The role of the target listing status in shareholder wealth effects for acquiring firms

Transcription

1 Master Thesis Finance The role of the target listing status in shareholder wealth effects for acquiring firms J.C. Dekker

2 Master Thesis Finance The role of the target listing status in shareholder wealth effects for acquiring firms J.C. Dekker Abstract This master thesis examines the determinants of cumulative abnormal returns (CARs) earned by acquiring firms in mergers and acquisitions (M&A). The focus of this thesis lies in the effect observed amongst others by Fuller, Netter, and Stegemoller (2002) that acquirers of public targets tend to underperform compared to acquirers of unlisted targets, either subsidiary or private firms, in terms of abnormal returns realised upon the announcement of the acquisition. I find positive wealth creation in M&A for the entire sample and for the subsamples of private and subsidiary target companies. For the subsample of public target companies wealth creation for acquirer shareholders is not found in the data. Evidence for this target listing status effect is persistent and present across various subsamples. The effect however is not identified for periods of economic crisis; during the dot-com bubble collapse and the subprime mortgage crisis M&A activity doesn t show signs of a presence of the target listing status effect. Keywords: Mergers and acquisitions in Europe, takeovers, target listing status effect, shareholder wealth effects Student Number Faculty School of Economics and Management Department Finance Thesis Supervisor Dr. F. Feriozzi Chairman of Graduation Committee Drs. J. Grazell Date 9 February, 2012 I am thankful for the guidance and support from my supervisor Dr. F. Feriozzi and the helpful comments made by P.M.M. Maas and N.T. van Vlerken. 2

3 Table of Contents I. INTRODUCTION... 5 II. CURRENT STATE OF THE LITERATURE... 8 A) Acquiring firm characteristics... 9 Size... 9 Overconfidence... 9 Tobin s Q Misvaluation Growth opportunities Leverage Cash holdings Ownership Corporate governance B) Deal characteristics Deal financing Deal structure Investment bank influence C) Target firm characteristics Cross border effect Within industry acquisitions Target listing status D) Relevance of research III. SAMPLE SELECTION, DATA, AND METHODOLOGY A) Sample and data sources B) Methodology IV. RESULTS A) Analysis of cumulative abnormal returns Persistence through time Persistence across nations of incorporation CARs by deal and acquirer characteristics

4 B) Multivariate regression analysis on determinants of CARs Regression variables and setup Regression results C) Subsample analysis Target listing status and matters of payment Western vs. Eastern Europe Pre-Crisis vs. Post-Crisis D) Correlations E) Robustness checks V. CONCLUSION VI. LIMITATIONS AND RECOMMENDATIONS REFERENCES APPENDIX

5 I. Introduction On August 15, 2011 Google Inc. announced the acquisition of Motorola Mobility Holdings Inc. Google was announced to purchase Motorola for $40.0 per share, a 63.0% premium to the prevalent share price of Motorola 1. The offer implied a purchase price of $12.5b. During the window of five days surrounding this announcement (two days before the announcement until two days after the announcement) the Google share (NASDAQ:GOOG) decreased by 10.2% or alternatively the market value of Google declined with $18.2b. During the same period the NASDAQ increased with 0.7% making market movement unlikely to have had any negative effect on the announcement return of the acquisition. Furthermore, there was no other company specific news on Google or the internet industry during the event window. Then how is it possible that the decrease in market value of Google was actually larger than the amount they announced to spend on the acquisition? To look at the event from a standalone point of view would mean that the management would have been better of handing out the money to charity or just make a bonfire out of it for that matter. Of course mergers and acquisitions do not result in wealth destruction to the shareholders of the acquiring company in all cases. A good example of a profitable acquisition is the purchase of Hollandsche Beton Groep (HBG) by Royal BAM Group in HBG was a subsidiary firm of Grupo Dagrados S.A. upon the announcement of the deal on June 11, After the payment of 710m in cash all shares were transferred to Royal BAM on October 25, Upon the announcement of the acquisition in June the equity market reacted positively and the stock (AMS:BAMNB) achieved a cumulative abnormal return (CAR) of 23.6% during the five day event window surrounding the announcement of the acquisition. Events like the above described acquisition of Motorola by Google raise many questions. For one, why did the management go through with this acquisition when it did not create any value? To extrapolate from this example the question can be asked whether mergers and acquisitions create value as a whole. The answer to this question remains ambiguous. Many studies conclude that the wealth effects for the shareholders of the target companies are clearly present 3. However, conclusions from recent research tend to vary and are not unanimous on the aggregate wealth effect for acquiring firm s shareholders. The combined wealth creation for target and acquirer shareholders seems positive but this is not without dispute. Acquisitions take place between all sorts of different companies; much research has been dedicated on determining common characteristics between deals that affect abnormal returns. These characteristics can be specific to the target company, the acquiring company, or the deal itself. The 1 Source: 2 Source: SDC Mergers and Acquisitions database. 3 Empirical findings on target CARs range from +9% (Goergen and Renneboog (2002)) to +31.5% (Higson and Elliot (1998)). 5

6 above described examples of M&A failure and M&A success bear many of the identified factors that affect the CARs of the acquiring firm. For example the acquisition of Motorola implied that Google intended to enter into a new market or in other words the acquisition was an industry diversifying acquisition 4. On the other hand the acquisition of HBG, a concrete manufacturing company, by building company Royal BAM resulted in clear synergies and enabled Royal BAM to achieve a strong position in the West-European construction industry. The theory behind the industry diversifying effect and other theories on determinants of CARs will be discussed elaborately in the literature review section. Next to this industry effect the example also clearly illustrates the effect that will be of interest to this master thesis. The bad acquisition of Motorola by Google is an acquisition where the target company is a publicly listed entity. For the good acquisition of HBG by Royal BAM the target is unlisted, in this case a subsidiary firm. Although these are one of a kind examples, and one would be able to find lots of deals where the reverse roles are found, an effect has been identified that acquisitions where the target company is a publicly traded company result in lower CARs for the acquirer compared to acquisitions where the target company is unlisted, i.e. a privately owned or subsidiary firm. This is found amongst others by Hansen and Lott (1996), Chang (1998), Fuller et al. (2002), and Faccio, McConnell, and Stolin (2006). Researchers have been trying to identify the fundamental factors that might explain this target listing status effect. However, the economic forces that drive this effect remain inconclusive. In this master thesis the focus will be on determining the target listing status effect in the data for European countries. The events used are mergers and acquisitions with European acquirers. Although mergers, the combination of two companies into a new company, and acquisitions, a takeover of a company by the acquiring company, differ in definition they do not differ in economic impact on the acquiring company. Therefore, the terms merger, acquisition and takeover are used interchangeably in this thesis. Previous research has not yet identified this target listing status effect for various fields to my knowledge. First, research conducted up until now mainly focuses on acquisitions with an acquirer from the United States, the research that has been done on European acquisitions focuses on the Western European countries. To expand the research Eastern European countries are added to the sample. It is interesting to see if the target listing status is present for this geographical region as both the cultural and economic characteristics of Eastern European countries differ sharply from those in the Western European and Northern American countries. Secondly the recent decade has presented us with two periods of economic downturn and in this research I will try to investigate whether established factors affecting the abnormal returns from an acquisition are still valid in such periods of crisis. In other words: does the target listing status effect still persist during the dot-com crisis and the subprime mortgage crisis. Finally, I will try to identify additional factors 4 Acquirer and target are classified as different industry companies when the first three digits of the SIC codes are different, these are 3663 and 7370 for Motorola Mobility and Google respectively. 6

7 that explain cumulative abnormal returns earned by acquiring companies, this by means of analysis of financial ratios for the acquiring company. These financial ratios will be based on reporting items from the acquiring company and connect various theories from other financial fields to the field of merger and acquisition research. After imposing various data limitations the resulting sample contains 10,252 events making this one of the largest studies on European M&A activity. The majority of acquisitions is done by an acquirer from the United Kingdom (U.K.), and the majority of target companies also is incorporated in the UK. The data illustrates the cyclicality of M&As, deal volumes follow a wave like pattern over the sample period with low volumes during times of economic downturn. From the analysis on the data in the sample for this thesis it can be concluded that on average M&A create value. A positive equally weighted CAR of 1.44% is found over an event window of five days surrounding the announcement of an acquisition. It also shows that the negative effect of a purchase of a publicly listed target is widespread, these acquirers underperform in all subsamples, years, and countries. On average cumulative abnormal returns are found to be roughly 2.30 percentage point lower for public target acquisitions. The effect is determined using both univariate and multivariate analysis, OLS regressions and fixed effect regressions. All show that this effect is highly significant. Control variables that control for previously identified factors that affect the CARs are included. These provide support for the size effect identified by Moeller, Schlingemann, and Stulz (2004), larger acquirers appear to be worse acquirers. The means of payment effect is also confirmed by the analysis; cash financing has a positive effect on the CARs earned by the acquirer, for stock financed deals no significant effect is detected. The size of the deal is marginally significant at best in absolute terms, however in relative terms to the size of the acquirer this variable becomes highly significant. This means that as the importance of the deal to the acquirer increases the acquirer tends to perform better. Financial reporting ratios from acquiring firms seem to be playing a relatively small role at best in explaining CARs. Leverage and operating performance are found to have a positive effect on CARs. Furthermore, periods of financial turmoil in the sample are examined and it can be concluded that years of crisis, be it the dot-com bubble collapse or the subprime mortgage crisis, seem to disrupt the target listing status effect, while other effects remain significant during these periods the negative effect of an acquisitions of a public target becomes insignificant. For Eastern European acquirers not only the acquisitions of public targets are associated with lower CARs but the acquisitions of subsidiary targets as well, or put differently only acquisitions of private targets result in higher CARs. The setup of this thesis is as follows; section II describes the current state of the literature and will elaborate on the various effects found to affect CARs. Section III describes the methodology used to analyse the returns of the acquiring firms and how and where I selected the data for this research. Section IV presents the results from the analysis of the data using both univariate and multivariate testing; these will illustrate the empirical findings on the determinants of cumulative abnormal returns 7

8 for the shareholders of acquiring firms. Section V and VI will present concluding remarks, limitations, and recommendations. II. Current state of the literature Shareholder value creation in corporate takeovers is one of the most discussed topics in the financial literature. The first distinction that needs to be made is how to define value creation? The most apparent measure of value creation is value added for the shareholder, hence an increase in returns for the shareholders of the company upon the announcement of the takeover. Here it is important to consider the effect of the announcement and not the actual merger taking place, since markets are considered to be efficient new information like the announcement of a merger will be incorporated in the share price immediately. For target firms it is quite apparent that the shareholders gain from a takeover, acquiring firms have to pay a premium over the current market price to convince current shareholders or management it is wise to tender the shares. Average short term wealth effects for target company shareholders range from 9% (Goergen, and Renneboog (2002)) to 31.5% (Higson and Elliot (1998)). When considering the acquiring firm abnormal returns the current literature becomes more ambiguous. Acquiring firms seem to overpay quite often. Moeller et al. (2004) start their paper with identifying that: The equally weighted abnormal announcement return is 1.1%, but acquiring-firm shareholders lose $25.2 million on average upon announcement (pp. 201). This observation shows that the measurement method used for the announcement returns already influences the answer to the question whether acquisitions are profitable for acquiring firms or not. In another paper written by Moeller, Schlingemann, and Stulz (2005) they further examine the effects of large loss deals, deals with shareholder wealth losses exceeding $1 billion. They find that a relative small number of acquisitions is responsible for large losses and creates negative announcement returns for multiple years in their sample period ( ). For 4,136 deals from 1998 through 2001, of which 87 are large loss deals, they observe the following: The aggregate wealth loss associated with these acquisitions (large loss acquisitions) is $397 billion, while all other acquisitions made a total gain of $157 billion. The large loss deals represent only 2.1% of the 1998 to 2001 acquisitions, but they account for 43.4% of the money spent on acquisitions. Moeller et al. (2005) pp The identified phenomenon is especially prevalent during the technology boom ( ). So for a large group acquisition, announcements do result in positive average abnormal returns, but due to a few large loss acquisitions (just over 2% of the total observations) the conclusion can be drawn that on average acquisitions result in negative announcement returns for the acquirer during this period. Previous studies are divided in their conclusions about CARs for the acquiring companies. 8

9 One part of the studies report negative announcement returns for the acquirers (e.g. Servaes (1991), Andrade, Mitchel, and Stafford (2001), Kuipers, Miller, and Patel(2003)). Others find insignificant differences in returns (e.g. Healy, Palepu, and Ruback (1992), Mulherin and Boone (2000), Lang, Stulz, and Walking (1991)), and there are the studies that find positive announcement abnormal returns (e.g. Asquith, Bruner, and Mullins (1983), Schwert (2000), Loderer and Martin (1990)). What are the factors influencing the returns to the shareholders of the acquiring firm? What kind of firms or acquisitions tend to perform better than others? In the following section previously identified factors are subdivided in three different areas of influence on the performance of an acquiring firm: Acquiring firm characteristics, Deal characteristics, Target firm characteristics. A) Acquiring firm characteristics Size Various acquiring firm characteristics have been well documented in the literature. In their research Moeller et al. (2004) find that small firms doing acquisitions tend to be better performing post-acquisition compared to large firms. This size effect is persistent across time and regardless of the deal financing or the listing status of the target. On average Moeller et al. (2004) find that large firms have a two percent lower cumulative abnormal return at the announcement of an acquisition than the small firms in their sample. Their findings are based on the fact that the firms in their sample earn an equally weighted positive announcement return but in terms of total dollar value there is a negative return. Overconfidence Roll (1986) and Malmendier and Tate (2005), provide a potential argument for the bad picks of larger companies. Following their reasoning management of larger firms may experience more hubris, their overconfidence exceeds that of their counterparts at smaller firms. Therefore, managers of larger firms might pay too much in an acquisition. These more behavioural approaches look at the risk attitude by the decision makers in firms. Malmendier and Tate (2005) measure overconfidence by looking at the exposure of CEOs to company specific risk in their own stock holdings, when these CEOs fail to limit this company specific risk by diversifying the exposure in their portfolio when this is possible they are qualified as overconfident. Malmendier and Tate (2005) find in fact that overconfident managers make worse investment decisions. Furthermore, overconfident managers are more dependent on internal funds, as external funds are viewed as too expensive. This causes cash rich firms with overconfident CEOs to spend internal funds too freely and firms with less internal funds but with overconfident CEOs tend not to undertake profitable investments due to a need of external financing. This could be related to the research by Roll (1986) where it is stated that managerial hubris causes firms to overpay in acquisitions. Managerial hubris is mainly present when 9

10 the management of a firm has had to struggle to get to the top, the case for larger firms and when ownership of the firm is dispersed. Tobin s Q Tobin s Q, the market value of the firm divided by the replacement value of the firm s assets, is a measure of the current valuation of the firm by the market compared to the current value of the assets a firm holds on its balance sheet. A Tobin s Q greater than one means that markets value the firm as having a higher worth than the worth of its assets. It is suggested that firms with a high Tobin s Q are performing well, these firms are able to add value to their assets. Therefore Tobin s Q is used as a measure of firm performance in models analysing CARs. Acquirers with a higher Tobin s Q, ergo better performing firms, are found to have higher announcement returns by Lang et al. (1991), for tender offer acquisitions, and by Servaes (1991) for public firm acquisitions, however Dong, Hirshleifer, Richardson, and Teoh (2002) and Moeller et al. (2004) do not find any evidence for this effect. The measure Tobin s Q is not only used as a proxy for firm performance, it is also used to measure growth opportunities of the firm e.g. Booth et al. (2001) this could explain the ambiguous effect found in previously mentioned research. The effect an acquisition can have on the perceived growth opportunities of the acquiring firm is explained below. In addition, Tobin s Q can vary over companies with certain characteristics without saying anything about either the growth options or the firm performance. For example, when a company has large intangible assets which are not measured in its book value Tobin s Q will have a higher value than for a company with mainly tangible assets, all else equal. Misvaluation Dong, Hirshleifer, and Teoh (2002) investigate whether or not the takeover market is driven by misvaluation of the acquiring company by the stock market. They hypothesize that overvaluation of the stock of the acquiring company causes them to acquire lower valued targets. They find evidence that acquirers have higher valuation compared to the targets and highly valued acquirers are more likely to finance a deal with stock than with cash, they pay more and they earn lower announcement premia. This indicates that the shares of the acquiring company are in fact overvalued by the market and management uses these overvalued shares to finance the acquisition. Rhodes-Kropf and Viswanathan (2004) support this theory by stating that waves of cash and stock financed transactions can be explained by the valuation of a company at a given point in time. They show that valuations can cause merger waves even when there are no fundamental changes like new regulation or innovation. This is contradictory to Harford (2005) where it is concluded that merger waves are caused by technical, regulatory, or economic shocks provided that there is enough liquidity in the market. This research rejects the market timing hypothesis where misvaluation would drive the merger waves. 10

11 Growth opportunities McCardle and Viswanathan (1994) and Jovanovic and Braguinsky (2002) argue that negative returns for acquiring firms are not necessarily caused by bad acquisitions in terms of overpayment by management. They note that acquisitions can improve efficiency and still cause negative announcement returns. The argument they provide for lower returns for the acquiring firm is the growth opportunities signalling effect from acquisitions, meaning that firms send a signal to investors when they announce an acquisitions that organic growth is no longer possible and acquisitions are the only way left to achieve promised growth. Investors adjust their valuations based on these lower growth prospects, hence the lower valuation in the stock market. This could explain why firms with high Tobin s Q values, hence with high growth opportunities, leading up to the announcement of an acquisitions do not achieve positive CARs. Leverage Maloney, McCormick, and Mitchell (1993) find that firms with high leverage are outperforming their low levered counterparts in acquisitions. Higher levels of debt have a constraining effect on the firm and its management. This ensures that the acquisitions undertaken by highly levered firms yield higher returns than with firms that can spend their cash freely. Malony et al. (1993) find that firms that undergo a restructuring following an acquisition achieve higher returns. Highly levered acquisitions are often hostile acquisitions, a reason for this is that management is not keen on high levels of debt since they reduce the managerial freedom in corporate spending. This supports the literature that states that high leverage improves managerial decision making. Cash holdings The agency theory put forward by Jensen (1986) strengthens describes principal agent problems where the shareholder is the principal with the management as its agent that should act on behalf of the principal. However, there can be conflicting interests between management and shareholders. It is argued that management could engage in empire building and would rather spend excess corporate funds on value destroying acquisitions than to redistribute them to the shareholders of the company. These problems are naturally found in firms with large amounts of cash that management controls. Correspondingly, cash rich firms are found to be poorer performing in acquisitions. Harford (1999) finds that these cash rich firms undertake more acquisitions, are poorer performing in terms of cumulative abnormal returns upon announcement of the acquisition, and furthermore they are more likely to acquire companies from outside their industry. Ownership As agents, management seems to be playing a negative role in the returns associated with acquisitions. On the other hand, when management does not only take the role of the agent but also of the principal, by managerial share ownership, the conflict of interests should be mitigated. Lewellen, 11

12 Loderer, and Rosenfeld (1985) look at managerial share ownership as a determinant of CARs upon the announcement of an acquisition. They conclude that there is a positive relation between the percentage ownership of management in the company and the performance of the company in acquisitions. Sudarsanam, Holl, and Salami (1996) look at the ownership structure of the acquiring firm and its influence on the abnormal returns. They find that the presence of a large shareholder like an institutional investor in the shareholder base of either the acquirer or the target company has a negative influence on the CARs during the announcement of the acquisition. When the acquiring firm has a toehold in the shareholder base of the target this reduces the gains for the target from the acquisition. Corporate governance Martinova and Renneboog (2006) examine the European takeover market and pay special attention to the level of corporate governance in different European countries. The level of corporate governance regulation differs strongly between the United Kingdom, Scandinavia and Continental Europe. According to their research a high standard of corporate governance regulation has a positive influence on the announcement abnormal returns since there is more shareholder protection. They also discover a spill over effect of corporate governance standards when target and acquirer have different standards in their country of origin. When the standards of corporate governance regulation are higher for the acquiring firm this is passed on to the target firm which then increases announcement returns. Masulis, Wang, and Xie (2007) examine the influence of corporate governance in terms of takeover measures on the announcement returns. They find that acquiring firms with more antitakeover measures achieve significantly lower returns than the firms without those measures. The antitakeover measures prevent the market for corporate control to play a disciplinary function on the management of the acquiring firm. B) Deal characteristics Apart from the characteristics of both the target and acquiring firm the orchestration of the deal plays a role in determining the announcement returns as well. Deal financing The matter of payment in a merger or acquisition is found to be a highly significant factor in explaining differences in shareholder CARs. Fuller et al. (2002), Goergen, and Renneboog (2002), and Chang (1998), amongst others, find strong evidence that cash deals appear to be rewarded with greater returns following the announcement. This is driven by the equity signalling effect (Myers and Majluf (1984)) where the fact that a firm finances the deal with equity will indicate that the firm believes its equity is overvalued. Since management has more information on the firm than investors, these investors re-evaluate their investment in the acquiring firm and this is reflected in the share 12

13 prices. Travlos (1987) shows that differences in cumulative abnormal returns between all cash and all stock financed deals are significant and irrespective of the deal structure used. Deal structure The style of the buyer s approach to the target is influenced by the degree of entrenchment of the target s management. Mergers typically have a friendly character, negotiated between the top managements of buyer and target firms. Hostile bids, on the other hand, are structured as take-it-orleave-it proposals made directly to the target firm shareholders, and are often viewed by the target management as coercive. Theoretically, the value gains from hostile takeovers should stem from the replacement of management where apparently management is not performing in the best interest of the shareholders. The rationale behind friendly mergers from a value perspective is the realisation of synergies in the combined firm. Schwert (2000) does not find a difference in shareholder returns for the announcement of hostile or friendly acquisitions. However, several studies report larger announcement returns to bidders in tender offers than in friendly negotiated transactions, with successful bidders in hostile takeovers estimated to earn positive abnormal returns of 2 percent to 4 percent, by Jensen and Ruback (1983). The higher returns from tender offers may reflect bargain prices as well as the expected economic benefits from replacing management and redirecting the strategy of the firm. On the other hand, Healy, Palepu, and Ruback (1997) found that hostile deals were associated with insignificant improvements in CARs, possibly due to the payment of higher acquisition premia which are necessary in hostile takeover to convince shareholders without the backing of management. Boone and Mulherin (2007) look at the effect the method of sale of the target has on the announcement returns. They distinguish negotiations from auctions but do not find a significant difference in announcement returns for these sale methods. Investment bank influence Rau (1999) focuses on the other parties sitting at the table during the takeover negotiations and looks at the effect the leading investment bank has on the announcement returns. According to the superior deal hypothesis it is suggested that acquiring firms advised by top-tier (bulge bracket) investment banks, with a large market share and strong reputation, should earn higher announcement period excess returns. However, Rau (1999) shows in his research that, inconsistent with the superior deal hypothesis, the acquiring firms earn lower abnormal returns when advised by top-tier investment banks than when advised by second- or third-tier banks. As a possible explanation for this finding it is suggested that perhaps the first-tier investment banks advice their clients to pay a higher premium to make sure that the deal is completed, larger banks have more contingent fees and therefore depend heavily on their completion rate. When the client actually pays this higher premium the existing shareholders will pay for this, hence the lower abnormal returns upon announcement. 13

14 C) Target firm characteristics Cross border effect The earlier mentioned study of the European takeover market by Martinova and Renneboog (2006) investigates the cross border effect in Europe. They find that legal origin of the acquirer and target firm affects the CARs. The highest returns are found for cross border bids where a European bidder acquires a U.K. target. The evidence is not clear with respect to the cross border effect. Eun, Kolodny, and Scheraga (1996) amongst others find significant wealth gains for acquiring firms in cross border mergers and acquisitions and find no significant difference between domestic (U.S. based) and cross border deals. Within industry acquisitions Value creation from mergers and acquisitions can stem from two sources: synergies from the strategic combination of two companies or financial synergies. Strategic synergies include economies of scale, scope and learning, benefits like cost reduction and increased efficiency. Financial synergies come from an optimization of the capital structure of the combined firm: increased leverage can create tax benefits and provide the right incentive for management. For a firm to benefit optimally from the strategic benefits an acquisition of a firm within the same industry is seen as creating more value than an acquisition of a firm outside of the bidder industry. During the third takeover wave ( ) this was viewed differently and companies started to acquire unrelated businesses forming large conglomerates. The reasoning behind the formation of large conglomerates was amongst others the spreading of risk over different divisions in the firm. This diversification of risk can be done much more cheaply by the investors themselves. Conglomerate corporations started trading at a conglomerate discount and in the next wave ( ) they started to divest unrelated business parts. This conglomerate discount is first described by Mueller (1969) where it is documented that acquisitions result in higher gains for the shareholders of the acquiring firm when the target is in the same industry. Morck, Shleifer, and Vishny (1990) show that returns to acquirers are lower for firms with poor performing management, firms that acquire high growth targets, and thirdly for firms that diversify by acquiring targets from different industries. Target listing status Fuller et al. (2002), describe a firm type effect where they investigated the effect of the public status of the target firm on the announcement returns of the bidder. They find that the announcement abnormal returns appear to be higher when the target firm is a private or subsidiary firm compared to when it is a publicly traded firm. Faccio et al.(2006) find this effect for Western European deals as well and come to the conclusion that the fundamental factors that cause this effect remain elusive. The research by Fuller et al. (2002) focuses on acquirers that made five or more acquisitions during a period of ten years from 1990 to 2000 in the United States. They find that across their sample the 14

15 equally weighted average bidder CAR is 1.77 percent. However, when making a distinction between the public status of the target they find that bidders acquiring public targets earn a significant negative return of percent and on the other hand when a private or subsidiary firm is acquired the CARs are 2.08 and 2.75 percent respectively. The above identified factors that influence the abnormal returns of the acquiring firm upon announcement of the acquisition are mostly identified using samples of U.S. firms. When we look at research that has been done on European takeover activities, the literature becomes sparser. Goergen and Renneboog (2002) give an excellent overview of the value gains for shareholders upon announcement of domestic and cross-border takeover bids. In their extensive research they provide an overview of the European (continental and UK) merger and acquisition activity. They focus both on target and bidder CARs and measure both the short- and long-run price effects of the announcement on the share price. For several earlier documented effects in the U.S. they find the same evidence in Europe. For example with the matters of payment effect they also find higher abnormal returns in cash deals compared to equity and mixed deals. However, they don t find the evidence for the suggested size effect by Moeller et al. (2004). A possible reason for this is the sample used, Goergen and Renneboog (2002) only take into account deals with a value greater than $100 million (or 90 million at that time), whereas Moeller et al. (2004) use deals with deal values greater than $1 million. The resulting sample in the research by Goergen and Renneboog (2002) is a total of 228 merger or acquisition announcements in the period of When this sample could be extended perhaps it is possible to find evidence for the size effect. D) Relevance of research The existing literature that identifies the target listing status effect has as main contributors: Hansen and Lott (1996), Chang (1998), Fuller et al. (2002), Moeller et al. (2004) for the United States and Faccio, McConnel, and Stolin (2006) for Western Europe. All research that considers the target listing status as a variable come to the conclusion that public targets have a negative effect on the CARs and unlisted, i.e. private and subsidiary, targets have a positive effect on the announcement abnormal returns. This research adds to this existing research by an enlarged geographic focus, not only does it take into account the acquisitions from Western European firms but also the Eastern European countries are considered. It is interesting to see whether the target listing status effect is also present in Eastern European countries as the state of the economy differs from the state of the economy in Western European countries. Furthermore, this research adds to the existing literature in terms of an enlarged sample. All research mentioned previously does not take into account the most recent financial crisis. The financial crisis of 2008 was an extraordinary event and also had its impact on the merger and 15

16 acquisition activity throughout the world. Companies got into distress and funds were less readily available, causing companies to be weary of large acquisitions. On the other hand, the financial crisis was not a crisis of non-financial corporations per se. It was a crisis of financial institutions and of course the downturn in the financial markets had its reflection on almost every firm, There were still a lot of firms that performed reasonably and had the opportunity to go cherry picking as soon as the dust settled. This turmoil in the financial markets might have had an impact on the target listing status effect. Following the reasoning shown above certain firms were highly punished by the stock markets, while others were spared and survived the crisis with enough funds to spend on those now lower valued. Unlisted firms are of course less dependent on the stock market and are more willing to withdraw from any acquisitions activity unless higher prices than the current valuations are paid. This could mean that in crisis years the effect of the listing status of the target could be opposite from what is previously found in the literature. This research will combine the research done on financial constraints and investment freedom by means of using asset tangibility of a firm as a factor that helps explain the cumulative abnormal returns for the acquiring firms. Almeida and Campello (2007) show that when a firm is financially constrained, asset tangibility of a firm is a determining characteristic in its investment decisions. Here asset tangibility is used as a proxy for the amount of pledgeable assets in a firm and it is argued that when a firm has more pledgeable assets it has higher investment cash flow sensitivity. To translate this to the acquisition activity of a firm, a firm with a high tangibility ratio should have a larger amount of its cash flow invested, this should than also imply a higher amount of spending on acquisitions. The result is that firms with a lower tangibility will have to be more careful in selecting an investment opportunity and therefore only select those that result in the highest value for the company which in turn will result in higher CARs for the acquiring firm. Hence a negative relation is expected between the asset tangibility of the firm and the CARs upon announcement of an acquisition. Next to the tangibility of the firm other financial ratios and their effect on the CARs will be explored. III. Sample Selection, Data, and Methodology A) Sample and data sources For this study I consider acquisitions in Europe announced between January 1 st 1996 and December 31 st I use the year 1996 as a starting year because SDC data from outside the U.S. is considered to be reliable as of this year. This corresponds to the reasoning used in Faccio et al.(2006). The geographic region Europe is used as it is defined in SDC, this includes the bidder countries listed in Table II. An initial dataset of acquisitions is obtained from SDC along with the dollar amount paid for the target, the percentage of shares owned prior to the announcement and the percentage of shares owned post acquisition, the Datastream and CUSIP code of the target and acquirer, the public status 16

17 of both target and acquirer, the deal status, the Standard Industry Classification (SIC) Code, the nations of incorporation, the primary exchange codes, and the method of payment in the acquisition. Following Faccio, McConnel, and Stolin (2006), Fuller et al. (2002), and others an acquisitions is only included provided that the acquirer did not have a controlling stake (<10%) in the target company previous to the announcement of the acquisition and has a majority stake (>50%) post acquisition. This is defined as a complete control acquisition. Furthermore, the acquisition is required to have a completed status in SDC conform to other studies. The acquiring company is required to be a public company with share price data available in Datastream. Different from Goergen and Renneboog (2002), who use deal values larger than $100 million, but in accordance with Faccio et al. (2006) and Fuller et al. (2002) acquisitions are included when the deal value exceeds $5 million. Including deals with lower deal values will allow for a better investigation of the size effect. Furthermore, the public status of the target company is required to be public, private, or subsidiary, excluding companies that have a classification as privatizations, joint ventures, and unidentified cases. These limitations result in a sample of 10,252 acquisitions. A distribution of acquisitions by year can be found in Table I. Deal values are denoted in dollar amounts corrected for inflation using the Consumer Price Index (CPI) 5 index to represent 2010 dollar equivalents. The U.S. CPI is used because there is no reliable data for European inflation prior to 1996 and these numbers are necessary to calculate accounts reported before For the years following 1996 the CPI and European Harmonized Index of Consumer Prices (HICP) 6 for the Euro Area are compared and there are no large differences found in the indices. For sake of uniformity the CPI is used for all inflation corrections. There is a large variation in the number of acquisitions over the years with evidence of merger waves in the periods and since the total number and dollar amount increased sharply in these periods. The cyclicality in mergers and acquisitions was empirically shown by Golbe and White (1993), following this research six waves of M&A activity have been established; those of the early 1900s, the 1920s, the 1960s, the 1980s, 1990s, and the early 2000s. The latter two waves are (at least partially) included in the sample period of this research and are of particular interest since European companies really started to engage in M&A activity only for these last two periods (Martynova and Renneboog (2006)). 5 Source: 6 Source: Eurostat inflation statistics. 17

18 TABLE I Distribution of Acquisitions by Year Deal values are expressed in millions of U.S. dollars corrected for inflation to reflect the dollar value in U.S. dollar amounts are calculated based on year end exchange rates, the inflation correction is based on the change in the U.S. Consumer Price Index (CPI). Deal value Year Number of deals Mean Median Total , , , , ,014, , , , , , , , , , , , ,455 All 10, ,937,587 Table II lists the distribution of the acquisitions by the countries of incorporation as listed by SDC, 51.3% of the acquirers is incorporated in the United Kingdom making this the largest contributing country to the sample. When we look at the difference between Western Europe and Eastern Europe 7 it is apparent that acquisitions in the latter represent a much smaller portion of the total number of acquisitions. It might be the case that Datastream and SDC data for these countries are less readily available compared to the first subsample. For the target companies the most frequently observed home country is also the United Kingdom (34.6%) followed by the United States (16.25%). It is important to recognise this presence of the Anglo-Saxon countries in the sample. Where relevant the acquisitions with a bidder country from the United Kingdom need to be considered separately or a dummy will be added for the United Kingdom to correct for the large presence of this country in the sample. 7 Where Western Europe includes the countries Austria, Belgium, Denmark, Finland, France, Germany, Greece, Ireland, Iceland, Italy, Luxembourg, Liechtenstein, The Netherlands, Norway, Portugal, Spain, Sweden, Switzerland, and United Kingdom. Eastern Europe includes the other European countries listed in Table II. 18

19 TABLE II Distribution of Acquisitions by Home Country of the Acquirer and Target Panel A. European Nations Bidder Country Target Country Country Observations Fraction Observations Fraction Austria % % Belgium % % Bulgaria % % Croatia % % Cyprus % % Czech Republic % % Denmark % % Estonia % % Finland % % France % % Germany % % Greece % % Hungary % % Iceland % % Ireland-Rep % % Italy % % Liechtenstein % % Luxembourg % % Malta % % Netherlands % % Norway % % Poland % % Portugal % % Romania % % Russian Fed % % Slovak Rep % % Slovenia % % Spain % % Sweden % % Switzerland % % Turkey % % United Kingdom 5, % 3, % 9, % 7, % Panel B. Other nations United States 1, % Australia % Canada % Brazil % South Africa % China % India % Mexico % Chile, Hong Kong, South Korea % Argentina % Singapore % Ukraine % Israel % Bermuda, Lithuania, Taiwan % Egypt, Japan % Colombia, Indonesia, New Zealand, Peru, Serbia & Mont % Utd Arab Em, Venezuela % Kazakhstan % Malaysia, Philippines, Thailand % British Virgin, Monaco, Morocco, Serbia % Cayman Islands, Latvia, Panama, Puerto Rico, Uruguay % Antigua, Bahamas, Belarus, Bolivia, Ecuador, Laos, Macedonia, Rep of Congo, Tunisia, Uzbekistan, % Albania, Armenia, Costa Rica, Dominican Rep, Guatemala, Honduras, Iran, Jamaica, Moldova, Multi-National, Nicaragua, Pakistan, Saudi Arabia, Vietnam, Yugoslavia % Algeria, Bangladesh, Bosnia, Botswana, Cuba, Dem Rep Congo, Falkland Is, Georgia, Ghana, Jordan, Kuwait, Kyrgyzstan, Lebanon, Liberia, Mauritania, Mauritius, Mozambique, Myanmar(Burma), Namibia, Nigeria, Reunion, Seychelles, Sri Lanka, Tajikistan, % Total Sample 2, % 19

20 When looking at the subdivision based on the listing status (Table III, Panel A.) of the target firm there are three subsamples defined: private, subsidiary, and public. In 46.04% of the cases the target is a private entity forming the largest group, the smallest portion is formed by the public targets (13.29%) and the remainder (40.67%) are acquisitions with a subsidiary target. Further analysis of these subsamples shows that acquisitions of public companies are financed more often with all stock or with all cash, respectively 8.73% and 31.69%, than acquisitions of the other targets, which are therefore more often financed with a mix of stock and cash. Next, the characteristics of both the deal and the acquiring companies are examined for the different subsamples, the subsample all unlisted is formed by the combined acquisitions of private and subsidiary targets. Continuous variables included in any analysis are winsorized at the 5% level to exclude outliers. In Panel B. of Table III the deal characteristics are shown. To test if the means of the various variables are significantly different from each other a two sample mean comparison test is performed. Here the means for characteristics of companies acquiring public targets are compared with the means of these characteristics for companies acquiring all unlisted targets, private targets, and subsidiary targets respectively. A test statistic is obtained to determine with what level of confidence it can be stated that the means are different for the relevant variable. This test statistic is measured as follows: The median values for the characteristics are shown below the mean values in Table III. To test if the median values are statistically significantly different a Wilcoxon-Mann-Whitney test is performed. For most characteristics the subsample of acquirers purchasing public targets differs significantly from the three other subsamples, below the differences will be discussed. As could be expected listed targets are bought by larger companies as measured by the market value of the acquirer. This market value is obtained from Datastream for the closing of the previous fiscal year. This is done to make it comparable with the other variables in the analysis which are based on the reporting done at the end of the applicable fiscal year. The average size of the acquirers in deals with a public target is $12,287 million compared to $8,437 and $3,980 million for respectively subsidiary and private target acquisitions. When going back to the fundamentals of why companies go public, Pagano, Panetta, and Zingales (1998) show that for a company to have reasons to undergo an Initial Public Offering (IPO) it has to have a certain size to bear the costs of listing, both direct and indirect. This is clearly visible in the sample if we look at the transaction value for the different subsamples, acquisitions of public targets have a much larger value than acquisitions of private and subsidiary targets. This also explains largely why companies acquiring these public targets are larger 20