Distance to Frontier and New Import Growth
| Published date | 01 September 2013 |
| Date | 01 September 2013 |
| DOI | http://doi.org/10.1111/sjpe.12016 |
| Author | Purba Mukerji |
DISTANCE TO FRONTIER AND NEW
IMPORT GROWTH
Purba Mukerji
ABSTRACT
In a panel-data study involving product-level import data for 48 developing coun-
tries that underwent substantial trade liberalization between 1989 and 2001, this
study finds that the growth of trade in new goods imported following major trade
liberalization is related to the state of technology that existed just before liberaliza-
tion. The study develops two new measures of the extensive margin. Findings indi-
cate that greater is the distance of a country from the world technology frontier,
the faster is its growth of new goods imports. This indicates a higher cost of trade
protection for countries further away from the world technology frontier.
II
NTRODUCTION
Following major trade liberalization, where there is a substantial reduction in
trade barriers creating the opportunity to increase imports across a wide range
of products, countries tend to disproportionately increase the share of their
imports (as well as exports) in products that were not traded before.
1
The
objective of this study was to investigate this trade fact further, by testing the
hypothesis that after major trade liberalization, countries that are lagging in
technology would have a significantly higher growth of new import trade
compared with other countries. This is because being further away from the
world technology frontier, these countries might have a pent-up demand for
new imports that they cannot replicate domestically.
This hypothesis is interesting as one possible method of measuring exposure
to new technology is the extensive margin of trade, which is defined as trade
in products that have not been traded previously by a country, as a propor-
tion of its total trade. If pent-up demand leads to faster growth of the exten-
sive margin of imports for technologically backward countries, it increases the
probability of catch-up in the sphere of technology for these countries. Cross-
country technology diffusion can occur through something as simple as inter-
national trade contact with importers, exporters, engineers, and researchers.
Connecticut College
1
E.g., empirical findings of Goldberg et al. (2009, 2010), Kehoe and Ruhl (2002), Mukerji
(2004).
Scottish Journal of Political Economy, Vol. 60, No. 4, September 2013
©2013 Scottish Economic Society.
390
There is evidence of spillovers particularly through technology embodied in
imports (Keller, 2004). The spillover could be sequential where the new tech-
nology is first only consumed as embodied in imported products, followed by
imitation in production, through, for example, reverse engineering.
The empirical literature on technology spillovers through imports does not
take into account the potential role that differences in the existing state of
technology could play.
2
The contribution of this study is to account for these
differences and use major national trade liberalizations and the extensive mar-
gin growth that follows, to test the hypothesis that the existing state of tech-
nology could impact the new goods import trade.
In models of international technology diffusion (e.g., Kortum, 1998;
Eaton and Kortum, 1999; Keller, 2004), the speed of technology diffusion
variable represents the idea that innovation in one country is eventually
made available to others. The diffusion can occur via the introduction of
new products to the technology recipient country or by raising the produc-
tivity of research in the recipient country through knowledge spillovers. In
practice, trade can be a channel of technology diffusion. In this study, we
study the potential of new goods trade to increase access to technology fol-
lowing substantial trade liberalization. There are at least two reasons for
considering new goods trade to have the potential to disseminate technol-
ogy. First, the distance-to-frontier literature indicates that imitation is a
more expedient method of catching up in technology for countries at ear-
lier stages of development (Acemoglu et al., 2006). The growth of the
extensive margin of imports is the focus of this study because the introduc-
tion of new products is a fertile starting point for imitation. Second, the
process of trade itself raises the level of contact of the recipient country
with experts in the technology such as engineers and researchers. This con-
tact can speed up the process of imitation, as well as raising the probabil-
ity that technology diffusion in the sense of knowledge spillovers will
ultimately occur (Keller, 2004).
Romer (1994) models the case of a technologically backward country that
suffers from the ‘disappearance’ of entire markets due to trade protection.
The country is unable to produce substitutes for potential imports due to its
technological backwardness. This results in an additional burden from trade
protection. Mukerji (2009) extends the Romer (1994) partial equilibrium
model to a general equilibrium setting and finds that the differences in the
costs of protection arise directly from differences in the ability to produce
close substitutes.
The added burden of missing products could lead to faster growth of the
extensive margin once trade is liberalized in technologically backward coun-
tries and trade liberalization is the catalyst that unleashes this growth. This
study seeks to address both these implications by estimating the determinants
of the extensive margin before and after liberalization.
2
The literature only controls for the existing state of technology to isolate the impact of
the technological progress of trade partners on technological spillovers (Coe and Helpman,
1995; Coe et al., 1997).
DISTANCE TO FRONTIER AND NEW IMPORT GROWTH 391
Scottish Journal of Political Economy
©2013 Scottish Economic Society
To continue reading
Request your trial