Draft Paper on MDB Leverage for Guarantees
March 1, 2005
Executive Summary
Proposal Concept
This paper recommends, given their excellent asset quality including on guarantees, that multilateral developments banks whose Articles of Agreement currently have gearing / leverage ratios of one dollar of capital to one dollar of loans and guarantees increase the gearing / leverage of capital for purposes of their guarantees to four to one.
Under this proposal, multilateral development banks could provide four dollars in guarantees (versus only one dollar of loan) for every dollar of capital. This will provide incentives for MDBs to promote guarantees, which currently account for a tenth at best to less then one hundredth of the activities of MDBs.
Benefits
The benefits of the proposal are several-fold:
· Alignment of the core competence of MDBs, which is risk mitigation, with a the corresponding product, namely, guarantees (not loans)
· Increased headroom, based on capital currently allocated to guarantees alone, of guarantees of $8.8 billion at the World Bank and the Asian Development Bank, the Inter-American Development Bank and the European Bank for Reconstruction and Development.[1]
· Increase MDB’s developmental impact through attraction of private capital without any increase in the current utilization of capital and with no capital increases
· Indefinitely deferred capital increases as MDBs shift gradually from being mainly lenders to being increasingly facilitators of private lenders through their guarantee operations
· Greater MDB collaboration with and participation of the private sector in funding activities in emerging markets with reduced risks
· Greater access to private funding for more emerging markets borrowers forcing them to pay more attention to factors private markets consider importing in lending and investment decisions such as
o Adherence to the rule of law
o Predictable regulatory regimes
o Political stability
o Macro-economic and fiscal stability
In time, if guarantees became more established as a product, the role of MDBs as lenders would become and their role as providers of risk mitigation products to foster private lending and investment in emerging market countries would rise. The private markets have become the largest source of external capital for emerging markets and this change is not likely to be reversed, despite the Latin American debt crisis and the Asian financial crisis. But private capital flows are erratic, ebbing and flowing with market supply and demand factors. A rising use of MDB guarantees could well increase the stability and size of such flows.
Obstacles to Date
· Market demand for guarantees is a function of the level of investment in emerging markets, which ebbs and flows
· Market demand for guarantees has also been limited by the limited level of risk guarantors are willing to take (markets prefer payment guarantees whereas they can only procure insurance with exclusions
· Better guarantee products need to be designed in partnership with the private sector
· Increasing the leverage for MDBs guarantees will provide incentives to MDBs to be more innovative in increasing the share of guarantees relative to loans
· Articles limit gearing / leverage to one-to-one and may need to be amended or reinterpreted in the case of guarantees
Next Steps
· A meeting of MDB Treasurers and officials in charge of guarantee operations to discuss the concept
· Persuading the stakeholders of MDBs, starting with key shareholder governments, that increase in leverage on guarantees should be pursued in place of capital increases
· A detailed study of the Article provisions on gearing / leverage to determine whether higher leverage on guarantees requires charter amendments
· Drafting such amendments of the Articles
· Discussions with the rating agencies to ensure that triple-A ratings would be preserved
· Start the process of amendments in the Articles
1. Introduction
The purpose of this paper is to argue that while one-to-one leverage on the part of MDBs may have been essential in their early days, when they had no track record, it is probably unnecessary given their long and successful track record, at least as far as their guarantee operations concerned, which remain very small and which should be increased as guarantees play to the core competence of multilateral development banks: risk mitigation.
Multilateral development banks (MDBs) have for more then fifty years played a key role as lenders and investors in emerging markets, typically providing long term loans at fixed or floating rates in a variety of currencies for developing country projects designed to develop economic and social infrastructure across all major sectors such as power, roads and ports, telecommunications, health, education, the environment in a wide range of programs and projects. MDBs typically fund their long-term loans in the capital markets at rates befitting their triple-A credit ratings. Under their respective Articles of Agreement (“Articles”), they typically follow conservative financial policies with respect to liquidity, asset quality, funding, gearing and leverage. The policy constraint that most commonly binds is that on loan concentrations.
MDBs tend to follow conservative policies on asset quality, lending only to governments of their developing members states or to public sector enterprises with counter-guarantees of such states and generally do not reschedule their loans, being considered preferred creditors not only by the borrowers but by other lenders.[2] IFC and EBRD do lend to the private sector, without government guarantees. MDBs have no deposit base but rely instead only on capital market borrowings for funding. They tend to hold substantial amounts of cash (derived not only from paid in capital but also to differing degrees of borrowed funds raised at very fine rates), which they tend to invest in liquid and highly rated securities, earning positive returns. They also tend to take little or no currency risk, passing it along instead to their investors. With respect to gearing or leverage, MDBs set aside against each dollar of loan or guarantee commitment at least as much capital, in a ratio loans and guarantees to capital that may not exceed one-to-one. Specifically, this ratio of the sum of loans and guarantees to the sum of callable capital, paid in capital and unimpaired reserves including loan loss provisions is often referred to as the gearing or leverage ratio, terms that we will use interchangeably for purposes of this article.
When initially established, it was assumed that MDBs would have capital market access only if they were designed to have a gearing or leverage ratio of one-to-one, as they were to make loans only for purposes of reconstruction of a Europe devastated by war or to improve the standards of living of developing countries. That the lowest possible gearing would be necessary for capital market access was probably a good assumption at the time but we believe it is time to reexamine that assumption. In order to address potential concerns of capital market investors, the concept of callable capital was devised under which only a portion of subscribed capital is paid in, with the rest being callable in the event it was needed to meet the MDB’s debt service or lending obligations. Thus, capital market borrowers could rest assured that debt service payments would be made from current operations and if not then from calls on capital of the shareholders of the MDB. With the implied shareholder support, the important political and economic purpose of the institutions and their conservative statutory and policy controls, MDBs achieved triple-A ratings with virtually no track record.
Other somewhat similar institutions – which we shall refer to as regional investment banks sponsored by OECD countries (RIBs) -- that aimed their lending and investment operations not at developing countries but at the industrially developed OECD members states, such as EIB and NIB, were able to achieve higher leverage of 2.5 to 1 from the outset. They could thus lend and invest up to two and a half times their capital plus reserves.[3] Such higher leverage was accepted to the markets for the European RIBs because their loans were probably of higher quality than those of MDBs.
In practice, both MDBs and the RIBs have had excellent asset quality, with only insignificant loan losses.[4] Through the post-World War II period, they have amassed an impressive track record in achieving their goals of fostering reconstruction and development as well as regional integration.
Rating agencies have rated both MDBs and the regional European investment banks on the basis of well-established rating criteria (see Appendix 1). While rating agencies typically require only a five-year track record to be able to rate a financial institution, the oldest MDB, the World Bank, now has a track record of nearly sixty years. One of the youngest MDBs, the European Bank for Reconstruction and Development (EBRD), which was founded in 1991, already has a track record of nearly fifteen years. Based on a successful track record along the various rating criteria, it may now be possible to achieve triple-A ratings without one-to-one leverage.[5] Indeed, IFC has achieved such ratings with a leverage of four times, with no callable capital, although it lends to the private sector in emerging markets, with no governmental guarantees.
2. MDB Guarantees are Vastly Underutilized
MDBs provide three main types of guarantees that can be important risk mitigants in infrastructure finance. First, partial credit guarantees (PCGs), which cover debt service defaults on a specified portion of a loan or a bond.Coverage is typically available for new investment and not generally offered by MDBs for refinancings or restructurings. Such guarantees could permit longer financing terms and lower spreads. This type of facility was used in the 1998 EGAT financing. Second, partial risk guarantees (PRGs) cover debt service defaults on a loan to a private sector project caused by a government's failure to meet its contractual obligations related to a private project. Third, policy based guarantees cover a portion of debt service on a borrowing by an eligible member country from private foreign creditors in support of agreed structural, institutional, and social policies and reforms. Such guarantees have been provided on the basis that the private markets must bear some of the risk of the loans as development partners of the MDBs. Therefore, the guarantees have been “partial” covering either only a part of the risks – such as political risk, risk of non-payment of some portion of the debt service obligation such as the last three interest payments and the final principal payment, etc.
Traditionally, the World Bank Group (including IBRD, MIGA, IFC and IDA) have been active as risk mitigators through a variety of instruments – direct guarantees with or without government counter-guarantees, A/B loan structures, partial credit and partial risk guarantees, etc.[6] The World Bank's guarantee instrument was formally mainstreamed in 1994 to address the growing need to offer political risk mitigation products to commercial lenders contemplating financial investment in the infrastructure sectors of developing countries.The World Bank's fundamental objective in offering guarantees is to mobilize private capital for such projects on a "lender of last resort" basis.
The ability of MDBs to offer guarantees is hindered by the leverage restrictions they apply to such guarantees. For example, the World Bank and the Asian Development Bank apply one-to-one leverage to guarantees as they do to loans, without adjusting for evidence that may show that guarantees may be less risky. This restrictive leverage for guarantees unnecessarily limits their guarantee capacity and increases the need for capital before the markets demand it.
By contrast, though members of the World Bank Group, MIGA and IFC use a leverage ratio of four to one, still enjoying the same triple-A ratings as the MDBs with one-to-one leverage such as ASDB and the World Bank. PRGs are also provided by bilateral investment guarantee agencies such as the US Overseas Private Investment Corp (OPIC) and bilateral export credit agencies (ECAs) such as US Export Import Bank and NEXI, whose ratings depend on the ratings of their respective countries.
3. Who provides guarantees?
A key multilateral risk mitigator of political risks for equity and debt investment within the World Bank group is MIGA.[7] In addition, IDA, the World Bank’s soft loan window, pursuant to a pilot project established in 1997 has extended two PRGs. IFC, another member of the World Bank group, also has a portfolio of loan guarantees and risk mitigation products and IFC and offers other forms of credit enhancement, such as the purchase of the junior tranche of an asset backed financings. Regional MDBs, like the Asian Development Bank, the Inter-American Development Bank and the European Bank for Reconstruction and Development, with their high ratings and substantial capital, have also become active in political and credit risk mitigation.
MDBs such as the World Bank and African Development Bank have traditionally been lenders first and a guarantor second. The volume of loans they make exceeds by far the volumes of guarantees (by multiples that range from a high of 148 for IDB and a low of 11.6 times for EBRD in 2003) except in the case of MIGA, which was established in as a specialized multilateral guarantee agency that provide no loans. Although a part of the World Bank Group, MIGA can leverage its capital at four times, unlike MDBs. The World Bank has provided some two-dozen guarantees since the late 1980s and currently has guarantees outstanding of about $2 billion.
MDBs act as though they are required under their Articles to set aside the same amount of capital for each loan as for a guarantee, though this may be more a matter of interpretation rather than a statutory requirement. Identical leverage for guarantees as for loans from the same institution is an impediment to expanded use of the guarantees. The rock bottom leverage for guarantees of one-to-one from the World Bank and the regional MDBs is one-to-one (one dollar of loan must be typically be backed by one dollar of capital) can lead them to price their guarantees out of the market. In comparison, private monoline and multiline guarantors and even MIGA apply much higher leverage.[8] We believe that the MDBs with one-to-one leverage could increase such leverage to a multiple as high as four times for purposes of guarantees, probably without jeopardizing their triple-A ratings. If they were to do so, and offer better guarantee products in collaboration with the private sector, they could help expand substantially the investments flowing into infrastructure development in the emerging markets.