Beyond a Free Trade Area: Does Mercosul Need Exchange Rate Coordination?

I - Introduction
The Mercosul was created in 1991 when Brazil, Argentina, Paraguay and Uruguay signed the Asuncion Treaty. Its purpose is to promote integration through free movement of goods, people, services and production factors, the adoption of a common trade policy, including a common external tariff, macroeconomic coordination, and harmonization of its members’ legislation.
Initially the plan was to create a common market by 1995, but until now this objective has only been partly reached. Free trade (realized since 1991) and a common external tariff (realized since 1994, Ouro Preto Treaty) have been officially realized, however many specific goods still enjoy high levels of protection. Macroeconomic policy coordination still seems far from happening.

Initially, tariffs were significantly reduced and intra-Mercosul trade increased 190% from 1990 to 1994. After a period of fast integration, at the end of the 1990s the process started to fall back, mainly because of the economic relation between Argentina and Brazil, the two biggest economies of Mercosul. Their relation started to set apart after the Brazilian crisis, which led to the real devaluation in 1999, flooding the Argentinean market with cheap Brazilian products. The Argentinean government, suffering from balance of payments deficits, increased the restrictions to Brazilian products in order to defend its fixed peg against the US dollar and to protect national producers. After the Brazilian crisis of beginning 1999 it was Argentina’s collapse in 2000-2001 that buried any chance of a more ambitious economic integration in the near future. In 2003- 2004, although the crisis was outmoded, the difference of strategy in dealing with their creditors kept them distant. Brazil had to renew around US$ 30 billion a year
of foreign loans, and at that moment it would not be helpful to straighten relations with its neighbor that had just defaulted on its debt and was confronting the international financial community. The worsening of the relations between the two biggest economies of Mercosul led the integration process to a retreat and nearly to a complete collapse. 4

Today, in 2007, the situation is different. Argentina completed negotiations of its debt and its economy has been growing at an average of 9% in the past four years. Brazil and Uruguay have successfully completed their stabilization programs. It seems that the Mercosul integration is starting to move forward with the recent creation of important institutions, such as the Mercosul Parliament. Venezuela has become a full member of the block in 2006, which also now
includes Bolivia (since 1996), Chile (since 1996), Ecuador (since 2004), Peru (since 2003) and Colombia (since 2004) with an associate member status.
The lack of policy coordination is still a big risk, especially concerning exchange rate fluctuations. Brazilian and Argentinean currency crises made it clear that consistent exchange rate policy between the two countries is a condition to deepen Mercosul economic integration. An attempt of coordination was made in 2000, but it was unsuccessful mainly due to the Argentinean crisis.
Exchange rate fluctuations were already an issue between Brazil and Argentina and are now again in focus due to a recent Brazilian real appreciation.
In order to be able to discuss if exchange rate coordination is the best strategy for the Mercosul countries we will use the Optimal Currency Area (OCA) theory. The theory was first suggested by Mundell in 1961 with the objective of determining whether a group of countries should either fluctuate the exchange rate, or if they should form a single currency area. Many authors contributed to this theory and nowadays it is the theoretical referential when discussing exchange rate systems.
The objective of this study is to analyze the relevance of exchange rate policy coordination in Mercosul answering the question if some kind of common policy in order to reduce exchange rate variability makes sense in the light of the Optimal Currency Area theory and how important would that step be for a deeper integration of the region. The study will focus mainly on Brazil and Argentina, using existing literature about the subject and empirical data analyses. Firstly, we will start with an overview of the Mercosul block and exchange rate policies in the main two countries. Secondly, we will discuss the importance of exchange rate coordination for the region, analyzing if Mercosul forms an Optimal Currency 5 Area. Thirdly, we will consider the first attempt of coordination and the reasons for its failure. The study will finish discussing the possibility of adoption of a single currency in the future.

II – Mercosul
The origin of Mercosul is the 1986 Argentine-Brazilian Economic Integration Program (ABEIP). Brazil and Argentina put aside long-time differences to gather in a formal program for economic and political cooperation. The agreement had a strong political reason: promote stability for the new democratic states, after prolonged years of military rule. In the economic field the treaty aimed expand, and diversify, bilateral trade, issuing protocols that emphasized trade, in specific
sectors, such as capital goods, agribusiness and automotive (Manzetti 1994).
However, few of the economic goals of the agreement were realized: Argentina’s exports to Brazil increased from 5.8% of total exports in 1985 to 11.4% in 1990, while Brazil’s share of exports to Argentina hardly changed. By the end of the decade the economic situation in both countries was not good, due to the failure of domestic macroeconomic policies (high inflation) and an increase of foreign debt. The economic situation slowed down any progress toward integration.
In order to save the integration project, the presidents of both countries signed a new agreement in 1989, which had the intention of creating a free trade area over a 10-year period. In the early 1990s, adopting the “Washington Consensus” policies, both countries chose a free-market orientated economic foreign policy, which culminated in the Buenos Aires Act. By this agreement both countries committed on having a common market established by the end of 1994. One month later Paraguay and Uruguay joined the integration agreement. In March 1991 the four countries signed the Treaty of Asunción, which6 established the creation of the Mercado Comun do Sul1 (Mercosul). The main points stated in the treaty were:
- Abolishment of cross border tariffs: Tariffs would be phased out reaching a zero tariff between the four countries (Argentina and Brazil by the end of 1994 and Paraguay and Uruguay by the end of 1995)
- Coordination of macroeconomic and sector policies: On trade, agriculture, industry, fiscal policy, monetary policy, exchange rate policy, services, customs, transportation and communication..
- Common external tariff
- Institutional framework to solve trade litigation
- Creation of the Council of the Common Market: Highest decision making organ of the Mercosul. With the objective of establishing general policy guidelines, formed by foreign affairs and economy ministers of the member countries. The presidency of the council will rotate every six months between countries.
- Creation of the Common Market Group: The executive institution of Mercosul. Takes care of the technical aspects of the negotiation process. Foreign affairs ministers of member countries form this organ.
The foundation of Mercosul was looked at with skepticism. The ambitious project of creating a common market in four years between countries characterized by political instability, low economic performance (in the 1980s), and delayed economic reforms seemed rather difficult. Nevertheless, through the Asunción Treaty there was intensification of intra-regional trade between the member countries. In the period between 1990 and 2000 the intra-regional average annual exports growth was of 16.4%, while the countries’ total exports grew at an average of 5.9% (Machinea 2004).
Despite natural conflicts between its member countries, the Mercosul evolved from a situation where trade was taxed, towards a free trade area, 1 Mercado Comun del Sur (Mercosur) in Spanish and South American Common Market in English7 though initially incomplete due to the existence of many goods that were still protected, and later a customs union was created. On January 1st 1995, after the Ouro Preto Treaty (1994), the free trade area became a customs union, after the adoption of a common external tariff of 0% to 20%. However, each country still had a long list of temporary exceptions, goods that are still not integrated in the common external tariff system, in order to protect national industry.
The initial large increase of the trading volume between the Mercosul countries is explained mainly by the low trading level on the starting point of the agreement (Eichengreen 1998). In 1991 trade between Mercosul countries was US$ 5.1 billion, representing 11% of Mercosul total trade, while in 1995 trade within Mercosul had jumped to US$ 14.3, 20.4% of total exports. In addition, during this period stabilization plans of Argentina and Brazil were held, boosting economic growth in these countries, which increased consumption and, therefore, imports. In the end of the 1990s this favorable scenario changed and economic integration slowed down. This change was mainly caused by the 1997 Asian crisis leading to a crisis on the exchange rate policies of the two biggest economies of the block. In the next section stabilization plans and shocks will be analyzed.

After the turmoil of the Brazilian and Argentinean economies and the consequent fall back of the development of Mercosul, both countries recuperated from their financial crises and the Mercosul started to advance again. In 2002 the Mercosul Permanent Tribunal of Revision was created, which is responsible for dispute resolution upon complaints about member states following the Mercosul rules. In 2007 the Mercosul Parliament started its activities. In 2006 Mercosul integrated Venezuela as a full member2, while Chile, Peru, Colombia, Bolivia and Equator gained associate members status. Associate members have specific trade agreements with Mercosul, which have the objective of reducing trade tariffs in order to, in the future, form a free trade area.
Today Mercosul is an intergovernmental agreement, with all the negotiations carried out between presidents. Representatives of the main 2 Not yet ratified8 institutions are part of each country’s government (not directly elected as a Mercosul representative). The intergovernmental relation was chosen in the beginning to speed up the process, avoiding slower bureaucratic institutions and the process of creating them as well. Mercosul already signs that it is changing to become a supranational institution. The creation of the Parliament and the Permanent Tribunal of Revision (high court of Mercosul) are already a sign that Mercosul is moving towards supranationalism. This sign is stronger in the case of the Parliament, which will have its representatives elected in 2010 (the other institutions still have representatives of each country’s government). The institutional structure by itself also gives great support to this idea. Mercosul is formed by an executive organ, the Common Market Group, and by a legislative organ, the Common Market Council. These two organs deal with not only trade matters, but have a wider role dealing with human rights, child labor, agriculture, women rights, energy, etc.

III – Argentina – Brazil: Exchange Rate and Trade
Exchange rate policy was an indispensable instrument for the Argentinean and Brazilian stabilization plans of the early 1990s, which had as a main objective reducing inflation, a chronic problem of these economies. Argentina based its stabilization plan on a currency board, which started in 1991, fixing the peso to the US dollar. The Brazilian government started in 1994 the “Real Plan”, also pegging its currency to the US dollar. Both policies were successful regarding
reducing inflation, however they led the countries to different and conflicting results.
In the early 1990s Argentina and Brazil had their exchange rates pegged to the dollar, Brazil had a crawling peg established in 1994, devaluating its currency by 7% per year, while Argentina instituted a currency board in 1991, fixing the value of the peso against the US dollar. Both countries kept their currencies overvalued with the purpose of holding inflation down: cheaper9 imports increased competition in the internal market, reducing prices (Grasel, 2005). This strategy increased trade deficits in both countries, which had to be financed by capital inflows, increasing external vulnerability.

Under Argentina’s convertibility plan, inflation came down and the economy entered a period of fast growth. The real exchange rate vis-à-vis Brazil appreciated sharply and Argentina’s trade surplus with its neighbor turned into a deficit. The real exchange rate between Argentina and Brazil is represented in figure 1 (appendix), an increase represents an appreciation of the peso and a depreciation of the Real3. In the same figure we can see the increase of the Brazilian exports to Argentina and the decrease of the Brazilian imports from Argentina, influenced by the change in competitiveness between the two countries. In 1992 Argentinean authorities responded by imposing anti-dumping duties and safeguarding measures against Brazilian exports of farm machinery, spark plugs, steel, refrigerators, paper, textiles and chemicals (Eichengreen 1998).
In July 1994 the situation changed. Brazil launched its stabilization plan, and now it was the Brazilian Real’s turn to appreciate against the peso. The effect was similar to the one experienced by its Mercosul partner: Inflation was reduced to the levels of Argentina, the economy boomed and, in October 1994, Brazil started to have trade deficits with Argentina (figure 1). By the end of the first semester of 1995 Brazilian authorities raised tariffs, imposed import quotas and restricted the availability of trade credit in order to reduce the increasing
deficit.
Despite macroeconomic divergences in the first five years of the Mercosul we can observe in figure 1 a sharp increase in trade. Trade between Brazil and Argentina went from US$ 2.045 million in 1990 to US$ 11.976 million in 1996. Eichengreen (1998) suggests that Mercosul survived this period because of a number of exceptional conditions. First, he points out that trade started from unusually low levels, and the competition levels were at the 3 To avoid confusion between “real” (as in real appreciation) and the Brazilian currency, I will spell the second one starting with a capital letter: Real10 beginning too low. Second, both countries when adversely affected by these surges of imports were experiencing unusual rapid growth in their economies. Third, the world economy was growing and foreign financing was easily available at low levels of interest. High availability of capital was the key to financing both countries’ current account deficits.
Capital availability was very high at that period of time, with huge amounts of capital being invested in developing countries. This capital availability, which enabled maintaining current account deficits in Argentina and Brazil, was drastically reduced after the Asian and Russian crises, in 1997 and 1998 respectively. In January 1999, pressured by huge capital outflows after the Russian crisis, Brazil had to devaluate its currency and started to float its exchange rate. We can observe in figure 1 (appendix) that during the period of exchange rate stability from the Real plan implementation in 1994 until the Brazilian crisis in 1998, there was a large increase in trade: By September 1998 Brazilian imports from Argentina reached US$ 854 million and exports to Argentina US$ 635 million. The Brazilian devaluation of 1999 had a negative impact on Argentina’s trade balance with respect to Brazil. The devaluated Real increased competitiveness of Brazilian exports, putting even more pressure on Argentina’s current account deficit and its fixed exchange rate. In 1998 Argentina had a trade surplus with Brazil of US$1.3 billion. In the months that followed the Brazilian Real devaluation Argentina had trade deficits with Brazil. Argentinean authorities responded with protectionist measures, trying to avoid current account deficits with Brazil.
The Argentinean currency board had the value of the peso fixed with the US dollar guaranteed by constitution, which made the change to a floating exchange rate more difficult and painful than in the Brazilian case. After the Brazilian crisis Argentina suffered from huge pressure to devaluate its currency. In December 2001, government defaulted on its debt, followed by a devaluation in January 2002, which put an end to the currency board. The value of the Argentinean peso, which was fixed to the U.S dollar since 1991, devaluated 64% until the beginning of April 2002. If we look at figure 1, the crisis brought a11 negative result to trade. However, in this case much of the change is concentrated in the decrease of imports of Argentina from Brazil. This can be explained by the strong decrease in consumption in Argentina, GDP shrunk 10.9% in 2002 and credit was very scarce. Soon, however, after a successful renegotiation of its debt the country managed to return to growth: 8.9% 2003; 9.0% in 2004; 9.2% in 2005 and 8.5% in 2006. The increase in consumption and exchange rate stability quickly increased Argentinean imports from Brazil.
Today, the exchange rate between the countries is rather stable, with a slow and constant appreciation of the Brazilian Real against the Argentinean peso. Until now the Brazilian Real appreciation has not had a clear impact on trade between the two countries. The main reason for that is Argentina’s strong growth rate in the past four years, which has pushed imports from Brazil despite the appreciation of the Brazilian currency. In the period 2003 – 2006 imports of Argentina from Brazil grew 152%, reaching US$ 11,872 million in 2006, while exports grew 77%, reaching US$ 8,152 million in the same year. The trade deficit of Argentina with Brazil was US$ 3,720 million in 2006 (Table 1 in the appendix).
In the past years Argentina has again been imposing barriers on Brazilian products, mainly quantitative restrictions. In July 2004, Argentina launched a set of emergency protectionist measures against many Brazilian products, especially household appliances, pork meat, cellulose, shoes and textiles. What can we expect from trade between Argentina and Brazil in the next years? Argentina’s economy already gives signs that it is slowing down (is expected to grow 7,5% in 2007).

The decrease in consumption together with the Brazilian Real appreciation will probably reduce Argentina’s trade deficit in the long run: trade deficit has decreased in 2006, and exports of Argentina to Brazil started to grow at faster rates than imports. Historically Argentina has a positive trade balance with Brazil. In case this scenario reappears, it is most likely that another round of devaluations or/and protectionist measures take place. Is that the best strategy for these two countries and for Mercosul? The use of devaluations and protectionist measures only exports the problem to its neighbor, which does not seem to be a good strategy when the neighbor represents 10% of12 its exports (in the case of Brazil), or 35% of its imports (in the case of Argentina).
Would some form of exchange rate coordination be a better strategy? In order to answer that question, we will use the Optimal Currency Area (OCA) theory. In the next two sections we will discuss the importance of coordinating and if coordination is a good strategy for Mercosul.

IV – Benefits of Exchange Rate Coordination
The optimal currency area theory points out that the elimination of exchange rate variability risk, lower transaction costs and welfare gains of policy coordination are expected to be the gains from monetary unification. The costs of monetary unification are related to the loss of an independent monetary policy and the use
of exchange rate as an instrument of adjustment.
Eichengreen (1998) points out three views linking the importance of
measures to stabilize exchange rate and regional integration. One view is that
exchange rate variability harms trade and market integration, making price
comparisons more difficult and increasing hedging costs for importers and
exporters. Firms engaged in international trade would not need to hedge when
participating in intraregional trade. Without monetary union, a firm will most
probably use the forward market in order to hedge against the threat of
depreciation of a foreign currency. Whether the foreign currency depreciates or
not, the firm will have to pay the cost of hedging. In the case of exchange rate
coordination between Mercosul countries, this cost is no longer present. In
addition, firms would also identify a potential for reducing their financial
department employees, which are in charge of handling these financial
instruments.
This view was the official position of the European Commission in order to
justify the argument that the European market needed the single currency.
According to Eichengreen there is an emerging consensus that there is a
statistically significant negative impact of exchange rate on trade. According to
13
Ferrando (2000), US states are more integrated to one another than the USA to
Canada, despite the absence of barriers and tariffs between these countries,
suggesting that different currencies harm trade. However, Eichengreen defends
that the maintenance of separated currencies fluctuating against each other is by
no means incompatible with regional integration.
A second view discussed by Eichengreen is that exchange rate variability
creates import surges, which go against interests of the integration process.
According to this view separate national currencies produce exchange rate
swings and import surges that lead to protectionist backlash. Support for this
view is widely present in the Mercosul experience. As discussed in the previous
section, since the creation of the Mercosul until now in many opportunities
exchange rate variability led to a protectionist backlash and to a step back in the
integration process. In this sense, exchange rate variability can have an impact
on the political economy of the integration process.
A third view discussed by Eichengreen is the one that denies any
incompatibility between regional integration and fluctuating exchange rates.
Supporting this argument is the example of the North American Free Trade
Agreement, the NAFTA, which member states’ (USA, Canada and Mexico)
exchange rates continue to fluctuate widely. Trade tensions may arise from time
to time, but without being a serious threat to integration. Canada and Mexico are
heavily dependent on exports of primary commodities, making exchange rate an
important instrument of adjustment within the free trade area, due to variability of
world commodity prices.
Eichengreen points out that there is no incompatibility between these
views. The effect of exchange rate movements on regional integration depends
on the depth of integration and the source of shocks to which the exchange rate
moves.
According to Machinea (2006) a high level of integration among countries is one
of the main reasons for increasing the coordination of macroeconomic policies. A
high level of interdependence between a group of countries suggests that each
member of that group is affected by what happens to the other countries. We will
14
discuss the level of interdependence of the Mercosul members in the next
section.
De Grauwe and Vanhaverbecke (1993) analyze the variability of the
bilateral exchange rate among EU countries. According to this study the average
absolute variation of the real exchange rate of the EU countries in the period
1977-1985 was 4.4%. Ferrando (2000) did the same estimation for Mercosul and
NAFTA; his results are presented in table 2. Excluding Mexico, NAFTA has
similar exchange rate variability as the EU had, while the Mercosul countries
show higher levels of variability. Higher levels of exchange rate variability show
that the benefits of stabilizing exchange rate can result in even higher gains to
the Mercosul countries, reducing the change in relative prices that affect trade
and investment (higher exchange rate variability can also represent higher costs
as we will see in section V). Stability, cohesion and the increase in integration
would be a catalyst for foreign investment in Mercosul, due to the gains of scale
that a truly unified and stable market will bring.
Ferrando (2000) also points out that stability and cohesion in the region
can be important for negotiating bigger trade agreements, for instance the Free
Trade Area of the Americas (FTAA), as well as negotiations involving the WTO
and other blocs, such as the EU (Lavagna and Giambiagi, 1998).
Another important issue concerning exchange rate coordination in the
Mercosul is related to its effect on reputation and credibility. Lavagna and
Gianbiagi (1998) argue that article 1 of the Assuncion agreement, which
establishes coordination of macroeconomic policy (trade, agriculture, industry,
fiscal, monetary, exchange rate and capital flows) between members, has not
been fulfilled. Lavagna and Gianbiagi affirm that by not demonstrating any intent
of respecting this point of the agreement, uncertainty of its own member
countries, as well as other countries in relation to Mercosul as a whole increases.
Ferrando (2000) highlights another perspective regarding credibility and
reputation, considering the positive effect for Mercosul countries of giving up their
exchange rate policy autonomy. This point is usually seen as a cost, but
according to Ferrando, after adopting coordination, the costs of changing this
15
policy (exit barriers) are too high, increasing credibility on macroeconomic policy
of the countries of the region. Increased credibility of exchange rate stability will
have a positive effect on risk. If Mercosul countries engage on this kind of
coordination it would reduce exchange rate risk and, consequently, the cost of its
debt, as well as reducing cost of capital.
In order to be able to answer the question if coordination is the best
strategy to Mercosul we have to analyze if the benefits of coordinating exchange
rate policy are higher than the costs. By analyzing the literature of Optimal
Currency Areas we will try to answer the question whether the Mercosul bloc
forms one.
V – Does Mercosul Form an Optimal Currency Area?
Mundell (1961) first questioned the use of a system of national currencies
connected by flexible exchange rates. He proposed a simple model to determine
if a group of countries would benefit from abandoning their flexible exchange
rates and advancing to a fixed exchange rate within that group or a single
currency area.
In his model he considers two countries, in our case Argentina and Brazil.
Suppose both countries, initially in full employment and balance of payments
equilibrium, are affected by an asymmetrical shock. We define here asymmetrical
shock as a shock that affects both countries in a different way. Suppose the
shock increases demand of factors from one country, Argentina, and decreases it
in the other country, Brazil. With flexible prices, a decrease in demand for labor
will increase unemployment and, at the same time, decrease wages, keeping
unemployment close to its natural level. When relative prices are not flexible and
the supply of labor reacts to the shock, the adjustment of relative prices through
the variation of the exchange rate is not needed. In that case workers would
migrate from Brazil to Argentina, reducing the impact of unemployment. Brazilian
unemployed workers would go to Argentina, where there is an excess demand
16
for labor. This mechanism would eliminate the need for a decrease in wages in
Brazil and an increase in Argentina, ending the problem of Brazilian
unemployment and Argentinean inflation.
If any of these adjusting mechanisms (flexible prices or labor mobility) are
not present the government of Brazil will have all the incentives to adjust by
devaluating its nominal exchange rate against Argentina. When a group of
countries is affected by shocks in an asymmetrical way a country only needs to
use its exchange rate to adjust relative prices within this group, if prices are fixed
and there is no labor mobility.
De Grauwe (2005) agrees with Mundell (1961) that if wages are rigid and
labor mobility is limited, countries that join a monetary union would have more
problems in adjusting to asymmetrical shocks (higher cost) than countries that
kept their autonomy. However, still considering an inflexible labor market, what
would happen in the case of symmetrical shocks? Would devaluation of the
countries’ currency be the best option? The answer is no. Suppose there is a fall
in demand in both Argentina and Brazil. If Brazil devaluates against Argentina,
aggregate demand in Brazil would be stimulated. In Brazil aggregate demand
increases, while the Argentinean one decreases. In other words, Brazil would fix
its problem by exporting it to Argentina. The harmed country could respond to the
neighbor’s policy with a devaluation, creating the danger of a devaluation spiral.
According to De Grauwe (2005) a monetary union is more attractive than a
regime of independent monetary authorities when the shocks that affect the
countries are symmetric. Countries with high levels of market integration will
have more similar production structures and, consequently, more symmetrical
shocks, therefore being better candidates for a currency area.
Kenen (1969) defends that the expected cost of an asymmetric shock is
negatively related to the level of diversification of the economies of the member
countries. Diversification ensures member countries against asymmetric shocks
through two effects that operate simultaneously. The first effect is related to the
fact that greater diversification suggests that the productive structures of the
countries are more similar, reducing the possibility of shocks that affect them in
17
different ways. The second effect can be considered as the effect of pure
diversification. If production in a country is diverse, representing a portfolio
composed of negative related risks, the presence of a shock in one sector can be
compensated through other sectors, eliminating (or at least reducing) the macro
effect of the shock. In that case joining a monetary area would not represent an
important cost to the country.
Kenen also discusses the possibility of reducing the effects of shocks
through fiscal integration. Countries with a high level of fiscal integration are able
to smooth diverse shocks by fiscal transfers from a low unemployment region to
a high unemployment region.
McKinnon (1963) points out that very open economies rather not use
exchange rate to adjust relative prices. The effect of exchange rate variation in
countries with open economies, which are characterized by having more tradable
goods, is small, because this effect is easily reflected in domestic prices,
reducing the impact on relative prices. The extreme case is a small and open
economy, where all goods are tradable, and all devaluations therefore have a
complete pass-through, affecting domestic prices immediately and to the full
extent. On the opposite extreme is the case of a large and closed economy. In
this case the effect of a devaluation is limited, given the substantial influence of
the economy on world prices.
Another important point for determining if a region forms an optimal
currency area is the similarity of their inflation rates (Fleming 1971, p. 476). If the
candidate countries have different inflation rates prices are not affected
proportionately, resulting in changes in relative prices that will lead to real
appreciation (depreciation) between the currencies. This would make undesirable
for countries to form a monetary union, as then exchange rate adjustments will
not be available. In that sense, it is also important to consider how important
exchange rate variability is to each country’s government as a form of
adjustment.
Political factors have to be analyzed, since they play a fundamental role in
the integration process. According to Manzanetti (1994, p. 108) the most
18
successful integration experiences had political purposes in the centre of its
objectives, rather than only economical. Moreover, Manzenatti suggests that
poor track records of integration schemes in Latin America are due to lack of a
clearly defined political agenda.
In a general sense, we can say that a region forms an optimal currency
area if the benefits of renouncing part of its monetary autonomy are larger than
the costs. How could we evaluate the theoretical conditions for the formation of a
monetary area? Which questions should be answered in order to evaluate if
Mercosul is an optimal currency area? The analyzed literature focuses a number
of criteria between the potential members of the OCA. These criteria are mainly
focused on the costs of joining a currency area. This study will deal with the
following: 1) correlation of shocks and cycles; 2) degree of Interdependence and
trade; 3) diversification and openness 4) degree of labor mobility; 5) price and
wage flexibility; 6) similarity of inflation rates; 7) the need for real exchange rate
variability; 8) political factors; and 9) fiscal integration. The greater these links are
between the countries the closer from forming an OCA they become. In addition
to the criterions just described, we will add an analysis on endogeneity (point 10)
of the OCA, which defends the hypothesis that monetary union itself will increase
the likelihood of satisfying the criteria. We continue this section by analyzing the
OCA criterion on the Mercosul case.
1. Correlation of Shocks and Cycles
The burden of adopting a policy of exchange rate coordination will be lower
depending on the degree of symmetry of the shocks between member countries.
The higher the symmetry, the lower is the necessity of adjustment between the
relative prices, and consequently, losing monetary independence does not imply
costs. Hypothetically, if the macroeconomic policies and the productive structure
of two countries are the same, there are fewer reasons for the exchange rate
19
between them to fluctuate. Therefore, it is fundamental to know the degree of
asymmetry of the shocks, in order to estimate the costs of coordination.
Berenboim (2004, pp. 53) describes two ways to quantify the correlation of
the shocks. The first one uses the synchrony of business cycles. The second one
tries to measure to what extent the intensity of trade influences the correlation of
the shocks. The intensity of correlation of the shocks will depend on the degree
of integration between the analyzed countries. The future correlation of the
shocks of the countries will be related to the extent of the trade integration and
the trade-shocks elasticity.
Bayoumi and Eichengreen (1994) quantify the shocks using the method
proposed by Blanchard and Quah (1989). In this methodology supply shocks are
considered permanent, while demand shocks are temporary, and only supply
shocks can affect the level of production. Bayoumi and Eichengreen measured
supply and demand shocks for several regions (Latin America, NAFTA, EU, Asia
and USA) in the period 1960-1990. The study concludes that the estimated
correlation of supply shocks for European countries were positive and significant;
the same did not stand for Mercosul. Although Brazil and Argentina had a high
positive correlation, it was not significant. Machinea (2004), finds similar results,
however he analyzes the period 1962 to 2002. For the period 1990 to 2002
Machinea finds an even higher degree of correlation, but still statistically not
different from zero.
Ferrando (2000) criticized the approach used by Bayoumi and
Eichengreen. He argues that the hypothesis that demand shocks are temporary
is not very likely, since changes in terms of trade have an impact on supply and
demand. Changes in consumer preferences, especially in the amount of savings,
have an impact on demand and long run production, which will be identified as
pure supply shocks under this methodology. In addition, demand shocks
analysis is in this case imprecise, due to contaminated data by the use of the
exchange rate as a policy to increase competitiveness by Mercosul countries.
The use of the exchange rate in order to gain competitive advantages against its
neighbors produces a negative correlation of the demand shocks, which can lead
20
to mistaken conclusions, because without interference of the governments the
results could be different. However, in a monetary union, this kind of policy
cannot be used. Therefore, the estimation of Bayoumi and Eichengreen (1994) is
not appropriate to measure the adequacy of a monetary union in the case of
Mercosul, where the exchange rate has been often used in order to increase
competitiveness.
Carrera and Sturzenagger (1998) observe the correlation of the business
cycles of Brazil and Argentina. The study of the economic cycles shows
increasing correlation from 1980, with the Argentinean cycle leading the
Brazilian. That is not explained by the influence of one economy on the other, but
by the stabilization plans that were firstly undertaken in Argentina.
In order to avoid the limitations of one particular methodology, Ferrando
(2000) uses four different methodologies, to reach a conclusion, using data of the
period 1970 to 1996. His research uses two different ways of identification of
supply and demand shocks. The first one is a parametric estimation of the supply
and demand curves and the second one is the structural approach used by
Eichengreen and Bayoumi. In addition, Ferrando also uses the stationary state
correlation model and, finally, the errors model of Johansen.
The results of the four different methodologies converged. The
correlations of the shocks are positive (indicating symmetrical shocks); however
not significantly different from zero, suggesting that in the period analyzed the
average of the asymmetrical shocks was offset by the symmetrical shocks.
Symmetrical shocks such as the debt crisis (1980s), or the more recent scarcity
on foreign capital (result from the Russian crisis) were strong enough to hide the
effects of asymmetrical shocks, such as increase of the price of coffee and
petroleum. In addition, in terms of intensity of shock correlation, Mercosul
presents the weakest results, behind Europe, which has on average stronger
positive correlations, and NAFTA, which presented similarity of supply shocks
between USA and Canada (Ferrando, 2000).
The biggest problem in those regression models is related to the fact that
data is contaminated by the stabilization plans; therefore the conclusions, derived
21
by the estimations, are misleading. In this sense, it is possible to say that the only
shocks that had greater impacts on the economy than the stabilization plans
were the debt crisis and the decrease of capital flows; all other shocks had a
smaller impact. In addition, the stabilization plans were characterized by fixed
and appreciating exchange rates, creating consumption booms (as a result of
cheaper import goods and, consequently, higher real wages) in the beginning
and recession in the end.
Another fact that weakens the results obtained in most of the studies
undertaken is that before the 1990s South American countries were
characterized by being very closed economies (even after strongly reducing
trade barriers in the beginning of the 1990s, we can see in figure 11 in the
appendix that South American markets still remained closed compared to
others), which contributed to keep the shocks idiosyncratic. Open economies are
more likely affected by the same external shocks, increasing symmetry.
For the problems stated above, the results obtained in the empirical
analysis available in literature can only be used as indicative and not as strong
evidence in relation to the adequacy of a monetary union to the Mercosul. One
interesting result, which must receive some attention, is the fact that correlation
of shocks has increased after the 1990s. This result is present in Machinea
(2004) and Ferrando (2000) (studies that use more recent data). As discussed
previously, greater degree of openness influenced this result. In the beginning of
the 1990s financial and trade liberalization, probably, had a role increasing the
symmetry of the shocks. The increase of integration would raise the degree of
correlation of shocks between the member countries. This is part of the
hypothesis known as the endogeneity of OCA (to be elaborated in section V.10).
2. Trade and Interdependence
How do we measure the degree of interdependence? This is usually measured
on the basis of the trade and financial links among the countries of the target
22
group. Looking at figures 2 and 3 we can see the increase in intraregional trade
among Mercosul countries both in terms of total exports and GDP. This tendency
is only not observable in Uruguay, where variation was slightly negative in the
period observed. The increase in these ratios suggests an increase in
interdependence among Mercosul countries. Intraregional trade grew steadily up
to 1998. This trend changed abruptly since the Brazilian crisis in 1999 until the
Argentinean devaluation in 2002. Looking at figure 4 (appendix) we can see the
decrease in the share of intra-regional exports in relation to total exports in
Mercosul after 1998. This change is mainly explained by the decrease in trade
between Argentina and Brazil after the Brazilian crisis in 1998-1999 and the
Argentina crisis in 2001-2002 (figure 1 in the appendix).
After the Argentinean crisis, trade between Brazil and Argentina started to
grow again (due to the recuperation of the Argentinean economy), resulting in
increase in the share of intra-regional trade in Mercosul (figure 4). It is important
to notice that intra-regional trade is increasing despite the boom in total exports
from Mercosul. The increase in total exports from Mercosul is explained by the
boom in commodity prices starting in 2003. Higher commodity prices led to the
increase in commodity exports to countries outside Mercosul, especially China
and USA.
Figures 5 and 6 compare Mercosul integration with other regional
agreements. Indicators show that integration in Mercosul, although increasing in
the past decade, is substantially less tight then other regions (Machinea and
Rozenwurcel, 2006). This conclusion is even more evident if we analyze
intraregional trade in terms of GDP. Despite considerable trade interdependence
in terms of intraregional exports, these exports account for a relatively small
proportion of GDP. This analysis suggests that the degree of openness in the
Mercosul countries is lower than in other regional agreements.
Machinea argues that the size of the bloc is a major determinant of the
relative volume of trade within the bloc. According to this view it is unthinkable
that trade interdependence in Mercosul, which accounts for only 1.4% of the
world trade, would reach the same quantitative levels as those seen in larger
23
blocs, such as the EU, which accounts for 37.2% of world trade. This approach
cannot answer the different levels of integration between the EU and the NAFTA.
Both blocs have very high levels of intraregional exports as a share of total
exports, meaning high interdependence between the countries, however the
degree of integration between the EU countries is a lot higher then in the NAFTA,
showing that interdependence between NAFTA countries is not as high as one
would expect. In that sense, what else explains interdependence?
In Eichengreen’s (1998) view the level of interdependence of a bloc is not
a matter of size or relative size of its intraregional trade, but a matter of policy. He
argues that greater interdependence is present when policy makers are seeking
to create a truly unified regional market, not when integration stops at the border,
just limited to a creation of a free trade area or a customs union. NAFTA has
minimal tariff barriers, but there are many other restrictions to cross border trade,
such as different public procurement rules, differential access to domestic
distribution system, different degrees of government subsidization for public
enterprises and national champions. In the European Union, on the other hand,
the goal is to eliminate these hidden obstacles to cross-border competition,
making governments to rescind measures favoring their domestic producers,
empowering the European Commission, a supranational organ. According to
Eichengreen, exchange rate variability is more likely to launch a political
backlash against regional integration when policy makers are trying to create a
truly unified regional market. In a truly unified market, removal of trade tariffs in
addition to harmonization of domestic regulations will result in more open
domestic markets and more intense competition, making exchange rate changes
more harmful.
The question to be answered here is what kind of integration are Mercosul
countries looking for? From its origin, on the Assuncion treaty, Mercosul already
demonstrated an intention of having an integration that would go beyond trade.
The treaty predicted not only trade integration, but as well political, social
(reducing asymmetries) and common goals.
24
When making a parallel with the NAFTA and EU, the first one being an
example of integration until the border and the second of a deeper one, at first
glance an observer can say that Mercosul has more similarities with NAFTA,
because both are intergovernmental agreements with a trade integration
objective and with the economic dominance of one big country (USA in NAFTA
and Brazil in Mercosul). But if we look deeper at the institutional structure and the
relation between the Mercosul countries the differences become clear. The
cultural, historical, social and economical similarities between the Mercosul
countries are incomparable, even when we think of other trade agreements in the
world. Figure 7 (appendix) shows the economic leadership of Brazil in Mercosul,
comparing it with EU and NAFTA. Brazil does not dominate the Mercosul to the
same extent as USA dominates NAFTA. Exchange rate fluctuations of the
Mexican peso and Canadian dollar have a small effect on the United States, due
to the relatively small size of the Canadian and Mexican economies compared to
the US. In Brazil’s case, its Mercosul partners are large enough to have an
impact on its economy (Eichengreen 1998). In addition, the number of members
in Mercosul is expected to grow, reducing Brazilian leadership. Venezuela just
moved out of CAN (Andean Community) and joined as a full member of Mercosul
in 2006, Bolivia already stated its intention to join (Mercosul has a working group
studying this possibility), Equator’s president has also declared its intention to
join in the short run; and Mercosul and CAN members (Colombia, Peru, Bolivia
and Equador) are also members of Unasul (Union of South American Nations),
which intends to unite both free trade agreements.
As discussed previously, Mercosul is today changing its institutional
structure. It is changing from an intergovernmental agreement to a supranational
institution. This step towards supranationalism and relation proximity between the
countries gives us strong signs that interdependence between Mercosul
countries is expected to grow stronger (the endogenous criteria argument;
section V.10).
25
3. Degree of Diversification and Openness
Keren (1969) emphasized that less diverse economies are more likely to
be struck by asymmetrical shocks than others. In that sense, it is important to
analyze the degree of diversification in the Mercosul economies. According to
Barenboim (2004) more developed economies are usually more diversified, while
more closed economies are more diversified than open ones. Barenboim’s study
uses data from 28 industries in order to develop a Herfindahl index. The higher
the index the lower is the diversification of the country’s production. It is possible
to infer from figure 8 (appendix) that Brazilian economy is very diversified, to the
same level as developed economies, while Argentina has become more
specialized throughout the years (the Argentinean surprisingly fast increase in
specialization was mainly due to a combination of rapidly increasing trade
openness with a Industrial Specialization Regime – ISR - established by the
government with the main objective of promoting export specialization by
industrial firms). Despite having data only until 1989, figure 8 gives us a good
insight on diversification in Mercosul (building a Herfindahl index for more recent
years was out of the scope of this study). Figures 9 and 10 (appendix) give us an
idea of how diversification evolved on Mercosul countries in more recent years.
Brazil’s high-technology exports as a share of manufactured exports increased
considerably (while manufactures exports remain the same), suggesting that
Brazilian diversification increased, because despite being mainly a commodity
exporter (around 50% of exports) its share of high-technology exports increased
significantly. The other Mercosul countries seem to have a stable degree of
diversification in the past decade.
Regarding the degree of openness, the Mercosul economies still remain
very closed. A common measure of openness is the total trade as a share of
GDP, which is the one used in figure 11 in the appendix. Mercosul countries had
a great improve on this ratio in the period 1990 – 2005: Brazil went from 11% to
22%; Argentina went from 12% to 38%; Paraguay went from 44% to 72% and
Uruguay went from 33% to 43%. However, figure 11 shows us how closed
26
Mercosul economies are in comparison to other countries. Argentina and Brazil
have a degree of openness close to the one of USA and Mercosul’s “small”
economies (Paraguay and Uruguay), which we expected to be very open, are still
far behind other small countries such as Netherlands and Belgium. We can
conclude that Mercosul economies are not yet open enough in order to join in a
currency area.
4. Degree of Labor Mobility
According to Mundel’s OCA theory (1961), a high degree of labor mobility can
facilitate adjustment, reducing the importance of exchange rate adjustments. A
demand shift from Brazil to Argentina, results in an increase demand for labor in
Argentina.
It is rather difficult to quantify labor mobility in Mercosul. Despite having
data concerning how many citizens moved from one country to another within
certain period, detailed information about labor skills, motives, etc. are not
available. However, it is possible to conclude from the high degree of
unemployment in these countries (figure 12 in the appendix) that labor market is
rather not flexible.
Policy harmonization and labor standards are required in order to promote
higher degree of labor mobility. Mercosul’s labor market is characterized by
national specificities: different labor rights, labor costs and labor policies across
Mercosul countries. These asymmetries in the labor market can be observed in
some important indicators, such as the Urban Unemployment Rate (figure 12 in
the appendix) and Average Real Income (figure 13 in the appendix). Other
indicators, such as non-wage labor costs still differ a lot: In Brazil it represents
81.4% and in Uruguay 46.9% of the costs of labor. We can infer from the
different numbers presented above that labor rules on Mercosul are not
compatible. Several formal and informal labor rules are not compatible: overnight
27
work, work shift, extra hours, unions, bargaining structure, firing costs, strikes,
maternity leave, etc.
Due to those specificities the implementation of free labor mobility has a
potential impact on wage determination, social security and unemployment. For
coordination of labor policies and integration of labor markets, some
compatibilization of labor rights, labor institutions and social security is required.
In order to achieve this objective Mercosul created labor institutions to coordinate
that process.
The main labor institutions of Mercosul are the SGT10 and the Social-
Labor Commission. The first one is responsible for all labor relations,
employment, migration, training, health and social security matters. The second
one is concerned with promotion of rights and the implementation of the “Social-
Labor declaration of Mercosul”. This declaration establishes that active labor
policy is a major aim: job creation programs, safety net, training programs, etc.
Recommendations of the Social-Labor declaration on labor protection and job
programs have been generally implemented by countries.
Significant recent initiatives to accomplish harmonization and mutual
recognition have been undertaken. Many adjustments have already been made
regarding regulation of labor on Mercosul countries: 1) provision on the
recognition of professionals through the development of mutually acceptable
criteria to determine the equivalence of licenses, certifications professional
degrees, and accreditations granted by other member countries; 2) protocols on
recognition of certificates, degrees and studies that cover all levels of education;
3) plans for adopting basic and technical syllabi.
Although labor mobility is not high among Mercosul countries, many
initiatives have been undertaken to improve it. Labor ministries of Argentina and
Brazil declared on April 2004 that harmonization of labor policies and integration
of labor markets are Mercosul’s major aim. Language and culture are other
factors that will facilitate this goal. Mercosul, differently from the EU, has only two
languages (Portuguese and Spanish), which are very similar, reducing migration
barriers. Besides, the cultural proximity also facilitates relations between people
28
(this observation is also valid for the other Latin American countries, which are
potential Mercosul members). For those reasons we expect that the degree of
labor mobility will increase, with potential to become higher than in the EU
(endogeneity criteria).
5. Price and Wage Flexibility
Prices and wages changes are an important form of adjustment between
countries in the absence of exchange rate adjustments. If prices and wages are
flexible, demand will be driven to higher unemployment regions (lower prices and
wages), smoothing the business cycles and shocks.
Wage flexibility is particularly important in that form of adjustment.
Mercosul countries have a rather inflexible labor market (as discussed in the
previous section). This characteristic is mostly due to strong unions and old labor
regulations. Strong unions increase wage bargaining power of workers, reducing
flexibility of wages. In addition, legislation emphasizes very strongly in protecting
the workers by employment protection mechanisms such as unemployment
insurance system and minimum wage provision. Mercosul members recently
approved changes in their labor regulations in order to increase flexibility.
Argentina introduced substantive labor rule changes, while the other countries
experienced minor changes.
However, when analyzing the labor market in Mercosul countries it is
important to look at the informal sector, which accounts for a large share of
workers in this region. For instance in Brazil informal labor market accounted for
29.1% of the workers in July 2007 (IBGE). For the other Mercosul countries this
number is of around 20% of the workers, which is lower, but still very significant.
If we take in to account the informal sector, wage flexibility in Mercosul is a lot
higher, as labor legislation do not apply for more than 20% of the workers.
29
6. Similarity of Inflation Rates
For the exchange rate coordination to be successful, member countries’
monetary authorities ought to have credibility, with a reputation of pursuing stable
inflationary policies, otherwise the time inconsistency problem will appear. Time
inconsistency problem refers to a central bank, which pursues a particular
inflationary rule; in the future it might be suboptimal to follow the rule, leading
central bankers to change their policy. As economic agents assume that the rule
will be modified (due to lack of credibility), changes will have no effect on
increasing output, only inflation.
South American central banks were for a long time known as non-credible.
However, after stabilization plans, especially in Argentina (beginning of the
1990s) and Brazil (mid 1990s), this characteristic seems to be changing. Both
countries have been working on increasing autonomy of their central banks,
which have as their main objective a stable monetary policy.
Looking at figure 14 in the appendix, we can observe how Mercosul
member countries’ inflation reduced in the 1990s and seems to be converging to
lower and more stable levels. Argentina had an expected strong increase in its
inflation rate after the end of the currency board in 2002. Inflation remains at high
levels due to high economic activity, which can be explained by successful
policies and recovering from the 2001 crises. However, high levels of inflation are
expected to be reduced, as economic growth is expected to drop in the coming
years, bringing inflation levels closer together. Although inflation levels are not
very close yet, increasing central banks’ autonomy and, therefore, credibility, was
a great step in favor of adopting exchange rate coordination. One country would
be less likely to join in a currency area with a country which central bank lacks
credibility, because there is a possibility that this central bank will create surprise
inflation, in order to reduce unemployment, causing at the same time a real
appreciation of the country’s currency.
30
7. The Need for Real Exchange Rate Variability
As seen previously, Mercosul countries are characterized by having high
variability of nominal exchange rates. This factor can be seen either as a reason
for adopting exchange rate coordination, or as a barrier. Higher exchange rate
variability can imply higher benefits from stabilization, because elimination of a
very volatile exchange rate would have a higher positive impact on trade and
investment. However, if real exchange rate variability has been high, it shows
that countries needed to use real exchange rate changes in order to adjust
relative prices, implying that loosing the exchange rate mechanism (by
coordinating nominal exchange rate) would be more costly.
Ferrando (2000) and Eichengreen (1998) analyzed the variability of the
exchange rate among Mercosul countries comparing it with other regions. Both
considered the higher exchange rate variability found between Mercosul
countries only as a benefit. Eichengreen (1998), analyzed if the level of
exchange rate variability was economically acceptable for Mercosul by
comparing the exchange rate variability between Mercosul, Europe and Asia. He
created a model which explains exchange rate variation by a number of factors
derived from the Optimal Currency Area theory, such as asymmetric output
disturbances, differences in the composition of production and trade between
countries, trade linkages and size. Ferrando (2000) compares exchange rate
variability between EU, NAFTA and Mercosul. As we can see in table 2
(appendix), variability within the Mercosul is higher, while NAFTA presents
similar results as the EU.
The exchange rate was widely used in the Mercosul countries for
adjustments during the high inflation period. At that time the impact of a currency
devaluation on the inflation rate was less important, because there was no
commitment of keeping inflation at low levels. After the stabilization plans in the
1990s, exchange rates in Argentina and Brazil were pegged to the dollar,
reducing exchange rate variability. The Brazilian and Argentinean crises led
31
again to an increase in variability, due to the Brazilian Real and the Argentinean
peso devaluations in 1999 and 2002.
However, we expect exchange rate adjustments to become less common.
Increasing concern with maintaining low inflation levels will make governments
avoid using the exchange rate as an adjustment instrument. In addition, as seen
previously, Mercosul countries have been opening their economies since the
1990s and greater degree of openness reduces the effectiveness of exchange
rate adjustments on relative prices. Further, increase in trade between Mercosul
countries, will also make exchange rate adjustments less attractive to these
economies (endogeneity).
In that sense, the need for exchange rate variability should be reduced,
increasing the importance of other adjustment mechanisms, and, most probably,
lowering variability levels, closer to the ones presented by the EU and NAFTA.
8. Political Factors
Tavlas (1993) points out that a successful currency area is more likely to occur in
the presence of a dominant country, which will use its influence to lead the
integration process, or in the presence of a strong institutional network, strong
enough to make the loss of sovereignty tolerable to each of the member
countries.
For instance, in the European case, the German Bundesbank (central
bank) assumed the leadership of the monetary unification. Germany represents
the largest economy of Europe and had power enough to intervene decisively
over the intra-regional exchange rate parity if needed. Analyzing the Mercosul
case, the biggest economy is Brazil. However, the Brazilian central bank does
not have the same strength and reputation as its German counterpart. In
addition, it does not seem that any of the Mercosul countries is willing to carry the
burden of being the leader of the integration process.
32
Mercosul countries seem to be opting for the second approach proposed
by Tavlas, a strong institutional network. Mercosul has been strengthening its
institutions as well as creating new ones, in order to move from an
intergovernmental to a supranational system. Despite recent advances, there are
still some political barriers to be transposed. Loss of sovereignty is always a
delicate issue in monetary unions. This problem is even more latent in Mercosul
countries, which are culturally adapted to strong leaders and weak institutions.
Another barrier towards supranationalism concerns differences in size of
the member countries. How are countries going to be represented in the
institutions? Brazil has more than 50% of GDP and population of the Mercosul;
however none of its partners would accept Brazil to have more than 50% of the
representatives. Paraguay and Uruguay are very small compared to its other
Mercosul partners; therefore they are afraid to have few representatives and that
all the decisions will be taken by Argentina and Brazil.
Stronger institutions, in order to make loss o sovereignty tolerable, make
supranationalism fundamental for the evolution of the Mercosul block. Although
many improvements have already been made there is yet a lot to be
accomplished and the advance of the integration in Mercosul will depend on how
long it will take to reach these objectives.
9. Fiscal Integration
Fiscal integration has the role of smoothing asymmetrical shocks trough fiscal
transfers between the member countries. It requires a degree of political
integration higher than the one reached by Mercosul so far. At this moment fiscal
transfers between Mercosul countries do not exist. Countries will accept to have
a join fiscal policy with fiscal transfers most probably when a certain degree of
political integration has been achieved.
If we again take the example of the European Union, there are only fiscal
transfers through the Common Agricultural Policy and the European Structural
33
Funds. The EU only created a commission for fiscal harmonization only after the
monetary union was established.
In addition, differences in size between Mercosul countries will be a barrier
to reach fiscal integration. As seen in figure 7 (appendix), Brazil represents more
than 60% of Mercosul’s GDP. Therefore, it will be difficult for the other Mercosul
countries to make transfers, which will be large enough to have a positive impact
in Brazil, in case of high unemployment in this country. This disparity is expected
to reduce with the expected increase in the number of members of Mercosul
(discussed in point 2 in this section).
10. Endogeneity
Many economists defend that using past data to analyze the OCA criteria can
drive to misleading conclusions, due to the endogeneity of each particular
criterion. The hypothesis is that monetary unification itself may lead to an
increase in linkages and hence more shocks symmetry. A country would be more
likely to satisfy the criteria for entering a monetary union ex post than ex ante
(Frenkel and Rose, 1997).
The European Commission defends that the higher the degree of
integration the lower the possibility of asymmetric shocks, therefore income
between member countries will tend to differ less. According to Frenkel and Rose
(1997) there is a theoretical ambiguity. Reduced trade barriers will probably
result in production specialization by a country, increasing the possibility of
asymmetrical shocks, due to the presence of industry-specific shocks. On the
other hand, increased integration may result in more symmetry of shocks, due to
common demand shocks and intra-industry trade.
Frenkel and Rose (1997) examine empirically the impact of the intensity of
trade on the correlation of the shocks. Their hypothesis was that after countries
join a monetary union, they experience dramatically different business cycles
than before. They analyzed this hypothesis empirically using a panel of 30 years
34
of 20 industrialized countries. The authors found a strong positive relation
between trade intensity and correlation of business cycles, meaning that greater
degree of integration has resulted in more synchronized business cycles.
Ferrando (2000) analyzed the effects of regional integration on the
symmetry of the shocks in the case of Mercosul. In order to do this he divided the
data in two periods: 1975-1989 and 1990-1996; and compared them. This study
shows that the second period had a stronger positive correlation of shocks than
the first one. It is difficult to establish to which degree the increase of correlation
of the shocks is a result of deeper integration. However, the results obtained let
us conclude that there is a tendency of increasing similarity of the shocks that
have an impact on the Mercosul countries. In this sense, in order to determine if
Mercosul forms an OCA, we have to analyze the degree of integration between
the members and its capacity to integrate in the future. From the analysis of the
criteria done previously we could see that Mercosul countries have been
improving in many important points and, therefore, similarity of shocks is
expected to increase more.
VI – First Attempt of Coordination in Mercosul: lessons for the future?
In 2000 an attempt of macroeconomic coordination was undertaken. Although
not successful, mainly due to the 2001 Argentinean crisis, it might be interesting
to analyze it.
Ministers of finance and central bank presidents of the Mercosul members
agreed in a meeting in the beginning of that year to set common targets of fiscal
deficit, public debt and inflation, start measuring and publishing economic
indicators based on a common methodology and establishing a macroeconomic
monitoring group. Harmonizing statistics for fiscal deficit, public debt and prices
were achieved during that year. In December 2000 another meeting was held to
set targets for fiscal deficit, public debt and inflation to be met starting in 2002.
The inflation rate established was a CPI of 5% until 2005 and 4% from then on.
35
Concerning fiscal policy, the maximum consolidated deficit was set at 3% and
debt/GDP ratio should decrease, reaching a 40% limit at 2010. Other goals, such
as current account and short-term external debt, were left to be set later.
The governments of the Mercosul countries tried to agree on a system of
incentives for compliance of the goals established. The only “penalty” agreed for
deviating from the targets was to report to the macroeconomic monitoring group,
informing which measures would be taken in order to carry out the targets.
Although these goals were unattainable, due to later events already
mentioned, Machinea (2006) points out three important lessons learned by this
experience. Firstly, this experience demonstrated that it is possible to quickly
achieve coordination goals, especially when the highest political decision-making
levels are involved and capable technical bodies exist. Secondly, despite the
high levels of inflation and fiscal deficit targeted, the levels selected represented
an effort for all the countries of the region. This choice left room for more
demanding criteria in the future, otherwise goals will be considered too difficult to
achieve. This preoccupation shows that macroeconomic responsibility has begun
to gain more importance in this region. Thirdly, concerning the penalties required
to ensure compliance of the goals, Machinea highlights that the lack of
mechanisms to secure compliance is a problem for the effectiveness of the
system.
This first attempt of coordination failed. The collapse of the Argentinean
economy and the worsening of the situation in the whole Mercosul led to new
tariff restrictions, an increase in exceptions for the common external tariff and
many trade disputes among member countries. By the result of this experience of
coordination of macroeconomic policy, Machinea (2006) concludes that
excessive instability and high exchange rate volatility, especially in one of the
larger countries, make macroeconomical coordination unworkable. We can infer
from Machinea that in order to coordinate macroeconomic policy, some kind of
exchange rate coordination should be present. In the next section we will discuss
what type of coordination is more suitable for the region in the future.
36
VII – Future: Single Currency?
After analyzing Mercosul, its recent history, costs and benefits of exchange rate
coordination, it is rather difficult to establish a direct answer whether coordination
is suited for Mercosul or not. Although not conclusive, the OCA analysis gives us
some theoretical support, which leads us to some important conclusions.
Mercosul economies have become relatively more stable in the past few
years. Member countries managed to keep inflation (figure 14 in the appendix)
and public deficit relatively under control (greatest problems of the 1980s) and,
after the crisis in the last decade, were able to keep positive flows on the current
account (figure 15 in the appendix), stabilizing their balance of payments. In the
second half of the 1990s, exogenous shocks prevailed.
Further, Mercosul has a lot to advance in order to fulfill the full set of OCA
criteria. Advances in integration, regulation and legislation are required.
However, we cannot deny that the endogeneity of the OCA criteria started to play
a role during the 1990s, and continues to do so. It seems that a historic approach
will be misleading in recently stabilized economies compared to economies that
are characterized by a less volatile environment. In other words, the historical
analysis looses credibility in countries that were very unstable a few years ago,
because trade integration was very weak. In addition, the higher amount of
shocks that affected the countries increased the probability of asymmetrical
ones.
As seen in section IV, reducing exchange rate variability would be very
important for Mercosul. If achieved, it would have a positive effect on trade and
integration, increasing intra-regional trade, stability and cohesion on the region.
There are enough elements to defend that some level of coordination is needed
in Mercosul, especially if we assume that member countries are aiming a
strategy of strengthening Mercosul. Machinea (2006) discusses two forms of
coordination: “soft” coordination and “hard” coordination. Hard coordination would
be represented by the EU, meaning the creation of a Monetary Union. Soft
coordination can be defined as what Lavagna and Gianbiagi (1998) called
37
coordinated convergence, meaning that Mercosul countries should define
common macroeconomical targets, which was their suggestion for the region
(unsuccessfully implemented in 2000, as seen in section VI).
Tavlas (1993) points out that there are various levels of integration on the
monetary side: Exchange rate unions, pseudo exchange rate unions, monetary
integration and monetary unification. In exchange rate unions, the exchange rate
between the union countries is irrevocably fixed with no margins of fluctuation.
There is no need for coordination of monetary policies, so some form of capital
control is needed. Pseudo exchange rate unions are characterized by fixed
exchange rate between the member countries, free capital movements and,
consequently, policy coordination, but no formal integration of monetary policies.
Monetary integration requires an irrevocably fixed exchange rate, full and
irreversible convertibility of currencies, financial market integration, common
monetary policy and liberalization of movements on current transactions.
Monetary unification is a monetary integration with a single currency and a
common central bank.
Supranational institutions would represent an essential ingredient for
successfully creating a monetary union. One of the crucial questions for Mercosul
to become a monetary union will be whether the Mercosul countries will cede
sovereign rights to supranational Mercosul entities. This change is vital, because
if the Mercosul countries engage in such a deep level of integration they cannot
be dependent on the risk of their leader’s relations go sour, which could easily
stop the evolution of Mercosul. In addition, supranationalism is necessary in
order to create institutions, which have power of enforcement, indispensable for
the success of a monetary union (Lavagna and Giambiagi, 1998). Some steps in
that direction have already been made, especially through the creation of the
Mercosul Parliament and the Permanent Tribunal of Revision, but there is a lot
still to be done in order to reach supranationalism. In addition, Mercosul has a lot
to advance on harmonization of financial legislation, institutional structure and
increase of labor market mobility. Although Mercosul countries are moving in
direction to deeper integration, we have to consider that these changes cannot
38
be made very easily and will still take some time and effort to be accomplished,
especially due to the gradual approach of evolution of Mercosul.
Considering the reasons mentioned above, “hard” coordination does not
make sense for Mercosul at this moment. Even the most ardent proponents of
this option see it as something that Mercosul countries can achieve in one or two
decades. Eichengreen (1998), points out that over a longer time span like 15
years, these countries’ characteristics are likely to change in response to the
exchange rate regime chosen today. This means that the option of a monetary
union cannot be labeled as impossible and excluded as an option for the future.
However, Lavagna and Gianbiagi mention a risk in not engaging in any
coordination. In case of no coordination, future exchange rate swings could lead
to another integration retreat, or even lead to the end of the integration process,
especially because Mercosul is still intergovernmental, making exit barriers
lower. In that sense, a softer coordination is more suitable for Mercosul. Let us
analyze the other levels of monetary coordination proposed by Tavlas
enumerated previously.
Monetary Integration requires an irrevocably fixed exchange rate, with no
margins of fluctuation. This arrangement would not leave space for any
exchange rate adjustments. It seems not to be the best option for Mercosul,
which still has to advance on increasing labor mobility and trade integration.
Besides, an irrevocably fixed exchange rate will only be credible if the countries
would defend it at any cost. This, as in a monetary union, would need institutions
with power of enforcement, not yet available in Mercosul. Moreover, we have to
consider that this form of integration needs a degree legislation harmonization,
not yet reached by Mercosul countries.
Forming an exchange rate union does not seem attractive for Mercosul.
This option also requires an irrevocably fixed exchange rate, which is not suitable
for Mercosul due to the same reasons stated in the case of a monetary
integration. In addition, exchange rate union requires capital controls, which
would represent a big change for the Mercosul economies, characterized by a
39
high degree of capital mobility. Therefore, it would represent a very controversial
measure and politically difficult to be implemented.
A pseudo exchange rate union seems to be the best alternative for
Mercosul. It has many elements of the coordination intended in 2000. Involves
policy coordination, but with no formal integration of monetary policy. Obviously,
establishing macroeconomic coordination mechanisms would be also necessary;
otherwise targets chosen will not be feasible in the long run. However, differently
from the other options of coordination, formal integration is not required and rules
could become stricter gradually, while institutions could have time to have their
powers increased. However, improvements towards empowering Mercosul
institutions are fundamental for deepening the integration and increasing
credibility, therefore empowering institutions should be simultaneously pursued.
Since this option does not involve capital controls, an irrevocably fixed
exchange rate is not possible (due to incompatible trinity: fixed exchange rate,
capital mobility and monetary independence). Exchange rates will fluctuate,
however variability will be lower than today. According to Machinea (2006) when
countries adopt a floating exchange rate system and steer monetary policy
towards a particular inflation rate (inflation targeting) variability of the exchange
rate is lower. In addition, policy differences among the countries are responsible
for much of the lack of synchronization in their cycles, so macroeconomic
coordination will not only reduce instability, but also eliminate one mechanism
that amplifies external shocks.
One important issue when establishing a macroeconomic coordination
mechanism will be if the targets set are feasible and if compliance can be
enforced. In that sense, it is important to have a good balance on how strict rules
should be, while enforcement of those rules should be linked to incentives. Too
strict rules could make them impossible to follow or that compliance would be
very costly. Therefore, some degree of flexibility should be present. However, too
much flexibility could lead to loss of credibility.
The main difference in relation to the attempt of coordination of 2000 is the
improvement of the situation of Mercosul economies nowadays. After the
40
collapse of the Argentinean currency board, Argentina and Uruguay managed to
stabilize their economies, adopting a flexible exchange rate regime, similar to the
one that was already being used in Brazil. Establishing goals for specific
macroeconomic variables has to be made in common agreement of member
countries, which becomes a lot easier when these countries already operate with
the same exchange rate regimes.
We expect that the increase in coordination will deepen the integration by
boosting the endogeneity effect of optimal currency areas. In that case, we
expect that “harder” forms of coordination will be able to be implemented and
maybe in the future a monetary union could take place.
VIII – Conclusion
Mercosul has been dealing with many problems derived from not reaching its
main objectives: form a complete free trade area and customs union, linked by
common tariffs and rules. These problems are not only a result of the
weaknesses of the block and its institutional fragility. Macroeconomical instability,
derived from the Brazilian and Argentinean crises, played an important role in
stopping Mercosul of fulfilling its goals. The Mercosul nearly collapsed after the
crises, due to the increase of trade restrictions between its member countries.
However, Mercosul economies managed to recover. Brazil was able to
stabilize its economy relatively fast after the 1999 devaluation. Argentina, on the
other hand, was in a worse situation, due to its currency board. However, in
2003, after a great recession, the economy started to recover. Intra trade started
to grow stronger again and, as a result, Mercosul started to advance towards
integration. Trade restrictions were reduced, new members were incorporated
(one full member and two associates) and new important institutions were
created.
Despite the advances of Mercosul, exchange rate instability is still present.
For instance, during the last four years the Brazilian Real has been appreciating
41
against the Argentinean peso. This movement could lead to another step back on
Mercosul’s integration process and, in that way, a worsening of the situation in
the member countries economies. In order to reduce instability and increase
integration, this study analyzes the possibility of reducing exchange rate
variability through coordination of policies.
For that analysis the Optimal Currency Area Theory was chosen as the
instrument. The analysis of traditional OCA theory tells us that Mercosul
countries do not fulfill many important criteria. The analysis showed a weak
degree of correlation of shocks and cycles, weak trade linkages, too closed
economies, low degree of labor mobility and absence of fiscal integration as main
weaknesses of the Mercosul in order to form an optimal currency area. We can
conclude that Mercosul integration still has to advance much in order to fulfill
many of the criteria. However, this conclusion does not give enough elements to
conclusively determine if Mercosul should coordinate policies or not. The single
application of these criteria is of limited value. The endogeneity argument plays
an important role in the case of Mercosul, due to the historic of instability in its
member countries economies. Until the beginning of the 1990s, shocks on the
Mercosul countries were highly asymmetrical. However, since the second half of
that decade, symmetrical shocks gained importance. In fact, some degree of
exchange rate coordination would increase integration and, consequently, move
Mercosul closer to form a theoretical optimal currency area. Complete exchange
rate coordination is not suitable for Mercosul at this moment: costs of monetary
union will be too high (did not fulfill many of the OCA criteria) and lack of
institutional development of Mercosul (lack of supranationality).
In addition, the risk of exchange rate variability inflicting a retreat on the
integration process, the benefits of reducing exchange rate variability and the
increase of intra-trade let us conclude that a soft form of coordination would be
beneficial. Analyzing the possible coordination mechanisms, we conclude that
coordination of macroeconomical targets, without formal integration of monetary
policy, similar to the one attempted in 2000, would be the best option today.
42
Finally, adopting coordination of macroeconomic targets would reduce
exchange rate variability, increasing Mercosul’s trade interdependence. If during
this period reforms in order to harmonize regulation and create supranational
institutions are made, the need and possibility of adopting a harder form of
coordination could be able to grow. If these advances are undertaken, it could
lead in the future to the creation of a monetary union.
43
References:
Banco Central del Paraguay (BCP). Informe Econômico Preliminar 2007.
www.bcp.gov.py
Banco Central del Uruguay (BCU). Informe de Poliítica Monetária 2006 - 2007.
www.bcu.gub.uy
Barenboim, I. (2004) União Monetária para Brasil e Argentina: Uma Análise
Empírica. INTAL/Banco Interamericano de Desarrollo. La coordinación
macroeconómica y la cooperación monetaria, sus costos, beneficios y aplicabilidad
en acuerdos regionales de integración. pp. 40-75
Bayoumi, T. (1992) A Formal Model of Optimal Currency Areas. International
Monetary Fund Staff Papers, v. 41, n.4, pp.537-554.
Bayoumi, T. and Eichengreen, B. (1994) One Money or Many? Analyzing the
Prospects for Monetary Unification in Various Parts of the World. Princeton Studies
in International Finance, 76.
ECLAC (Economic Commission for Latin America and the Caribbean). www.eclac.cl
(access July/2007)
Eichengreen, B. (1998) Does Mercosur Need a Single Currency?. NBER Working
Paper No. 6821
Fanelli, J. M. (2003) Monetary Regimes and Macroeconomic Policy Coordination in
Mercosur. CEDES Buenos Aires, Argentina
Ferrando, G. L. (2000) Un Área Monetária para el Mercosur? IN: Carrera, Jorge;
Sturzenegger, Federico. Coordinación de Políticas Macroeconômicas en el
Mercosur. Buenos Aires, Argentina: Fondo de Cultura Econômica. pp. 183-222.
Fleming, J. (1971) On Exchange Rate Unification. Economic Journal, v. 81, n. 323,
pp. 467-488.
Frankel J. and Rose A. (1998) The Endogenity of the Optimum Currency Area
Criteria. The Economic Journal , v. 108, n. 449, pp. 1009–1025.
FUNCEX site (Fundação Centro de Estudos do Comércio Exterior)
www.funcex.com.br (access July/2007)
Giambiagi F. and Barenboim I. (2005) Mercosul: Por uma Nova Estratégia Brasileira.
Revista do BNDES (Banco Nacional de Desenvolvimento Econômico e Social), v.
12, n. 24, p. 77-110
Grasel, D. (2005) Brasil: Plano Real e a Estabilização Econômica
Inacabada.;Informe Gepec, v. 9, p. 73 - 89
44
De Grauwe, P. (2005) Economics of Monetary Union. 6th edition, New York: Oxford
University Press.
De Grauwe, P. and Vanhaverbeke, W. (1993) Is Europe an Optimum Currency
Area? Evidence form Regional Data. In: P.R. Masson and M.P. Taylor (eds). Policy
Issues in the Operation of Currency Unions, Cambridge: Cambridge University
Press.
International Monetary Fund (2007) World Economical Outlook Database.
www.imf.org
Instituto Brasileiro de Geografia e Estatistica (IBGE). www.ibge.gov.br
Instituto Nacional de Estadisticas y Censos (INDEC). www.indec.gov.ar
Instituto de Pesquisa Econômica Aplicada (IPEA). www.ipea.gov.br
Kenen, P.B. (1969) The Theory of Optimal Currency Areas: An Ecletic View. In
Mundell and Swoboda, eds. Monetary Problems in the International Economy,
Chicago: University of Chicago Press.
Lavagna, R. and Giambiagi F., (1998), Hacia la Creación de una Moneda Común.
Uma Propuesta de Convergência Coordinada de Políticas Macroeconômicas em el
Mercosur.
Manzetti, L. (1994), The Political Economy of Mercosur. Jornal of Interamerican
Studies and World Affairs, v.35, n. 4, pp. 101-141.
Machinea, J. L.(2002) Mercosur: In Search of a New Agenda. Exchange Rate
Instability in Mercosur: Causes Problems and Possible Solutions. Inter American
Development Bank. Buenos Aires,
Machinea, J. and Rozenwurcel, G. (2006), Macroeconomic Coordination in Latin
America: Does it have a Future?. UN Economic Commission for Latin America and
the Caribbean. Santiago, Chile.
McKinnon, R. (1963) Optimal Currency Areas. American Economic Review, v. 53, n.
4, pp. 717-24.
Mercosul Site. www.mercosur.int (access May/2007)
Mundell, R. (1961), A Theory of Optimum Currency Areas. American Economic
Review, v. 51, n. 4, pp. 657-664.
Tavlas, G. S. (1993) The “New” Theory of Optimal Currency Areas, World Economy,
v. 16, n. 6, pp. 663-685.