Premium
Intra‐industry Trade and Industry Distribution of Productivity: A Cournot–Ricardo Approach
Author(s) -
Song E. Young,
Sohn ChanHyun
Publication year - 2012
Publication title -
the world economy
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.594
H-Index - 68
eISSN - 1467-9701
pISSN - 0378-5920
DOI - 10.1111/j.1467-9701.2012.01446.x
Subject(s) - cournot competition , economics , productivity , monopolistic competition , distribution (mathematics) , sample (material) , index (typography) , total factor productivity , conjecture , competition (biology) , bertrand competition , econometrics , microeconomics , international trade , oligopoly , mathematics , macroeconomics , monopoly , mathematical analysis , ecology , chemistry , chromatography , biology , world wide web , computer science , pure mathematics
This study constructs a general equilibrium model of trade by incorporating Cournot competition in a Ricardian model of trade with a continuum of goods. In this model, the share of intra‐industry trade (IIT) in bilateral trade volume decreases as trading partners become more different in terms of the industry distribution of labour productivity. We show that this relationship is a special case of a general accounting identity: the intensity of IIT is decreasing in a specialisation index. The specialisation index is equal to the difference in the industry distribution of labour productivity in our Cournot–Ricardo model, while it is equal to the difference in factor endowments in the well‐known monopolistic competition model. As both technologies and factor endowments would play important roles in shaping specialisation patterns, we conjecture that the share of IIT is decreasing in both measures of specialisation. Using the data from Nicita and Olarreaga, we test this conjecture. Our pooled ordinary least squares (OLS) estimations strongly support this conjecture. However, in fixed‐effect panel estimations, only the industry distribution of productivity is significant in the entire sample. Subdividing the sample, we find that the difference in the industry distribution of productivity is significant in explaining North–North IIT, while the difference in factor endowments is significant in explaining North–South IIT.