Subcommittee on Economic Policy
Subcommittee on International Trade and Finance

Hearing on Official Dollarization in Latin America

Prepared Testimony of Dr. Liliana Rojas-Suarez(1)
Chief Economist for Latin America
Deutshce Bank Securities, Inc.

9:00 a.m., Thursday, July 15, 1999

Dollarization in Latin America?

My position on the current debate about dollarization in Latin America is that, while the goal of a currency union, with the US dollar as the lead currency, could be a good idea in the long run, a jump into "official dollarization" is not advisable for most countries in the region in the immediate future. Dollarization should be the last step taken in the ongoing market-led process of regional financial integration. The current proposal calling for "quick" dollarization as the means to help Latin American countries deal with highly volatile international capital flows, suffers from several fundamental deficiencies, the most important is that it fully ignores the main risk facing countries in the region, the so-called "country" or "default" risk. This risk has a lot more to do with the underlying fundamentals of an economy than with a specific exchange rate regime. Two additional considerations lead to my skepticism about dollarization in the immediate future. (I) Flexible exchange rate regimes better equip countries facing large terms of trade shocks to minimize the costs of a shock and (II) key pre-conditions for dollarization are often taken lightly by most adamant defenders of the common currency proposal.

1. The need to focus on country risk. In today's global financial world, emerging markets have found it increasingly convenient to meet some of their financing needs in the international bond markets. In contrast to bank lending, where syndication and other concerted arrangement mechanisms are possible, including default, bond holders are too distant and fragmented to make a coordinated arrangement. It is, therefore, not surprising that any economic policy, or political news affecting investors' perceptions about a country's capacity to service its debt, is immediately reflected in the yield spread between bonds issued by a particular country and comparable US Treasury instruments of corresponding maturity. Since both yields are expressed in US dollar terms, the spread is free of exchange rate risk and is considered a typical measure of the "country" or "default" risk.

During the 1990s, especially during the most recent emerging markets crisis, "country" risk perceptions have usually been accorded greater importance than exchange rate risk for determining investors' attitudes toward an individual country credit. Deterioration in investors' perceptions about a country's risk characteristics, as reflected in higher spreads increases that country's external financing costs. These increasing costs, in turn, translate to rising domestic interest rates, as existing financing needs press against the limited supply of domestic funds. In a number of the recent emerging market crisis episodes, large and increasing stocks of short-term debt, fueled by equally large and increasing fiscal deficits, increased doubts about a country's capacity to service its debt obligations.

Dollarization would not have prevented the deterioration of country risk from arising due to these policy inconsistencies. Indeed, I would even argue that, contrary to what is often advanced by supporters of dollarization, increased country risk leads to increased devaluation risk and not the other way around. As perceptions of country risk deteriorate, countries find it more difficult to rollover maturing external debt. Net external amortization payments and the consequent reduction in international reserves follow, thus calling into question the sustainability of the exchange rate. Indeed, in some cases such as Mexico in 1994 or South Korea in 1997, devaluation was part of the solution. Improved competitiveness allowed countries to acquire needed resources to restore credit worthiness in the international investor community.

To conclude this point, let me add that it is no coincidence that the few Latin American countries that have been awarded the classification "investment grade" by international rating agencies - Chile, Colombia, El Salvador and Uruguay -- share an important characteristic: these countries have never restructured their external debt.

2. Facing Terms of Trade Shocks. An important feature of a number of Latin American countries is that they are subject to large terms of trade shocks. These shocks represent a sudden reduction in the net transfer of real resources from abroad. As such, they require an adjustment in relative prices, implying a reduction in the price of non-tradable goods relative to the price of tradable goods. This needed adjustment can be characterized as a "real", as opposed to a "monetary", phenomenon. As the relative price of non-tradable goods declines so do the real revenues of producers in this sector and, therefore, their capacity to service debt obligations also falls.

Dollarization can not prevent this adjustment, but it can influence the form in which the adjustment takes place. If the exchange rate can not adjust (the definition of dollarization), then the adjustment will need to take place through a more severe contraction in output growth than would be required under a flexible exchange rate system which can, in part, compensate for lost competitiveness. Consistent with my view that "country risk" is the most important factor influencing investors' attitudes towards emerging markets, long and deep recessions do nothing but exacerbate the perception of a country's reduced capacity to service its debts.

3. Pre-requisites for Dollarization. From my point of view, supporters of dollarization often ignore the necessary pre-conditions for an effective dollarization process. More often than not, dollarization's defenders conclude that dollarization can be implemented very quickly regardless of the state of economic and financial conditions. I argue that this is a faulty notion. While this is not the place to present a full list of what I view as the pre-requisites for even considering the transition to dollarization, I want to stress the one pre-condition that I consider essential: a sound domestic banking system.

To exemplify my point, imagine if Mexico had attempted to dollarize its system in 1995 in the middle of its severe banking crisis. In Mexico, a run on deposits was contained partly because of the presence of full deposit insurance in pesos. While the problems with this system are well known by now, and is entailing a significant fiscal cost and the persistence of relatively high inflation in Mexico, sudden dollarization when banks were facing loan problems would have resulted in a bank run as depositors realized that neither the banks, nor the authorities, had sufficient dollars to back up their deposits. Of course, this problem could be avoided if dollarization were accompanied by an arrangement with foreign banks for large capital inflows or by a monetary agreement with the Fed, including assurances for access to the discount window and the FDIC. I would argue, and even the staunchest defenders of dollarization would no doubt agree, that neither foreign banks nor the Fed would be willing to allow access to their deposit insurance if a country had a severe banking crisis.

4. Concluding Remarks. Proponents of dollarization often lay blame on exchange rate risk and the absence of appropriate monetary policy for causing major financial fragilities in emerging markets, in general, and in Latin America, in particular. Based on my own assessment, the truth is that the drastic reduction of inflation is among Latin America's most important achievements in recent history and it resulted precisely from central banker's favorable management of monetary policy.

Most of the remaining problems in Latin American countries, as reflected in high international spreads, are due more to fiscal and banking mismanagement, for which dollarization is far from being a useful tool, than to monetary issues. Indeed, flexible exchange rates combined with large reserve holdings seem to have served well those countries that faced large external shocks. The flexibility of the exchange rate allows countries to compensate for competitiveness losses following the shock, while foreign exchange liquidity, both in the financial system and in the central bank, help countries to assure investors with that debt obligations will be met on time despite large unanticipated shocks. The effects of exchange rate fluctuations on the banking system could also be minimized if provisioning requirements were done ex-ante, rather than ex-post a crisis. If, as we all agree, sudden external shocks affect the non-tradeables sector the most, larger provisioning requirements for loans to this sector "in good times" could be the least costly crisis prevention mechanism.


1. The author is Managing Director and Chief Economist for Latin America at Deutsche Bank Securities.

The views presented here are solely those of the author and not those of Deutsche Bank Securities.

Home | Menu | Links | Info | Chairman's Page