The behavior of the real exchange rate, measuring movements in the relative consumer price indexes between countries, remains a prominent puzzle in international macroeconomics. Two key theories of the real exchange rate differ in the role played by goods not traded internationally. On one hand, the theory of Balassa-Samuelson, on the other hand, models with sticky prices. This study provides new empirical evidence on nontraded goods importance in real exchange volatility by using more highly disaggregated data than used in previous literature on prices and trade between the U.S. and Mexico for the period 2002-2009. The main results suggest that the nontraded component accounts for between 69 and up to 84 percent of the real exchange rate volatility. In addition, the results show that the nontraded component is negatively correlated with the traded component despite both countries being in a flexible exchange rate regime contradicting previous literature. These results generally support the Balassa-Samuelson theory.