Working Paper: NBER ID: w27957
Authors: Diego A. Comin; Robert C. Johnson
Abstract: Did trade integration suppress inflation in the United States? Conventional wisdom says “yes,” based on the disinflationary supply-side impacts of trade. We argue that these supply-side arguments are incomplete, because trade integration also influences aggregate demand. Our analysis leverages two facts: trade integration was a long-lasting, phased-in shock, and offshoring accounts for a large share of it. Given these facts, we show trade integration is inflationary in conventional New Keynesian models. This result continues to hold when we account for US trade deficits, the pro-competitive effects of trade on domestic markups, and cross-sector heterogeneity in trade integration.
Keywords: Trade Integration; Inflation; Offshoring; New Keynesian Models
JEL Codes: E5; F1; F15; F4; F6
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| trade integration (F15) | aggregate demand (E00) |
| aggregate demand (E00) | equilibrium inflation (E31) |
| anticipated increases in trade (F19) | aggregate demand (E00) |
| anticipated increases in trade (F19) | equilibrium inflation (E31) |
| increased offshoring (F69) | unit production costs (D24) |
| increased offshoring (F69) | inflationary pressures (E31) |
| procompetitive effects of trade (F13) | domestic markups (D49) |
| procompetitive effects of trade (F13) | inflationary impact (E31) |
| rise of trade in late 1990s and early 2000s (F19) | inflationary (E31) |
| trade integration (F15) | lower long-run output price inflation (E31) |