Finance and Growth: The Neglected Role of the Business Cycle
A canonical cross-country/time-series literature argues that financial development, most commonly measured as private credit in percent of GDP, leads to higher growth. This literature typically aims to control for the business cycle by averaging the data over non-overlapping five years periods. We show that averaging over fixed-length intervals of five years does not eliminate business cycle fluctuations. We find that growth and some of the most commonly applied proxies for financial development are strongly and positively correlated with various measures of the output gap for annual as well as five years averaged data. Studies explaining growth with measures for financial development and not adequately controlling for the pro-cyclicality of both variables therefore likely overstate the long-run impact of finance on growth. We illustrate the significance of this bias by explicitly sweeping out low-frequency variations in the data-sets of Beck and Levine (2004) and Beck et al. (2013), two recently published articles whose authors claim to have found a positive effect of finance on growth. Once output gap fluctuations are purged, we find that the impact of private credit on growth is no longer significantly different from zero. We conclude that there is no significant positive correlation between the most commonly applied proxy for financial development and growth in these data sets in the long-run, once the short-run pro-cyclicality of credit and growth is addressed. Important contributions to the empirical literature on the finance and growth nexus are therefore the outcome of a flawed and spurious research design. As five years averaging of data has become a standard procedure to control for cyclical fluctuations, our findings might also have implications for the robustness of other results in the macroeconomic cross-country/time-series literature.
SEARCH PERI >>>>>>>>>>>>>>>