Introduction

From an empirical point of view economic growth is usually seen as a long-run process. Growth is conceived of as the trend in GDP over a substantial time span. Therefore, to explain growth by econometric techniques, one needs data for a large number of countries to apply cross-section analysis. The equation estimated contains the growth rate of GDP either in total or per capita as the dependent variable and a number of explanatory variables based on economic theory. There are basically two strategies that can be followed. First, the estimated equation may be derived from a theory of economic growth (e.g. Dowrick and Nguyen, 1989; Barro and Sala-i-Martin, 1992; Mankiw et al., 1992). Second, a pool of explanatory variables, which come from different macroeconomic theories, may be considered, assuming that they can be entered independently and linearly (e.g. Kormendi and Meguire, 1985; Grier and Tullock, 1989; Barro, 1991; Levine and Renelt, 1992). In the latter case it is useful to sort out variables that really matter. As shown by Levine and Renelt (1992), by applying cross-section analysis for the period 1960–89 to a sample of about 100 countries, the number of robust explanatory variables with respect to real per capita GDP is rather limited.

This chapter looks at economic growth as a process of the medium as well as the long run. In studying growth, one has to eliminate the business cycle, but there is no compelling reason to assume that differences in growth rates across sub-periods must be averaged out to get the right picture.