Credit supply shocks, network effects, and the real economy

We consider the real effects of bank lending shocks and how they permeate the economy through buyer-supplier linkages. We combine administrative data on all firms in Spain with a matched bank-firm-loan dataset incorporating information on the universe of corporate loans for 2003-2013. Using methods from the matched employer-employee literature for handling large data sets, we identify bank-specific shocks for each year in our sample. Combining the Spanish Input-Output structure and firm-specific measures of upstream and downstream exposure, we construct firm-specific exogenous credit supply shocks and estimate their direct and indirect effects on real activity. Credit supply shocks have sizable direct and downstream propagation effects on investment and output throughout the period but no significant impact on employment during the expansion period. Downstream propagation effects are quantitatively larger in magnitude than direct effects. The results corroborate the importance of network effects in quantifying the real effects of credit shocks and show that real effects vary during booms and busts.