The central question addressed in this chapter is how the pattern of trade specialization can affect long-term economic growth performance. This issue has received little attention in theoretical analysis and empirical studies. Standard trade theory has been interested on the question of how growth - through the changing composition of factor endowment - affects comparative advantages and thus the trade pattern. It has had little to say, however, on the causal links we are interested here, i.e. those running from trade specialization to growth. Neither has modern growth theory paid much attention to this issue. Neoclassical growth theory focuses on the role of factor accumulation, i.e. investment rates in physical and human capital as well as labor force growth. Moreover, its assumption of a constant returns to scale technology leaves little room for economic structure to affect the growth rate. Recent endogenous growth models have brought increasing returns into growth theory but the level of aggregation assumed in these models has led them to focus on factors other than the pattern of specialization. Growth empirics, inspired by these two brands of growth theory, has at most looked (without much success) at how trade openness, rather than trade specialization, may affect differences in growth rates among countries. There is thus a theoretical as well as an empirical gap to fill in this important area.
All this is not to say that there has been complete neglect. In the old literature on trade and in new trade theory, there is a significant list of contributions that are all relevant to our topic - on infant industry protection, the Prebisch-Singer thesis on the terms of trade of primary products, immiserizing growth, and multi-sector growth models with different rates of learning in new trade theory. In looking at how the pattern of specialization can affect growth, a common feature of these contributions is that the pattern of specialization of an economy, as determined by comparative advantages associated with the current factor endowment or locked in by historical accident, may be different from the pattern of specialization that yields the largest long-term economic benefits. This may occur because the economy's static comparative advantage does not coincide with its dynamic comparative advantage in the presence of, for example, technological externalities due to learning by doing, or because the evolution of the terms of trade is such that the