Does trade really reduce unemployment? The empirical analysis in Treffler (2004) highlights ambiguous effects in the short and long run of the US–Canada free trade agreement. In line with public sentiments against globalization, he documents employment losses shortly after the establishment of the agreement. Those negative employment effects stand in stark contrast to the positive long-run effects documented in the same study, as well as most of the more recent analyses in this field. Models with heterogeneous firms and search frictions give rise to a channel in the Melitz (2003) model through which trade liberalization fosters firm selection but reduces unemployment in the long-run. Felbermayr et al. (2011) show that more productive firms are relatively less efficient in recruiting workers, so that the net-effect on job-creation is positive. Helpman et al. (2008, 2010) show that the results may also be negative depending on the model setup. The crucial assumptions in those theoretical approaches, however, are an exogenous job-separation rate and no channel for job-to-job transitions, which is likely to mask important adjustment processes in the labor market after trade liberalization. Recent contributions aim to overcome this shortcoming, albeit they focus on introducing on-the-job search into theoretical trade models (see, e.g., Larch and Holzner, 2011; Suverato, 2013). We contribute to the literature by empirically analyzing the impact of trade on labor market dynamics. Our approach goes beyond the established literature as we identify both the short- and longrun effects within a structural VAR approach.