Faculty Scholarship 1994 - Present

Risk-Arbitrage Spreads and Performance of Risk Arbitrage

Although the announcement of a takeover attempt generally causes the target stock?s price to rise, it usually trades below the offer price, thereby producing an arbitrage opportunity. The appropriate arbitrage positions to undertake are closely linked with the deal consideration structure. With cash offers, arbitrageurs usually only need to buy the target?s stock. In merger attempts involving stock payments, risk arbitrageurs generally offset their long position in the target firm?s stock with a short position in the acquiring firm?s stock. These trading positions are designed to earn the positive spreads between the target stock?s market price and the consideration offered. If the merger is successfully consummated, risk arbitrageurs realize this spread. However, when a merger fails, the spread will generally widen instead of converging to zero causing risk arbitrageurs to incur huge losses. The risk-arbitrage spread, defined as the percentage difference between the offer price and the target?s post-announcement market price, represents the expected gain for arbitrageurs if a deal succeeds and thus plays an important role in the risk-arbitrage process. By analyzing a comprehensive sample of 1,223 announced takeover attempts occurring between 1995 and 2005, we find risk-arbitrage spreads evidence considerable cross-sectional variation. The sample?s arbitrage spreads have a mean of 6.10% and a standard deviation of 11%. About 6% of the arbitrage spreads are negative (target?s price above the offer level). These interesting behaviors of risk-arbitrage spreads give rise to a number of questions: What drives the cross-sectional variation in arbitrage spreads? Can we link arbitrage spreads to a variety of deal-specific factors and develop a prediction model? Such a prediction model may help risk arbitrageurs identify which deals are most attractive. For example, deals with arbitrage spreads above their model predicted values may be mispriced (i.e., market may have overestimated the risk and the required compensation) and thus be attractive risk-arbitrage candidates. A number of studies (Larker and Lys [1987]; Dukes, Frohlich, and Ma [1992]; Karolyi and Shannon [1999]; Mitchell and Pulvino [2001]; Barker and Savasoglu [2002]; Jindra and Walkling [2004]; and Branch and Yang [2003,2006]) have explored the risk and return characteristics of risk arbitrage and in some cases developed models designed to help risk arbitrageurs improve their performance. However, only Jindra and Walkling [2004] studied initial arbitrage spreads. They analyzed a sample of 362 cash tender offers from 1981 to 1995 and found arbitrage spreads to be greater for failed deals than for successful deals, negatively correlated with future price revisions and positively correlated with the length of time from deal announcement to deal consummation (also known as deal duration). Jindra and Walking [2004] focused on the market?s ability to anticipate a deal?s timing, outcome, and future revision. This article differs from their work in the following aspects: 1) we restrict our tests to relations between risk-arbitrage spreads and a variety of ex-ante variables (i.e., firm characteristic variables and market information); 2) we seek to develop prediction models that could be used to enhance the profitability of risk-arbitrage strategies. Our analysis reveals that the following variables have a significant role in determining the spreads for cash offers: target?s growth opportunity (measured by target?s market-to-book ratio), target?s price run-up, target termination fees, target resistance, transaction costs, arbitrageurs? activity, relative size of the target, and bid premium. For stock offers, the important determinants are: target?s price run-up, bidder?s return volatility, bidder?s systematic risk (beta), transaction cost (measured by a dummy variable indicating low share price), arbitrageurs? activity (measured by abnormal trading around merger announcement) relative size of the target and bid premium. In short, we find that we can use factors that are relevant to the probability of deal success (i.e., target?s price run-up, target termination fees, target resistance, arbitrageurs? activity, and relative size of the target), bid revision (i.e., target?s growth opportunity), potential loss when a deal fails (i.e., bidder?s systematic risk, bid premium) and transaction costs for risk arbitrageurs (i.e., bidder?s return volatility, low-priced shares) to develop a prediction model for risk-arbitrage spreads. Moreover, we create risk-arbitrage portfolios by comparing predicted arbitrage spreads with actual arbitrage spreads. We find deals whose actual spreads exceed the predicted spreads tend to be more attractive investments.