The profitability of risk arbitrage in the related literature applies two lines of analysis:
time series analysis and cross sectional analysis. Time series analysis involves the construction of a risk arbitrage index portfolio, and focuses on investigating abnormal returns.
Baker & Savasoglu (2002) reported annual abnormal returns of 7.2% and 10.8% on risk arbitrage for stock swap and cash offers. Mitchell & Pulvino (2001) analyzed stock swap and cash offers during the period 1963-1998, and reported that, after controlling trading restrictions, an annual abnormal return of 4% for cash offers was realized. On the other hand, cross sectional analysis is used to explore certain factors that help to explain the variation on risk arbitrages returns. Jindra & Walkling (2001) examined cash offers on risk arbitrageurs during the period 1981-1995, and found that the links of arbitrage spread returns were negative with the magnitude of price revision, and positive with duration. Baker & Savasoglu
4
(2002) tested the cross-sectional implications of limited arbitrage2, and found that risk arbitrage yielded abnormal returns for cash offers, generating annual returns of 10.8%.
Those previous papers about risk arbitrage did analysis on the general mergers and acquisition or cash offer acquisitions in the U.S, but less of them just focus on the leveraged buyouts (LBOs), which have had a great impact on the acquisitions in decades. In this paper, we just concentrate on the leveraged buyouts deals. Besides, we also extend the previous research on risk arbitrage profitability by exploring how information asymmetry influences the profitability of risk arbitrage. Branch & Yang (2006) reported that information asymmetry in an acquisition is determined from the spread between the offer price and market price of target firm’s stock on the announcement date. Their findings indicated that the attributes of cash offers may entice arbitrageurs into the bidding process, thereby pushing up the market price of a target firm’s stock on the announcement date, and consequently reducing risk arbitrage profits3. This fact also provides us with the important implication that higher spread returns could result in higher risk arbitrage returns, and investors were certain of spread returns on the announcement day. Therefore, for leverage buyouts, most of target firms reverse their stock price during the period of deals; this attribution is quite different from the general acquisitions. Investors can earn spread returns in a successful deal, but they may get great losses in a failed deal. The primary objective of our research is to investigate the relationship between spread returns and key variables such as bid premiums and holding
2 Limited arbitrage means that arbitrageurs’ risk-bearing capacity is constrained by the deal completion risk and the target size. Most risk arbitrageurs are passive investors, who do not influence the acquisitions process and outcome. They buy the target firms’ shares and face completion risk. More details are given in “Research design”.
3 When more arbitrageurs are involved in buying the shares of target firms it can boost the price of the target firms’ shares, and spread returns will decrease. Investors usually buy shares at the closed market prices on the announcement date, as a result, they earn relatively lower spread returns and total risk arbitrage returns.
5
period (duration). These determinants actually affect risk arbitrage returns. We explain them as follows.
2.1 Target size
A firm’s size has two possible effects. First, most large firms can boast high liquidity. If we hold risk and arbitrage capital constant, larger dollar amounts sold lead to higher returns.
However, smaller firms may lead to higher risk arbitrage returns, because arbitrageurs are probably interested in smaller firms where transaction costs are lower. Some authors found that the risk arbitrage returns tend to be positively related to the target firm’s size (e.g. Jindra
& Walking, 2001; Baker & Savasoglu, 2002). In this paper, a target firm’s size is its target equity market value, which is calculated by multiplying the total number of target shares outstanding by the target stock price at the announcement date. We also take one relevant variable into consideration: the price-to-book ratio (P/B), which represents a growth measurement for a firm. Furthermore, P/B anomalies due to mispricing are well known in finance. If target firms have a lower P/B, it will result in higher risk arbitrage returns. P/B is a new and unique factor we discuss, because most previous research investigates the relationship between firms’ size and arbitrage returns.
2.2 Bid premium
Bid premium is the ratio of the closing price one day prior to the announcement day scaled by the offer price. The positive link between bid premium risk arbitrage returns is straightforward. Most authors reported that risk arbitrage returns tend to increase with the magnitude of the bid premium (Jindra & Walking, 2001; Mitchell & Pulvino, 2002; Baker &
Savasoglu, 2002).
2.3 Liquidity
Some research on the link between liquidity and risk arbitrage returns is inconsistent.
Agrawal, et al. (2004) reported that the higher bid-ask spread is, the greater the proportion of investors, including internal shareholders, hedge funds and other institutional investors, who
6
are likely to have better information than general investors. If these informed investors hold a higher proportion of the target firm’s shares than general investors, acquirers would be inclined to make a higher offer price (a higher offer price means larger spread returns).
Therefore, targets firms with less liquidity may tend bring higher risk arbitrage returns.
However, Jindra & Walkling (2001) found that spread returns are significantly negatively related to a target firm’s liquidity, which is measured as abnormal volume (the ratio of event trade volume relative to pre-announcement volume). Chen & Kan (1995) did not find any reliable relationship between excess arbitrage returns and bid-ask spreads. In this paper we infer that most informed investors will sell off their shares during deals, and that general investors who buy less liquid target firms may sustain considerable losses. To determine the effect of liquidity on risk arbitrage returns, we needed to replace trade volumes with bid-ask spreads. Our observations, however, did not have sufficient trade volume information in the CRSP database, because our sample consisted of large NASDAQ-listed stocks.
2.4 Investment cost
Two components involved in all costs are transaction costs and holding costs. In practice, direct transaction cost include bid-ask spreads and brokerage commissions. Ali, et al. (2003) used historical bid-ask spread as an additional measure of direct transaction cost and bid-ask spread had been indicated in the previous section. In addition, for successful offers, arbitrageurs generally generate higher prices on the completion day, and the holding period is usually more than half a year; brokerage commissions could be ignored. On the other hand, investors buy their target shares and then hold them to the completion day. This holding cost, which is also an opportunity cost of any other investments, should be proportional to the duration of the offer (Ali & Hwang & Tromble, 2003; Jindra & Walkling, 2004).
2.5 Risk
7
Risk is composed of systematic and unsystematic risk. It is important to assess the relationship between risk and the profitability of leveraged buyouts. First, some authors found that the beta of private equity funds is not a significant driver of performances (e.g. Zollo &
Ludo, 2006; Ljunqvist & Richardson, 2003). Their evidence showed that the relationship between stock systematic risk and stock performance is weak. If our research supports this evidence, it indicates that market performance does not affect target stock price movement during the period of risk arbitrage. Second, a cross section analysis on limited risk arbitrage revealed that idiosyncratic risk is a determinant of expected returns (Baker & Savasoglu, 2002). For systematic volatility, arbitrageurs got compensated and could eliminate it by hedging; however, idiosyncratic risk cannot be hedged and were not well diversified. Ali, et al.
(2003) reported idiosyncratic volatility of stocks in the portfolio is greater of concern than systematic risks. They computed the variance of the residual term to obtain the idiosyncratic risk or unsystematic risk in the market model.