References and Literature

This chapter covered a variety of models. Section 19.1 focused on the implications of international financial flows on economic growth. This topic is discussed in detail in Chapter 3 of Barro and Sala-i-Martin (2004), both with and without limits to financial flows.

Section 19.2 is motivated by Lucas’s classic (1990) article. There is a large literature on why capital does not flow from rich to poor countries. Obstfeld and Taylor (2004) contain a survey of the work in this area. The work by Caselli and Feyrer (2007) discussed above provides a method for estimating cross-country differences in the marginal productive capital and argues that differences in the return to capital are limited. This work supports models that explain the lack of capital flows based on productivity differences, such as the model presented in Section 19.2. See also recent work by Alfaro, Kalemli-Ozcan and Volosovych (2005), which also emphasizes productivity differences and links these to institutional factors which we discussed in Chapter 4. Recent work by Chirinko and Mallick (2007) argues that the Caselli and Feyrer (2007) procedure may lead to misleading results because they do not incorporate adjustment costs in investment in their calculations and that once these costs are incorporated, returns to capital differ significantly across countries.

The rest of the chapter relies on some basic knowledge of international trade theory. Space restrictions preclude a detailed review. The reader is referred to a standard text, for example, Dixit and Norman (1990). Section 19.3 provides a slight generalization of the model in Ventura (1997) (it considers a general costs and returns to scale production function rather than CES production function is in Ventura, 1997), though it omits some of the more detailed characterization of transitional dynamics in the paper. A similar but less rich model was first analyzed by Stiglitz (1971). Stiglitz did not feature labor-augmenting productivity differences across nations and assumed exogenous saving rates. Other papers that combine Heckscher- Ohlin trade with models of economic growth include Atkeson and Kehoe (2000) and Cunat and Maffezoli (2001). Section 19.4 builds on Acemoglu and Ventura (2002). The importance of terms of trade effects are well recognized in the theory of international trade (see again Dixit and Norman, 1990), but their growth implications had not previously been recognized. The model presented in Acemoglu and Ventura (2002) is a much simplified Ricardian model, exploiting the structure of preferences first introduced by Armington (1969), but in the production of the final good rather than in preferences. Richer Ricardian models typically build on the seminal article by Dornbusch, Fischer and Samuelson (1977), though this richer setup has not yet been integrated with growth models. Ventura (2005) provides a survey of international trade and economic growth, focusing on the models in Sections 19.3 and 19.4.

The model in Section 19.5 builds on Krugman’s (1979) seminal article on product cycle. As noted in the text, Vernon (1966) was the first to formulate the problem of the international product cycle, emphasizing the economic forces modeled in Krugman (1979) and in Section 19.5 here. Grossman and Helpman (1991b) provide richer models of the product cycle with endogenous technology, similar to the economy discussed in Exercise 19.25.

There is a large empirical literature on the impact of trade on growth. Many of the best- known papers in this literature were discussed at the beginning of Section 19.6. The rest of Section 19.6 builds on Romer and Rivera Batiz (1991) and Grossman and Helpman (1991b), but uses the formulation from Section 13.1 in Chapter 13. Grossman and Helpman (1991b) assume that R&D requires labor and introduce competition between the R&D sector and the final good sector. In this case, the nature of the knowledge spillovers becomes important for the implications of trade on the pace of endogenous technological progress. Romer and Rivera Batiz (1991) also discuss the implications of the form of the innovation possibilities frontier for the effects of trade on technological change. This point, which is developed in Exercise 19.29, also features in recent work by Atkeson and Burstein (2006). Grossman and Helpman (1991b) also present much richer models with multiple sectors and factor proportion differences across countries, leading to Heckscher-Ohlin type trade. Another potential effect of international trade on technological change would be by influencing the direction of technological change. This topic is analyzed in detail in Acemoglu (2003b), where I show that trade opening, with imperfect intellectual property rights, can make new technologies more skill-biased than before trade opening. Similar models are also analyzed in Thoenig and Verdier (2002) and Epifani and Gancia (2006).

Section 19.7 presents a model inspired by Young (2003) and Matsuyama (2002). Lucas (1988) and Galor and Mountford (2006) also present similar models, which feature interaction between specialization and learning-by-doing. Other models where international trade may be costly rely on differences in the amount of rents generated by different sectors because of imperfections in the labor market or institutional problems. Levchenko (2008) and Nunn (2007) present models in which trade leads to the transfer of rent-creating jobs from countries with weak institutions to those with better institutions and may be harmful to countries with weak institutions. Davis and Harrigan (2007) present a model in which trade leads to the reallocation of high-rent jobs to some countries and can be harmful to the economies that are losing these jobs.

19.10.