Essays on the effects of financial and sovereign crises in macroeconomics

This dissertation consists of three chapters that study the macroeconomic effects of financial and sovereign crises. Chapter I studies how the probability of bailout affects banks' portfolio decisions in normal times and during a sovereign debt crisis. Theoretically, an increase in the probability of bailout implies a change in the expected "bailout rents" delivered to bank owners, where the sign of the change depends on the sovereign risk and government's own borrowing costs. The model's main predictions are that an increase in the probability of bailout induces banks to decrease lending to firms and purchase sovereign bonds when the government's borrowing costs are sufficiently low, while the effect reverses during a sovereign debt crisis. Empirically, data on Italian banks between 2007 and 2014 provides evidence consistent with these predictions. Chapter II assesses the impact of the main unconventional monetary measures adopted by the European Central Bank in 2011-2012 on the Italian economy. Together with co-authors, I first estimate the direct effects on financial and credit markets and then map these effects onto their macroeconomic implications. The results suggest that all operations have, to varying degrees, contributed to counteract the increase in government bond yields and to improve credit supply and money market conditions. From a macroeconomic perspective, the measures have had a large positive effect, mainly through the credit channel, with a cumulative impact on GDP growth of 2.7 percentage points over the period 2012-2013. Chapter III builds a dynamic partial equilibrium model in which heterogeneous firms have access to two different sources of external financing, namely bank loans and credit lines. Under a parameterization that corresponds to the U.S economy in the first quarter of 2008, the model matches the empirical counterpart for aggregate levels of debt and credit line drawdowns. I simulate the impact of the financial crisis by varying some key parameters and studying the response of aggregate variables. I find that the model can replicate the change in firms' financing policy behavior observed during the financial crisis, and in particular the interaction between access to bank lending and use of credit lines.