A More Focused IMF
Outline of a presentation at a conference on "The Future of the Bretton Woods Institutions: The European Contribution to the Debate on the New Financial Architecture"
Organized by the American Council on Germany and the Reinventing Bretton Woods Committee
© Peterson Institute for International Economics
The author is indebted to Morris Goldstein for comments on a previous draft.
One of the few propositions that almost all of the participants in the debate on the future of the International Monetary Fund seem to agree on is that the Fund ought to reverse the mission creep of recent years and return to focus on the areas of its core competence. Failure to do this threatens to make it as unwieldy as the World Bank has tended to be, and without the excuse that its mandate is so wide as to leave it with no alternative.
The first part of the Fund's first Article, and indeed its very name, say that it is concerned with international monetary cooperation. Since exchange rates result from the interface between the monetary systems of different countries, they are clearly the Fund's business. Since monetary policy is a part of macroeconomic policy, one needs to cover the other main component, namely fiscal policy, as well. And since the monetary system is merely one part of the financial system, it is advisable to include that within the Fund's remit too. Thus I agree with the delineation of where its core competence ought to lie, as sketched in the third bullet of the background note. With the exception of the financial system, where the Fund's deep involvement is quite recent, one can say that these are indeed the areas in which the Fund's competence has traditionally been concentrated.
Its present activities, namely surveillance, lending, and governance, are also accurately summarized in the first bullet of the background note. Some of us think it unfortunate that the provision of international liquidity does not rate mention as another activity. Let me say a few words about each of these activities, dividing surveillance into systemic surveillance and country surveillance, and explaining under the former why I consider it unfortunate that the Fund never developed a significant role in providing international liquidity.
The Fund's role in systemic surveillance is at present essentially a reporting role, with the reports being made public in the biennial World Economic Outlook and the annual report on International Capital Markets, plus occasional ad hoc studies. The reason that these reports have little role as inputs into policy coordination is that there is very rarely anything worthy of the name of policy coordination, and in the rare exceptional cases the action is within the G-7 rather than in a wider framework. While the IMF prepares background papers for such meetings and an IMF representative participates in much of the discussion, it is a long time since any identifiable result has emerged from such meetings. Moreover, the IMF representative does not report back to the IMF board on the proceedings, so there is no link between the G-7 talks and the broader world community.
It is often objected that policy coordination will not happen because no country is willing to subjugate its national economic interests to those of some other country or group of countries. This mistakes the nature of what policy coordination seeks to achieve: not persuading one country to sacrifice itself to the general good, but searching for mutually beneficial policy adjustments (including beneficial inter-temporal tradeoffs). One example concerns international liquidity creation. At the end of 1999 total reserve holdings of IMF members were $1319 billion, an increase of $522 billion in the preceding 6 years, of which the bulk ($391 billion) was accounted for by developing countries.1 That may be compared with a cumulative US current account deficit of some $1202 billion over the same period. The desire to accumulate reserves on the part of the developing countries has been reinforced by the Asian crisis, and this was presumably a significant factor underlying the recent magnification of the US current account deficit. If one worries that the size of this deficit is building up vulnerability for the future (admittedly as unfashionable a worry as Thailand's current account deficit was prior to late 1996), then one might think there could be mutual advantage in large-scale SDR allocations that satisfied the desired buildup in developing country reserves without requiring a US deficit. That is doubtless inconceivable unless and until the developing countries demonstrate a desire to hold a major part of their reserves in SDRs rather than just to receive allocations that they promptly pass on to the OECD countries, but at present there is not even an attempt to scope out the possibility of mutual gains of this character.
The Fund's surveillance of individual countries is rightly focused on helping countries avoid crises. It has recently been expanded beyond the traditional macroeconomic concerns of Article IV consultations, to include assessment of whether member countries are implementing the wide-ranging standards and codes that have been developed, largely under the aegis of the Financial Stability Forum, in response to the Asian crisis. The result is seen in the Reports on the Observance of Standards and Codes (ROSCs) that the Fund is now publishing on the internet. One has to be impressed by the speed with which agreement has been reached on these minimum international standards, and the rapidity with which the Fund has got around to initiating surveillance of them.
Surveillance is concluded without any attempt to rate a country's performance. Indeed, in summing up the IMF board's review of Fund surveillance of international standards and codes on 8 September 1999, the chairman went out of his way to note that a number of Directors "had indicated that there was a risk that reports could be perceived as providing ratings of countries, and could create the impression that the Fund was assuming the role of a rating agency". He reassured them that this was not the intention. This seems to me misguided. The proposals for a new capital adequacy framework issued by the Basel Committee on Banking Supervision in June 1999 sought a way to eliminate use of the crude proxy of OECD membership as a measure of the riskiness of a borrower, and to replace this by a more satisfactory measure. The main such possibility discussed was use of the rating given a country or bank by the credit rating agencies. That proposal has attracted much criticism, and rightly so, given the weak record of those agencies. An alternative would be to use the process of IMF surveillance to generate ratings. This might cover both the traditional macroeconomic surveillance under Article IV, and the new surveillance based on the ROSCs. To permit these to be used to generate ratings, it would of course be necessary to have surveillance end by the Executive Board agreeing to some measure of a country's performance. That would indeed mark a sharp break with tradition, but it seems altogether preferable to relying on the rating agencies.
The Fund has already made a number of worthwhile changes in its lending facilities since the latest round of crises. One is introduction of the Supplementary Reserve Facility (SRF) in the midst of the Asian crisis, a facility designed to lend larger sums for shorter periods at higher interest rates than has been traditional in Fund lending. In my view those changes were exactly what was needed in a facility designed to meet capital account crises. In addition, the Fund earlier this year abolished a number of facilities that had outlived their usefulness and were virtually unused in recent years.
A more difficult issue concerns introduction of the Contingency Credit Line (CCL). This was intended to provide semi-automatic access to credit for countries hit by contagion that have good macro policies, are complying with the agreed international financial standards embodied in the codes discussed above, and have good relations with their private creditors. So far no country has applied for a CCL (although both Mexico and Argentina have recently shown some interest in applying). Two reasons have been advanced to explain this failure to use the facility. One is that the financial terms are too demanding, with a commitment fee payable when the application for coverage is approved, and the same interest rate as that on the SRF payable when money is drawn. That gives no financial incentive to pre-qualify. Allied with the fact that the entitlement to draw is not quite automatic even if contagion hits, that leaves no obvious positive incentive for applying. However, a financial incentive is quite likely to be provided soon, with abolition of the commitment fee and a lowering of the interest rate to less than that on the SRF. But there is a second reason that may explain the lack of interest in the CCL, which is the ambiguous signal that applying for it would send to the markets. These might see an application as signifying weakness (a need to establish a line of defense) rather than strength. Reinforcing that concern is the certainty that if the IMF subsequently felt obliged to disqualify a country, that would trigger a crisis. If this is the true explanation, a working CCL will require replacement of the provision that countries apply for the CCL by one that they are automatically entitled to access it in case of need if they satisfy certain standards, without ever going through the process of applying and being approved. (But even this would leave the problem of an adverse signal being sent should the IMF find it necessary to declare a country no longer qualified to draw.)
Even more fundamental than the issue of how to design a CCL is whether any such facility can really be expected to head off capital account crises.2 The danger is that a decision to draw on such a facility would be taken as a sign of weakness by the financial markets, which would react by staging a run. The result would be that money drawn from the CCL would be frittered away on allowing some creditors to get out while the going is good. I sometimes doubt whether there is a realistic intermediate solution between providing unlimited official loans to a country judged solvent (i.e., turning the IMF into a real lender of last resort) and restructuring a country's debt. If that is right, and we agree that there is no chance of gaining political approval for a lender-of-last-resort capacity for the Fund (specially since this would surely raise the problem of moral hazard), then the focus has to turn to involving the private sector in debt reconstruction. One would hope that this would normally involve simply extending maturities, as in Korea, rather than reducing the present value of debt service payments, and accordingly that it should prove possible to negotiate debt restructuring relatively rapidly. The simple way to provide an incentive to private creditors to enter such negotiations is to allow a country to impose, and the IMF to sanction, a standstill to debt service payments (or at least amortization payments) until the debt profile has been restructured to one that the country can be expected to service according to the revised contractual terms.
Another contentious issue is whether the IMF should retain the Poverty Reduction and Growth Facility (PRGF), rather than hand this over to the World Bank. To do justice to its name and spearhead the international dimension of a country's growth program, the Fund would need to establish expertise in a range of issues going far beyond its traditional mandate. Hence various people have suggested that this was the natural place to trim the Fund back to size. No one has denied that the Fund's traditional macro concerns have a legitimate place in PRGF programs, in that macro stability is a key precondition for rapid growth which in turn is essential for progress in reducing poverty. Those suggesting moving the PRGF to the Bank have argued, rather, that the Fund could adequately assert its concern for macro stability if its sign-off were required before the Bank approved a program, just as the Bank's sign-off is required under current arrangements. The Fund has responded that it doubts its ability to influence a country's policies if it is not directly bringing money to the table, which makes one worry that the Bank must in that case be impotent under the existing system. The Fund has also argued that it wishes to continue to provide services to all of its members, including the poorest, who could not afford to borrow under the regular facilities. The obvious response is to allow the Fund to lend to its poorest members under similar circumstances to those under which it lends to others, namely when they confront an external crisis, but to grant them a concessional interest rate. Such arguments appear to fall on deaf ears, which makes one worry that either the Fund will go on expanding into new fields, or that poverty reduction and growth will be ill-served by lop-sided programs.
Governance of the Fund
Three issues are of key importance: quotas and the composition of the Executive Board, the US veto on major decisions, and selection of the managing director. The figures below offer a big-picture comparison of output, population, and IMF quotas as a percentage of global totals:
Percentage of GWP3
|of which, Euroland|
If one takes output as the relevant criterion, it is clear that the big legitimate grumble of under-representation in the Fund is that of the Asians (whose complaint would have even more force if one thought population should be a relevant factor as well). Europe is unambiguously the most over-represented region, from which the major concessions will therefore have to come. An elegant way of resolving this problem would be if Euroland saw fit to treat itself as a single country for purposes of IMF membership. Once the UK has adopted the euro, it would be reasonable for the expanded Euroland to contemplate a quota equal to that of the United States, which would avoid the problem of relocating IMF headquarters that would arise if it were to obtain a larger quota. In return for releasing a large block of quotas for redistribution, it would be reasonable for Europe to raise the question of the United States abandoning its veto on major decisions, by reducing the qualified majority needed from 85 percent to (say) 75 percent.
The other big governance issue is how the managing director is selected. No one was happy with the way this was conducted earlier this year. We are probably not yet ready to go beyond resolving that in future (a) the procedure should be different and more dignified, and (b) that the aim should be to seek the best qualified candidate in the world, irrespective of nationality. But agreement on that much would suffice to initiate the search for a new procedure.
2. I define a capital account crisis as one that results from a reversal in capital flows, as opposed to the traditional current account crisis, initiated by a current account deficit that grows to exceed the size of the capital inflow. Doubtless the dividing line is ambiguous, but the concepts are nonetheless worth distinguishing.