Date of Award
Doctor of Philosophy (PhD)
Dr. Dave O. Jackson
Dr. Andre V. Mollick
Dr. Thanh N. Ngo
This dissertation includes three essays on CEO networks and banking. I built a social networks matrix among the U.S. bank CEOs based on their biographical information. I investigate how social networks of CEOs impact bank risk-taking, bank performance, and bank acquisitions during financial crisis. Essay one, "CEO networks and bank risk taking", investigates the relevance of bank CEO networks to bank risk during the credit crisis of 2008. If social networks provide a CEO an enhanced flow of information, a better understanding of risk embedded in industry innovation, and more accurate marketing timing and forecast, then the expectation is that CEO networks have a negative impact on default risk In addition, I expect to find that well-connected bank CEOs are more likely to reduce their equity holdings at the onset of the crisis. Taken together, the results are expected to suggest that CEO networks expose banks to less risk, and that CEOs networks enable CEOs to reduce personal wealth losses in the anticipation of the crisis. Alternatively, if herding dominates CEO networks, I expect to find a positive relation between CEO networks and bank risk, as well as a positive association between CEO networks and CEO insider trading in 2006.
Essay two, "CEO networks and bank performance", focuses on the association between bank CEO networks and bank performances in terms of both stock returns and accounting return on equity (ROE) during the crisis. If CEO networks provide a CEO more efficient information and better business opportunities, then banks with well-connected CEOs are expected to enjoy better performance. On the contrary, if well-connected CEOs are more vulnerable to group thinking bias, then I expect the opposite results. Furthermore, I classify banks into two groups depending on whether a bank is regarded as "too big to fail" (TBTF). Government protection entitled to banks that are perceived as TBTF may lead to less pronounced social network effects. As such, this test may shed light on whether bank networks are a substitute for protection from downside market risk, as herding theory suggests.
Essay three, "CEO network and bank acquisition", examines whether the presence of CEO social networks is related to the abnormal returns to the acquirers, targets and the combined entities around acquisition announcements during the crisis. In addition, I test whether the target bank's CEO, who is socially connected with acquirer bank's CEO, is more likely to stay in the executive team after the acquisition. If CEO networks provide a better opportunity to reduce information asymmetry in the acquisition, especially in such an extremely volatile market as in a financial crisis, then I expect higher abnormal returns surrounding acquisition announcements along with a higher retention rate of target banks' CEOs. On the other hand, if market view a such type of acquisition as a collusion between the acquirer's CEOs and the target's CEO with high likelihood of weakening critical thinking and increasing judgment errors in managerial decisions, then I expect to find the opposite results.
University of Texas-Pan American