Product Variety, Firm Entry, and Terms of Trade Dynamics

In this paper, I revisit the problem of the anomaly of terms of trade dynamics. First, I empirically analyze the effect of a US aggregate labor productivity shock on the US terms of trade using a VAR and Maximum Forecast Error Variance identification. I find that the shock appreciates the terms of trade of the US. Next, using a non-homothetic preference a lá Melitz and Ottaviano (2008), I explain the dynamics of the terms of trade in response to a positive aggregate productivity shock theoretically. Using a model with endogenous markup and heterogeneous firm-specific productivities, the appreciation of the terms of trade can be generated even under a complete asset market assumption. Unlike previous studies, I explain the dynamics of the terms of trade through a new channel, which is the channel of relative cutoff firm-specific productivity that determines the optimal export decisions of the firms. Depending on the asset market structure, two competing effects, i.e., the income effect and the markup effect, have different implication to terms of trade dynamics. Under the assumption of financial autarky, the income effect is bigger than the markup effect and the terms of trade depreciates in response to a positive aggregate productivity shock. However, if we allow for the trade of state-contingent or non-state contingent bonds, the income effect is mitigated and the markup effect appreciates the terms of trade, which is in line with the empirical findings. ∗Department of Economics, Duke University. E-mail: [email protected] †I would like to thank my dissertation supervisor, Craig Burnside for his support and many discussions. I would also like to thank my dissertation committee members, Pietro Peretto, Barbara Rossi and Juan Rubio-Ramirez for useful comments and suggestions. I am also grateful to Kent Kimbrough and George Tauchen as well as seminar participants at Duke University for their helpful comments.