Testing the Pricing-to-Market Hypothesis: Case of the Transportation Equipment Industry

Pricing-to-market (PTM) theory suggests that monopolistic firms which export adjust their destination-specific markups in reaction to exchange rate shocks. These adjustments limit changes in the price of their exports. Thus, important movements in the bilateral nominal exchange rate between two countries that trade are not necessarily fully reflected in the price of imports. Evidence in favour of PTM has been mostly obtained through hypothesis testing on the OLS, instrumental variable (IV), or single-equation error-correction estimates of partial-equilibrium models. However, we know from the recent econometric literature that Wald tests applied to some of these estimates may give erroneous results in the presence of endogeneity and weak instruments. In this paper, we examine the reliability of the evidence supporting the hypothesis of pricing-to-market using LIML-based LR Monte Carlo tests. These tests, developed by Dufour and Khalaf (1998), have good power and, unlike the Wald test, also have the correct test size. We find that the size-correct Monte Carlo LR-based test reverses half of the results obtained from the popular Wald test, indicating that PTM may not be as widespread as previously believed. In addition, our results support the view that PTM behaviour is likely to be present in the same industry across different countries and that pass-through is possibly higher with a larger market share of exports. The above findings are illustrated using the model developed by Marston (1990) and our analysis is conducted for export pricing firms in the transportation equipment industry for three country pairs: Canada exporting to the United States, the United States exporting to Canada, and Japan exporting to (mainly) the United States.