The lender of last resort function under a currency board: The case of Argentina (Q1367837)

The role of a central bank as the lender of last resort for financial institutions has a long history. The evolution of dominant commercial banks, such as the Bank of England and other European commercial banks, from private institutions to government agencies, and the establishment of the Federal Reserve system in the United States, were strongly influenced by the perceived inability of private institutions to stop financial panics. The analytical argument for this government function has been refined considerably in recent years. There is now widespread, but not universal, agreement that asymmetric information, which accounts for the existence of financial intermediaries, also makes them prone to self-perpetuating declines in asset values during a ``panic''. Modern, fractional reserve banking systems are inherently unstable in that, although an incentive compatible regulatory framework will encourage bankers to insulate their portfolio from diversifiable risk, common shocks -- systemic risks -- reduce the market value and banks' assets. Any system in which the value of bank liabilities do not also decline is viable only if a lender of last resort is willing and able to transfer wealth to depositors. This implicit liability taken on by such a lender of last resort (LOLR) can be reduced in a number of ways but not entirely eliminated. An LOLR steps in because bank failures entail economic costs related both to a break-down of the payments system and the special role banks play in evaluating credit risks. It is also widely accepted that the existence of a lender of last resort is itself a distortion that can reduce the probability of declines in asset values but at the cost of subsidizing private investment decisions that reinefficient and potentially very costly. The solution currently in place in industrial countries is that institutions benefiting from lender of last resort facilities are constrained both in their investiment decisions and in their capital and liability structures. While solutions other than regulation have a long history as logical arguments, there is little recent experience with how they might work in practice. The classic description is to do away with banks as we know them by requiring that insured institutions hold highly liquid reserves against most of or all of their monetary liabilities. While these alternatives are discussed below, it should be kept in mind that attempting to operate the financial system according to such a scheme would be a considerable step into the unknown. For this reason we focus on the traditional trade-off between insurance and regulation. The case of Argentina provides an interesting and unique case to review the role of the Central Bank, a quasi-currency board, in attempting to both provide limited LOLR functions and defend the unique Argentine convertibility plan -- a combination of fixed exchange rate and full convertibility between the peso and the US dollar. This system was put to the test during the 1995 post-tequila crises, and although the system survived, it revealed the inherent difficulty of the current arrangement. The plan of the paper is as follows. In the next section we first put the issues into the Argentine context and examine the operation of the quasi-currency board in the recent banking crisis of 1995 and look at the analytics of the problem. Section III discusses the role of reserve of liquidity requirements and the option of full dollarization of the banking system. Section IV considers the risks associated with changes in the exchange rate, which any fixed rate arrangement faces. The final section (V) explores a range of policy options to augment or reduce the need for the central bank to exercise its lender of last resort function in the Argentine context.