Growth and institutions in Latin America: A pooled and cross-time series analysis (1951-1999)

Abstract

During the last two decades, economists have developed more interest in incorporating institutions as a variable for economic growth models. Such a task is not an easy one. It is hard to generate instruments that measure the development of institutions that are referred to different problems such as political instability, rule of law, trust, democracy, business risk and the like. The main objective of this paper is to develop a growth-model for Latin American countries that incorporates not only the usual economic
variables, such as investment, human capital, government expenditure and trade, but also variables that can be used as proxies of institutions for these countries, covering the period of the second half of the twentieth century. The emphasis is more oriented to empirical issues rather than theoretical ones. The results show that the institutional variables do play a role, but due to limitations of the data base in the area of non-economic variables and the nature of this research, the evidence is somewhat weak. However,
after the 80s, there seems to be stronger empirical evidence about the role of non-economic variables in economic growth. One merit of this work is that it is an initial attempt to build such models based on pooledcross- time series and applied only to Latin America . Usually, most of the models we have seen are just cross-country regressions. Some techniques that are easily applied in cross-country regressions are more difficult in pooled and cross time series analysis.