Working Paper: CEPR ID: DP5850
Authors: Richard Baldwin; Daria Taglioni
Abstract: This paper provides a minimalist derivation of the gravity equation and uses it to identify three common errors in the literature, what we call the gold, silver and bronze medal errors. The paper provides estimates of the size of the biases taking the currency union trade effect as an example. We generalize Anderson-Van Wincoop?s multilateral trade resistance factor (which only works with cross section data) to allow for panel data and then show that it can be dealt with using time-varying country dummies with omitted determinants of bilateral trade being dealt with by time-invariant pair dummies.
Keywords: country dummies; gravity equation; pair dummies
JEL Codes: F10; F33; F4
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| gravity model (naively applied) (C59) | significant biases in trade estimates (F14) |
| omitted variables correlated with trade costs (F12) | upward bias in estimated impact of currency unions (F36) |
| correcting for biases (C83) | alters estimated effects of the euro on trade (F17) |
| common empirical practices (C90) | biases in estimates of trade effects (F14) |
| currency unions (F36) | distort estimated effects on trade (F14) |