAFOD, whose comments were used in this briefing.