رشد، ترکیب بخشی، و تکامل سطح درآمد

We assert that the endowments of production factors cause cross-country differences in GDP by generating disparities in the sectoral composition. We characterize the dynamic equilibrium of a two-sector endogenous growth model with several consumption goods that are subject to minimum consumption requirements. In this model, economies with the same fundamentals but different endowments of capitals will end up growing at a common rate, although the long run sectoral composition of GDP will be different. Because the total factor productivity (TFP) in multisector models depends on sectoral structure, these differences in capital endowments will also generate sustained differences in TFPs.

New growth theory has provided increasing evidence suggesting that the accumulation of production factors alone cannot explain the observed cross-country differences in GDP per capita (see, for instance, McGrattan and Schmitz, 1999 and Parente and Prescott, 2004). Authors like Klenow and Rodriguez-Clare (1997) and Hall and Jones (1999) argue that differences in GDP per capita are mainly explained by differences in total factor productivity (TFP, henceforth). Simultaneously, another branch of development literature explains international differences in the growth rates of GDP as the result of differences in the sectoral composition of GDP (see Echevarria, 1997 and Laitner, 2000). Recently, Caselli (2005), Chanda and Dalgaard (2008), and Cordoba and Ripoll (2009) unify these two lines of research by showing that changes in the sectoral composition contribute not only to output growth, but also to productivity growth without any true technological change. By using multisector growth models as the basis of growth accounting exercises, these works demonstrate that the aggregate level of TFP can be decomposed into a contribution from sectoral composition and a contribution from the level of technology. Furthermore, the empirical evidence shows that there are meaningful differences in the sectoral composition across countries. Therefore, the composition effect can explain a substantial amount of the observed differences in aggregate TFP levels across countries.

In this paper we have analyzed the dynamic equilibrium of an extended version of the two-sector, constant returns to scale and endogenous growth model, in which both sectors produce consumption and investment goods. We have shown that the introduction of a second consumption good modifies the patterns of growth both in the long run and during the transition if we assume that preferences are non-homothetic. Under these assumptions, two economies with the same fundamentals but different initial endowments will converge into a BGP with the same relative prices and growth rates, although the capital ratio, the output-capital ratio and the sectoral composition will be different. Given that in this model the aggregate TFP depends on the sectoral structure, it follows that this TFP is then endogenous because a rise in the capital stock affects it by altering the sectoral structure.