Policy Brief 00-5

The Role of the IMF: A Guide to the Reports

by John Williamson, Peterson Institute

May 2000

This brief is a preliminary draft of a paper commissioned by the Commonwealth Secretariat for presentation to a Conference on Developing Countries and the Global Financial Architecture organized by the Commonwealth Secretariat and the World Bank in London on June 22-23, 2000. The author is indebted to C. Fred Bergsten, Morris Goldstein, Stephany Griffith-Jones, and Kevin Morrison for comments on a previous draft.

© Institute for International Economics. All rights reserved.

 

Introduction

Many of the discussions on a new international financial architecture that were spawned by the east Asian crisis have dealt with the future role of the International Monetary Fund (IMF). This policy brief starts by summarizing the recommendations of five recent reports and one speech, and the reasoning that lies behind them. These recommendations are divided into four major areas: 1) the scope of Fund activities, 2) surveillance, 3) lending, and 4) governance (on which topic I also summarize the views of a recent academic paper). I offer my own verdict on the first three of these topics in the last section of this brief.

The first of the five reports considered was jointly published by the International Center for Monetary and Banking Studies in Geneva and the Centre for Economic Policy Research (CEPR) in London. The authors were Jose De Gregorio, Barry Eichengreen, Takatoshi Ito, and Charles Wyplosz (1999). The report also contains brief accounts of alternative reform proposals, by Kiichi Miyazawa, Jeffrey Sachs, Sebastian Edwards, France, the United Kingdom, and Italy, and of the idea of regional funds. It was discussed at a conference held in Geneva in May 1999, which is also reported in the volume. This will be referred to as the Geneva Report.

The second report is that of an Independent Task Force sponsored by the Council on Foreign Relations (1999). This task force was jointly chaired by Carla Hills and Peter Peterson, with my colleague Morris Goldstein as project director and 23 other luminaries of the American internationalist establishment (including C. Fred Bergsten, director of the Institute for International Economics) as members. This will be referred to as the CFR Report. It contained eight statements of Dissenting Views, but all members signed the main report.

The third report was commissioned by the G-24 and written by Montek Ahluwalia in 1999. It is published by the United Nations Conference on Trade and Development in the latest volume of its series of publications of G-24 studies. It will be referred to as the Ahluwalia Report.

The fourth report is that of the International Financial Institution [sic] Advisory Commission established by the US Congress and chaired by Allan Meltzer, with an additional ten members who included academics Charles Calomiris, Jerome Levinson, and Jeffrey Sachs, businessmen, politicians, and think tank Directors C. Fred Bergsten of the Institute for International Economics and Edwin Feulner of the Heritage Foundation. Despite the singular "Institution" in its title, the terms of reference covered the World Bank, the three regional development banks, the World Trade Organization (WTO), and the Bank for International Settlements (BIS), in addition to the IMF. This was issued in March 2000, and will be referred to as the IFIAC report. It was accompanied by two "Supporting Statements" arguing that it did not go far enough in gutting the international financial institutions (IFI), a Joint Minority Statement by four members (including Bergsten and Levinson), three of whom did not sign the main report, and two additional dissents by two members of the minority (one of whom was Levinson) who did not sign the main report.

The fifth report is that of a task force established by the Overseas Development Council in Washington which reported in April 2000. This was cochaired by John Sewell and Sylvia Saborio, directed by Kevin Morrison, and comprised a further 11 members from academia, think tanks, and NGOs, "who agreed with the overall direction and recommendations of the report, but not necessarily with all statements and emphases." I was a member of this Task Force, as were Nancy Birdsall and Joe Stiglitz. This will be referred to as the ODC Report.

---

---

The speech included is, of course, that given by US Secretary of the Treasury Lawrence Summers at the London Business School last December (Summers 1999).

These six works include the views advanced by an international group of academics, by a collection of Americans who qualify as the great and the good, by a leading developing country official writing on behalf of the Group of 24 developing countries, by a mixed group of Americans writing a report for the US Congress, by another mixed group of predominantly American composition concerned with the problems of developing countries, and by the US Secretary of the Treasury. Although there is obviously some bias toward American sources, I believe this provides a reasonable cross section of informed thought on which to develop a set of proposals.

 

The Scope of the Fund

Not all of the six documents being surveyed address all four of the topics into which I am dividing the discussion. The desirable scope of Fund activities, for example, is not touched on by the Geneva Report.

The CFR Report also treats the topic relatively lightly, but it does urge the Fund (and, for that matter, the World Bank) to go "back to basics." It argues (p.115) that the Fund is still needed to help countries resolve payments problems in an internationally responsible way, to address liquidity crises, and to act as a crisis manager or convenor, and elsewhere it emphasizes the Fund's role in crisis prevention. It also argues that "the IMF is losing its focus and reducing its effectiveness by doing too much. Specifically, the IMF should limit the scope of its conditionality to monetary, fiscal, exchange rate, and financial-sector policies" (p.116). But its surveillance needs to be concerned with monitoring compliance with financial standards, as well as macro fundamentals.

The Ahluwalia Report (p.22) dismisses the case for a merger between the Fund and the Bank on the grounds that there is an important and distinctive role for the Fund in dealing with crises, both in prevention (via surveillance) and in management (via financing). The report also argues that "such financing does not have to be long-term and certainly not concessional." It also states that the "Fund should focus more sharply on sources of instability in the international financial system and on handling balance-of-payments problems which are either short-term or systemic in nature. It could even be argued that financing operations related to chronic balance-of-payments problems of low-income countries, e.g. the Enhanced Structural Adjustment Facilities (ESAF) and…Heavily Indebted Poor Countries (HIPC) initiative…should perhaps be shifted to the Bank, with cooperation from the Fund being available on technical matters."

The Executive Summary of the IFIAC Report declares that "the IMF should continue as crisis manager under new rules that give member countries incentives to increase the safety and soundness of their financial systems" (p.6). The Report identified three roles that this implied: 1) in serving as quasi-lender of last resort to emerging economies; 2) in collecting, publishing, and disseminating data on member countries; 3) and in providing advice (as opposed to imposing conditionality) relating to economic policy (pp.42-43). It urged an end to long-term loans, and specifically called for closing what it called the "Poverty and Growth Facility" (p.43). It also called for the replacement of conditionality by prequalification, according to principles outlined in the section on lending below.

---

---

The ODC Report identified the IMF's core competence as macroeconomic policy, and hence sees its central role as crisis avoidance and, when that fails, promoting speedy recovery from crisis. It argued that this implied that its lending should be restricted to short-term liquidity lending in macroeconomic crises, and called for moving the Poverty Reduction and Growth Facility (PRGF) to the Bank. It should retain a role in the poorest countries, but only in the context of emergency lending, as in other member countries. The Fund should maintain its role in surveillance, geared toward providing advice that would minimize the probability of crises. But the report argued that statistics should be collected and disseminated by an independent statistical agency, rather than the Fund (or Bank).

The Summers speech also urged a focus on core competence (p.5), but it interpreted this rather more widely than the preceding reports. The IMF should promote financial stability within countries, a stable flow of capital between them, and rapid recovery following any financial disruption (p.3). Summers asserted that this points to six critical areas: 1) promoting the flow of information from governments to markets and investors, 2) giving attention to financial vulnerability as well as macroeconomic fundamentals, 3) developing a more selective financing role focused on emergency situations, 4) catalyzing market-based solutions, 5) focusing on growth and poverty reduction in the poorest countries, and 6) modernizing the IMF as an institution.

The degree of consensus reflected in these five sources is rather remarkable. All reflect a concern with mission creep, and urge the Fund to focus on its core competence. All see the Fund as having a central role in aiming to prevent financial crises, and in managing them when they nevertheless occur. All wish the Fund to continue to lend in crisis situations. All concur in wishing to maintain surveillance, and none challenge the proposition that this should focus on financial standards and vulnerability as well as traditional macroeconomic fundamentals.

Despite the apparent agreement, there is a profound gulf between the IFIAC majority on the one hand, and the other five (and the IFIAC minority as well) on the other, about the value of having an IMF at all. Everyone except the IFIAC majority emphasizes the need for an international institution dedicated to building collaborative macroeconomic policies among countries, to helping avoid crises, and to aiding countries cope with crises that nonetheless occur. They all appear to agree that the world is a lot better off for having built that degree of international collaboration. The IFIAC majority start instead from a concern that IMF lending may promote moral hazard (a phenomenon whose importance "cannot be overstated," according to the report¹), and concede only reluctantly (and to the dismay of two of their number) that there may after all be a limited role for the Fund.

On more concrete issues the disputed topics would seem to be:

· whether the Fund should maintain the PRGF (Summers, yes), whether this should be closed (IFIAC majority report), or whether it should be moved to the Bank (Ahluwalia and ODC); and

· whether the collection and dissemination of statistics should be moved to a separate agency, as urged by ODC.

 

Surveillance

Fund surveillance takes two forms: 1) general surveillance of the world economy, as reflected in the biennial publication World Economic Outlook and the annual International Capital Markets Report, and 2) surveillance of individual countries, as undertaken primarily in Article IV consultations. No one seems to have challenged the utility of the former exercise, or to have offered significant suggestions for improving what the Fund does, except the Ahluwalia Report which urges that the Fund should draw on this information in introducing developing-country interests into G-7 discussions. The debate focuses rather on surveillance of individual countries, and how this could be improved to diminish the probability of crises occurring.

The Geneva Report suggests that surveillance should seek to identify country vulnerabilities in areas like the banking system, exchange rate policy, reserve levels, or accounting standards, and give countries confidential warnings of these vulnerabilities. It recognizes that the Fund lacks expertise in many of the fields where standards are needed and being developed, and urges the Fund to accept that the standards will be designed by others, with its own role being confined to monitoring, with the use of experts from other institutions in its missions.

The CFR Report also suggests that the Fund should focus on each member country's compliance with international financial standards like the Fund's Special Data Dissemination Standard, the Basel Committee's Core Principles of Effective Banking Supervision, and international accounting standards, along with a "viable" exchange rate regime, prudent debt management, etc. Rather than the results being communicated confidentially, however, the report proposes that the Fund should periodically publish a "standards report" that details each country's performance, along with the Article IV reports assessing policies and prospects. It sees the incentive to comply with these standards as being provided by the likelihood of a lower cost of market borrowing, cheaper access to Fund credit when a country has to borrow, and lower capital requirements for bank loans to those countries (pp. 93-97). The Fund should encourage countries with fragile domestic financial sectors and weak prudential frameworks to adopt Chilean-style capital inflow taxes (p.98).

---

---

The Ahluwalia Report declares that surveillance is a core activity of the Fund and recommends that it be strengthened, primarily by increasing the disclosure of key information to financial markets.

The IFIAC Report has rather little to say about surveillance, except that the Fund should abandon Article IV consultations for the OECD countries (on the grounds of avoiding costly duplication of effort) and should publish promptly all Article IV consultation reports for other countries (pp. 43-44). It also sees a major function of the Fund as being the collection and prompt publication of data, with a view to keeping market participants well informed (p.43). The Fund should encourage countries either to hard fix their exchange rate, or to float, since intermediate regimes are more subject to crises.

The ODC Report also approves of data collection (though arguing that this should be moved to a separate agency) and transparency, but argues that improvement in these directions is unlikely to end crises. It sees a unique role for the Fund in advising countries on macro policy, aimed inter alia at avoiding macroeconomic crises, and normally not based on financial arrangements. It too endorses Fund monitoring of a wide range of standards, while cautioning that the Fund does not have in-house expertise on all of them. It recommends that discussion of Article IV reports be moved from the full Executive Board to the sub-boards comprised of the executive directors (EDs) from particular geographical regions of the world, so as to diminish the work load on the full Board, but with regular reports from the sub-boards to the full Board. This report cautions against enthusiasm for the two-corners exchange rate fad (the notion that every country ought to have either a currency board or a floating exchange rate, but nothing in between).

Secretary Summers also favored a Fund role in collecting and disseminating information to investors and markets. Countries should be encouraged to adopt the Special Data Dissemination Standard and the various international codes for sound policies that are being developed, and the Fund's assessment of their compliance with those standards should be released publicly. Surveillance should cover financial vulnerability as well as macro fundamentals, and it should be recognized that this vulnerability is a function of the level of short-term foreign debt and of excessive government granting of guarantees. The Fund should focus on the strength of national balance sheets, for example, by developing a more meaningful measure of reserve adequacy than the traditional reserves/imports ratio. It should draw attention to "the dangers of opening up to short-term capital in the presence of too many domestic guarantees," and should highlight the risks posed by unsustainable exchange rate regimes.

 

Once again, the degree of consensus exhibited is quite significant. There is general enthusiasm for data collection (if not necessarily by the Fund), transparency, publication, and continued surveillance. This is rather remarkable if one considers how secretive an institution the Fund has traditionally been. Several of the sources explicitly endorse focusing attention on vulnerabilities in the financial system, foreign debt, the various areas in which international standards are being promulgated, and the exchange rate regime, and no one opposes going down that road. There was far more recognition of a possible constructive role for capital inflow taxes than one would have expected to find in these places prior to the East Asian crisis. There remains a disagreement as to whether all intermediate exchange rate regimes are to be condemned as unsustainable, but everyone recognizes that the issue of sustainability is an important one.

 

Lending

The Geneva Report expressed skepticism about the proposal to "include some form of 'prequalification' for financial support by the IMF" (p.44).² It saw the appeal of prequalification as lying in a resolution of the problem of moral hazard, since a government could no longer be expected to be bailed out if it ignored warnings of imprudent behavior. But it questioned whether government moral hazard is a real problem (governments suffer enough when they engulf their countries in a crisis to eliminate any incentive to flirt with danger), and also argued that the criteria for prequalification would be arbitrary and the policy would be time-inconsistent (the threat to withhold help from countries that have not prequalified is not credible). The report also declared that the Contingent Credit Line (CCL) created by the IMF in April 1999 suffered from the same drawbacks, plus the danger that dequalifying a country previously qualified could precipitate a crisis. It noted that no country had so far been induced to apply for a CCL (which remains true).

---

---

The Geneva Report also argued that the Fund's facilities had proliferated excessively and need to be streamlined with a view to making the Fund's emergency lending more transparent, simple, and effective (p.48). It applauded the new Supplemental Reserve Facility (the SRF, which can lend exceptionally large sums at a penalty interest rate and was first deployed in South Korea) as a step in the right direction (p.53). It argued that the capital account crises that are now dominant are essentially caused by a lack of liquidity rather than bad fundamentals, and therefore require temporary financing with front-loaded disbursements (though it questioned whether the support need always be large). The Fund support will need to be accompanied either by "cofinancing with the private-sector rollovers and rescheduling" or by a restructuring of external debt obligations in order to keep the size of financial packages within reason. The experience of South Korea in 1997 showed that a standstill could be a useful instrument in bailing in the private sector, but the experience of Mexico in 1982 should also stand as an awful warning that standstills are not a panacea.

The CFR Report proposed to draw a sharp distinction between "country crises" and "systemic crises." Finance for the former would be limited to normal access limits (100 percent of quota per year and 300 percent cumulatively), and would be financed from the Fund's existing resources. Systemic crises might be financed from the General Agreement to Borrow (GAB) and the New Agreement to Borrow (NAB) or from a proposed new Contagion Facility that would replace both the CCL and the SRF. The Contagion Facility would be used for victims of contagion in which the payments deterioration reflects developments largely beyond their own control, and would not require a Fund program (p.110). It would be financed by a one-time Special Drawing Rights (SDR) allocation in which all Fund members would donate their newly-allocated SDRs to the Contagion Facility. The report also declared that in extreme cases, where the debt profile is clearly unsustainable, the IMF would require debtors to engage in "good-faith" debt restructuring negotiations with their creditors as a condition of its support (p.102). Those discussions might be facilitated by declaration of a temporary standstill by the debtor. Interest rates on borrowings from the Fund would be lower for countries that made a series of efforts to forestall crises, by complying with the international codes being developed, following sound macro policies, maintaining a viable currency regime and a prudent debt profile, and establishing contingent sources of liquidity support (p.94).

The Ahluwalia Report is sympathetic to the Fund acting as a lender of last resort in response to capital account crises, but worries about how such lending is to be financed, and also suggests resorting to a special SDR allocation (p.14) as well as to bigger quotas. In terms of conditionality for such lending, it is sympathetic to prequalification but points to the problem that performance criteria judged adequate prior to a crisis may not still appear to be so after the crisis has erupted (p.16). It suggests a compromise solution in which prequalification would entitle a country to a first tranche almost automatically, but subsequent drawings would require conditionality. It notes the danger that withdrawal of cover before a crisis could precipitate a loss of confidence that would provoke the very crisis that the arrangement was designed to avoid.

The IFIAC Report identifies the first of the Fund's responsibilities as being "to act as a quasi-lender of last resort to solvent emerging economies" (p.42). The first point to note is that this is the only lending window that the report discusses: it appears to preclude not just lending to industrial countries, but also to the poorer developing countries that are not included in the term "emerging economies." Indeed, it specifically calls for closing the PRGF, currently the main instrument for lending to those countries (p.43), and it rules out lending for nonfinancial emergencies, such as famines (p.47). The report goes on to state that "except in unusual circumstances, where the crises poses [sic] a threat to the global economy, loans would be only to countries in crises [sic] that have preconditions that establish financial soundness" (p.43). Preconditions would replace conditionality. The preconditions proposed (pp.44-45) are:

• freedom of entry and operation for foreign financial institutions;
• well-capitalized commercial banks, preferably with part of the capital in the form of uninsured subordinated debt;
• regular and timely publication of the maturity structure of outstanding sovereign and guaranteed debt and off-balance sheet liabilities;
• "a proper fiscal requirement," the nature of which is not specified.

Countries that need to borrow before they have been able to fulfill these conditions should be entitled to do so at a "super penalty rate" (all borrowing would be at a penalty rate), and countries that choose not to fulfill the conditions should be ineligible to borrow (p.46). These loans would be of short maturity (e.g. a maximum of 120 days) with only one allowable rollover. The report also goes to considerable length to ensure the priority of IMF claims over all other claims, in analogy with the requirement of collateral in traditional last-resort lending, but in my view this is fussing over a non-problem so I do not propose to waste time on the issue.

Perhaps the most important criticism voiced by the minority who did not sign the IFIAC Report concerned the proposed limitations on borrowing from the Fund. They questioned whether it would be possible to fashion a fiscal prequalification requirement that would dispense with the need for conditionality, and also worried that the prequalification approach might preclude lending to countries of great systemic importance (pp.121-22). They commended instead the CFR proposal to grant preferential lending terms to countries that have adopted the Basel Core Principles to strengthen their domestic banking systems (pp. 123-24).³

The ODC Report also sees the Fund's lending role as driven by crisis management, but it states explicitly that all countries should be eligible to borrow from the Fund in times of macroeconomic crisis (p.6). In furtherance of the objective of cutting back the IMF to its core competence, which excludes structural issues, it proposes to abolish the Extended Fund Facility and to transfer the PRGF (and hence also responsibility for the Heavily Indebted Poor Countries program) from the Fund to the Bank. The Fund should advise the Bank on the macro conditions to be required for PRGF loans, though without a veto. The report argues that crisis lending should be done through the use of normal standby arrangements, which should be accessible by any member country, with a subsidized interest rate when one of the low-income members borrows. Conditionality should revert to focusing on the basics of macro policy, without the addition of numerous structural conditions such as adorned the East Asian programs and that in the event proved almost entirely irrelevant to nurturing the region's rapid recovery from crisis. The report calls for an effort to make ex ante assessments of the impact of IMF programs on the poor, with a view to trying to reduce their adverse impact. It expresses skepticism about the CCL but calls for maintenance of the Compensatory Financing Facility.

---

---

Secretary Summers also called for the Fund to focus its financing on emergency situations. It should be a last, not a first, resort; a backstop, not an alternative, to private finance. Longer term lending would be phased out and the core instruments would become the CCL, short-term, standby arrangements for countries with nonsystemic problems, and the SRF, for systemic capital account crises. The penalty rate on SRF lending was a precedent on which to build, although the CCL might have a lower interest rate to encourage countries to qualify and apply. Conditionality will have to fit the specific country circumstances, but it should not intrude in areas irrelevant for restoring stability and growth. However, "the stability of banking systems, issues of social cohesion and inclusion, and the capacity to enforce contractual arrangements" may all be relevant (p.6). He also urged that the official sector should help creditors to recognize their collective interest in maintaining exposure, even when their individual interest is in withdrawing funds, though it will occasionally be necessary to seek less voluntary debt restructuring, and in exceptional cases the IMF should be prepared to lend into arrears.

In his remarks on the poorest countries (p.8), Summers lauded the progress made in developing the HIPC as "a fundamentally new framework for the international community's efforts to combat poverty, one that gives the World Bank the lead and the IMF a more tightly focused role." He did not hint at the desirability of moving the PRGF.

The common theme of these six sources is the central role of IMF financing in managing crises. There are clearly a number of other ideas that have appealed quite widely without achieving unanimous support: some form of prequalification (though with a strategic difference as to whether failure to prequalify would disqualify a country from borrowing or simply stiffen the terms); shifting the PRGF from the Fund to the Bank; and accompanying IMF crisis lending by some form of payments standstill, at least in certain circumstances.

 

Governance

The Geneva Report argues that the IMF needs greater transparency and more accountability. More decisions should be taken by vote rather than consensus, and the minutes and the votes should be published. Programs should be evaluated both by staff and by outside panels, and again the findings should be published. Above all, the Executive Board should become independent in the same sense that the boards of many central banks are now independent: they should be appointed for multi-year terms and should not receive instructions from the governments that appointed them, the Board should be given an explicit mandate such as promoting economic and financial stability, and the Board should periodically report to what is now the International Monetary and Financial Committee (the IMFC, formerly known as the Interim Committee). A country under discussion should send a representative to sit with the Board. In order to increase its independence from governments, the Fund should borrow in the market rather than acquire its resources from its member governments.

The CFR Report also urged more transparency, but Fund governance was not prominent among its concerns. Likewise the Ahluwalia Report said little on this topic, except for arguing against a merger of the Fund and the Bank, and suggesting the establishment of an overarching ministerial committee to supervise them both. The IFIAC Report called for the Fund to be restructured as a smaller institution (p.42), and for more transparency in its accounting (pp.50-51).

The ODC Report called for a realignment of the voting power in the Fund to reflect the current weight of economic power, which everyone knows would involve Asian representation growing and that of Europe diminishing. The report recommended reducing the super-majority needed for certain key decisions so as to eliminate the US veto. It also called for a more neutral and transparent process to select the managing director. It recommended that the links between member countries and the Fund should be broadened, so that it might interact with a prime minister's office or a planning ministry (or, in developed countries, an aid ministry) rather than just the finance ministry and central bank. It urged establishment of a small external evaluation unit to report to the IMFC. (The Fund announced the establishment of such a permanent evaluation office just days before the report was published, but reporting to the Executive Board rather than to the IMFC.) The report also urged that data collection and dissemination should be relocated to a separate statistical agency.

---

---

It is difficult to detect much common ground between these proposals, beyond the general desire to continue to advance in the direction of greater transparency and openness, although it might be possible to find a widespread desire to reform the process of selecting the managing director after the recent fiasco. Perhaps it is premature to try to reform the Fund's governance before we have decided what we want the Fund to do.⁴

 

An Agenda for Reform of the Fund

At this stage let us turn to discussing what an agenda for the reform of the Fund might look like. I will not address the issues of governance—we need to decide first what the Fund should do.

I find myself in full agreement with the consensus views on the scope of the Fund that were noted at the end of that section. In particular, I endorse the view that the failure to resist the mission creep imposed on it by the G-7, and most specifically by the conditionality that the US Congress attached to the most recent increase in Fund quotas, threatens to undermine the effectiveness of the IMF. The Fund should indeed return to concentrate on its core competence and there seems to be unanimous agreement about what that is (at least among those who do not dismiss the IMF as irreparably incompetent). For example, the G-7 communique of 15 April 2000 stated that "Crisis prevention and response should be at the core of the IMF's work." And everyone seems to agree that that involves both maintaining surveillance with a view to avoiding crises and helping to manage such crises as nonetheless occur.

What strikes me as quite extraordinary is that the official world, including Secretary Summers as well as the decisions endorsed at the spring meetings of the IMFC, regard these principles as consistent with the maintenance of the PRGF (and therefore HIPC, whose conditionality is tied to the PRGF) in the Fund. It is one thing to oppose the recommendation of the IFIAC majority to close the PRGF,⁵ which would imply reducing the resource transfer to the poorest countries. But both the Ahluwalia and the ODC reports suggested an alternative: not closing it, but rather transferring it to the Bank. The argument for this is that the PRGF is not concerned with crisis lending, the area of the Fund's core competence, but with poverty reduction and growth. No one doubts that growth, and the poverty reduction that flows from it, is critically dependent upon disciplined macro policies, and that these lie within the core competence of the Fund. But macro policy is, as emphasized in the Bank's Comprehensive Development Framework, merely one of a number of areas that it is essential to get roughly right if an economy is to grow at anything close to its potential rate. Since the Bank has the core competence in most of these fields, it seems quixotic to place the PRGF in the Fund rather than the Bank. This is an anomaly that can be explained only by history.⁶

Quixotic it may seem to some of us, but the location of the PRGF in the Fund has been vigorously defended by Stanley Fischer in his capacity as acting managing director. As quoted in the Financial Times on 14 April 2000, he said, in response to a question about the recommendation to move the PRGF in the ODC report:

there was no reason poor countries should not be able to benefit from the IMF's expertise in macroeconomic policy. The argument strikes me as one which imagines there is a different macroeconomics for poor countries and rich countries… Inflation and economic [in]stability is bad for all people.

This misinterprets the ODC report, which explicitly argues (p.5) that the Fund has "a unique role in the international system, including in poor countries: to advise countries on how to avoid macroeconomic crisis and to restore stability in the midst of such crises." The ODC report also says baldly: "Stability is an essential condition for growth." In fact the argument is one for eliminating the differential treatment of poor and rich countries in all respects except one: the interest rate at which they are entitled to borrow should they need to borrow in the event of a macro crisis. The Fund would retain the same role in both types of countries, namely surveillance directed to crisis avoidance and short-term lending when avoidance fails. Perhaps there is a good argument for retaining the PRGF in the Fund, but, if so, Fischer did not articulate it.⁷ Of course, he would have jeopardized his reputation as an outstanding bureaucrat had he acknowledged the logic of transferring a substantial part of his organization's responsibilities to its sibling institution. One can hardly expect him to spearhead this particular reform, but that does not make it an undesirable change.

The danger of locating the facility in the Fund is that its traditions will prevent it from treating macro policy as merely one among a number of critical areas. On past experience, one has to expect that the Fund will always make macro stability primus inter pares, whether it deserves to be or not. However, when countries are in crisis, macro stability ought not to be accorded primacy. Once macro policy is reasonably adequate, it is more important for the limited human capital available in the Finance Ministry of the typical poor country to devote its attention to other topics, rather than fussing about inflation targeting or free floating or creating a treasury bill market or whatever the latest macro fad happens to be. If the Fund is in charge, there will be no one to countermand an excessive emphasis on macro perfection at the expense of getting public expenditure priorities right and reforming corporate governance and building up the education system…. If the Bank is in charge, the Fund will still have the duty to examine macro policy and will be able to make a case if it sees problems, and if the Bank agrees that macro stability is in jeopardy then it would have the duty to hold up disbursement until policy has been adjusted appropriately. This will ensure both that the Fund cannot be ignored and that countries cannot be deprived of the advice of the Fund. But since another agency will have to agree that macro stability is indeed at risk, the country will be safeguarded against an excessive emphasis on macro stability at the expense of other priorities.

---

---

The other argument for relocating the PRGF concerns the time horizon of Fund programs. We know that poverty reduction requires decades rather than the three years allotted to a PRGF program, implying that one must look forward to a succession of such programs and a long-term IMF involvement in development finance under present arrangements. In the past it has always been assumed that Fund involvement should be occasional and episodic rather than a continuing matter, and one may wonder whether confusion between these two roles may not prejudice the Fund's ability to act effectively in the event of a crisis.

Supporters of the status quo are likely to argue that the Fund is a more effective vehicle for dispensing conditionality than is the Bank. As one who has served in both organizations, I can confirm that they have a valid point. There is no question that the Fund's hierarchical organization is more effective in producing timely and coherent action than the loose organization and pluralism in ideas that characterize the Bank.

However, two counter arguments deserve attention. One is that the way a bureaucracy develops is in part a consequence of what it is asked to do. Since the Bank has not in the past had any responsibility for organizing a PRGF-type program, it is not surprising that it does not have up and running the capacity to do so. The question is whether there are convincing reasons for supposing that the Bank would be incapable of developing such capacity should it be given the responsibility for the PRGF.

The other counterargument is that the PRGF is not intended to replicate the pattern of past conditionality. On the contrary, lending under this program is to be guided by a Poverty Reduction Strategy Paper (PRSP), which is to be prepared by the borrowing government in consultation with its civil society and private sector (as well as the Fund and Bank). The intention is to ensure that the program has local ownership, something that recent research has demonstrated conclusively to be of key importance if reforms are to deliver. The danger is that the PRSP will prove to be a mere figleaf, which the Fund dictates as it has so often dictated conditionality in the past (which is also supposed to be the borrowing government's own program). Some Fund staff members regard ownership and conditionality as antithetical—conditionality ought to require, in their view, a country to do things that it does not want to do. The logic is that conditionality is used as a device to ration access to cheap IMF credit; make the conditions the perpetuation of sensible past policies, and the IMF would soon be flooded by requests to borrow. One may have serious doubts about the ability of an organization where such attitudes exist to make the intellectual leap to lending on the basis of programs that enjoy ownership. The Bank, in contrast, has worked quite hard in recent years to foster local ownership, and would therefore be much better placed to initiate a program in which ownership is key.

Another change suggested by the ODC report concerns the collection and dissemination of statistics (p.12). It urges that both the IMF and World Bank should hive off their statistics operations and that these be placed in a separate and independent agency devoted exclusively to collecting and publishing economic data. I share the view that this would be a useful change, which would preclude the potential danger that a conflict of interest could corrode data, as well as centralizing statistical expertise.

The surveillance section above also noted the substantial measure of consensus on the Fund's role in surveillance, particularly concerning the desirability of increasing transparency. In fact the Fund has already come a long way in this direction. I recall my pride in being so subversive as to publish the text of a Letter of Intent in Williamson (1983): today the IMF routinely publishes the text of Letters of Intent, and much more, on its Web site. I endorse the widespread consensus that the Fund could usefully focus attention on vulnerabilities stemming from weaknesses in the financial system, the level and maturity structure of foreign debt, and progress in implementing the sundry international standards currently being developed. I also endorse the CFR view that the Fund should actively encourage potentially vulnerable countries to impose appropriate capital inflow taxes.

---

---

There remains one major area where I (like the ODC report) am out of sympathy with the current conventional wisdom. This concerns the question as to whether all intermediate exchange rate regimes should be discouraged by IMF surveillance (for emerging markets and industrial countries, if not necessarily for low-income countries where capital mobility is still low) in favor of one or other of the two "corner solutions," either a currency board or a floating rate. I have discussed this extensively elsewhere (Williamson 2000), but let me just clarify that I would not deny that intermediate regimes are probably more crisis-prone than the corners. The point is that they also offer benefits that the corners do not, namely, the possibility of resisting the misalignments that are so often generated by both fixed and floating exchange rate regimes. If one judges that a seriously misaligned exchange rate jeopardizes the possibility of rapid and sustained growth, this is serious, and suggests one should resist the temptation to focus surveillance exclusively on crisis avoidance. Important as that is, one wants also to encourage countries to make the most of their growth potential.

Nevertheless, the major differences about the future of the Fund are concerned with its role as a lender rather than with surveillance. Everyone agrees that the Fund should have a central role in any financing that may occur in the context of a macro crisis, but that is about the extent of agreement. The disputed issues are:

• the range of facilities under which the Fund should lend;
• the role, if any, of prequalification in the Fund's lending operations;
• the role, if any, of a payments standstill in accompanying Fund crisis lending;
• the terms on which the Fund lends.

The Range of the Fund's Facilities. At present the Fund is able to lend under six different facilities: 1) traditional standbys, 2) the high-interest Supplementary Reserve Facility (SRF) introduced in 1998, 3) the Contingency Credit Line (CCL) announced in 1998 but so far unutilized, 4) the Extended Fund Facility (EFF) introduced in 1975 with the objective of allowing the Fund to make longer-term loans to developing countries experiencing a payments problem with a structural origin, 5) the PRGF through which the Fund makes low-interest loans to low-income members, and 6) the Compensatory and Contingency Financing Facility (CCFF) that dates from the 1960s and makes low-conditionality loans to countries experiencing an exogenous and temporary shortfall in export proceeds, or surge in the cost of cereal imports, or increase in interest costs. This already represents a significant rationalization against the situation prevailing before the spring 2000 meetings of the IMFC, which eliminated the Currency Stabilization Fund, the Buffer Stock Financing Facility, and support for commercial bank debt reduction (i.e., the Brady Plan). I have already argued that the PRGF should be transferred to the World Bank. One needs also to ask whether further streamlining would be appropriate.

To start at the end, there is a strong logical case for retaining the CCFF. This is a mechanism whereby the international community helps primary-producing countries to cope with shocks that are truly exogenous with respect to their own behavior, without requiring them to devote their own real resources to building up reserves ahead of time. It economizes on the need to build up reserves.

An equally persuasive case cannot be made for retaining the EFF. When this was introduced, in the 1970s, many middle-income countries were only just establishing access to the international capital market. The World Bank did not have a capacity to lend for adjustment, its lending was all project-directed. Hence such countries could not rely on being able to borrow in order to adjust to a payments shock with a structural origin, so it seemed reasonable for the Fund to provide a facility to respond to this need. But times have changed, in three ways. One is that most middle income countries can now borrow on the international capital market. Another is that the World Bank has since moved into structural adjustment lending, which is able to cope with very much the same type of situation. The third is that the low-income countries that are excluded from the international capital market are now accommodated by the Fund through the PRGF, which will still be available to them even if it is relocated to the World Bank. Bulgaria perhaps provides the strongest recent case for arguing that the EFF still has a role, but it is not obvious that Bulgaria could not have been accommodated through a World Bank structural adjustment loan instead.

---

---

That then brings us to the CCL. This was introduced with the hope that countries that felt themselves liable to be exposed to contagion would be able to fortify their liquidity to a point that would deter any speculative attack. But, as noted, no country has so far applied for a CCL, and one needs to ask why. It is not difficult to find a plausible explanation. Application is in itself liable to be interpreted by the market as an admission that the country fears a speculative attack, an interpretation that is liable to provoke the very attack that it is hoped to deter. Even if that danger is circumvented, there is a similar danger that an attack could be induced if the Fund ever found it necessary to withdraw a country's eligibility to borrow. Then there is the fact that the Fund has judged it necessary to avoid a completely automatic right-to-draw even after a country has been declared eligible, and envisages an attenuated but nonetheless substantive process of review that could end with the imposition of additional conditionality. Thus it is not difficult to see why the CCL has so far failed to appeal to potential candidates. It is difficult to see this lack of interest ever changing by fiddling about with interest incentives. The conclusion I draw is that it would be sensible to abolish the CCL, while absorbing some of the features of its design into the other facilities designed to allow the Fund to respond to crises.

Those other facilities are standbys and the SRF. Everyone, even the IFIAC majority, agrees that the Fund needs to be able to lend in a crisis situation. The questions are: under what conditions, on what terms, and in what quantities. While it seems quite sensible to envisage larger lending (relative to quota) carrying a higher interest rate, it is difficult to see what advantage is gained by having a separate window for the higher-interest lending. Accordingly, I recommend consolidating both facilities into a single window, which one might call the Crisis Facility, since its purpose is precisely to allow the Fund to help countries deal with crisis situations.⁸

Thus the Fund would be streamlined down to two facilities. The CCFF would provide low-conditionality loans in response to shocks that were clearly outside a country's own control, such as shortfalls in the value of primary commodity exports. It would seem logical to include also other exogenous shocks, including natural disasters (such as Bangladesh's floods of 1998). The Crisis Facility would make loans in situations of macroeconomic crisis.

Prequalification. While I have argued above that the CCL is unattractive to potential borrowers for very basic reasons, it would be a mistake to dismiss the line of analysis that motivated its creation. I would question the wisdom of the IFIAC majority recommendation that (after a transitional period) the Fund should lend only to countries that had prequalified. Nevertheless, the idea that countries should be able to borrow more, and/or more easily, and/or more cheaply is one with considerable merit if they have pre-satisfied certain conditions.

---

---

The attractions are most obvious with respect to the Crisis Facility. One wants to encourage countries to take actions that will minimize their vulnerability to crisis, and it seems natural to reward those that do by giving them assured access (or at least semi-assured access) to a lender of last resort (or at least to a quasi-lender of last resort). The key question is then, what actions should be required to prequalify? The majority IFIAC report suggests four:

• freedom of entry and operation for foreign financial institutions;
• well-capitalized commercial banks, preferably with part of the capital in the form of uninsured subordinated debt;
• regular and timely publication of the maturity structure of outstanding sovereign and guaranteed debt and off-balance sheet liabilities;
• "a proper fiscal requirement."

The first of these is problematic. Traditionally countries have been allowed to decide for themselves whether or not they wish to allow entry of foreign banks. There is a legitimate economic reason why countries may under some circumstances hesitate to allow foreign banks to enter, namely, that this can erode the franchise value of existing banks, and therefore precipitate "gambling for redemption." Nor does there appear to be any empirical evidence that foreign banks can be relied on to stand by a country in times of crisis by increasing their exposure: indeed, the foreign banks in Argentina froze their exposure during the tequila crisis in 1995. Thus this proposal would seem at best premature.

In contrast, a requirement of a solvent, well-capitalized, and well-supervised banking system would appear entirely appropriate. The worst crises happen when a weak banking system deters a central bank from raising interest rates as needed, so that a currency crisis and a banking crisis occur simultaneously. And the suggestion that a part of bank capital should be required to take the form of uninsured subordinated debt held by third parties is also compelling: its attraction is that the holders of such debt have no possibility of upside gains from risky bank behavior, so that they can be relied on to monitor and penalize any gambling behavior by bank management. In asking how one might go about implementing this requirement, an attractive possibility would be to adopt the suggestion of the CFR report (and the IFIAC minority) that the test should be whether a country has adopted and implemented the Basel Core Principles. One might also ask whether it might not make sense to extend the principle to reward observance of some of the many other codes of standards currently being prepared, but, at least initially, it may be best not to overburden the system by making too many demands on surveillance.

The requirement for regular and timely publication of statistics regarding the maturity structure of sovereign debt (and off-balance sheet sovereign liabilities) is also sensible and unobjectionable. The obvious question it raises, however, is whether data on sovereign debt will suffice. None of the East Asian countries had a serious problem with sovereign debt: the problems arose with private sector debt, incurred either by banks (e.g. South Korea) or the corporate sector (e.g. Indonesia). Accumulating accurate and timely data on private sector debt raises altogether more formidable difficulties than are posed by sovereign debt. It happens that the Fund has already established a standard on this topic, which embodies a judgment as to how much data it is reasonable to expect a country to collect. The criterion for a country to get preferential treatment should be that it subscribes to the Fund's Special Data Dissemination Standard.

The IFIAC report does not attempt to spell out the nature of the "proper fiscal requirement" that it suggests including as a prequalification requirement, presumably because this was added at the last moment in response to the objections of the minority that its absence would expose the Fund to supporting countries with runaway budget deficits. It is nonetheless not difficult to imagine the form that such a requirement might take. Perhaps it would be like the Maastricht fiscal requirement for joining EMU: a budget deficit no greater than 3 percent of GDP and a ratio of public sector debt to GDP of under 60 percent (or trending down). Or perhaps it would be expressed in terms of the primary balance, to avoid the objection that a criterion expressed in terms of the total deficit could act as an inappropriate deterrent to tightening monetary policy (though this creates the problem that the necessary primary balance varies across countries depending on their level of public sector debt). Or perhaps it would be expressed in terms of the cyclically-adjusted balance, to avoid the objection that a criterion expressed in terms of the crude deficit could act as an inappropriate deterrent to an anticyclical fiscal policy. Or perhaps it should be expressed in terms of the operational deficit, to avoid making unreasonable demands on countries that have still not eliminated a high inertial inflation. Or…perhaps it would be better not to try to lay down a universal requirement to be met by all countries, but instead to have the Fund make a regular judgment on a country's fiscal position.

---

---

That then raises the question as to why the Fund's judgment should be restricted to the fiscal dimension. Why not have the Fund's Article IV consultation end with the award of a rating of the country's overall macroeconomic policy? One would surely want this rating to be more like those of the ratings agencies⁹ than the simple yes/no rating embodied in the CCL, so that countries can be downgraded when they deserve it without automatically provoking Armageddon in the markets. Having such a rating awarded regularly by an official institution would also resolve the problem of what to use to determine the risk ratings used in calculating bank capital adequacy requirements.

One other idea merits inclusion, in addition to the conditions suggested by the IFIAC report. This is the suggestion in the Geneva report (p.71) that the IMF should provide an incentive by lending on more attractive terms to countries that include appropriate provisions in bond covenants to make their bonds renegotiable under crisis conditions. These provisions "would include majority representation, sharing, nonacceleration, minimum legal action threshold and collective representation clauses, where these last provisions allow an indenture trustee to represent and coordinate the bondholders."

• Hence my suggestions for a list of prequalification criteria that would entitle countries to draw from the Fund under the crisis facility on enhanced terms:
• adoption and implementation of the Basel Core Principles for the domestic banking system;
• subscription to the Special Data Dissemination Standard;
• a good rating for macroeconomic policy in the most recent Article IV consultation;
• inclusion of collective action and allied clauses in its foreign bonds, specially sovereign bonds.

How about drawings from the other facility that I have argued the Fund should retain: the CCFF? Many of the countries that are most likely to suffer strong variations in commodity prices are unlikely to have banking systems that have advanced to the point of implementing all the Basel Core Principles. Similarly, they may not be able to afford a statistical service sufficiently sophisticated to be capable of subscribing to the SDDS. It would be unfair to penalize them for not meeting the full standards expected for crisis borrowers. They should nonetheless face the same requirement of good macroeconomic policy as any other borrower, and they might also be rewarded for any of the other conditions that they meet.

Payments Standstills. Three of the six sources being analyzed in this paper—the Geneva report, the CFR report, and Secretary Summers's speech—saw a role for standstills in dealing with at least some capital account crises. (The subject is not taken up in the other three reports.) All of them regard standstills as something to be deployed as a last resort rather than embodied as a regular element of crisis management.

This is a topic on which I believe the conventional wisdom to be deeply unrealistic. The world tried for many years after 1945 to deny that sovereign debts ever needed to be restructured, but the Brady Plan finally acknowledged that this is not tenable. Not all contingencies are foreseeable, and hence, no matter how conscientious the debtor, contingencies may arise in which it is something between unreasonably costly and totally impossible for the debtor to maintain debt service according to the original contractual terms. This is now widely acknowledged, but its corollary is not. That corollary is that any creditor that suspects restructuring to be a possibility has an incentive to liquidate its claim while that remains possible. Limited official loans will simply allow more creditors to get out rather than encourage them to stay in. The choice is between unlimited official loans (a real lender of last resort rather than a quasi lender of last resort) and restructuring the debt. In any single instance the provision of unlimited liquidity may well be the most attractive option, provided at least that the country really has got its fundamentals in order so that its problem is indeed one of illiquidity rather than insolvency. But, even if one is not sure that past IMF loans have been a major source of moral hazard in the way the IFIAC majority believe, it seems quite implausible that promulgation of such a policy would not crea

RELATED LINKS

Book: Strategy for IMF Reform February 2006

Op-ed: How the Fund Can Help Save the World Economy March 5, 2009

Policy Brief 07-1: The IMF Quota Formula: Linchpin of Fund Reform February 2007

Op-ed: The Fund Appears to Be Sleeping at the Wheel October 3, 2005

Working Paper 01-4: IMF Structural Conditionality: How Much Is Too Much? April 2001

Paper: What Next for Argentina? February 2004

Working Paper 03-1: Debt Sustainability, Brazil, and the IMF February 2003