This paper analyzes trade among and the convergence of per capita income for India, Pakistan, Bangladesh, and Sri Lanka. The extent of trade and its relationship with the magnitude of income convergence is studied among these countries and their trading partners. We use intra-trade convergence and the difference-in-differences approach for the estimations. The results demonstrate that an increase in trade between the groups decreases the per capita income differential. Our results suggest that trade liberalization policies could be effective in achieving convergence. More importantly, we find that the per capita income of our source countries converged more rapidly under post-liberalization regimes than pre-liberalization regimes.
Keywords: Intra-trade, income convergence, per capita income, South Asia.
JEL Classification: C21.
(ProQuest: ... denotes formulae omitted.)
I. Introduction and Literature Review
An extensive body of literature recognizes the link between trade and income convergence and divergence among countries (see studies based on cross-country growth regressions such as Baumol (1986), Dowrick and Nguyen (1989), Barro (1991), and Levine and Renelt (1992); studies that are based on beta convergence [regression toward the mean] such as Barro (1984), Barro and Sala-i-Martin (1992), Baumol (1986), and De Long (1988); and other sets of studies based on sigma convergence [concerning cross sectional dispersion] including Barro (1984), Dowrick and Nguyen (1989) among others). The theoretical relationship between trade and convergence is examined by incorporating the role of international trade and liberalization. The argument here is that trade liberalization causes a convergence in per capita income-since trade liberalization increases competition and domestic firms' absorption capacity for knowledge and ideas, knowledge levels among countries converge to a common level, leading to per capita income convergence (see Sachs and Warner (1995), Ben-David (1993, 1994a, 1994b, 1996, 2000) and Ben-David and Kimhi (2004) among others). Ben-David and Loewy (1998) posit a model that demonstrates how moving toward free trade increases trade volumes and reduces income differentials among liberalizing countries.
The evidence indicates a higher incidence of income convergence in some subsets of countries but no convergence tendencies among other subsets of wealthier countries (see Baumol, (1986), Baumol, et al (1989), Baldwin (2003), and Ben-David (1993, 1994b) among others). The literature based on endogenous growth demonstrates a lack of income convergence,1 [Romer (1986) and Lucas (1988)], but strong evidence of conditional convergence in studies such as Barro (1991), Mankiw, et al (1992), and Levine and Renelt (1992).2 These studies point toward a number of factors such as human capital and government policies that help account for convergence. Other studies such as Grossman and Helpman (1991) suggest that trade can contribute to the local knowledge stock and new ideas. Baldwin, et al (2001) argues that the exogenously falling cost of trade has resulted in technological externalities in the world's northern countries. However, studies based on endogenous models, such as Eicher (1999), result in income convergence. Young (1991) presents a static trade model based on five different equilibriums, most of which lead to convergence, and supports the idea that trade should generate convergence. Kravis (1970) argues that trade is only one of the various contributors to growth, and may not necessarily emerge as the dominant factor. In an earlier analysis, Corden (1971) combines the traditional theory of gains from trade with the growth models of Solow (1956) and Swan (1956), who argue that trade not only produces static gains but also increases capital accumulation and leads to higher growth of per capita output. Corden (1971) implies that a country that moves from autarky to free trade attains a higher steady-state income and as a result grows faster during the transition period. …