Financial Development, Financial Fragility and Growth Norman Loyza & Romain Rancière The Worldbank CREI

What do we know about Long and Short run effects of Financial Development? Empirics: –The long run growth enhancing effect of financial intermediation. King and Levine (1993); Beck, Loayza and Levine (2000) –Lending boom, banking crisis and growth slowdown. Demirguc-Degatriache (1997), Gourinchas and alli (2000) -> integrated view: [ Kaminsky-Schmukler (2002)] Theory: –Financial Development fosters growth by reducing liquidity risk and allocate efficiently savings. Greenwood-Jovanovic (1990),Bencivenga and Smith (1991), –Lending Boom and Growth Slowdown: financial accelerator effects, limited monitoring and enforcement, bailout guarantees. Aghion and alli (1999), Schneider-Tornell (2004)

What do we do? Provide an empirical framework to help reconcile the apparent contradiction between two strands of the literature on the effects of financial intermediation on economic activity. 1.Short and Long Run growth effects of financial intermediation on Growth 2.Link with other aspects of financial intermediation: Volatility and Banking Crises 3. Discussions: –Theory: Can we rationalized those results? –Policy Implications

A Direct evaluation on short and long run growth effects of financial development using an annual data panel Application of the pooled estimation methodology of long run relationship in heterogeneous panel proposed by Pesaran-Shin and Smith Error correction model allowing for short run dynamics to differ from long run equilibrium and to be represented by an Autoregressive Distributed Lag Model (ARDL): Long run equilibrium Short Run Adjustment

ModelPooled Mean Group Mean GroupDynamic Fixed Effects Long Run Coefs HomogeneityHeterogeneityHomogeneity Short Run Coefs Heterogeneity Homogeneity InterceptHeterogeneity Homogeneity Annual Panel Data Financial indicators: liquid liabilities / GDP ; private credit/ GPP Control set of variables : Initial Income; government size, trade openness, Inflation rate Control for time effects Optimized selection of ARDL process by information-based criteria Model choice: Hausman test of homogeneity restrictions

Short Run Dynamics: Is there a link with Banking Crises and Volatility? A cross-country distribution of short run coefficient from PMG estimation A cross-country distribution of volatility of credit/gdp over A cross-country distribution of frequency of Systemic Banking Crises Short-run coefficientsNumber of crisisFinancial volatility Short-run coefficients (0.008)(0.001) Number of crisis (0.0195)(0.000) Financial volatility (0.0094)(0.0005)

What did we learn? Results from Pooled Mean Group estimation Result 1: An homogenous positive long term relationship between financial development and growth Result 2: An heterogeneous on average negative short run adjustment. Result 3: A mapping between countries with a negative short run dynamics between financial development and growth, and countries with an experience of systemic banking crisis and financial volatility

Theoretical Discussion Dell’Arricia and Marquez (2004): bank’s incentive to screen project are reduced in the aftermath of financial liberalization. Adverse Selection among banks. Rajan (1994): strategic complementary among lenders in credit policy –> finance bad project in good times and squeeze credit in bad times-> the role of supervision. Ranciere-Gaytan (2003). Liquidity Role of Banks and Effects on Growth at different level of economic development.

Final Remarks Negative short run effects of financial development coexists with a positive long run relationship for a large set of countries including most of the countries with an experience of financial crisis. More work to integrate the dual effect of financial liberalization and financial development on growth and to draw policy implications.