Journal Issue

Consideration of a New Liquidity Instrument for Market Access Countries

International Monetary Fund
Published Date:
April 2006
  • ShareShare
Show Summary Details

I. Introduction

1. As part of his report on implementing the Fund’s Medium-Term Strategy, the Managing Director suggested that a new vehicle for the provision of high access financing for crisis prevention be considered, targeting emerging market countries that have strong macroeconomic policies, sustainable debt, and transparent reporting and that are making progress in addressing remaining vulnerabilities to shocks.2 The key objective in considering such an instrument would be to ensure that the Fund continues to meet the needs of this dynamic and globalized segment of the membership.

2. The objective of this paper is to consider key issues in the design of a new liquidity instrument for market access countries. Section II reviews the motivations and challenges of creating such an instrument. Section III lays out options and issues related to four key design elements. Recognizing the need to consider the relationship between these elements, Section IV sets out an example of how they might be combined in a lending instrument - called here the Reserve Augmentation Line, or RAL. Section V discusses risks. Section VI lists issues for discussion.

II. Key issues

A. Benefits of a New Liquidity Instrument

3. A number of members, both emerging market countries and others, have called for consideration of a financing instrument specifically designed to support crisis prevention efforts by members active in capital markets. These members have stressed that, for such an instrument to be of value, the availability of large-scale financing from the Fund should be predictable, while also recognizing that such assistance can only be effective if it supports strong policies by the member that are well understood by the market. As such, any new instrument should be designed to provide a commitment device for policies directed at reducing vulnerabilities, to send strong signals to markets regarding policy momentum, and to reinforce confidence that substantial financing is available if needed.

4. A new instrument would aim to benefit members and the financial system as a whole by fortifying strong policies and reducing the likelihood of crises.3 In particular, it could lessen the risk that a negative shock or confidence test would explode into a full-blown crisis, by giving the member breathing room to demonstrate commitment to addressing underlying issues. Recent Fund research found that the availability of large scale Fund financing (drawings or accumulated drawing rights) can significantly reduce the probability of crisis for members active in capital markets, noting that the marginal effect of such financing is small where underlying fundamentals are weak. This analysis recognized the importance of strong policies and the signaling effects of Fund financing.4 In addition, the instrument could provide a supportive framework for “exit” during the transition period when a member pays down outstanding Fund credit. It also reduces the need for costly reserve build-up in cases where the desire for liquidity is the primary motivator for such build-up.

5. Does the Fund need a new instrument to provide these benefits? Some argue that Fund membership in and of itself provides a degree of assurance that financing is available if needed. A specialized instrument, on the other hand, could reduce uncertainty regarding the scale of available financing and conditions for drawings. Similarly, some observers argue that the purposes of the new instrument can be met through existing precautionary structures. Precautionary arrangements within normal access limits do provide important benefits as a signaling and commitment device, but they do not provide assurances that adequate financing will be available. To date, the Board has been hesitant to formalize rules for the use of high access precautionary arrangements, contributing to some uncertainty regarding their availability, although they have been used effectively on an ad hoc basis in a few cases (for example, Brazil, 2003).

6. Emerging market members, particularly those with large reserves, may not seek this form of financing from the Fund in the near future. Global interest rates are low, liquidity is strong, commodity prices are high, and there is robust global growth. These conditions have provided scope for important advances by many emerging markets in addressing vulnerabilities and, in some cases, in building sizable reserves. But the circumstances of emerging markets vary, with many countries still facing important vulnerabilities and new countries gaining access to capital markets. The level of reserve cover also varies widely, and, as recent volatility in equity, debt, and currency markets demonstrates, it is prudent to consider the potential needs of members in periods when market access may be more constrained. The emergence of new pooling arrangements, some of which the Fund is engaged in, also points to the desire by members for access to additional liquidity as a cushion against capital account driven shocks.

7. Finally, questions have been raised regarding moral hazard, as well as financial and reputational risks, that may be associated with such an instrument. These are important issues in the design of any Fund financing and carry special importance in context of the provision of large-scale access. They are discussed in Section V.

B. Key Objectives and Design Challenges

8. To be successful, a new instrument would need to address a number of objectives:

  • First, it would need to provide users (and markets) with predictability that financing is available if needed on a scale that is relevant to their risks, so long as the policy program remains on track.
  • Second, it would need to provide a framework for policy commitment and monitoring both to reinforce incentives for sound policies and provide an effective signal to markets.
  • Third, it would need to address legitimate concerns of prospective users regarding the potential for unintended negative signals that could exacerbate a crisis.
  • Fourth, it would need to provide adequate safeguards for the use of Fund resources, appropriate to the high access involved, while seeking to minimize members’ perceptions of excessive or unnecessarily intrusive conditionality.
  • Fifth, it would need to minimize moral hazard and reputational risks to the Fund through selectivity, transparent decision-making, an appropriate structure of charges, and ongoing monitoring.

9. Meeting all of these objectives implies tensions that would need to be carefully addressed in the instrument’s design. In particular, the benefits of automaticity of access would need to be balanced against the need for strong safeguards. Likewise, a vehicle that provides strong positive signals may also provide negative signals when circumstances deteriorate. While no design can fully eliminate these tensions, it would be important to structure an instrument that balances them to minimize potential risks and downsides. The experience with the Contingent Credit Line (CCL) is instructive in this context (see Box 1).

Box 1.The Contingent Credit Line (CCL)

The CCL was approved in 1999 in response to global financial market volatility during the Asian crisis. Under the facility, members meeting demanding eligibility criteria could make a large drawing if hit by a financial crisis caused by factors outside of their control. Specifically, members seeking to use the CCL had to meet four criteria:

  • No expected need for Fund resources. Financing was only to be provided where a balance of payments difficulties resulted from circumstances largely beyond the control of the member and stemming primarily from adverse developments in international capital markets consequent upon developments in other countries.
  • A positive assessment of policies during the last Article IV consultation and thereafter, including progress toward adherence to internationally accepted standards.
  • Constructive relations with private creditors and progress towards limiting external vulnerability.
  • A satisfactory macroeconomic and financial program and a commitment to adjust policies.

In the fall of 2000, the CCL was amended to provide greater predictability in drawing the first portion of the resources. The rate of charge was also lowered and the structure of commitment fees revised. However, the facility was not used and it expired in 2003 when support for extension fell short of the necessary 85 percent majority of total voting power.

Several problems may have contributed to the lack of demand for the CCL, though it is possible that these issues might not have been decisive if a suitable “first-mover” had requested a CCL:

  • The very strict first criterion under the CCL limited potential use to a narrow set of members with extremely strong fundamentals; these members had largely addressed external vulnerabilities and were unlikely to see large benefits in use of the instrument.
  • The lack of prior experience with use of the instrument led to uncertainty as to whether a request for a CCL would be viewed as a sign of weakness.
  • Members were concerned that losing eligibility for, or access under, the CCL would send a negative signal, particularly if this came during a sensitive period. In this context, the requirement for an activation review to confirm the availability of resources was perceived by potential borrowers as contributing to an unacceptable level of uncertainty.
  • The small scale of the initial tranche raised further questions regarding the instrument’s usefulness given the generally front-loaded needs associated with capital account driven pressures.

III. Design Elements and Options

A. Overview

10. A framework for a new liquidity instrument could incorporate different options for four key design elements: (1) qualification standards and procedures, (2) monitoring structure, (3) scale and access policies and (4) financial terms. It is important to consider the relationship between these elements, given potential trade-offs between them in meeting the objectives described in Section II. A number of proposed frameworks for a liquidity instrument from the Fund have been drawn from variations on these four elements, taking on the lessons from the CCL, as well as earlier proposals for a Fund-based credit line. These proposals can be grouped into two simplified models, those that focus on “ex ante” safeguards, tying access to a member’s ability to meet a qualification standard, and those that focus on “ex post” safeguards, tying access to ongoing performance and implementation of policy commitments through a monitoring structure, such as under a traditional precautionary stand-by arrangement.

11. Both models have merits in promoting strong policies including through effective signaling, and both also present challenges.

  • In ex ante models, qualification acts as the primary mechanism both to safeguard Fund resources and to reinforce incentives for sound policies.5 Challenges include the need to define criteria that provide sufficient safeguards without narrowing the list of eligible members excessively and to address blunt negative signals when a member loses access.
  • Ex post models, including formalized rules for use of high-access precautionary stand-by arrangements, provide for safeguards based on ongoing monitoring of a policy program.6 Such safeguards would allow the instrument to be available to a less narrow, though still selective, group of members, though they do imply some tradeoffs in terms of the degree of automaticity. Some members have expressed concern that insufficient differentiation from existing Fund instruments under precautionary models may reduce the clarity of positive signals and limit demand for the instrument.

12. The following sections looks at each of the four design elements highlighted above, drawing out the distinctions between ex ante and ex post models, where relevant.

B. Qualification

13. A new liquidity instrument would need to be designed to meet the specific needs of emerging market countries that are performing well, but still face some vulnerabilities in their national balance sheet. In an ex ante model, qualification standards would serve as the primary safeguard, implying the need for a very strict standard, as noted above. Qualification standards could also be used in parallel with ex post monitoring, in a model that combines ex ante and ex post elements. Qualification should give strong weight to the member’s current position and the credibility of the member’s commitment to implementing strong policies. They should also verify that current economic policies are focused on addressing remaining vulnerability, with strong ownership.

Nature of criteria

14. One key issue is whether qualification criteria are defined objectively or in a manner that provides for more judgment. Objectively-defined criteria are often, though not always, associated with ex ante frameworks. The key benefit of such rule-based criteria is that they support predictability, transparency, and even-handed application of standards. However, defining such criteria is extremely challenging in the context of diverse and changing country circumstances and the rapidly evolution of capital markets. A standard that is not well matched to the changing circumstances of members may be viewed as arbitrary and one that is modified frequently may lead to accusations of capriciousness.

Application-based models

15. Under some frameworks, qualification would be assessed only at the member’s request. The key benefit of this structure is its voluntary nature. Concern has been raised, nonetheless, that the request itself may send a negative signal. Maintenance of a selective qualification standard is important in this context. Experience with precautionary arrangements, which are seen as providing a valuable commitment device and signaling vehicle by many members, suggests that such concerns may diminish over time as experience with the instrument’s use is gained.7

Prequalification-based models

16. Prequalification through automatic assessments, for example under the Article IV process, has been proposed in some variants of ex ante models. Under this structure, members meeting the criteria would have access to Fund resources upon demand should the specific shock arise, either for a specified period or so long as they continue to meet the criteria.8 The key benefits of prequalification is to reinforce the effectiveness of the surveillance framework through transparent and systematic assessments of members and to eliminate the “costs” associated with requesting an arrangement.9

17. Members have in the past been hesitant to support automatic assessments under the surveillance framework, given concern about the Fund shifting too far in the direction of a rating agency. Public “ratings” under such a structure would imply strong “on/off signals, and as such it may be difficult to avoid political pressures on the system.10 A system where prequalification is not made public could be considered to reduce such negative signals, but this would compromise the objectives of providing a commitment device and signaling vehicle, as well as reinforcing market confidence that large-scale liquidity support is available.

Graduated Ratings/Access

18. To lessen the potential impact of the negative signals when access is lost, graduated levels of access based on differential qualification thresholds have been proposed. Under such a model, negative signals may be less blunt, given their stepped nature. Such models may also provide scope to expand eligibility to members in a much broader range of circumstances. However, implementation would be challenging. Graduation would require fine judgments regarding the relative position of members. With downgrades or low ratings also carrying potentially strong signals, the risk of political pressure may still exist. At the same time, it would be important to consider whether providing lower access in some cases is consistent with the instrument’s objective of providing large scale access to reinforce confidence.

C. Monitoring

19. Under ex ante models, qualifying members have access to resources without additional conditions, providing strong automaticity. The lack of additional conditionality recognizes the qualifying country’s strong position and the Board’s confidence that the member will take needed actions if confronted with balance of payments pressures. As noted earlier, the member’s continued qualification against the instrument’s criteria would be assessed, either periodically (for example, in the context of the annual Article IV consultations) or on a continuous basis.

20. The monitoring structure plays a central role in ex post models that would seek to ensure that macroeconomic policies remain appropriate and to verify that the member is making progress in reducing vulnerabilities that might trigger a capital account crisis. Key benefits of ex post conditions are that they allow for a less stringent qualification standard and provide additional safeguards through the member’s commitments under their policy framework; these commitments can also be adapted to changes in underlying circumstances. The formal presentation by the member of its policy plans, including through a quantified program framework, also supports the instrument’s effectiveness as a commitment device. While exit signals cannot be fully eliminated, they would be less blunt than under strict ex ante models, as failure to complete a review would potentially represent a delay (as the member makes needed adjustments) rather than a termination of access.

21. The specific modalities for a monitoring structure under an ex post framework can be defined in a number of ways. High access precautionary arrangements have used a traditional structure for Fund conditionality. The CCL used a structure centered on the provision of an economic and financial program, including a quantified framework by the member. The framework did not include performance criteria or call for a technical memorandum of understanding, but staff-monitored implementation on a continuous basis, with the member required to provide regular and timely provision of relevant data. At the time of a request for a disbursement, a formal activation review was required to confirm that the circumstances underlying the balance of payments need were consistent with the instrument’s purposes and that the member’s BOP difficulties were largely beyond the member’s control. As such, the activation review was an additional safeguard, although some members have argued that it implied costly delays and raised uncertainties regarding the availability of access.

D. Scale and Access


22. To meet its objectives, the instrument would need to provide large-scale access to reinforce confidence that sufficient liquidity is available, if needed. That said, determining the appropriate level of access is also constrained by the difficulty of measuring a potential need, particularly in a capital account context. At the same time, access decisions need to recognize potential effects on the Fund’s liquidity and broader financial risk position.

23. Various mechanisms for setting the scale of access could be considered. In particular, access could be based on a uniform scale, measured by quota, or through a case-by-case approach. In the case of the CCL, there was no general access limit, but it was expected that commitments under the facility would typically have been in the range of 300-500 percent of the member’s quota.


24. Frontloading of access is important to the instrument’s effectiveness, given the specific nature of capital account-driven balance of payments needs. Indeed, concerns were raised regarding the small scale of the initial tranche under the CCL (which was to be one-third of total access). Some members have also raised concerns in the CCL context that the conditions for access to resources beyond the first drawing were not set up front.

Access rules

25. One issue that would need to be addressed in designing the instrument would be whether resources would be available for any balance of payment need or only a specifically defined balance of payments requirement. Both the CCL and the SRF were designed to allow their use only in the context of a specific balance of payments need, deriving from a sudden and disruptive loss of market confidence reflected in pressure on the capital account and the member’s reserves, and in the case of the CCL, that these resulted from contagion.

26. An important issue for some ex post frameworks is the need to address “black out periods” when access is interrupted due to data lags between test dates for performance criteria and the receipt of data and scheduled reviews. This has special relevance for this type of instrument, since periodic delays in access work against a key purpose of enhancing confidence that substantial liquidity is available, if needed.

E. Financial Terms

27. As with all Fund instruments, the maturity and charges structure for the new instrument should support the revolving nature of Fund resources consistent with the instrument’s objectives, reinforce safeguards on Fund resources, and help to ensure that Fund financing receives appropriate remuneration. For example, countries drawing under the CCL would have been expected to repay within one year to 18 months of the date of each disbursement. The surcharge over the IMF’s normal rate would have begun at 150 basis points, rising to 350 basis points, depending on the duration of the drawing; these rates were reduced from original levels in the course of the review of the CCL in 2000.

28. The Fund’s commitment fee structure would also be relevant to a new liquidity instrument. The existing commitment fee is 25 basis points for the first 100 percent of quota that could be purchased in each 12-month period, and 10 basis points for additional access.11 This fee is charged from the date that resources are committed and is refunded if resources are drawn. From the member’s perspective, the fee is one of many factors that may influence the choice to use such an instrument versus other alternatives, including holding reserves.

IV. An Example—The Reserve Augmentation Line

29. A lending framework would need to consider options for the four elements together as a package to ensure that the design imperatives described in Section II are fully met. This section lays out one option for doing so, a Reserve Augmentation Line (RAL), that brings together features of both ex ante and ex post models. Its key features are highlighted in Box 2, and the procedures associated with the instrument are outlined in Figure 1.

Figure 1.Process for the RAL

Box 2.Summary of the RAL

  • Qualification: The instrument would be available to members with strong macroeconomic policies, a sustainable debt situation, and a proven credibility in policy implementation that are still faced with, but making progress to reduce, balance sheet weaknesses and other vulnerabilities to capital account crises. Members qualifying for the instrument would be in a position where significant tightening of macroeconomic policies was not needed. Qualification would be assessed at the request of the member (i.e., the RAL does not include automatic assessments).
  • Monitoring: The member would present a policy framework, including quarterly quantitative indicators, that meet the standards of upper credit tranche conditionality. There would be no performance criteria (except the standard criteria on exchange and trade restrictions). The program would focus on policies to maintain macroeconomic stability and reduce balance sheet-related vulnerabilities. Performance would be monitored by staff, with a formal Board review semi-annually. Access would be interrupted after a fixed date if this review cannot be completed; as with a precautionary stand-by, such access would be restored, however, immediately upon completion of the review at a later date.
  • Access: Access would be 300 percent of quota. The full amount of access would be available from the outset if a capital account crisis emerged and the program was on track. As an exception, access could be lower for members exiting from large use of Fund resources; in such cases, the Fund should be open to member’s requests to allow the size of access to decline over time where the member wishes to avoid a sharp drop in access upon exit.
  • Terms: The RAL would be established as part of the SRF, as it is intended to address the same type of special balance of payments need, with specific design features consistent with its special purpose. It would also have the same charges and maturities structure as the SRF. The arrangement would be one year in duration and a series of several RALs would not be discouraged. The decision effecting the relevant SRF amendment could be taken by a majority of the votes cast, as it would not involve a change in any of the aspects of the SRF that require a higher majority (i.e., the rate of charge or the repurchase period).

A. Qualification

30. Country-specific assessments would be based on a discrete set of qualification standards, allowing for a transparent and disciplined, but also tailored assessment of the member’s position. The standard is somewhat broader than under the CCL, but still selective. Specifically, financing under the instrument would only be recommended where a country meets the following criteria:

  • At the time of approval, the member is not expected to need to use Fund resources; 12
  • The member has pursued and remains committed to strong macroeconomic management and to policies directed at reducing remaining vulnerabilities, including as they relate to balance sheets and the financial sector, giving confidence that the member will react appropriately in the event of crisis. These policies are described in a forward looking economic and financial program prepared by the member, that would include a quantified framework for the period covered by the RAL;
  • A rigorous and systematic analysis indicates that there is a high probability that the debt will remain sustainable; and
  • The member has demonstrated a commitment to transparent reporting of economic data, including through subscription to the Special Data Dissemination Standard.

31. Qualification would be assessed at the member’s request, with no automatic assessment. If management agrees that there is a basis for moving forward, discussions of a possible RAL could begin. In such a case, the Board would be consulted promptly in an informal meeting, with staff providing a note setting out as fully as possible the basis for the judgment that a RAL may be appropriate, as stipulated under the exceptional access procedures. Formal Board consideration of the request would be based upon presentation of the member’s forward looking program and a staff report that includes an assessment of the member’s qualification.

32. A selective qualification standard serves as an important, but not the sole, safeguard under this example. Instead, this role is reinforced through focused ex post monitoring (Section C, below). Nonetheless, a selective qualification standard supports frontloaded access (Section D, below). In effect, a recommendation to provide a member with an arrangement under this example implies that staff believe that the member has met the standard for the RAL disbursement based on the strength of its existing position and the credibility of its forward looking commitments. A selective qualification standard for the RAL would also support its signaling value and, as noted earlier, help to mitigate potential concerns with negative entry signals.

33. A member requesting access under the instrument would be subject to existing procedures for exceptional access, including a report on the implications of the arrangement for Fund liquidity.13

B. Monitoring

34. The monitoring framework would focus on meeting the RAL’s objectives of supporting macroeconomic stability and reducing vulnerabilities to capital account crisis. It would take into account the selective qualification standards.

35. As with the CCL, the member would present a policy framework that would be expected to be of sufficient quality and strength that it would meet the standards required of drawings in the upper credit tranches. A substantial change in the authorities’ macroeconomic adjustment plans would not be expected to be needed for members qualifying for the instrument. Quarterly indicative targets for macroeconomic parameters would demonstrate policy continuity and provide timely information if there is a material adverse change in underlying conditions. The authorities would commit to a policy agenda focused on addressing remaining balance sheet vulnerabilities that could trigger a capital account crisis. A key issue in this regard will be measures to improve the resiliency of the debt position, including in terms of the composition of the debt, and to address balance sheet vulnerabilities in the financial sector. It will be important, however, to also recognize the need for flexibility that allows the authorities to be responsive to market conditions.

36. Performance against this program would be monitored by staff on a continuous basis, with any material changes reported to the Board in a timely fashion. A semi-annual review would be conducted in which the Board would confirm that the member is implementing its program successfully and is adjusting policies appropriately in response to any significant economic changes. In this context, the review would include an assessment of the member’s performance with respect to their policy framework, including the quarterly quantitative indicative targets, focusing on changes that indicate when policies have gone off track and not on minor deviations from projections. As such, there would be a presumption that a review would be completed where there are no material deviations from the member’s policy framework and a forward-looking assessment indicates that policies remain supportive of the underlying objectives of macroeconomic stability and reducing vulnerabilities to a capital account crisis. As was the case under the CCL, there would be no specific performance criteria and a Technical Memorandum of Understanding would not be required; however, the basis on which the authorities compiled and reported data would need to be clearly presented in the context of the policy framework to allow for effective monitoring. The Managing Director would not recommend that the Executive Board approve a request for access under the RAL unless the authorities had agreed to publish the policy framework and staff reports related to progress under the instrument, including post-drawing review documents (see below).

37. Were a drawing made, a press release would be issued promptly, followed by an early post-drawing review report to the Board; this report would also be published along with a press release following the Board discussion. The post-drawing review report would analyze the member’s economic situation, discuss the circumstances leading to the drawing and the authorities’ response, and provide a forward-looking assessment.14

38. To the extent that the member faces a continuing balance of payments need following a drawing under the RAL, additional access could be made available through a traditional arrangement (rather than a follow-on RAL) that would include phased purchases and understandings on the policies and financing framework to be pursued in response to the crisis. The role of other creditors would also need to be assessed.

C. Scale and Access

39. The RAL would provide the member with access equivalent to 300 percent of quota. The member would have the right to draw the full amount available at any time a qualifying balance of payments need occurred and scheduled reviews (i.e., those with dates that had already passed) had been completed. As such, the structure provides a strong degree of predictability, though one that recognizes the need for some trade-offs to achieve the safeguards and positive signaling effects afforded by ongoing monitoring. At the same time, it recognizes the concerns of many emerging markets regarding the small scale of the initial tranche under the CCL, as well as the uncertainty created by the activation review and the fact that conditions for access to the remaining resources were not set up front.

40. This structure also avoids the concerns associated with black-out periods, as there would be no test dates for performance criteria and thus no interruption based on data lags. Access could only be interrupted in the event that a review is delayed beyond its scheduled date or in the event of nonobservance of standard continuous performance criteria on exchange and trade restrictions.

41. The proposal of an access level of 300 percent of quota, available in a single tranche has four key aspects:

  • Frontloading: The unique, single-tranche structure under the arrangement recognizes that provision of significant up-front access is important to the confidence effects of the instrument. Frontloading of access is based on the safeguards provided by the strong qualification standard, indicating that Board agrees the member has met the standard for a disbursement of large scale resources.
  • Link to quota: As with all Fund financing, the basic metric for access is set in terms of the member’s quota. The efforts in the context of the medium term strategy to address misaligned quotas will help to rationalize potential access under a RAL.
  • Scale: The choice of 300 percent of quota access level derives from several factors:15
    • > It seeks a balance between the need for substantial access to reinforce confidence and the importance of a prudent limit to the Fund’s exposure.
    • > It is consistent with the overall scale of access (300 to 500 percent of quota) that had been available under the CCL. Access at the lower end of this range recognizes the fact that resources under the RAL would be made available in a single tranche.
    • > It is in line with the findings in Fund Supported Programs and Crisis Prevention that a substantial scale of financing can materially reduce the likelihood of a crisis for countries experiencing heightened vulnerability.1617
  • Uniform access as a percent of quota: While access needs will vary across members and over time, the proposal for a uniform access level recognizes the difficulty of measuring a potential need, particularly in a capital account context. A uniform access level would provide a simple, transparent mechanism for setting access in this context.

D. Financial and Other Terms

42. The RAL would be established as part of the SRF, would be available for the same type of special balance of payments need, and would carry the same charge and maturity structure.1819 Arrangements would have a one-year length, implying a single mid-year review, but a series of several RALs would not be discouraged where the member continues to meet the qualification criteria. The one-year length is consistent with existing SRF terms and minimizes the risks that there will be a material change in policies or fundamentals during the period of the arrangement. The SRF decision would need to be amended, inter alia, to allow for the approval of a RAL for qualifying members with a potential need. This amendment would allow for the commitment of resources in such cases, though members would continue to be required to have an actual qualifying need at the time of a drawing. Depending upon the final design features and modalities, adoption of the RAL policy could require amendments to (or an explicit carve-out from) certain other UFR decisions and policies.

43. A commitment fee would be charged to the member using the RAL, as with current policies for the use of Fund resources.

44. In order to limit the Fund’s liquidity risk associated with provision of resources under this instrument, a global cap could be set, for example at SDR 50 billion, to ensure that sufficient resources remain available to be provided under traditional arrangements.

V. Dealing with Risks

45. Any financial engagement by the Fund carries risks that must be carefully assessed and controlled. Risks associated with a new liquidity instrument include possible reputational risks associated with exposure of Fund resources, the impact on Fund liquidity, and possible moral hazard. Any design for a new liquidity instrument would need to contain those risks within an acceptable range, considering its benefits for crisis prevention.

A. Predictability of Access

46. A new liquidity instrument would need to be designed to limit the possibility that the Fund will be called upon to provide financing in a situation where the member is unlikely to take the appropriate policy action to address the shock, while preserving an adequate degree of predictability so that members and markets can rely on access being available when needed. The RAL framework described in Section IV, for example, includes important elements that limit the risks to the Fund’s resources and reputation that could arise if the Fund disburses in adverse circumstances:

  • The qualification criteria, including debt sustainability, demonstrated policy credibility, and specification of a strategy to reduce remaining vulnerabilities are important elements to limit both the likelihood and the severity of a potential crisis.
  • The one-year period of the arrangement would lower the risk that policies or fundamentals deteriorate during the time of the arrangement. The limits on access would also help to control risks.
  • The ongoing relationship and dialogue with the authorities in the context of a RAL would help prevent the adverse situation where the member postpones a closer relationship with the Fund as a crisis approaches and the Fund lacks real-time information needed to provide sound advice.
  • Ex post discussions with the Executive Board would be required after a drawing. Also, experience suggests that in a deep crisis the member would likely request a successor arrangement, at which time policy actions and conditionality would be agreed.

47. Ultimately, any Fund financing involves risks. The risks are particularly large when the Fund is called in to help a member deal with an ongoing capital account crisis. Access is often very large, and heavily concentrated in the first purchase. Policy-making under such conditions is difficult and subject to disagreements. To the extent that a new instrument helps members to reinforce their resiliency to crisis, it would reduce the probability that the Fund would need to provide financing in such situations.

B. Fund Liquidity

48. If a new instrument generated substantial demand, it could tie up the Fund’s liquidity, potentially limiting the ability of the Fund to respond to a new crisis. On the other hand, to the extent to which it helps prevent broader or deeper crisis, it could limit the amount of Fund resources outstanding under traditional arrangements. The size of these effects is difficult to predict, particularly at the outset. The proposal in Section IV for a global cap on the use of RALs would allow several large emerging market members to draw under the instrument, while leaving substantial resources to be provided under traditional arrangements.

C. Moral Hazard

49. Moral hazard in the context of Fund financing is the risk that the availability of Fund financing may encourage reckless behavior among borrowing members and their creditors. It can arise if Fund involvement shields either the debtor or the creditor from facing possible negative consequences of their actions.

50. While Fund financing will always involve a degree of moral hazard, the extent of such risks can be mitigated and should not be exaggerated. Several factors limit adverse incentives for debtors and their creditors. Most importantly, a crisis involves enormous costs to the debtor country—even if substantial Fund financing softens its impact. Authorities thus have strong incentives to limit vulnerabilities and, in addition, they would have to have demonstrated a commitment to such policies under the new instrument. With respect to private creditors, it is important to note that the availability of Fund financing does not protect them from losses in cases of unsustainable debt. Experience shows that private creditors have taken a hit in past emerging market solvency crises. This includes the recent cases of Russia and Argentina. The risk of reckless lending (and borrowing) fueled by the expectation that the Fund’s involvement would prevent private sector loses in solvency-driven cases is thus likely to be limited. Strong selectivity is important, nonetheless, including through the inclusion of ex ante based qualification criteria to ensure the instrument would only be available to countries that are performing well and have sustainable debt, further reducing moral hazard risks. In crises driven primarily by liquidity concerns, temporary Fund financing can reinforce market confidence and provide the member time to take necessary actions to address the underlying pressures. As a result, spreads would likely be lower. However, this would be the welcome reflection of the reduced risks of liquidity-driven crises developing into default, rather than a moral hazard problem. Also, surcharges for large-scale access alleviate moral hazard concerns associated with subsidized financing. 20

VI. Next Steps And Issues For Discussion

51. Following the informal discussion of this paper by Executive Directors, staff and management will consider how to proceed in the context of preparing the work program following the Annual Meetings in Singapore.

52. Directors may wish to comment on options for various features of framework for lending for crisis prevention, including by drawing on the example examined in Section IV. Specifically, do Directors consider that:

  • Qualification criteria which offer scope for some judgment or those that are objectively defined are preferable?
  • Qualification assessments should be done at the member’s request or automatically through the Article IV process?
  • Models for graduated ratings should be explored?
  • An ex ante model (with strong automaticity) or some degree of ex post monitoring (to reinforce safeguards and avoid too narrow a qualification standard) is preferable?
  • Access levels should be set uniformly, on a case by case basis, or within a range?

Regarding the example for a RAL framework in Section IV, do Directors consider that:

  • The framework finds a reasonable balance to address potential reputational, financial, and moral hazard risks, as well as the need for a strong degree of predictability to support the value of the instrument to potential users?
  • The four qualification criteria described in the example would form a strong basis for ensuring the instrument is used selectively?
  • The monitoring structure is appropriate, recognizing the unique purposes of the instrument in support of crisis prevention by members who are performing well?
  • The access structure is appropriate (normally 300 percent of quota, available continuously through a single tranche, as long as the program remains on track)?
  • The use of the SRF, using a one-year arrangement, would be appropriate?
  • A series of RALs should not be discouraged, so long as the member continues to meet the qualification criteria?
  • A global cap would be appropriate to limit risks to Fund liquidity?

This paper was prepared by a staff team consisting of Ms. Shannon, Mr. Reichold, and Mr. MacArthur (all PDR).


The Managing Director’s Report on Implementing the Medium-Term Strategy, April 5, 2006.


See Country Insurance—The Role of Domestic Policies, June 19, 2006, for a discussion of shocks and their costs.


Fund-Supported Programs and Crisis Prevention, March 23, 2006. See also Sudden Stops and IMF-Supported Programs, Eichengreen, Gupta, and Mody, IMF WP/06/101, April 1, 2006. The authors of this second study also find that sudden stops are fewer and generally less severe when a Fund arrangement is in place and that this form of financing works best for countries with strong fundamentals. They find no evidence that a Fund-supported program attenuates the output effects of capital account reversals if these nonetheless occur.


See, for example, Tito Cordelia and Eduardo Levy-Yeyati, A (New) Country Insurance Facility, IMF WP/05/23, January 2005 and Jonathan D. Ostry and Jeromin Zettelmeyer, Strengthening IMF Crisis Prevention, IMF WP/05/206, November 2005.


See, for example, Adapting Precautionary Arrangements to Crisis Prevention, June 11, 2003.


Precautionary Arrangements - Purposes and Performance, March 23, 2006. Empirical work in the context of this study found no evidence to suggest that use of precautionary arrangements negatively affected borrowing country spreads.


While assessment during regular Article IV consultations could reinforce the surveillance process, more frequent or continuous assessment would reduce the potential that access would remain available in the event of a marked deterioration of performance against the qualification criteria during the interim period.


Prequalification is generally associated with pure ex ante models (i.e., there is no further conditionality), although conceptually it could be used to define a list of members eligible to apply for use of an instrument that also incorporates ex post conditionality. In such a case, the eligible member would need to make a request for access to the instrument.


Advocates of objective criteria argue that they could mitigate against such pressures, though they too could be subject to challenges given the difficulties associated with their design.


Rule l-8(a) of the Fund’s Rules and Regulations governs commitment fees for all use of GRA resources. The commitment fee structure was revised in the context of the 2000 review of the CCL to lower the fee on amounts over 100 percent of quota that could be purchased in each 12 months. Any change to the Fund’s commitment fee structure requires a decision taken by 70 percent of the total voting power of the Executive Board.


This criteria is designed to ensure that the RAL is not used when a traditional Fund arrangement is more appropriate. However, unlike the first criteria under the CCL, it does not exclude members that are performing well, but still face some vulnerabilities.


In his report on Implementing the Fund’s Medium-Term Strategy, the Managing Director indicated that the criteria under which a request for large-scale access is assessed should be reviewed and modified to ensure that they provide a predictable gauge to the range of circumstances in which exceptional access may be provided. In this context, it is important to consider that the absence of an immediate balance of payments need would not only be expected, but required for members requesting this form of financing.


Drawings under the RAL are proposed to available in the event of an SRF need arising from a “large short-term financing need resulting from a disruptive loss of market confidence reflected in pressures on the capital account and the member’s reserves.” While in existing arrangements, there is normally no formal ex post verification of the existence of need, the post-drawing review under the RAL would provide the opportunity to confirm formally that the requisite need existed at the time of drawing.


Access could be lower for members exiting from large use of Fund resources. In such cases, the Fund should be open to member’s requests for the size of access to decline over time to avoid a sharp drop upon exit.


Specifically, the study looked at the level of access that would have been needed in the period prior to the crisis to lower the probability of crisis (from an average of 85 percent) to a specific threshold for the 11 crisis cases in the study. While there is a strong degree of variability across countries, the average level of access needed to reduce this probability to 25 (5) percent was 345 (460) percent of quota at the time of the arrangement’s approval. This average would be 270 (360) percent based on current quota levels.


Differences in ratios based on actual and current quota reflect the fact that some crises occurred prior to the 1998 quota increase. Current efforts to address misaligned quotas would also lower this ratio for affected members and on average. The calculations in the study keep constant policies (and other variables) although, in practice, policies can be assumed to be strong under the RAL, contributing to a lower likelihood of crisis.


As is currently the case under the SRF, SRF terms would apply only to the portion of a member’s access that is in excess of the annual or cumulative access limits in the credit tranches, with normal credit tranche terms to apply to the portion within the limits.


The upcoming review of charges, including the proposal to align charges for all exceptional access, would also cover access under this new window. See Review of Charges and Maturities, May 23, 2005, and PIN No. 05/101, August 1, 2005.


For a helpful overview see Olivier Jeanne and Jeromin Zettelmeyer, “The Mussa Theorem and Other Results on IMF-Induced Moral Hazard.” IMF WP/04/192, October 1, 2004.

Other Resources Citing This Publication