Theses Doctoral

Three Essays in Banking

Antoniades, Adonis

This dissertation consists of three separate essays which address questions in the field of banking. The first two essays are motivated by the Great Recession, and study key aspects of the experience of commercial banks during this period. One is the impact of liquidity risk on credit supply, and the second is the effect of portfolio choices on the probability of bank failure. The third essay shifts the focus from commercial banks to M & A transactions, and studies the impact of a key provision in merger agreements on the initial offer premium and target firm value. In the first essay, titled "Liquidity Risk and the Credit Crunch of 2007-2009", I document the connection between liquidity risk and the credit crunch experienced during the financial crisis of 2007-2009. Using extensive micro-level data on mortgage loan applications, I construct a measure of the supply of credit that is free from demand-side bias. I then use this measure of credit supply to estimate the effect of cross-sectional differences in unused lines of credit and core-deposit funding on the supply of mortgage credit moving through the crisis. I find that lenders with higher liquidity risk contracted their supply of mortgage credit more. The channel of contraction was significantly stronger for larger lenders, which had the largest exposure to liquidity risk. The first phase of the contraction was due to liquidity risk arising from high exposure to lines of credit and was immediately followed by further tightening due to the collapse of the markets for wholesale funding. I estimate that the total contraction of mortgage lending due to liquidity stresses experienced by lenders during 2007-2009 was $41.5 billion - $61.9 billion, or 5.2%-7.8% of total mortgage originations during that period. In the second essay, titled "Commercial Bank Failures During The Great Recession: The Real (Estate) Story" I identify the channels through which shocks to the real estate sector contributed to the wave of commercial bank failures during the Great Recession. I focus on the banks' loan, marketable securities and credit line portfolios, and consider how choices which shifted the composition of each portfolio towards real estate products impacted the probability of bank failure. I find that augmenting a baseline model of failure with variables that capture the composition of these three portfolios improves the fit of the model by approximately 70% for small banks and 230% for large banks. I find no evidence that banks which held more of their loans in traditional closed-end mortgages suffered a higher probability of failure. Rather, it was investments in loans for multifamily properties and other non-household real estate loans, as well as off-balance sheet exposures to credit lines issued to non-household real estate borrowers, that are robustly identified as precursors of bank failure for both small and large banks. Exposure to open-end residential real estate loans contributed to the failure rates of small banks only. Exposure to private-label MBS is strongly associated with a higher probability of failure for large banks, but not for small ones. On the other hand, high holdings of agency MBS are associated with a higher probability of failure only for smaller banks, but this result is less robust. The third essay, titled "No Free Shop: Why Target Companies in MBOs and Private Equity Transactions Sometimes Choose Not to Buy 'Go Shop' Options" is joint work with Charles W. Calomiris and Donna M. Hitscherich. In this essay, we study the decisions by targets in private equity and MBO transactions whether to actively "shop" their initial acquisition agreements prior to the shareholders' approval of those contracts. Specifically, targets can insert a "go-shop" clause into their contracts, which permits them to use the agreement to solicit offers from other would-be acquirors during the "go-shop" window, during which the termination fee paid by the target is temporarily lowered. We consider the "go-shop" decision from the theoretical perspective of value maximization under asymmetric information, and also consider conflicts of interest on the parts of management, bankers, and attorneys that might affect the decision. Empirically, we find that the decision to retain the option to shop an offer is predicted by various firm attributes, including larger size, more fragmented ownership, and various characteristics of the firms' legal advisory team and procedures. These can be interpreted as reflecting a combination of informational characteristics, litigation risk, and attorney conflicts of interest. We employ legal advisor characteristics as instruments when analyzing the effects of go-shop decisions on target acquisition premia and value. We find, as predicted in our theoretical framework, that go-shops are not a free option; they result in lower initial acquisition premia, ceteris paribus. Our theoretical framework has an ambiguous prediction about the effects of go-shop choice on target firm valuation. Consistent with theory, we find no significant effect on abnormal returns from choosing a "go-shop" option.


  • thumnail for Antoniades_columbia_0054D_11500.pdf Antoniades_columbia_0054D_11500.pdf application/pdf 2.24 MB Download File

More About This Work

Academic Units
Thesis Advisors
Chiappori, Pierre-André
Ph.D., Columbia University
Published Here
June 28, 2013