The determinants of merger arbitrage return: an impirical analysis in the UK context

This thesis explores the magnitude and the determinants of the return to the merger arbitrage strategy in the UK context. We perform empirical analysis of the three hypotheses namely the risk-based hypothesis, the limited arbitrage hypothesis and the arbitrageurs' role hypothesis. First, in the risk-based hypothesis, using a sample of 1105 UK cash and stock mergers from 1987 to 2007, we find that the strategy generates significant positive return in excess of the systematic risk adjustment benchmark. The result is robust to a range of methods to control for systematic risk. The finding is consistent with the existing evidence from other markets. As for the risk-return characteristics of the strategy, in contrast to the US evidence, we find little evidence supporting the non-linear pattern. This finding is in line with the restrictions on bidder's ability to abandon the bid imposed by UK Takeover Code. This finding, combined with the evidence in the US market (strong non-linearity) and the Australian market (no nonlinearity), demonstrates the impact of takeover regulation on the risk-return characteristics of the strategy. Second, in the limited arbitrage hypothesis, we test the impact of different types of risks, costs and constraints (other than systematic risk) on the arbitrage return. We find that transaction costs are one of the important drivers of the cross-sectional variation of the arbitrage return. The result is robust to 4 different proxies for transaction costs, that is, firm size, price level, dollar trading volume, and frequency of zero return days. Holding costs are found to be an important determinant of the return. Idiosyncratic risk, the most important type of holding costs, contributes significantly to the source of the arbitrage return. We find that short-sale constraints appear to be another important holding cost that the arbitrageurs concern about. The result about the impact of shortsale constraints is, however, still inconclusive due to the small sample size. We also test the agency-based model of limited arbitrage hypothesis proposed by Shleifer and . Vishny (1997) but find no supporting evidence. Third, in the arbitrageurs' role hypothesis, utilizing a manually collected dataset to identify arbitrageurs and their holding of target stocks, we examine how different roles that arbitrageurs play in the takeover process help explain the source of the return to the strategy. We find that arbitrage holding is significantly related to arbitrage return after a host of factors that can determine the bid outcome and the market's assessment of the bid outcome are controlled for. This finding shows that the arbitrageurs are better than the average investors in the market in picking better takeover bids, the investment in which yields higher risk-adjusted return. In contrast to the US evidence, arbitrage holding is found to be negatively related to bid premium and has no impact on the probability of bid success. The difference between this finding and the US evidence may be attributable to the much more stringent UK disclosure rule during the takeover period compared to the US counterpart. Overall, our study, while providing evidence broadly in support of significant return to the merger arbitrage strategy, also highlights the importance of recognizing the impact of the takeover regulation on such return.