Institutions and parameter heterogeneity

The role of institutions has taken center stage in the empirical growth literature recently, with many innovative approaches and suggestive results. Most of the previous literature has focused on the direct correlation between institutions and growth or its determinants. In this paper, I investigate the indirect effect that institutions may have on growth through parameter heterogeneity. In particular, the quality of a country's institutions could affect the relationship between growth and its other determinants, such as human and physical capital accumulation, geography, and government policies. After allowing for several types of nonlinearities, the results are generally supportive of Glaeser et al. [Glaeser, E.L., La Porta, R., Lopez-de-Silanes, F., Shleifer, A., 2004. Do institutions cause growth? Journal of Economic Growth, 9 (3), 271-303] in that most of their preferred measures of institutions are not correlated with growth. However, there is some evidence that institutions do matter for policy variables, particularly those associated with trade policy.