Finance Causes Growth: Can We Be So Sure?

Abstract

In recent years, a number of studies have found in favour of a causal link between finance and growth. This paper examines the empirical foundations of such studies, performing some additional econometric tests using the datasets and methodologies of Rajan and Zingales (1998a) and Levine and Zervos (1998). First, we consider the stability of published results across countries at different stages in their economic development, re-estimating Rajan and Zingales specification to allow separate coefficients for OECD and non-OECD member countries. We find that finance has a greater impact upon growth in non-OECD countries, with bank finance of particular importance. Second, we explore the issue of identification, concluding that it is difficult to disentangle the effect of financial development from that of other correlated factors. In particular, we show that the results of these authors depend heavily on the strong performance of the Tiger economies during the 1980s.