Financing balance of payments adjustment: options in the light of the elusive catalytic effect of IMF-supported programmes.

Bird, Graham ; Rowlands, Dane

INTRODUCTION

Countries are presumed to turn to the International Monetary Fund (IMF) when they have an unsustainable current account balance of payments deficit. The circumstances that lead to such a situation have been studied extensively and have been found more often than not to involve a combination of domestic macroeconomic mismanagement and external shocks. (1) Crises during the 1990s, however, demonstrated that referral to the Fund could have as much to do with sharp capital reversals, implying that current account deficits could no longer be financed by capital inflows, as with any marked deterioration in the current account itself. The IMF responded to these events as well, despite the presence of Article VI of the Fund's Articles of Agreement precluding it from providing resources merely to finance capital outflows.

When approached after the onset of unfavourable economic events, the basic purpose of the Fund's involvement is to restore balance of payments viability, thereby enabling the borrowing country to eventually disengage from the Fund. The programmes supported by the Fund combine balance of payments adjustment with external financing. Financial assistance from the Fund comes with policy strings attached; the so-called IMF conditionality. An IMF programme is supposed to balance the adjustment process with external finance in an optimal way. If financing inflows on the capital account fall short of the expected and desired amount, the speed of adjustment on the current account will be greater than expected. There is, of course, the opposite problem of diluted adjustment when financing is excessive. As two IMF researchers put it, 'the availability of external financing, the first component of the strategy, determines the magnitude and pace of the necessary adjustment period' (Mussa and Savastano, 1999, p. 20). The tradeoff between external financing and the speed of adjustment is likely to be a source of tension between the Fund and the country concerned, and striking the right balance requires to reflect both economic and political realities.

The Fund recognises that the resources it provides may be far from adequate on their own to finance the adjustment programmes that it approves, even when it has violated its original quota-based lending limits. The difference between the resources required to finance an adjustment programme and those which the IMF provides is supposed to come from other public and private sources. Specifically, the Fund has traditionally claimed that its programmes perform a catalysing role, encouraging others to lend more than they would otherwise have done. The view of the Fund is represented by the following quote:

In most cases, the IMF, when it lends, provides only a small portion of a country's external financing requirements. But because the approval of IMF lending signals that a country's economic policies are on the right track, it reassures investors and the official community and helps generate additional financing from these sources. Thus, IMF financing can act as an important lever, or catalyst, for attracting other funds. The IMF's ability to perform this catalytic role is based on confidence that other lenders have in its operations and especially in the credibility of the policy conditionality attached to its lending. (IMF, 2002)

Similarly in the Prague Framework launched by the IMF's Managing Director, Horst Kohler, in September 2000, the catalytic effect of IMF lending was seen as an important element in enhancing crisis resolution. The above quote from the IMF's website is useful because not only does it say something about the assumed quantitative significance of the catalytic effect but it also explains the mechanism through which the effect is intended to work. (2) This paper reviews what we know about the catalytic effect of IMF lending.

The next section briefly explores the analytical basis for believing that there will be a catalytic effect, and reviews the empirical evidence. The subsequent section then examines ways in which IMF programmes might be strengthened in terms of achieving the desired balance between financing and adjustment. A final section places the discussion within the broader context of current trends in terms of capital flows to emerging and developing economies.

THE CATALYTIC EFFECT OF IMF LENDING

In the above quote, the Fund argues that catalysis is based on the 'credibility of the policy conditionality attached to its lending'. Expanding on this idea, Cottarelli and Giannini (2002) suggest that in principle the catalytic effect works through five channels. First, IMF programmes may improve policy design. Second, IMF programmes may be based on transmitting information to other lenders and investors. Third, signing an IMF agreement may indicate a government's commitment to reform. Fourth, the process of negotiating an IMF programme may screen out governments not committed to reform. Finally, markets may perceive IMF programmes as a form of insurance for their capital. (3)

The first four of these point to two primary catalytic functions for IMF programmes. First, the programme's policy conditions must represent a significant improvement from the lenders' or investors' perspective relative to the policies that the government would otherwise have adopted. Second, the signing of a programme must imply a credible commitment to implement these improved policies. Although clearly connected, the question of policy design and implementation are considered in turn.

The IMF's policy conditionality has long attracted criticism. Nowzad (1981, p. 9) suggests that '[n]o aspect of the Fund's activities has been more controversial, more persistently criticized, or more routinely misunderstood than ... conditionality.' The primary irritant from the perspective of many critics has been the presumed overemphasis on reducing aggregate demand, which can be achieved relatively quickly, versus the expansion of aggregate supply. The need to emphasise rapid adjustment via aggregate demand contraction is itself seen as symptomatic of insufficient financing. Helleiner (1983, p. 1) summarises the traditional arguments of IMF critics:

 In particular, the IMF has been attacked for its overemphasis on
 demand management, blunt monetary-policy instruments, and 'shock'
 treatment to reduce or eliminate inflation and balance of payments
 disequilibria; its relative neglect of supply-side policies, longer
 term development, and income distribution; and its traditional
 aversion to controls, selective policy instruments, and 'gradualist'
 approaches

While these older criticisms are still repeated today, they have been updated. The IMF has been accused of exacerbating the trauma of the Asian crisis, with contractionary monetary policies especially singled out as interacting adversely with weak financial sectors, a policy dilemma highlighted in recent research on 'twin crises' (Kaminsky and Reinhart, 1999). More recently, commentators have put more emphasis on the IMF's apparent inability to encourage meaningful adjustment of any kind (for example, IFIAC (2000) also referred to as the Meltzer Commission). (4) Are the critics correct?

The IMF's own research provides mixed reviews regarding the outcome of its programmes. Haque and Khan (1998) review most of the relevant research and conclude that IMF programmes do, in fact, generally succeed at reducing the current account deficit. They go on to suggest, however, that other targets are often not met. Thus, the effects of programmes on inflation, while generally favourable, do not seem to be statistically significant. Further, they conclude that economic growth tends to decline in the early period of a programme, (5) although it picks up over the medium-to-long term. Unfortunately, it is in the early adjustment period that positive signals are required, and it is not clear why other lenders would be motivated to provide financing during the initial period of contraction when it is needed the most. (6) One IMF study goes so far as to suggest that there may be 'some inherent inconsistency in the design of IMF programmes' (Baqir et al., 2003, p.15), though it argues that this conclusion needs to be treated tentatively.

The evidence from research conducted outside the IMF is even less sanguine. Several studies have concluded that IMF programmes have adverse consequences for investment and economic growth, or are otherwise ineffective (Conway, 1994; Killick, 1995; Hutchison, 2001; Przeworski and Vreeland, 2000; Barro and Lee, 2001; and Evrensel, 2002). Why should the prospect of declining growth attract external capital? In short, the evidence on the effectiveness of IMF programmes is not unambiguously favourable, and agreements thus appear to provide a weak signal upon which capital markets may base their financing decisions, especially, early on in the adjustment phase.

Of course, one of the reasons for the relatively weak results regarding IMF programmes may not be the conditions themselves, but their execution. Signing an agreement with the IMF does not necessarily signal commitment to carry through the agreed policies. (7) The relationship between implementation and catalysis may be important because of its potential circularity. If a good record of implementation fails to generate a positive catalytic effect, then the burden on adjustment will be greater than originally envisaged. The increased adjustment costs may then make further implementation less likely, having a negative effect on capital flows and imposing a further burden on current account adjustment. A history of past agreements may be taken as indicating entrenched problems or little commitment to reform.

According to the IMF's own research (Mussa and Savastano, 1999) the reality is that less than a half of the programmes that are signed appear to be successfully implemented. Although the determinants of implementation are not fully known, they seem to include a range of domestic political variables, (Ivanova et al., 2001; Bird, 2002a). The credibility of Fund programmes is also eroded by apparent serial over-ambition. The Fund is regularly too optimistic about the prospects for economic growth, export earnings, and tax revenue (IEO, 2002).

Further reason for doubting the credibility of programmes is recidivism in IMF lending (Bird et al., 2004). About a half of the recent programmes have been with countries that had made prolonged use of IMF resources (IEO, 2002). The fact that some programmes that are cancelled due to failed implementation are immediately replaced with a new agreement, seems inconsistent with the signalling of commitment. Empirical studies confirm that the existence of past programmes is a significant factor in the econometric equations explaining current programmes (Joyce, 1992; Conway, 1994; Knight and Santaella, 1997; Bird and Rowlands, 2001). Therefore, negotiating a programme with the IMF for many countries will indicate a high probability that they will continue to face severe economic difficulties in the short-to-medium term, hardly a signal that will foster confidence in capital markets. When interviewed, money managers claimed that they formed their own independent views and that they relied on these even when they disagreed with those of the IMF (Bird and Rowlands, 2000).

Finally, the ability of the Fund to create more credible signals is confounded by the appearance of political interference in its programme structure and allocation. Thacker (1999) and Barro and Lee (2001) provide large sample evidence to back up the considerable anecdotal accounts of IMF activity unduly reinforcing the direct political interests of the United States. Bird and Rowlands (2001), while failing to find similar evidence, identify other political influences that systematically or idiosyncratically bias the allocation of Fund agreements. Political noise will tend to interfere with the signal that might otherwise be transmitted by an IMF programme.

Of course, capital markets are neither uninformed nor uniform, and their response to the different elements of IMF programmes will undoubtedly vary. Political interference, for example, may be seen as a positive feature by official lenders and donors seeking to increase their leverage. Private markets may be able to sort through the mixed signals of different IMF agreements and respond differently to their various effects over the course of the adjustment process. These responses may also depend on the specific characteristics of the recipient country, such as income level, economic structure, or governance capacity. These potentially diverse reactions of the different types of capital flows, which are already difficult to model and estimate, will make it impractical to generalise about any catalytic effect for the purposes of developing adjustment programmes.

What does the empirical evidence tell us about the catalytic effect, and is this evidence consistent with the channels of transmission outlined by Cottarelli and Giannini (2002)? Comprehensive surveys of the empirical research pertaining to the catalytic effect already exist (Bird and Rowlands, 2002; Cottarelli and Giannini, 2002). The methodologies used have included some survey work (Bird and Rowlands, 2000), case studies (Bird et al., 2000; Ghosh et al., 2002) and large sample econometric investigation (Rodrik, 1996, Bird and Rowlands, 1997, 2000, 2002; Edwards, 2000; Mody and Saravia, 2002). There are several difficulties with which this research has attempted to come to terms. First and foremost is the sheer number of possible dependent, independent, and antecedent variables that need to be examined in order to reflect the potential reactions to IMF programmes. Second, there are problems with establishing the counterfactual and dealing with selectivity bias: what would capital flows, policy reform, and the path of adjustment have been in the absence of a Fund programme? Third, it is important to determine the time horizon in which the catalytic effect should be observed, and thus the lags between programme and response that should be reflected in the models. Finally, it is important to acknowledge the implications of balance of payments accounting, since an improvement in the current account of the balance of payments will be reflected by either a change in international reserves or a reduction in net capital inflows. An advantage of using a range of approaches for measuring the catalytic effect is that, while each may have its own particular methodological problems, it is reasonable to assume that common inferences are robust.

Early anecdotal reports seemed to take the catalytic effect as a stylised fact, though there were dissenting voices (Bird, 1978; Killick, 1995). However, subsequent case study evidence suggested that the catalytic effect was not general, but instead limited primarily to official bilateral flows and to cases where there was supporting and independent evidence of the recipient government's commitment to a sound policy agenda (Bird et al., 2000). Research from the IMF was even more pessimistic. Ghosh et al. (2002) concluded that in all of their eight cases, private capital inflows were lower than predicted by the IMF. Even in the case of Argentina, where the shortfall was the smallest, they concluded that there was no evidence of the programme having a strong catalytic effect. As a consequence, the size of the current account adjustment was larger than originally envisaged in all of the countries they had studied. In the cases of Korea and Thailand, the discrepancy amounted to 13-15 per cent of GDP.

Large sample econometric investigation of the catalytic effect has taken on a number of forms and is still evolving and becoming increasingly sophisticated. Recent research is helping to specify the parameters in which catalysis may operate. The strongest catalytic effect is generally found for official flows (Rowlands, 2001), though this appears to be more a question of coordination than traditional catalysis. Bird and Rowlands (2002) focus on examining the range of possible effects by differentiating between IMF agreements, recipient country income levels, and various types of capital flows. This study also attempts to control for IMF programme history and degrees of programme implementation. No general catalytic effect is observed. Instead, some programmes have a positive effect on some flows, but only for certain groups of countries. In other instances, the effect is insignificant or negative. The relationship between IMF programmes and capital flows, therefore, appears as highly idiosyncratic. These results do not appear to be related to problems of simultaneity, the direction of causality (which appears to run in both directions), or selection bias.

Selection bias is treated in a more sophisticated fashion in Edwards (2000). If countries receiving IMF programmes are significantly worse off than non-recipients, selection bias might lead to the estimated catalytic effect being biased downwards. But no significant selection bias is found. Edwards also examines the extent to which catalysis is affected by the degree of compliance with IMF conditionality. His results are even starker than Bird and Rowlands (2002), reporting that good implementation does not seem to lead to increased capital inflows. Worse still, poor implementation seems to lead to reduced inflows. Edwards thus concludes that he finds 'no evidence that the Fund serves as a seal of approval to foreign investors and lenders.'

Capital flows may be intrinsically unstable and a decline in net capital inflows may reflect a reduction in the current account deficit. An alternative approach to estimating the catalytic effect is to examine the effect of IMF lending on the interest rate charged on a country's external debt and current borrowing, and on gross flows. Early research that examined programme effects on interest spreads, at least indirectly, is reported by Ozler (1993) and Haldane (1999). In both cases spreads were found to widen alongside IMF programmes, suggesting that signing a programme signals, or is correlated with, payments difficulties. A more recent study by Mody and Saravia (2002) examines both spreads and bond issuance. Because the data can be measured with a higher frequency, they are able to improve on past studies in terms of handling the complex timing issues and questions of causality in a better way. While they find little to suggest that IMF programmes on their own transmit a significant positive signal, a positive catalytic effect can be discerned in bond markets when they are accompanied by credible signals of commitment to policy reform, and when the economic circumstances of the countries have not deteriorated too far. They also find that precautionary programmes have more powerful positive effects, and that longer term Extended Fund Facility programmes have stronger effects than stand-by agreements.

Therefore, the research to date appears to suggest that the catalytic effect does not operate in its traditional, general form. Rather the direction and magnitude of catalysis is highly sensitive to a variety of antecedent conditions and is not consistent across programmes or capital flow measures. It should perhaps be a relief that capital markets do not condition their decisions on simple signals such as the announced signing of an agreement between a government and the IMF. One of the secondary results of the research on catalysis is the conclusion that explaining capital flows is inherently difficult; this being the consequence of complex interactions between 'push' and 'pull' factors (Taylor and Sarno, 2000). Indeed, no recent study of capital flows has felt compelled to include a variable that directly captures the role of the IMF.

The complexity of the response to IMF agreements imposes serious constraints on those attempting to construct a balance between external financing and economic adjustment within the context of a Fund programme. Evidence of a positive relationship between IMF lending and official flows and debt rescheduling fails to provide much support for the conventional concept of catalysis, but rather of concerted lending. While new research may continue to discover additional nuances, it seems unlikely that it will discover a strong and reliable catalytic effect, as it has been traditionally portrayed. If true, this conclusion has important and far-reaching implications for the design of policy. Basically, it is unwise to rely on something that is unreliable.

POLICY OPTIONS FOR IMF PROGRAMMES

The policy problem faced by the IMF and its clients is how to adjust to an unsustainable balance of payments structure. The basic parameters of this process are defined by the level of external financing that becomes available, which is inversely related to the speed of adjustment that needs to take place. This inverse relationship has led some to argue for less IMF financing in order to encourage more credible and rapid restructuring (Boockmann and Dreher, 2003; Vaubel, 1994). Others, however, including the IMF, would argue that such an approach implies that the optimal rate of restructuring is necessarily the fastest in all cases, and that the benefits of rapid adjustment always outweigh any associated costs. The preference of the IMF (and others) appears to be to assess this trade-off on a case-by-case basis in an attempt to identify an optimal rate of restructuring that is both politically and economically feasible. The result is supposed to be an IMF programme that balances total external financing with adjustment.

The implication of an illusory catalytic effect is that IMF programmes will tend to be under funded relative to their desired plan. Consequently, more emphasis will be placed on swifter balance of payments adjustment of the current account, as was reported across the eight episodes examined by Ghosh et al. (2002). Consequently, countries will have to bear higher welfare losses than initially conceived within the agreement, especially as the shortfall will have to be made up through a more aggressive contraction of aggregate demand. Such a response is inconsistent with the Fund's Articles of Agreement. (8) The political costs of adjustment will be higher, generating greater problems for commitment and implementation and further compromising any catalytic effect. There is little point in attempting to calculate the required financial support for an adjustment programme if the failure to generate it is regarded as a matter of no concern, nor does it enhance the Fund's credibility if it claims to have a catalytic effect that fails to materialise.

The previous analysis suggests two ways of addressing the detrimental effects of a weak or absent catalytic effect. First, efforts can be made to strengthen catalysis, using the information that has emerged from the studies of both catalysis and its channels of operation. Second, programmes may seek to limit the extent of any necessary adjustment. These two broad approaches are examined in turn.

If catalysis is designed to work via conditionality, credibility and time consistency, it may be strengthened by improving the effectiveness of conditionality via programme design and implementation. This is not an easy task. It is difficult to separate out the components of current programmes that enhance catalysis from those that do not, or to distinguish between programme content and implementation. Considerable faith has been put into the notion of streamlining conditionality and improving country ownership of programmes (Khan and Sharma, 2001), and these may indeed be useful changes. However, the evidence is less clear. Ivanova et al. (2001) suggest that the key reasons for implementation failure are fundamental domestic political economy problems, such as ethno-linguistic fractionalisation and political disunity. These characteristics are difficult or impossible to influence in the long term, let along in the compressed space of time available for setting up an IMF agreement. While improved ownership might make programmes more realistic, these implementation constraints will impose severe limitations on the set of feasible adjustment paths.

Even if general policy initiatives fail, however, traditional catalysis may still be useful in certain cases. Current research has identified some circumstances in which catalysis may be present in a sufficiently consistent and significant manner to allow the IMF to rely on the leveraging of other finance to supplement its own resources. Further research may allow these circumstances to be defined more precisely and the magnitude of catalysis estimated more reliably. Ideally, an IMF programme could even hope to bring about a more desirable structure of external financing, using instruments that encourage one form of capital inflow such as long-term bonds when these are deemed to be both susceptible to catalysis and valuable for diversifying a country's debt portfolio. We are, unfortunately, a long way from being able to deploy IMF programmes with this degree of precision.

A better understanding of the influence of different IMF agreements will nonetheless help programming, as well as rationalising and differentiating optimally the variety of facilities that the IMF currently has on offer. Of course, caution needs to be exercised. The Fund's experiment with its Contingent Credit Lines, which is specifically built around the idea of precautionary programmes, has been unsuccessful. Having already made one attempt to make the CCL more user-friendly the Fund then embarked on a second review (November, 2002) and eventually abandoned the facility in 2003. Yet in some ways the CCL conformed well to the conditions that Mody and Saravia (2002) identify as conducive to catalysis in bond markets: precautionary in nature, and thus signed prior to the onset of serious problems or the need to draw on it. Ultimately, countries have been reluctant to absorb the costs of negotiating a programme if there is no guarantee of resources from the Fund, as well as to carry the risk of spooking markets by transmitting what might be interpreted as a negative signal. Stand-by agreements were also initially seen as largely precautionary in nature, but have subsequently evolved into a standard programme for financial emergencies. The potential for programme restructuring, therefore, might be quite limited.

If IMF programmes do not generally catalyse private lenders in the traditional sense, modified versions of catalysis may still be relevant. In principle, an alternative to catalysis is coercion, or to use a more attractive term 'concerted lending'. The IMF adopted this policy in the aftermath of the debt crisis over the period 1982-86. The advantage of this approach is that the IMF will be in a relatively stronger bargaining position vis a vis private lenders in the immediate aftermath of a crisis, when the need to supplement their own resources is the strongest. This bargaining power does erode, however, as illustrated in the mid-1980s by the eventual abandonment of the Baker Plan and its focus on setting unenforceable lending targets. In its place, the international community adopted the Brady Plan, emphasising debt relief and seeking to shift the bargaining power in favour of debtors by having the IMF agree to lend to countries in payments arrears. Today, private lenders and investors are more diverse, and the legal framework and institutional imperatives faced by foreign investors may make it difficult for them to be concerted. Despite these difficulties, however, concerted lending was reactivated in Korea in 1998, suggesting that it might still be relevant under certain conditions.

The ultimate form of concerted lending is the explicit coordination of financing among official lenders and donors. Fewer players make coordination easier. Moreover, aid donors may possess greater freedom of manoeuvre, being subjected as they are to political constraints rather than commercial imperatives. Although aid donors will be looking for assurances about the conduct of economic policy in the same way as private investors will be, they will have different objective functions. They may indeed be attracted to lend to countries where the economic fundamentals are weak. They may view IMF resources and their own as largely substitutable, and thus potentially be willing to reduce the resource burden on the Fund while still ensuring adequate programme financing. While many forms of official development assistance (ODA) are limited to poorer countries, the coordination of official creditors may be relevant for a larger group of countries through the terms of Paris Club rescheduling. These reschedulings, which effectively require an IMF agreement to be negotiated, offer an opportunity to alter the debt payments of a country in a manner that can mimic catalysis, even though it is not really catalysis in the conventional sense. Not surprisingly, it is in the context of debt rescheduling that evidence supporting a positive association with IMF programmes has been discovered (Marchesi, 2001).

The final method of dealing with shortfalls in finance is to have the Fund provide the money directly. Of course, this approach would presumably impose a significant resource cost on the IMF. There are at least two potential drawbacks. The most obvious problem is that such a policy may require member countries to increase the resources of the IMF. The reluctance of some of the Fund's major shareholders to endorse the increase in quotas associated with the Eleventh General Review may be indicative of a general reluctance to do so. Indeed, it could be surmised that the attention paid to catalysis in the Prague Framework in part resulted from dissatisfaction among some powerful shareholders of the Fund with enhancing the Fund's own lending capacity. Second, there may be some understandable hesitation if members anticipate that a larger resource base might also encourage the Fund to assert greater independence. Others may view a downgrading of catalysis as the elimination of private oversight considering the efficacy of the IMF. In short, more resources might be seen as permitting the IMF to ignore the views of other key players in the international financial system. (9)

Therefore, a range of catalytic or catalytic-type effects might be marshalled in specific circumstances to enhance financing, and hence the viability of IMF programmes. If these are unavailable, what alternative does the IMF have? One approach would be to alter the degree of adjustment needed by a country through other means. These have already been alluded to in the context of improving catalysis, and involve the timing of IMF programmes and situations of crisis.

The stated preference of the Fund is to become involved with countries prior to the emergence of severe economic difficulties. This 'lender of first resort' function also seems consistent with the evidence on catalysis provided by Mody and Saravia (2002). By heading off difficulties, however, the Fund's conditions would be less onerous and the Fund's resource commitments could also be smaller. The long-stated desire by the Fund, however, has not yet translated into early intervention. Encouraging countries to turn to the IMF earlier in the evolution of their balance of payments problems will remain a challenge for as long as conditionality is seen by countries to impose significant costs. Moreover, if at this stage countries continue to have access to private capital markets, some of the Fund's shareholding countries may argue that they should not be seeking to use IMF resources. After all, governments have to have a balance of payments need before they are eligible to draw resources from the Fund. Ultimately, and in extremis, strong fundamentals might power up the catalytic effect but they would also eliminate the need for Fund involvement and perhaps also for its catalytic effect. The empirical evidence on catalysis suggests that private markets will be more prepared to lend to countries that have less need to borrow.

While some programme structure might yet be discovered that will encourage voluntary early resort to the IMF, the more likely route is the more aggressive use of Article IV consultations and a re-emphasis of its traditional surveillance functions (Pauly, 1997). Reliance on this mechanism might be more credible if it were applied more evenly across the Fund's membership as well, particularly to its wealthier members. Aside from moral suasion, however, there are no obvious enforcement mechanisms the IMF could rely on to compel countries to seek early intervention. In addition, it is hard to imagine nations providing the IMF with a means of enforcement that would jeopardise members' sovereignty.

If the scope of adjustment cannot be modified before economic crisis sets in, it may be possible to do so immediately afterwards. As noted in the discussion of coerced lending, crises may provide the IMF and its key members with considerable leverage over private lenders. Thus, for example, proposals for a Sovereign Debt Restructuring Mechanism (SDRM) may be viewed as an alternative to catalysis. (10) There are also more general forms of intervention such as IMF-endorsed standstills on debt repayments and capital controls. In the longer run, and if such mechanisms were to enable countries to move towards debt sustainability, they could even themselves exert a positive catalytic effect on future capital flows. The most reliable effect would be in the short run, however, where such policies may be useful in moderating capital flight. Consequently, the need for adjustment would be reduced. In principle, what might be achieved by an SDRM for middle-income countries could be achieved by the Heavily Indebted Poor Country (HIPC) initiative for low-income countries. Of course, HIPC-related negotiations need not take place in the same environment of financial crisis as might be relevant for the other forms of intervention, and might thus be more generally applicable. There are current doubts, however, about whether HIPC will be successful in allowing eligible countries to exit from their debt problems. Since debt problems appear to impede catalysis, it may be unwise to regard these initiatives as substitutes for the catalytic effect. Resolving them may contribute towards stronger catalysis in the future, and may even be close to a pre-condition for effective catalysis.

The outline of an adjustment policy for the IMF may be traced from the discussion in this section. There appear to be some instances where the traditional catalysis of private capital markets might work. Although considerably more research is needed to specify these instances with reasonable certainty, it may be possible eventually to build some agreements around catalysis in its traditional sense. In situations where such catalysis is unreliable, the IMF may wish to make arrangements to make up any apparent shortfalls from its own resources or those of other official lenders. These agreements may also be made more effective if the IMF were to push for earlier adoption. Continuing to experiment with innovations, such as streamlining conditionality and improving country ownership, may also help to identify programmes that have more desirable effects and better chances of implementation. This approach seems most relevant to middle-income countries with access to capital markets that are not in the midst of crisis.

When crisis does hit emerging market countries, the best alternative to catalysis is likely to be coerced lending to maximise new inflows in conjunction with some form of controls or standstills to limit outflows. Laying out the guidelines for crisis management of this sort will allow the market to work with clearer information, and will hopefully have the salutary effect of reducing political interference by wealthier nations. Of course, it may be the case that wealthier nations will also need to provide additional finance either directly on a bilateral basis, or indirectly via lending to and through the IMF.

Official financing will also play the most significant role in poorer countries, for which traditional catalysis is probably irrelevant. For these countries the emphasis needs to be on increasing programme performance and implementation, more generous debt forgiveness to restore debt sustainability, and improved official financing in the context of adjustment. Official financing needs to be coordinated and made explicit for agreements while at the same time minimising its impact on future debt sustainability. In the end it needs to be recognised that many of the poorest countries do not use the IMF in the manner in which it was originally meant to be used. Rather than facing temporary balance of payments problems, these countries face longer term structural problems associated with low levels of economic development. Programmes need to reflect this, and they should be distinguished from those deployed in emerging market middle-income countries in an effort to preserve some signalling capacity for the latter.

Thus by differentiating member characteristics, adjustment circumstances, and programmes, the IMF might be able to achieve the balance between financing and adjustment that catalysis is not currently providing on a consistent basis. By formalising these distinctions, the Fund may be able to reduce the noise-to-signal component of its agreements, and thus permit catalysis to function properly.

CONCLUDING REMARKS

Capital flows to developing and emerging economies have been falling since the East Asian financial crisis in 1997/98. Combined with the desire to replenish international reserves, this has implied that countries have had to substantially strengthen their current account balance of payments. Such adjustment carries a cost. In a period of declining capital inflows it has become important to consider alternative ways in which foreign capital can be mobilised. This issue is likely to become increasingly important as tighter regulation and prudential control in capital markets and an increased use of risk management techniques tend to reduce international capital flows still further.

This paper has examined the role of the IMF in mobilising capital flows to emerging and developing economies. The nature of the relationship between the IMF and private capital markets and official finance is complex. There are aspects of both complementarity and substitutability in it. The topic has received high-profile attention, and the associated discussion has raised very fundamental questions about the role of the IMF. Some observers argue that IMF lending creates severe moral hazard problems and delays necessary adjustment. Others, however, suggest that adjustment has taken place without concern for its adverse consequences, political feasibility, or sustainability, and without sufficient adjustment-related financing. In trying to strike the right balance between adjustment and external finance, the Fund has placed much emphasis on the catalytic effect of its lending.

The review and analysis in this paper casts doubt on the traditional notion of catalysis as a passive but reliable mechanism for increasing a wide range of external capital inflows for countries with an IMF agreement. Consequently, it is crucial to consider alternative policy approaches by which the Fund may attempt to mobilise capital and strengthen the financial position of emerging market and developing economies. By differentiating between low- and middle-income countries, between private and public capital flows, between crisis and non-crisis lending contexts, and between different IMF facilities, it might be possible to restore the balance between financing and adjustment in a more optimal manner. Some of the proposed policies are likely to prove controversial, while others might be technically impractical. However, the failure to compensate for an elusive catalytic effect in a more meaningful way will relegate many IMF recipients to a cycle of programme failure and continued economic unsustainability.

The next few years may be dominated by the problem of helping emerging and developing countries to gain the access to the foreign capital they need in order to complement their own developmental efforts. In this regard, the problem has changed markedly from that of the early 1990s when there were excessive capital inflows and the focus of the Fund was on stabilising them. Capital markets tend to exhibit patterns of feast and famine. In the middle of the famine years reliance on the largely illusory catalytic effect is unlikely to be sufficient. More radical action will be required. Without it global welfare could be significantly and adversely affected.

Acknowledgements

While the views expressed in the paper should be attributed only to us, we gratefully acknowledge the financial support of the British Academy.

(1) There are various studies that examine the economic circumstances in which countries turn to the 1MF for financial assistance (Bird, 1997; Conway, 1994; Killick, 1995; Knight and Santaella, 1997).

(2) There are various other statements by politicians in both the US and UK and by academics which suggest that the catalytic effect has been widely believed to exist (Bird and Rowlands, 1997: Cottarelli and Giannini, 2002).

(3) In this paper, we ignore the insurance role because it is directly related to the problem of moral hazard, a topic that has received considerable recent attention in this own right and is best dealt with separately. Moral hazard is generally discussed in terms of the potential for lenders to provide excessive credit in anticipation of being bailed out by a subsequent IMF programme. An alternative version of moral hazard is that the promise of subsequent IMF assistance causes governments to delay needed adjustment and engage in inefficient policy-making. For recent discussion and evidence, see Dreher (2004); Dreher and Vaubel (2004); Rogoff (2002); Evrensel (2002); Dell'Ariccia et al. (2002); Lane and Phillips (2000); and Nunnenkamp (1999).

(4) Mussa and Savastano (1999) provide an interesting review of, and response to, the main criticisms.

(5) This initial negative effect is hardly surprising and may indeed be a necessary element of adjustment. The debate is over the extent to which it is relied on in order to restore external equilibrium.

(6) Khan and Sharma (2001, p. 12) suggest that recent evidence is more favourable, but provide no information.

(7) Bird (2002b) concludes that programmes are unlikely to have strong credibility and that the signalling effect is as likely to be negative as positive, and generally quite weak.

(8) Article 1, sub section v, states that the purposes of the IMF include giving 'confidence to members by making the general resources of the Fund temporarily available to them under adequate safeguards, thus providing them with the opportunity to correct maladjustments in their balance of payments without resorting to measures destructive of national or international prosperity.'

(9) Although, as noted earlier, diminishing the extent to which political interference affects Fund behaviour might allow it to send clearer signals to financial markets.

(10) A useful summary of the evolution of the SDRM may be found in Rogoff and Zettelmeyer (2002). For a critical assessment, see Eichengreen (2002) who basically claims that such a mechanism would tip the balance too much in favour of debtors vis a vis creditors. For a long time, Eichengreen has argued that modifying bond contracts to include collective action clauses would be a superior way of organising orderly debt workouts. This view seems to reflect the current views of the US Treasury as well.

REFERENCES

Baqir, R, Ramcharan, R and Sahay, R. 2003: The consistency of IMF programs. IMF Research Workshop 'Macroeconomic Challenges in Low Income Countries'.

Barro, RJ and Lee, J-W. 2001: IMF Programs: Who is chosen and what are the effects? Paper presented at 2nd IMF Research Conference.

Bird, G. 1978: The International Monetary System and the less developed countries. Macmillan: London.

Bird, G. 1997: External financing and balance of payments adjustment in developing countries: Getting a better policy mix. World Development 25(9): 1409-1420.

Bird, G. 2002a: The completion rate of IMF programs: What we know, don't know and need to know. The World Economy 25(6): 833-847.

Bird, G. 2002b: The credibility and signalling effect of IMF programs. Journal of Policy Modeling 24(9): 799-811.

Bird, G, Mori, A and Rowlands, D. 2000: Do the multilaterals catalyse other capital flows? A case study analysis. Third World Quarterly 21(3): 483-503.

Bird, G, Hussain, M and Joyce, JP. 2004: Many happy returns? Recidivism and the IMF. Journal of International Money and Finance 23(2): 231-251.

Bird, G and Rowlands, D. 1997: The catalytic effect of lending by the international financial institutions. The World Economy 20(7): 967-991.

Bird, G and Rowlands, D. 2000: The catalyzing role of policy-based lending by the IMF and the World Bank: Fact or fiction? Journal of International Development 12: 951-973.

Bird, G and Rowlands, D. 2001: IMF lending: How is it affected by economic, political, and institutional factors? Journal of Policy Reform 4(3): 243-270.

Bird, G and Rowlands, D. 2002: Do IMF programs have a catalytic effect on other international capital flows? Oxford Development Studies 30(3): 229-249.

Boockmann, B and Dreher, A. 2003: The contribution of the IMF and the World Bank to economic freedom. European Journal of Political Economy 19(3): 633-649.

Conway, P. 1994: IMF lending programs: Participation and impact. Journal of Development Economics 45(2): 365-391.

Cottarelli, C and Giannini, C. 2002: Bedfellows, hostages, or perfect strangers? Global capital markets and the catalytic effect of IMF crisis lending. IMF Working paper no. 02/193.

Dell'Ariccia, G, Schnabel, I and Zettelmeyer, J. 2002: Moral hazard and international crisis lending: A test. IMF Working paper WP 02/181.

Dreher, A and Vaubel, R. 2004: Do IMF and IBRD cause moral hazard and political business cycles? Evidence from panel data. Open Economies Review 15(1): 5-22.

Edwards, MS. 2000: Re-evaluating the catalytic effect of IMF programs. Paper presentation at the Annual Meetings of the American Political Science Association. Washington, DC (September).

Eichengreen, B. 2002: Financial crises and what to do about them. Oxford University Press: Oxford.

Evrensel, A. 2002: Effectiveness of IMF-supported stabilization programs in developing countries. Journal of International Money and Finance 21(5): 565-587.

Ghosh, A, Lane, T, Schultze-Gattas, M, Bulir, A, Hamann, J and Mourmouras, A. 2002: IMF supported programs in capital account crisis. IMF Occasional paper no. 210. International Monetary Fund: Washington.

Haldane, A. 1999: Private sector involvement in financial crises: Analytics and public policy approaches. Financial Stability Review, November.

Haque, NUI and Khan, MS. 1998: Do IMF-supported programs work? A survey of the cross-country empirical evidence. IMF Working paper no. 98/169.

Helleiner, G. 1983: The IMF and Africa in the 1980s. Princeton Essays in International Finance No. 146. Princeton University: Princeton.

Hutchison, MM. 2001: A cure worse than the disease? Currency crises and the output costs of IMF-supported stabilization programs. In: Dooley, M and Frankel JA (eds). Managing Currency Crises in Emerging Markets. University of Chicago: Chicago.

Independent Evaluation Office. 2002: Evaluation of the prolonged use of IMF resources. IMF: Washington, DC.

International Financial Institutions Advisory Committee. 2000: Report of the International Financial Institutions Advisory Committee. Washington, DC.

International Monetary Finance. 2002: What is the International Monetary Fund? www.imf.org Updated August 2002.

Ivanova, A, Mayer, W, Mourmouras, A and Anayiotas, G. 2001: What determines the success or failure of fund-supported programs? Paper presented at 2nd IMF Research Conference.

Joyce, JP. 1992: The economic characteristics of IMF program countries. Economics Letters 38: 237-242.

Kaminsky, G and Reinhart, C. 1999: The twin crises: The causes of banking and balance-of-payments problems. American Economic Review 89(3): 473-500.

Khan, M and Sharma, S. 2001: IMF conditionality and country ownership of programs. IMF Working paper no. 01/142.

Killick, T. 1995: IMF programs in developing countries--design and impact. Routledge: London.

Knight, M and Santaella, JA. 1997: Economic determinants of IMF financial arrangements. Journal of Development Economics 54(2): 405-426.

Lane, T and Phillips, S. 2000: Does IMF financing result in moral hazard? IMF Working paper no. 00/168.

Marchesi, S. 2001: Adoption of an IMF program and Debt rescheduling. An empirical analysis. Development Studies Working papers no. 152. University of Oxford/Queen Elizabeth House: Oxford.

Mody, A and Saravia, D. 2002: Catalyzing private capital flows: Do IMF programmes work us commitment devices. IMF Working paper.

Mussa, M and Savastano, M. 1999: The IMF approach to economic stabilization. IMF Working paper no. 99/104.

Nowzad, B. 1981: The IMF and its critics. Princeton Essays in International Finance No. 146. Princeton University: Princeton.

Nunnenkamp, P. 1999: The moral hazard of IMF lending: Making a fuss about a minor problem? Kiel Institute of World Economics Discussion paper no. 332.

Ozler, S. 1993: Have commercial banks ignored history? American Economic Review 83(3): 608-620.

Pauly, L. 1997: Who elected the bankers? Surveillance and control in the world economy. Cornell University Press: Ithaca and London.

Przeworski, A and Vreeland, JR. 2000: The effect of IMF programs on economic growth. Journal of Development Economics 62(2): 385-421.

Rodrik, D. 1996: Why is there multilateral lending? In: Bruno, M and Pleskovic, B (eds). Annual World Bank Conference on Development Economics. World Bank: Washington, DC.

Rogoff, K. 2002: Moral hazard in IMF loans: How big a concern? Finance and Development 39(3).

Rogoff, K and Zettelmeyer, J. 2002: Bankruptcy procedures for sovereigns: A history of ideas, 1976-2001. IMF Working paper no. 02/133. IMF: Washington, DC.

Rowlands, D. 2001: The response of other lenders to the IMF. Review of International Economics 9(3): 531-546.

Taylor, M and Sarno, L. 2000: Capital flows to developing countries: Long term and short term determinants. World Bank Economic Review 14.

Thacker, SC. 1999: The high politics of IMF lending, World Politics 52(1): 38-75.

Vaubel, R. 1994: The political economy of the IMF: A public choice analysis. In: Bandow D and Vasquez I (eds). Perpetuating Poverty: The World Bank, the IMF, and the Developing World. The Cato Institute: Washington, DC.

Graham Bird (1) & Dane Rowlands (2)

(1) Surrey Centre for International Economic Studies, University of Surrey, Guildford, Surrey GU2 7XH, UK E-mail: g.bird@surrey.ac.uk;

(2) The Norman Paterson School of International Affairs, Carleton University, Ottawa, Canada