Globalization along with the resultant increase in international competition has led to the view that exports have become more sensitive to costs. International cost competitiveness is often measured in terms of unit labor costs, defined as labor cost per unit of output. This hinges on the idea that increases in unit labor costs are passed on in the form of higher export prices, and consequently hurt export sales. While there is a clear policy concern about rising unit labor costs in many European countries, there is little conclusive evidence about the impact of unit labor costs on export performance. The goal of this paper is to examine how well this measure of unit labor costs is related to export performance at the firm level. Specifically, we use a confidential Belgian firm-level dataset with detailed information on costs, productivity and exports for the period 1999-2010. Taking a microeconomic perspective with these disaggregated data, we are able to consider various dimensions often ignored in the earlier literature. More precisely, we study not only the export performance of continuing exporters, but also entry and exit of firms into export. Further, we allow for heterogeneous effects of unit labor costs on export performance according to firm and sector characteristics.

Our results show a negative relationship between unit labor costs and exports at the firm level: a 10 percent increase in unit labor costs implies a drop in exports of between 2 and 4 percent. We find that this elasticity varies between sectors and firms, and notably that labor-intensive firms have a higher elasticity of exports with respect to unit labor costs. In addition, the evidence shows that higher unit labor costs reduce the probability of starting to export for non-exporters and increase the probability of exporters stopping.

Our paper helps to evaluate the use of unit labor costs as a competitiveness indicator. While our results show that unit labor costs do actually have an impact on firm-level exports, the effect is rather small, suggesting that pass-through of costs into prices is limited or a low-elasticity demand for exported products. The latter indicates that other factors such as taste and quality may be just as important to incorporate into competitiveness indicators, as suggested by recent trade models focusing on quality and taste parameters. A major challenge for constructing indicators of competitiveness is therefore to identify proper measures for quality and taste.