How do credit supply shocks propagate internationally? A GVAR approach

We study how credit supply shocks in the US, the euro area and Japan are transmitted to other economies. We use the recently-developed GVAR approach to model financial variables jointly with macroeconomic variables in 33 countries for the period 1983-2009. We experiment with inter-country links that distinguish bilateral trade, portfolio investment, foreign direct investment and banking exposures, as well as asset-side vs. liability-side financial channels. Capturing both bilateral trade and inward foreign direct investment or outward banking claim exposures in a GVAR fits the data better than using trade weights only. We use sign restrictions on the short-run impulse responses to financial shocks that have the effect of reducing credit supply to the private sector. We find that negative US credit supply shocks have stronger negative effects on domestic and foreign GDP, compared to credit supply shocks from the euro area and Japan. Domestic and foreign credit and equity markets respond clearly to the credit supply shocks. Exchange rate responses are consistent with a flight to quality to the US dollar. The UK, another international financial centre, is also responsive to the shocks. These results are robust to the exclusion of the 2007-09 crisis episode from the sample.