Our currency risk measure is designed as a measure of "external vulnerability". Conceptually, it captures the interaction between a country's external net indebtedness and leading indicators for emerging markets' terms of trade on world markets (export prices relative to import prices). Both the level and the change of external debt are relevant. It is similar in spirit to the concept of a "critical threshold" of official reserves relative to short-term foreign currency liabilities (instead of the traditional measure of official reserves in months of merchandise imports) as suggested in Chairman Greenspan's presentation at the World Bank Conference on Recent Trends in Reserves Management in Washington D.C., April 29, 1999.

Our model does not assume that currency risk always arises from "within", i.e. from a single country's domestic conditions including its domestic policies. For instance, one potentially powerful factor affecting our indicator of emerging market risks are exchange rate shifts at the G3-level (US-Europe-Japan). Another potentially powerful factor are oil and commodity price changes on world markets, similarly beyond emerging market governments' control. Our model further suggests that with fully globalized financial markets and somewhat contrary to assumptions often made in official policy declarations, real (i.e. inflation-adjusted) exchange rate depreciation has not an expansionary, but a contractionary effect on a net debtor country's macro-economy. This has recently been acknowledged by a growing number of empirical researchers (see among others, Steven B. Kamin and Marc Klau, "Some Multi-Country Evidence on the Effects of Real Exchange Rates on Output" (in PDF), Board of Governors of the Federal Reserve System, International Finance Discussion Papers No. 611, Washington D.C., May 1998, see also Ramon Moreno, "Depreciation and Recessions in East Asia" (in PDF), Federal Reserve Bank of San Francisco Economic Review, No.3, October 1999.

According to our analysis, a negative effect of exchange rate depreciation on overall economic activity comes about through various financial sector channels such as tightening terms of international credit, reduced real balances and a decrease in net aggregate wealth. Earlier conclusions suggesting a positive wealth effect from exchange rate depreciation have recently been proved wrong in the professional economic literature (see Lance Taylor, "Exchange Rate Determination in Portfolio Balance, Mundell-Fleming, and Uncovered Interest Parity Models" (in PDF), Center for Economic Policy Analysis Working Paper Series II, No.8, New York, April 2001, sponsored by the MacArthur Foundation and by the Ford Foundation). Fortunately, also official international organizations have now woken up to this empirical reality. We commend, in particular, the Inter-American Development Bank, IDB, for its analytical leadership on this important issue (see "Financial Turmoil and the Choice of Exchange Rate Regime" (in PDF), by Ricardo Hausmann, Michael Gavin, Carmen Pages-Serra and Ernesto Stein, IDB, March 1999).

Exchange rate depreciation tends to raise real interest rate levels and, as a result, to lower aggregate output, because they weaken a country's capital account by stimulating private net capital outflows in the short run. In other words, exchange rate depreciation tends to feed on itself like a downward spiral (in technical jargon often referred to as "multiple equilibria"). Also a floating exchange rate policy chosen to protect official reserves cannot be expected to keep a country's overall debt service capacity in foreign currency constant. As private capital flows respond negatively to exchange rate depreciation in the short run, the relationship between local currency risk and foreign-currency credit risk is more likely to be complementary, not substitutional (there is no trade-off). In practice one therefore observes a given country's foreign currency credit spreads and its domestic currency interest rate spreads to move in parallel, which is inconsistent with the theoretical proposition, a floating exchange rate would be an effcient instrument to stabilize a country's real debt service costs. (At best, one may observe some short-term trade-off between home currency risk and transfer risk.) Since our currency rating is based on the maturity composition of a given country's net foreign liabilities, its value added is twofold: one, to allow international investors to compare market price measures (local currency interest rate spreads) with independent quantity measures of currency risk; two, to complement conventional measures of foreign currency solvency risk with a separate measure of foreign currency liquidity risk.

As an early warning system, our model places a premium on "data speed" meaning on data which are available at high frequency and relatively fast. We therefore abstract from "slow" and lagging indicators such as fiscal policy measures or actual inflation. Of the eight variables which for each country drive our currency risk measure, only one needs to be estimated for the last two months. The other seven are essentially same-month outturns. Our data sources are BIS (Bank for International Settlements), IMF (International Monetary Fund), IFC (International Finance Corporation, a World Bank affiliate), Standard & Poor's DRI, daily market quotations and national central banks. 

[Trial password for our internal member area.]

©  1998-2001. Atlantica Associates LLC. All Rights Reserved.