Monetary union in Latin America: an assessment in the context of optimum currency area

ABSTRACT

The article applies the optimum currency area (OCA) theory to Latin America to assess the potential of a monetary union in Latin America and in its major existing regional trade agreements (RTAs). According to OCA criteria we find that Latin America is far from being an optimum currency area, as its countries’ exposure to asymmetric shocks is high and their capacities to adjust in response to macroeconomic disturbances are limited. Using a panel of 20 Latin American countries from 1990 to 2014, we apply the dynamic OLS estimation techniques to estimate the costs and benefits of a potential monetary union in Latin America and in its various RTAs. to estimate the costs and benefits of a potential monetary union in Latin America and in its various RTAs. We find that the costs are high, because Latin America’s economies are vulnerable to severe macroeconomic disturbances and its RTAs differ significantly in their response to negative demand shocks. Most of the monetary efficiency gains are shown to be the result of a common restrictive monetary policy which would result in higher FDI inflows and, to a more limited extent, increased GDP, both overall and per capita. Although Central American countries are shown to be most suitable for further monetary integration, we conclude that Latin American countries should head first towards greater economic and political integration.

KEYWORDS:

• Optimum currency area
• Economic Integration
• regional trade agreements
• Latin America

JEL CLASSIFICATION:

• C23
• E50
• F15
• O11
• O54

Acknowledgements

The authors thank three reviewers whose comments have improved the quality of this study. We are grateful to Jennifer Jager for her efforts on a previous part of the OCA theory. We are, of course, responsible for all remaining errors.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ According to the IMF (2018), the general government gross debt ratio as a percentage of GDP add up to 86.6% for the Euro area and 60.5% for Latin America and the Caribbean in 2017, while their compound annual growth rates are −0.63% and 4.65% for the period 2012–2017, respectively.

² Latin America and the Caribbean has received a 3.6% share of inward FDI to GDP compared to a global average share of 2.6% in 2016. Although Europe, as a block, accounts for 53% of its total inward FDI, the United States is the leading (single) investor with an average share of 20% across the region, where U.S. firms invest on average more in Mexico, Central America and the Caribbean than in South America. Given that inward FDI remains the most stable component of foreign capital flows (i.e. more than 70%), both the absolute importance of inward FDI and the predominance of U.S. investments justify the choice of FDI and its division to assess costs and benefits of potential monetary union in Latin America. (Data source: CEPAL (2017a)).

³ Table A.1 in the appendix provides descriptive statistics for Latin American countries and RTAs.

⁴ Baltagi (2008) provides an excellent overview of issues in non-stationary panels and their different testing procedures for unit roots and cointegration.

⁵ Detailed estimates of the costs and benefits across RTA memberships for both sub-periods are available from the authors on request.

⁶ Stripping out inflation rates above 10% annually reduces the number of observations by more than third (i.e. 159 out of 425). In particular, Argentina, Brazil, Nicaragua, Peru and Uruguay had inflations rates above 100% in the early 90s (some of them even higher than 1000%). If we had chosen to eliminate inflation rates above 5% annually, the number of observations would have decreased by more than 300. Alternately, stripping out inflation rates above 20% annually would have reduced the number of observations by almost 80. However, a restrictive monetary policy by a common central bank, if successful, should lead to credibility gains and therefore to moderate inflation rates, ruling out inflation rates above 20% annually. So the reason for this setup was partly data restriction and partly the credibility gains argument from Alesina, Barro, and Tenreyro (2003).