Working Paper: CEPR ID: DP4468
Authors: Kristian Behrens; Andrea Lamorgese; Gianmarco I.P. Ottaviano; Takatoshi Tabuchi
Abstract: We extend the model by Krugman (1980) to a multi-country set-up and show that the ?home-market effect? highlighted with two countries does not readily extend to such a general setting. In particular, we prove that the most important result, namely the disproportionate causation from demand to supply, generalizes only under the fairly implausible assumption of pairwise symmetric trade costs between all countries. We argue, therefore, that the implications of product differentiation for the structure of world trade are better characterized in terms of spatial (?accessibility?) and non-spatial (?attraction?) effects, and we provide a theory-based specification that suggests how to test the home market effect in such a general setting.
Keywords: economic geography; home market effect; hub effect; market potential; new trade theory
JEL Codes: F12; R11; R12
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Demand (R22) | Supply (J20) |
| Increase in a country's expenditure share (H59) | Output share (D33) |
| Increase in a country's expenditure share (H59) | Industry share (L19) |
| HME in two-country model (F29) | HME in multicountry model (F12) |
| Output response to increasing demand (J23) | Product differentiation and imperfect competition (L13) |