Essays on Banking.

Summary This thesis focuses on the economics of banking and consists of three chapters. Chapter 1 explores the effects of monetary policy changes on the liability structures of U.S. banks and their funding costs. Banks obtain most of their funding from a combination of demand deposits - which are at zero interest - and interest-bearing deposits. Using local demographic changes as instruments for bank liability structures, I measure the impact of monetary policy changes on each bank's interest-bearing deposit rate as a function of its initial liability structure. I find that when monetary policy tightens, each bank faces an outflow of demand deposits. It responds by issuing interest-bearing deposits, but pays an interest rate that increases with the amount of demand deposits being substituted. This finding confirms the existence of the "bank lending channel" of monetary policy transmission. I also provide evidence that large banks can substitute funding sources at a lower cost than small banks, and that demand deposits are less sensitive to changes in monetary policy when the local banking market is more concentrated. Chapters 2 and 3 focus on mutual banks, which differ primarily from banks under the joint-stock system in their ownership. In the second chapter, Richard Meade and I provide an assessment of the effect of "demutualization" events on depositor welfare. Over the past few decades, many U.S. savings banks have opted to demutualize, converting their ownership from mutual banks to corporations. We first estimate a random parameter logit model of bank deposit choices, using data on commercial banks and savings banks from 1994 to 2005. After recovering depositors' preferences for bank characteristics, we measure the effect of a demutualization of all savings banks on depositors' welfare. We find that, on average, depositor welfare would increase. In particular, if demutualized savings banks offered a deposit rate in line with investor-owned savings banks, each depositor would earn $1.14 per year, for a total of $22 million for each state and year. Our results cast doubt on the ability of U.S. savings banks to serve their customers' interests well, and offer a new explanation for the demutualizations observed in the United States. In the third chapter, Thierry Magnac, Karine Van der Straeten, and I first present evidence that U.S. mutual banks are less risky, lend at lower rates, and have closer ties to their local communities. We propose a model of competition between mutual banks and banks under the corporate regime, adapting the model of Martinez-Miera and Repullo (RFS, 2010). We assume that the difference between the two types of banks is that mutual banks, given their stronger local reach, impose a greater "social cost" on failed policyholders. We simulate the model and evaluate the effect of competition on the probability of failure in both types of banks. Our results indicate that, consistent with the empirical evidence presented, mutual banks lend at lower rates and have a lower probability of default at any level of competition. Moreover, similar to what happens to banks under the joint stock regime, their probability of default is U-shaped in the level of competition.