No abstract

We examine the macroeconomic implications of bailing-in banks' creditors after a systemic financial crisis, whereby bank debt is partially written off. We do so within a RBC model that features an endogenous leverage constraint which limits the size of banks' balance sheets by the amount of bank net worth. Our simulations show that an unanticipated bail-in effectively ameliorates macroeconomic conditions as more net worth relaxes leverage constraints, which allows an expansion of investment. In contrast, an anticipated bail-in will be priced in ex-ante by bank creditors, thereby transferring the bail-in gains from banks to creditors. Therefore the intervention has zero impact on the macroeconomy relative to the no bail-in case. The effectiveness of the bail-in policy can be restored by implementing a temporary tax on debt outflows once creditors start to anticipate a bail-in.

We investigate the effectiveness of fiscal stimuli when banks are undercapitalized and have large holdings of government bonds subject to sovereign default risk. Deficit-financed government purchases then crowd out private expenditure and fiscal multipliers can turn negative. Crowding out increases for longer maturity bonds and higher sovereign default risk. We estimate a DSGE model with financial frictions for Spain and find that investment crowding out indeed leads to a negative cumulative fiscal multiplier. When monetary policy is exogenous, like at the ZLB or in a currency union, fiscal stimuli become more effective but multipliers are reduced when banks are undercapitalized.