Does financial development precede growth? Robinson and Lucas might be right

Abstract

This paper studies whether there is any causal link between financial development indicators and economic growth, using Sims–Geweke causality tests performed in the large panel data set provided by Levine, Loayza and Beck. In sharp contrast to their findings, no evidence was found of any positive unidirectional causal link from financial development indicators to economic growth. On the contrary, a substantial indication that economic growth precedes subsequent financial development was found. As argued by Robinson, financial development might primarily follow economic growth, as a result of increased demand for financial services. Although the present result does not quite imply that the role of financial development in the development process is not important, the bottom line is that a more balanced approach to studying the relationship between growth and finance needs to be adopted. As termed by Lucas, the importance of financial development in economic growth might be very badly ‘over-stressed’. Robinson and Lucas might be right.

Acknowledgements

The views expressed are our own and do not necessarily reflect those of the World Bank. This research was supported by the research fund of Hanyang University (HY-2003-G).

Notes

¹ There have been a few country case studies exploring the time-series relationship between financial development indicators and growth such as Bhattacharya and Sivasubramanian (2003) for India, Mazur and Robert (2001) for New Zealand and Chang (2002) for China. The fundamental weakness of these individual country studies, albeit interesting and insightful, would be the potential difficulties associated with drawing an internationally comparable conclusion.

² In a direct causality tests utilizing the Sims–Geweke framework, Blomström, Lipsey and Zejan (1996) found that growth induces subsequent investment more than investment induces subsequent growth. In a related subject, Carroll and Weil (1994) showed that growth Granger-causes savings, but savings does not Granger-cause growth, which implies that previous estimates of the effect of savings on growth may be overstated.

³ All the sensitivity analyses results are available upon request.