Latin American Reserves are Largely Borrowed

It is widely believed that the recent Latin American economic boom represents a significant break with the past: the region has been growing rapidly while running current account surpluses versus a tradition of running large deficits. This, it would seem, has made Latin America more similar to the East Asian economies. If so, accumulated international reserves, which have also been booming in recent years, would reflect true savings.

This impression is partly correct. In the 1950s and 1960s, Latin America grew rapidly while running small current account surpluses (small deficits if Venezuela is excluded). From the 1970s, however, rapid growth was always accompanied by current account deficits. This was the story of the 1970s and 1990-1997, with the latter period showing less impressive growth performance. In both cases, however, the need to correct the current account deficits eventually led to recession: the “lost decade” of the 1980s and the “lost half-decade” that followed after the Asian crisis.

Fast growth came back in 2004 and has been accompanied since then by current account surpluses. Although such surpluses are associated with booming commodity prices rather than competitiveness, it is still a major gain in relation to the past. The 1970s were also years of booming commodity prices but accompanied by current account deficits. Current account surpluses have been used to reduce external debt ratios and accumulate international reserves. As a result, Latin America is better prepared than in the past to undergo the current turbulence in international financial markets.

The recent story is only partly correct, however. To start, current account surpluses are a characteristic of a few (indeed, increasingly fewer) economies. If we take out Venezuela, the regional current account was basically in balance, not in surplus, in 2007. The countries running current account surpluses were mainly mineral economies –Bolivia, Chile, Ecuador, Peru and Venezuela—, which have benefited from booming mineral prices on the world market. The only exception to the association with mineral price booms is Argentina. Brazil was until recently also an exception, but ran a deficit in the last quarter of 2007 and is likely to run one for the whole of 2008.

Furthermore, the large accumulation of international reserves is associated with capital flows, not to current account surpluses. This is clearly indicated in the attached graph, which shows the aggregate current and capital account balances of the six largest Latin American economies (the other one is Venezuela, which displays very different behavior, more akin to that of the oil exporting economies of the Middle East). The pro-cyclical performance of capital flows is clearly shown: although foreign direct investment helped to generate a small recovery of capital flows in 1999 and 2000, after the initial effect of the Asian crisis, net capital flows fell in 2001 and were essentially zero from mid-2002 till mid-2004.

In mid-2004 capital flows came back, with strong winds. Two episodes of “exuberance” in international financial markets explain such force. The first started in mid-2004, when Latin American spreads in international financial markets fell below the pre-Asian level and capital flows started to generate a boom in Latin American stock markets. This was temporarily interrupted by the disturbances in international financial markets in May 2006, which were centered in China. This first episode was followed by an even stronger capital account boom from mid-2006 to mid-2007, which was interrupted, in turn, by the sub-prime financial crisis in the United States.

Viewed as a whole, it is the latter period that generated the largest accumulation of international reserves in Latin America, and their origin was the capital, not the current account. About three-fourths of the accumulation of reserves since 2004 (slightly over $210 billion from the first quarter of 2004 to the third quarter of 2007, for which we have data for all six economies) is the result of capital flows, and the proportion has been closer to 90% since the last quarter of 2006. So, Latin American reserves are largely “borrowed”. The source is different from the past: there has been a larger proportion of portfolio capital inflows invested in Latin American stocks and bonds denominated in domestic currencies. This again has several advantages, as it has reduced the risks associated with the increasing domestic burden of debts after exchange rate devaluation, which were devastating during past crises.

Therefore, it is neither competitiveness nor high commodity prices that explain the large amount of international reserves that the region has accumulated, but pro-cyclical capital inflows. The future will show whether these “borrowed reserves”, with their new features –i.e. the composition of the inflows—, are more resilient to an international financial crisis than in the past.

Current vs. captial accounts in six major Latin American economies (Argentina, Brasil, Chile, Colombia, Mexico and Peru; million dollars)

0802_ocampo.jpg

2 Responses to "Latin American Reserves are Largely Borrowed"

  1. Otaviano Canuto   February 7, 2008 at 10:43 pm

    Welcome, Jose Antonio. It is nice to have another high-standing fellow blogger like you.Your point on the role of (scaringly high) capital inflows as a driver of tremendous accumulation of reserves since 2006 is convincing. But I think you went too far by minimizing the change in foreign liabilities since the 90s (after all, it was not a simple swap for local currency-denominated debt). Even as most economies are marching toward current-account deficits – what is to be expected under current circumstances of growth differentials – one cannot deny that the scope for twin crises has shrunk substantially, and that has made a hell of a difference!

  2. Vitoria   February 12, 2008 at 6:23 pm

    Dear Jose Antonio,I am so glad your piece emphasize such an important point. The fact that latam reserves are mainly driven by the capital account reduces the whole argument that Latin countries are better prepared from a ‘sudden stop’ than before! Thanks for the great piece!Vitoria