Where at home do exporters produce and export?
Faculty of Economics, Kyushu Sangyo University, Higashi-ku, Fukuoka, Japan
Faculty of Economics, Shiga University, Hikone, Shiga, Japan
If international trade costs are the same for all regions, ceteris paribus, it does not matter where a
home firm chooses for production and exports. Once asymmetric trade cost is considered, however,
a firm prefers to produce and export from the region where trade costs can be minimised. Produc-
tion facilities move across regions, and as a result, some regions become agglomerated bases for
export platforms. This is one of the factors that cause some regions to grow faster than other
regions in an internationally open economy. Nevertheless, if the foreign market is relatively small,
meaning the domestic market is relatively more important, this mechanism of regional structural
change does not come into play. On the other hand, foreign markets growing beyond the threshold
or substantial decline in trade costs initiate the mechanism to work.
Making this framework applicable for an empirical analysis requires two prerequisites. First, we
need to look for the regions across which the factor of production is highly mobile. The regions in
the theoretical model can be interpreted as regions within a country or countries in close proximity,
but in any case high mobility for at least one factor of production is essential. Close proximity of
countries with relatively free labour migration clauses makes the EU one of the most applicable
cases for this study. However, it is more desirable to consider regions within a single developed
country because restriction on labour migration within a developed country is absolutely free. Sec-
ond, the empirical analysis requires export as well as production data disaggregated at industry as
well as regional level. The US state data set meets this requirement, and important empirical works
on international trade are conducted by McCallum (1995) and Yilmazkuday (2012).
The Japanese data set also provides some advantages over those of other countries. First, the
details of Japanese international trade are collected thoroughly by Japan Customs at the port
level. Manufacturing exporters have relatively easier access to ship their products from the near-
est international ports because each of 47 prefectures owns multiple international ports except
for a few cases.
Second, to match with the disaggregation level of trade data, production and
wage data in Japan are also readily available at the prefecture level. Third, mainly led by China,
foreign growth in the Asian region has been phenomenal to cause structural change in regional
production and exports within Japan. We therefore use a data set for Japan which is a relatively
geographically medium-sized country with rich disaggregated data available at the domestic
Equivalent to a state in the United States and a province in Canada, the largest political unit in Japan is a prefecture.
© 2017 John Wiley & Sons Ltd wileyonlinelibrary.com/journal/twec World Econ. 2018;41:1596–1639.