Traditionally most countries in Latin America have only considered pegs to the USdollar.
Only relatively recently has attention been devoted to the (for now mostly
theoretical) option of some kind of regional monetary integration.11
Up to the collapse of the Argentine currency board in December 2001, there was a clear
tendency in most Latin American countries to have a formally declared or de-facto peg
to the US-dollar, some going even so far as to dollarize completely (Ecuador, El
Salvador).12 In their assessment of the performance of exchange rate regimes in Latin
America, Hausmann et al. (1999) accordingly stress the revealed preference of most
countries to have fixed exchange rates, even if they had formally declared floating rates.
They attribute this to the fact that flexible rates tended to be accompanied by higher
interest rates, smaller financial systems, a higher sensitivity of domestic rates tointernational rates, and a higher tendency for wage indexation.13 Therefore, before 1998
most countries would not allow for large exchange rate movements, even in the
presence of considerable shocks.14 Hausmann et al. (1999) conclude that most of the
countries that had nominally declared flexible rates actually operated as if being on
fixed rates (but without their benefits). Hence the currency board solution that
Argentina had chosen before its collapse in 2001 was generally considered as a success
before 2001 (Ghosh et al. 2000).15
To conclude that fixed exchange rates have been useful for many countries in the region
does not necessarily imply that a regional arrangement should be advocated. In fact, it is
quite obvious that Latin American economies do not fulfill the standard criteria
developed in the optimum currency area literature. Systematic studies that compare the
region to the European example notice huge differences between Europe and the region.
For instance, in a comprehensive study Temprano Arroyo (2003) looks in detail at all
the traditional criteria, such as trade openness, asymmetric shocks, interregional
financial integration and labor mobility, to conclude that the countries of Central
America, the Andean Community and of Mercosur do not constitute an optimum
currency area. Trade interdependence and intra-regional financial and labor market
integration are low, implying as well that the potential for asymmetric shocks is high.
Similarly, Larrain and Tavares (2003) show that the countries in Latin America have a
higher degree of real exchange rate variability than could be expected, given their
geographical closeness and economic characteristics. In comparison with other regions
such as Europe or Asia, they conclude that the region is the least integrated and thus
most vulnerable to asymmetric shocks and real exchange rate fluctuations.
Focusing especially on Mercosur, Belke and Gros (2002) show that these do not form a
regional (or sub-regional) bloc. They also show that intra-regional exchange ratemovements are less important than those against the dollar and the euro and that those
exchange movements, and the interest variability that accompanies them, have a
negative influence on employment and investments in the region.
Temprano Arroyo (2003) in addition points out that any regional monetary integration
arrangement would suffer from the absence of an established and credible low inflation
anchor. Without a stable anchor, the danger of speculative attacks against the anchor
currency and satellites would be high. Nevertheless, taking account of the high degree
of financial dollarization and the crucial credibility aspect, he concludes that, at least for
the Central American countries, a case can be made for a peg to the dollar if not full
dollarization. For the Andean Community, this conclusion can hardly be drawn, given
the lower degree of trade integration with the dollar area. If, however, the Andean
countries should decide to peg to an outside currency, the case for a dollar peg would be
stronger than that for a euro peg, simply because large parts of the sub-region are
already dollarized.
Exclusive intra-regional pegs in Latin America therefore apparently make not much
sense, nor do full monetary unions between the countries in the region, because they are
too little integrated among themselves and rely too much on trade relations with third
countries.16 At the same time, one might conclude that financially dollarized countries
are well off with a dollar peg. Accordingly, there appears to be no particular exchange
rate regime that fits all Latin American countries, and especially Mercosur countries
might not benefit from forming a regional monetary union among themselves.
Nevertheless, it remains a fact that dissimilar exchange rate regimes may not only have
a significant impact on trade flows, they may also act as an impediment to deeper
regional integration, as countries gain price competitiveness at the expense of their
regional partners. This has become obvious within Mercosur, as the Brazilian real
strongly depreciated by around 40 per cent in early 1999, whereas the Argentine Peso
was pegged to the (then rising) US-dollar. As a consequence, Brazil’s firms enhanced
their price competitiveness and the trade balance improved significantly, though there
has been a time-lag of one to two years before imports and exports responded to
movements in the (real) exchange rate (Figure 2). Overall, Brazil’s trade balance
switched from a deficit of US$ 12.2 billion in 1998 to a surplus of US$ 2.1 billion in
2001.The Argentine peso, on the other hand, bound by its dollar-peg rose during that period,
which contributed to a erosion of Argentina’s export performance in third markets and a
loss of competitiveness of domestic firms vis-à-vis imported goods (Figure 3). Unlike
Brazil, Argentina’s trade balance remained roughly unchanged in the period 1998 to
2000, indicating that its exporters were not able to compete with Brazilian firms. Yet in
2001, even before the collapse of the currency board, Argentina’s trade balance
improved, as economic growth and thus imports collapsed drastically. Overall, during
the severe economic crises, total imports declined by two thirds from 2000 to 2002,
thereby boosting Argentina’s trade balance.The loss of relative price competitiveness becomes more visible if we take a closer look
at exports to the United States, an important trading partner of both Argentina and
Brazil. As can be seen from Table 3, Brazil’s exports to the United States rose by one
third during the period 1999 to 2001, whereas the same figure for Argentina is much
lower with an increase of only 7.4 per cent.Moreover, Argentina’s export to Brazil fell by 22 per cent from 1998 to 2001 (ITC
2004), thereby confirming the concerns of exporters in Argentina. The surge in
Argentina’s real effective exchange rate was compounded by a recessionary
environment and a further reduction of tariffs in some sensitive sectors. As a
consequence, a considerable number of disputes emerged between the two Mercosur
12
countries. Shortly after the Brazilian devaluation, Argentine firms were actively seeking
protection, such as a compensatory tariff mechanism, to diminish the full impact of the
exchange rate appreciation. Though the Argentine government did not fully comply
with these requests, they encouraged various protectionist measures, for instance,
voluntary export restraints for meat and iron and steel products or import license
requirements for shoes (IDB 2002). In sum, Brazil’s sharp devaluation of its currency in
1999 enhanced economic turbulence in Argentina and created economic and political
tensions between the two Mercosur member countries.