Trade costs, import penetration, and markups

The rise of market power and the decline of labor's share of GDP in the United States in recent decades is well documented and have critical macroeconomic implications, but the determinants of such trends remain unclear. This paper asks how and to what degree increasing import penetration contributes to the more concentrated market structure and the associated rise of mark-ups. We provide a general equilibrium framework linking the change of markup with the extensive margin of foreign-input imports. In the model, a reduction of importing costs induces non-importers to start importing intermediates and existing importing firms to increase the share of imported inputs. But the capability of importing more varieties of inputs depends on productivity as it requires fixed costs to select cost-efficient intermediate inputs to import. We then combine firm-level micro panel data, sector-level trade data and input-output table to present empirical evidence on the relationship between the rise of market power and the increase of imported inputs penetration. At the 6-digit sector level, the rise of imported input penetration induced market concentration, implying that only the most productive firms benefit from trade liberalization. We further test our predictions of heterogeneous firms' decisions on intermediates importing and the implications on the market structure using transaction-level custom data: decreasing trade costs induce non-importing firms to start to import intermediates and allow the existing importing firms to charge higher markups than before.