The Dodd-Frank Act has produced a new wave of bank M&As. This consolidation trend is mainly driven by mergers of small banks, since small banks feel the need to merge in order to absorb the compliance costs of the new regulation. We document that the $10 billion asset-size threshold has become the ceiling of the optimal scale for bank combinations, given that banks below this $10 billion mark avoid several regulatory hurdles imposed by the Dodd-Frank Act. Results for these "less than $10 billion mergers" suggest significant value creation for both firms' shareholders: Bidders experience large anticipated wealth gains during the passage of the legislation since the market had ex-ante identified these bids. Consequently, at the deal announcement date, bidders experience insignificant returns, targets experience large abnormal returns and the combined abnormal returns are statistically positive. Finally, bidders experience positive abnormal returns at the deal completion date. On the contrary, results for larger bank mergers indicate a redistribution of wealth from the bidder to the target firm.

In this paper, we use the sentiment of annual reports to gauge the likelihood of a bank to participate in a merger transaction. We conduct our analysis on a sample of annual reports of listed U.S. banks over the period 1997 to 2015, using the Loughran and McDonald's lists of positive and negative words for our textual analysis. We find that a higher frequency of positive (negative) words in a bank's annual report relates to a higher probability of becoming a bidder (target). Our results remain robust to the inclusion of bank-specific control variables in our logistic regressions.

We extend the U.S. bank M&As literature by examining announcement returns for acquisitions of both listed and unlisted targets by U.S. banking firms for a long period of time from the eighties till to date. Over these decades there have been implemented several regulation changes, notably the Dodd-Frank Act that would be of interest to examine whether they have any impact, and if indeed they have to which direction, on value creation in M&As in the U.S. banking industry. Contrary to the conventional wisdom that bidding banks lose upon the announcement of a merger, we find positive abnormal returns for these firms that choose to acquire privately-held targets. Further, returns for acquirers in private offers do not depend on the method of payment, legislative changes, size, or geographical scope. However, we find that the use of a financial advisor on the part of the bidder can better explain the variation in abnormal returns for such offers. Our results are not influenced by any unobserved bidder-specific component or sample selection issues.JEL Classification: G14, G21, G34.

We extend the U.S. bank M&As literature by examining bidder announcement abnormal returns in deals involving both public and private targets over a 32-years examination period. Our main findings document the existence of a listing effect in our sample. Banks gain when they acquire private firms and lose when they acquire public firms. Gains in private offers are even higher when bidders employ financial advisors, whereas the opposite is true for public deals. We argue that this adverse advisor effect relates to the different levels of information asymmetry between public and private targets. Our results remain robust when we control for usual determinants of bidder abnormal returns, such as the method of payment, size, or relative size and when we control for sample selection and endogeneity problems.

This study examines the predictive power of textual information from S-1 filings in explaining IPO underpricing. Our empirical approach differs from previous research, as we utilize several machine learning algorithms to predict whether an IPO will be underpriced, or not. We analyze a large sample of 2,481 U.S. IPOs from 1997 to 2016, and we find that textual information can effectively complement traditional financial variables in terms of prediction accuracy. In fact, models that use both textual data and financial variables as inputs have superior performance compared to models using a single type of input. We attribute our findings to the fact that textual information can reduce the ex-ante valuation uncertainty of IPO firms, thus leading to more accurate estimates.