D DDD Bidder Gains and Losses of Firms Involved in Many Acquisitions Antonios Antoniou,* Dimitris Petmezas, and Huainan Zhao Durham Business School University of Durham Durham, DH1 3LB, UK March 2005 ABSTRACT We examine shareholders wealth effects (both in short- and long-run) of UK frequent bidders acquiring public, private, and/or subsidiary targets with alternative methods of payment between 1985 and 2004. We find that, in the short-run, bidders lose when acquiring public targets and gain when purchasing private and subsidiary targets. This result is robust after controlling for bidder’s book-to-market ratio (value/glamour), core-industry (diversified/non- diversified), and target origin (domestic/foreign). Our long-run evidence, however, reveals that acquirers experience significant wealth loss regardless of the target type acquired indicating that markets may overreact at the acquisition announcement. As a consequence we argue, in contrary to Fuller et al. (2002), that a fruitful conclusion on the wealth effect of bidders acquiring private and subsidiary targets can only be drawn under both short and long run investigation. JEL Classification: G11; G14; G34. Keywords: Mergers & Acquisitions, Corporate Takeovers, Frequent Bidders, Method of Payment, Public/Private/Subsidiary Targets, Abnormal Returns. D We would like to thank Abhay Abhyankar, George Alexandridis, Phil Arbour, John Doukas, Krishna Paudyal, and seminar participants at the Durham Business School for their useful comments on previous drafts of this paper. * Corresponding author. Tel: +44-(0) 191-334-5290; Fax: +44-(0) 191-334-5289. Email address: [email protected] 1. Introduction The examination of shareholders wealth effects (value creation or destruction) of Mergers and Acquisitions (M&A) is one of the most coveted research areas in finance. To date, a large amount of research has focused on examining the short-window stock returns earned by targets and bidders around merger announcements. The stylized fact emerging from this strand of studies is undivided in that target firm shareholders earn a significant and positive abnormal return in a few days surrounding the takeover announcements, a finding that is rather unsurprising given the hefty premiums paid to the targets. Acquiring firms, on the other hand, are found to break even while the combined entity (target and acquirer) earns a positive abnormal return around the announcement date2. Given these findings, a simple but interesting question arises: Do these observed abnormal returns solely reflect the expectation of future cash flows resulting from the takeover events? Hietala, Kaplan, and Robinson (2000) argue that acquisition announcement reveals not only the value of the acquisition itself but also the stand-alone value of the bidders, the potential synergies of the combination, and possibly the bidder overpayment. Hence, it is often impossible to isolate the above effects from the observed abnormal returns. Fuller, Netter, and Stegemoller (2002) apply a sophisticated research design to control for (much of) the information about bidder characteristics contained in stock returns at the acquisition announcement.3 They investigate the returns to US frequent bidders making five or more bids within a three-year time horizon. As they argue, the sample of frequent bidders 2 For evidence on acquirers’ short-run stock returns see, for example, Dodd and Ruback (1977), Asquith, Bruner and Mullins (1983), Dennis and McConnell (1986), Bradley, Desai, and Kim (1988), Franks and Harris (1989). For evidence of combined firms see, for example, Bradley, Desai, and Kim (1988), Mulherin and Boone (2000), Andrade, Mitchell, and Stafford (2001). 1 allows to hold bidder characteristics constant when examining the pattern of announcement returns4. In general, the authors conclude that bidders experience significant wealth loss when buying public targets, while earn substantial gains when private and subsidiary targets are purchased. This is, however, a premature conclusion as short-run event study conclusions rely strictly on the assumption of market efficiency. Nevertheless, it is possible that stock prices temporarily deviate from their fundamental values due to investors systematic over or under-reaction to acquisition announcements. In such case, serious doubts arise towards short-run window’s ability to distinguish real economic gains from market inefficiency. Accordingly, Healy, Palepu, and Ruback (1992) posit that: “From a stock price perspective, the anticipation of real economic gains is observationally equivalent to market mispricing”. This view indicates that, indeed, short-run systematic under- or over-reaction to an event has gradually become accepted in the literature. Fama himself, the father of the efficient market hypothesis, has recently conceded that stock prices could become “somewhat irrational”.5 In a nutshell, the voluminous literature related to behavioural finance emphasizes that results generated by short-run event studies need to be interpreted with further skepticism. We thus believe that Fuller et al.’s (2002) conclusion needs to be braced with certain caution. In this case, we argue that a complementary long-run analysis in this context is considered essential in order to reach a relatively thorough investigation of shareholders’ wealth effects. If the long-run results mirror the short-run findings, we can then be more confident to accept their short-run conclusions. However, if the short-run evidence is not supported by the long- 3 Fuller et al. (2002) is the first major attempt in examining takeover announcement returns of multiple bidders involved in acquisitions of public, private, and subsidiary targets with alternative methods of payment between 1990 and 2000. 4 Fuller et al. (2002, p. 1792) argue “Since we control for acquirer characteristics in that the same bidder will often choose to acquire targets with varying ownership status, and with different payment methods, we can examine the variation in acquirer returns as a function of these bid characteristics”. 5 “As two economists debate markets, the tide shifts. Belief in efficient valuation yields ground to role of irrational investors Mr. Thaler takes on Mr. Fama”. The Wall Street Journal, October 18, 2004. 2 run results, we can then cast doubt on whether Fuller et al.’s (2002) suggestion is economically sound and intuitive or merely a potential product of short-run market inefficiency. In addition, of course, such findings have not been tested in other counties apart from the US. We therefore conduct, in this paper, a UK study by examining the stock returns (both in short- and long-run) of frequent bidders that successfully acquired three or more public, private, or subsidiary targets using alternative methods of payment within a three-year period between 1985 and 2004. Our comprehensive sample constitutes of 4173 UK takeovers taking place over a 20-year period. A point that is worth mentioning is that a significant proportion of UK firms appear to engage in multiple acquisitions over this period (more than 40% of the entire population) while, most importantly, private targets and subsidiaries are major components of the UK takeover market (approximately 90%), a fact that very few studies have taken into account. In general, our results demonstrate that positive abnormal returns are present only in the short-run (i.e., the takeover announcements). Bidders gain when buying private or subsidiary targets and lose when purchasing public targets. This finding is fully consistent with Fuller et al. (2002). In addition, we add further evidence to the short-run study by taking into account bidder’s book-to-market ratio (value glamour), core-industry (diversified/non-diversified), and target origin (foreign/domestic). On the other hand, our long-run results show that bidders experience significant losses regardless of the type of target acquired. This finding implies that the stock market may overreact in the short-run and its prices are gradually corrected in the long run. Hence, our evidence raises a big question mark towards Fuller et 3 al.’s (2002) conclusion as the short-run economic gains (i.e., the reflection of the acquisition synergies) of buying private and subsidiary targets cannot be materialized in the long run. The remainder of this paper is organized as follows: Section 2 describes the data and the methodology. Sections 3 and 4 report and discuss the empirical findings. Section 5 concludes our analysis. 2. Data and Methodology 2.1. Data We examine a sample of successful takeovers by U.K. public companies that acquired both domestic and foreign targets, announced between January 1, 1985 and May 6, 2004. The sample acquisitions are drawn from the Securities Data Corporation’s (SDC) Mergers and Acquisitions Database while the period selected is driven by the total availability of the database and the definition of multiple bidder we have set (acquiring 3 targets within a 3-year period).6 The following criteria are used in selecting our final sample: 1. Acquirers are U.K. firms publicly traded on the London Stock Exchange (LSE) and have five days of return data around the takeover announcement and one to three year return data listed on the DataStream Database. 2. The acquirer completes three or more bids in any three-year window during the sample period. 3. The bidder acquires at least 50% of the target’s shares as a result of the takeover. 6 SDC is a commercial database that includes information on U.K. Takeover Bids since 1980. However, the first multiple bidder appears to do the first bid in 1985. 4 4. The target is a public, private, or subsidiary firm.7 5. The deal value is one million dollars or more.8 6. We omit financial and utility firms (following Fama and French 1992) for both bidders and targets. We also exclude clustered acquisitions where the bidder acquires two or more firms within five days in order to isolate the overlapping effect among the bids. Our final sample consists of 618 unique acquirers proceeding to 4173 bids. The full sample is then divided into three groups based on the method of payment for the acquisition, i.e., pure cash, pure stock, and combined. The combined payment sub-sample includes all acquisitions in which the payment method is neither pure cash nor pure stock. As we use Dimson, Nagel, and Quigley (2003) UK 3-factors to account for UK book-to-market peculiarities, we include in our long run analysis bids carried out between 1985-1998 for 3-year analysis (2607 firms), bids up to 1999 for 2-year analysis (2995 firms) and takeovers from 1985-2000 (3383 firms) for 1-year analysis respectively. Table 1 presents the summary statistics for acquirers making multiple acquisitions and their targets. Panels A, B, C, and D report the annual mean and median acquirer and target size for all bids, only public bids, only private bids, and only subsidiary bids, respectively. The mean and median size for each acquirer and each target is the firm size at the year the deal was announced. The acquirer’s and public target’s market capitalization equals the price per share one-month prior to the bid announcement times the number of common shares outstanding For private and subsidiary targets, the firm size is measured as the deal value of the bid. The 7 We examine subsidiary targets, as they are one of the three main categories of the market for corporate control. All subsidiary targets are unlisted companies after checking the Target Public Mid Code from the SDC database. 5 final row of each panel presents the mean and median size for each unique acquirer and target (i.e., counted only once for each firm). Accordingly, for the entire sample in Panel A, the mean (median) size of the acquirer is 488 million pounds (77 million pounds) for 618 unique acquirers, while for 4173 unique targets the mean (median) size is 37 million pounds (6 million pounds). Table 1 also presents a general upward trend in both merger activity and size of acquisitions for public, private, and subsidiary targets, dropping slightly by 2000.9 Panels B, C, and D present the distribution of mean and median size of firms based on target ownership status, i.e., Public (Panel B), Private (Panel C), and Subsidiary (Panel D). Panel B illustrates that the mean (median) size is 159 million pounds (42 million pounds) for 195 unique public targets. Panel C shows that the private targets mean (median) size is much smaller than that of public targets, 15.8 million pounds (4.75 million pounds) for 2459 unique private targets. Panel D reports that the mean (median) size of 1519 unique subsidiary targets is also smaller (56 million pounds (8.7 million pounds)) than that of public targets. In sum, Table 1 shows that the size of public acquisitions is significantly greater than private and subsidiary ones. 2.2. Methodology We calculate Cumulative Average Residuals (CARs) for the five-day period [-2, +2]10 around the announcement date supplied by SDC. More specifically, we estimate the abnormal returns by using a modified market-adjusted model: 8 We employ a one million dollars cut-off point to avoid results being generated by very small deals. Similarly, studies like Fuller, Netter, and Stegemoller (2002), Moeller, Schlingemann, and Stulz (2004) in the US use a cut-off point of one million dollars. 9 Despite the decrease in number of deals after 2000 the total value of transactions has significantly increased. As an indication of the latest data evidence, the total value of takeovers in the first quarter of 2004 is almost double than of first quarter of 2003. 10 We choose the five-day period because Fuller et al. (2002) find that a five-day window around the merger announcement given by SDC is wide enough to capture the first mention of a merger every time for a sample of about 500 announcements. 6 AR = R - R (1) it it mt where, R is the return on firm i [ln (P )- ln (P )] and R is the value-weighed market it t t- 1 mt index return (i.e., the FT-All Share measured as the first difference of the log of the Market Index). We then calculate the CAR for each firm over the 5-day event window. The t- statistics are estimated using the cross-sectional variation of abnormal returns.11 It is obvious that in our long run analysis a subsequent acquisition will occur within less than 36 months after a preceding acquisition, since our sample consists of multiple acquirers. We therefore use Calendar Time Portfolio Regressions (CTPR) to sidestep the problem of aggregating daily returns to obtain a long-term return, and allow inferences that are not biased by cross-sectional dependence.12 In each calendar month, a portfolio is formed by including all stocks with an acquisition event during the past 12, 24, or 36 months. The portfolio is rebalanced every month by including new event firms executed a transaction in the previous month and dropping the ones whose latest acquisition event falls out of the one to three-year holding period. The average monthly abnormal return during the one to three- year post-event period is the intercept from the time-series regression of the calendar portfolio return on the Fama and French three-factor model. The FF 3-factor model are 11 We do not estimate market parameters based on a time period before each bid, since for frequent acquirers, there is a high probability that previous takeover attempts would be included in the estimation period, hence making beta estimations less meaningful. Additionally, it has been shown that for short window event studies, weighting the market return by the firm’s beta does not significantly improve estimation. (Brown and Warner, (1980)). 12 Cross-sectional dependence caused by overlapping observations leads to downwards-biased standard errors and therefore causes t-statistics to be biased upwards. In addition, according to Mitchell and Stafford (2000), due to the number of firms being different for each month, heteroskedastic residuals are likely to be present when regressing calendar time average portfolio returns in excess of the risk free rate against the factors of an asset-pricing model. Hence, we use Andrews (1991) heteroskedasticity and autocorrelation consistent standard errors so as to realistically assess the validity of our results. 7 estimated by using the UK 3-factor of Dimson et al.’s (2001) to account for the UK B/M ratio peculiarities :13 R - R =a +b (R - R )+s SMB +h HML +e (2) pt ft i i mt ft i t i t it whereR is the average monthly return of the calendar portfolio, R is the monthly risk free pt ft return, R is the monthly return of the value-weighted market index, SMB the value- mt t weighted return on small firms minus the value-weghted returns on large firms, and HML t the value-weighted return on high book-to-market firms minus the value-weighted return on low book-to-market firms. In addition, b , s and h are the regression parameters and e is i i i it the error term. The a (intercept) is interpreted as the average of the individual, firm-specific intercepts. 3. Empirical Results for the Short-run Analysis 3.1. Bidder Abnormal Returns by Target Type and Method of Payment In Table 2, Panel A, we present five-day CARs for the full sample classified by target public status and method of payment. For all bids, the CAR is positive (0.74%) and statistically significant at 1% significance level. When focusing on public targets we obtain a significant negative CAR of –1.95%. This is consistent with UK studies of Firth (1980), Draper and Paudyal (1999, 2004), Sudarsanam Holl and Salami (1996) and Sudarsanam and Mahate (2003) who find negative and significant bidder abnormal returns surrounding the announcement. When we further differentiate on the basis of method of payment CARs are, surprisingly, all negative irrespective of the mode of financing used, with stock payment 13 Dimson, Nagel, and Quigley (2003) use different breakpoints to those of Fama-French (1993) to construct Size and Book-to-Market portfolios mainly due to size and B/M ratio being negatively correlated in the UK and large firms (small firms) being concentrated in the low (high) BE/ME quartile. 8 generating the largest negative and highly significant CAR of –4.05%. This is consistent with Myers and Majluf (1984) hypothesis and Moeller, Schlingemann and Stulz (2004) finding, suggesting that the greater information asymmetry associated with stock payments leads to more negative performance.14 For cash payments CARs are still negative but marginally significant.15 This can be attributed to higher offers (premium) for cash exchanges to compensate target shareholders for the immediate payment of taxes. For private targets, CARs are positive (0.73%) and in most cases significant. This is in line with Chang (1998)16 and Ang and Kohers (2001) who document substantial gains for acquisitions of privately held firms. According to the explanations offered, private firms exhibit concentrated ownership, which leads to less agency conflicts, alleviate the public pressure from outside investors and therefore have the opportunity to avoid hubris-motivated takeovers. The nature of private targets itself ‘auto-protects’ the acquiring company from managers’ empire building incentives, since such acquisitions do not offer, in most cases, the prestige they pursuit. Private firms confront the problem of liquidity, meaning that they cannot be bought and sold as easily as public firms. Therefore, in order to create an attractive image for their company and a plausible incentive as a profitable investment opportunity for potential acquirers, they offer their shares at a discount (liquidity). This strategy of liquidity 14 Myers and Majluf (1984) argue that the premise of information asymmetry raises the proposition that managers with private information that their firm’s shares are overvalued offer these shares as consideration in takeover bids. Outside investors, recognizing the adverse selection problem, consequently revise their estimate of the offer’s value downwards, a plausible explanation for the negative performance of stock deals. Moeller et al. (2004) test this hypothesis and find that higher growth uncertainty due to informational asymmetry in the case of equity financing leads to more negative returns. 15 Better performance of bidders used cash financing is consistent to the literature. See for example, Travlos (1987), Fishman (1989) and Martin (1996) who find that bidders making cash offers have greater abnormal returns at the bid announcement than do those making stock offers. 16 Chang (1998) finds positive abnormal returns for stock offers to private targets and attributes this finding to the fact that privately held firms are not obliged to release value relevant information to the public, thus generating a high cost of obtaining information (information hypothesis). Such cost is very likely to be associated with larger returns for acquiring firms since they capture a greater proportion of the expected gains, particularly when there are only few firms with which the target may reap synergistic gains. It is finally likely that positive returns are attributed to the limited competition hypothesis that predicts high likelihood of underpayment. 9