Body
Conceptual Framework and Background
The concept of a lender of last resort (LLR) stems from a central banking principle dating back to Bagehot more than a century ago. The principle holds that in a financial panic, the central bank should stand ready "to lend freely…whenever the security is good" (Bagehot 1873, 48, 51). This precept implies that in a financial crisis there is a multipleequilibrium situation, and that the good outcome can be secured and unnecessary real economic damage avoided by providing temporary liquidity to those entities that are fundamentally solvent. Application of the LLR concept to sovereign financial crises requires the judgment that the same principles apply when lending is cross-border and largely devoid of tangible collateral. This difference from domestic LLR lending underscores the importance of making the right judgment that the sovereign in question is politically willing and able, given enough time, to secure the resources that ensure it is solvent based solely on full faith and credit. In both the domestic and international contexts, a central assumption of LLR lending is that after confidence is restored, a reflow of private lending will materialize and the central bank (or IMF) will be repaid promptly. International experience suggests that this process can take much longer for sovereign cross-border crises than for domestic financial crises.
The LLR concept, and the idea of providing new lending to avoid default when countries are deemed solvent but illiquid, were important in the international strategy initially adopted to deal with the Latin American debt crisis in the 1980s. Concerted lending by major international banks was the main mechanism, rather than large loans by the IMF (which instead played a catalytic role supported with moderate lending). By the late 1980s the strategy transited to Brady Plan forgiveness of debt overhangs. Those governments seeking forgiveness (most governments in Latin America, excluding those of Colombia and Chile) implicitly acknowledged that they had been insolvent after all, typically because of the "internal transfer problem" of fiscal insufficiency rather than the more tractable "external transfer problem" of insufficient foreign exchange. By the early 1990s, success of Brady restructurings and cheap money internationally set the stage for a renaissance of the international capital market for developing country borrowers, based increasingly on bonds rather than bank loans.
The first major crisis in the resurgent emerging capital markets of the 1990s was the Mexican "tequila crisis" of late 1994 and 1995, which was largely the consequence of an outsized current account deficit but whose timing reflected the rise in US interest rates and two political assassinations. Concerted bank lending was not an option because the debt in question was short-term Mexican treasury paper held widely in the market. In a massive LLR operation, the US Treasury provided $20 billion in support, the G-10 central banks $10 billion through the Bank for International Settlements, and the IMF $17.8 billion in standby support (IMF 1995), enabling Mexico to avoid default and carry out fiscal and external adjustment.
As the new emerging capital markets survived this key test, and as global credit conditions again became highly favorable (as shown by a decline in US high-yield bond spreads), by 1996 and early 1997 large financial flows went to a long list of middleincome countries. However, excessive short-term borrowing combined with weak domestic banking systems ignited the East Asian financial crisis in 1997–98, and Russia's default in August of 1998 marked a more general deterioration that helped sweep Brazil into crisis by year-end. The period 1997–98 was an intense phase of international LLR operations, this time centered much more heavily in the International Monetary Fund, in part because of the revealed political unpopularity of the US bilateral LLR operation in the Mexican case.2 Even the institutional machinery changed to accommodate the new reality, with the creation of the Supplementary Reserve Facility in the IMF, allowing much higher volumes of lending relative to a country's IMF quota. The interest rates were higher and maturities shorter in this new facility, as befitted an LLR instrument. A subsequent surge of LLR lending in even larger volumes but concentrated in fewer countries occurred in 2001–02 as Argentina, Brazil, and Turkey experienced financial crises associated in considerable part with political developments, including the prospect of election of a leftist government in the case of Brazil.
LLR Lending and Repayment
It is sometimes argued that the IMF cannot be a lender of last resort because, unlike domestic central banks, it cannot print money and hence does not have unlimited resources. In practice this distinction has not been relevant so far. The question of adequacy of IMF resources to carry out the LLR function did become increasingly germane, however, as a large share of the IMF's resources came to be concentrated in loans to just Argentina, Brazil, and Turkey. This phenomenon in turn has reflected the difficulties in ensuring that one of the LLR principles—prompt repayment upon revival of private flows—is observed in international sovereign LLR operations. The longer the repayment is delayed, the more the IMF's lending capacity is tied up in outstanding loans.
Commentary Type