Latin American countries have lost competitiveness in world markets in comparison to China over the last two decades. The main purpose of this study is to examine the causes of this development. To this end an augmented Dornbusch-type Ricardian model is estimated using panel data. The explanatory variables considered are productivity, unit labor costs, unit values, trade costs, price levels, and real exchange rates; all variables are evaluated in relative terms. Due to data restrictions, China's relative exports (to the US, Argentina, Japan, Korea, the UK, Germany, and Spain) will be compared to Mexico's exports for a number of sectors over a limited period of eleven years. Panel and pooled estimation techniques (SUR estimation, panel Feasible Generalized Least Squares (panel/pooled FGLS)) will be utilized to better control for country-specific effects and correlation over time. A simulation underlines the positive impact of relative real exchange rate advantages on relative exports for the textile sector. Standardized ß-coefficients identify relative real exchange rates, relative cost levels, and relative unit values as the drivers of competitive advantage in the textile sector.
