Abstract
This paper demonstrates that observable trade cost measures in logs are not linearly related to the overall log trade costs nor to the conditional mean of log bilateral trade flows. This is shown using a simultaneous quantiles regression model and data on bilateral exports in 2008. This is modeled as a function of geographical, cultural, and historical observables and a host of unobservable trade cost measures in a structural model of bilateral trade. In this model, trade costs differ not only statistically but also quantitatively across the quantiles of the conditional distribution of bilateral exports. As a consequence, comparative static effects of these trade costs vary as well.
Original language | English (US) |
---|---|
Pages (from-to) | 5-15 |
Number of pages | 11 |
Journal | Empirical Economics |
Volume | 50 |
Issue number | 1 |
DOIs | |
State | Published - Feb 1 2016 |
Keywords
- Gravity models
- Quantile regression
- Structural estimation
ASJC Scopus subject areas
- Statistics and Probability
- Mathematics (miscellaneous)
- Social Sciences (miscellaneous)
- Economics and Econometrics