How adverse selection of cash holdings affect corporate and investor behavior in M&A cases: Evidence from fully stock financed European acquisitions Finance Master's thesis Stefan Söderström 2012 Department of Finance Aalto University School of Economics Powered by TCPDF (www.tcpdf.org) AaltoUniversity Aalto University School of Economics Abstract Master’s Thesis 19.1.2012 Stefan Söderström HOW ADVERSE SELECTION OF CASH HOLDINGS AFFECTS CORPORATE AND INVESTOR BEHAVIOR IN M&A CASES: Evidence from Fully Stock Financed European Acquisitions PURPOSE OF STUDY This thesis examines how excess cash holdings affect firm and investor behavior in equity financed mergers and acquisitions. The theoretical framework for the study has been set based on research on firm capital structure and M&A outcomes. It therefore builds on a solid theoretical base, where optimality is seen as a basis for rational behavior. Equity financing has strong signaling effects, which have been reported to diminish shareholder value in the short-term especially if irrationally exercised. Reflecting on the selection of cash holdings, firms are subject to stronger signaling effects as essentially they provide a less expensive form of financing. In light of the prevailing theory, I intend to provide results that capture these effects in European context, while enabling comparison of domicile and cross-border M&A cases and the implications of different legislations. DATA AND METHODOLOGY I gather mainly two different samples in order to first measure excess cash reserves and secondly to inspect how they are translated into a sample consisting purely of equity financed acquisitions in the EU-15 countries. I use ThomsonReuters and SDC to retrieve the data for M&A cases occurring in 1999-2010. I primarily use OLS multivariate and Logit regressions to test the impact of excess cash reserves on M&A announcement effects, probability of using equity, and post- acquisition performance. Further, account for outliers and sampling bias by winsorising and the Heckman two-stage model respectively. RESULTS The results provide strong support for my hypothesis. Excess cash tends to have strong signaling effects on bidder cumulative abnormal returns, where the relation is significantly negative. I also provide evidence that lowering levels of information asymmetry off-sets this effect, as uncertainty about the bidder valuation levels is lowered. The probability of using equity in M&A cases is similarly negatively related to excess cash reserves. This effect is highlighted in market conditions that are more favorable, i.e. valuation levels are higher. Differentiating between country legislation builds on the finding, as higher investor protection increases the probability of using equity but simultaneously the adverse selection cost is emphasized. Post-acquisition operating performance is shown to be improved with higher excess cash ratios. Cash reserve levels are shown to decrease during the first years after the transaction, converging therefore towards normal levels. KEYWORDS M&A, merger, acquisition, adverse selection, asymmetry of information, capital structure, cash holding, precautionary motive, method of payment. Aalto-yliopisto Aalto-yliopiston kauppakorkeakoulu Tiivistelmä Pro gradu- tutkielma 19.1.2012 Stefan Söderström HOW ADVERSE SELECTION OF CASH HOLDINGS AFFECTS CORPORATE AND INVESTOR BEHAVIOR IN M&A CASES: Evidence from Fully Stock Financed European Acquisitions TUTKIELMAN TAVOITTEET Tutkielmassa tarkastellaan ylimääräisten kassavarantojen vaikutusta yritysten ja sijoittajien käyttäytymiseen, kun kyseessä on omalla pääomalla rahoitetut yrityskaupat. Teoreettinen viitekehys pohjautuu tutkimuksiin, jotka käsittelevät yrityksen pääomarakennetta ja yritysostojen tuloksia. Taustalla on laaja teoreettinen pohja, jossa nähdään optimaalisuuden olevan perustana rationaaliseen käyttäytymiseen. Pääomarahoituksen muodoista oman pääoman ehtoisella rahoituksella on voimakas signalointivaikutus, jonka on raportoitu vähentävän omistajien arvoa lyhyellä aikavälillä, erityisesti mikäli sitä on käytetty epärationaalisesti. Kassavarantojen tasojen valinnan myötä yritykset omaksuvat vahvempia signalointivaikutuksia, jotka tulee huomioida rahoituspäätöksissä. Vallitsevan teorian huomioonottaen esitän tuloksia, jotka käsittelevät näitä vaikutuksia Euroopassa. Vertailen myös maiden sisällä tapahtuvia ja maantieteelliset rajat ylittäviä yrityskauppoja ja siten tutkin eri lainsäädäntöjärjestelmien vaikutusta niihin. LÄHDEAINEISTO JA MENETELMÄT Käytän pääosin kahta eri otosta, jotta ensiksi pystyn arvioimaan määritelmän ylimääräisistä käteisvaroista ja toiseksi tarkastamaan, miten nämä heijastuvat puhtaasti omalla pääomalla rahoitettuihin yritysostoihin EU-15 maissa. Tiedot on kerätty ThomsonReuters ja SDC - hakemistoista kattaen yritysostot aikaväliltä 1999 - 2010. Olen ensisijaisesti käyttänyt OLS monimuuttuja ja Logit regressioita testatakseni ylimääräisten kassavarantojen vaikutuksia yrityskaupan ilmoitusvaikutukseen, todennäköisyyteen käyttää omaa pääomaa, sekä yrityskaupan jälkeiseen suorituskykyyn. Lisäksi huomioidakseni vieraiden havaintojen osuuden ja otosmenetelmän virheen käytän winsorising-menetelmää and Heckman kaksi-vaihe mallia. TULOKSET Tulokset tukevat vahvasti hypoteesejani. Ylimääräisillä kassavarannoilla on vahva signalointivaikutus ostavan yrityksen kumulatiivisiin epänormaaleihin tuottoihin, joissa suhde on huomattavan negatiivinen. Osoitan myös, että alentamalla tiedon epäsymmetrisyyden määrää voidaan tätä negatiivista vaikutusta vähentää, sillä epävarmuus ostajan arvostustasosta laskee. Ylimääräisten kassavarantojen kasvaessa oman pääoman käytön todennäköisyys yrityskaupoissa laskee. Tämä vaikutus korostuu edullisessa markkinatilanteessa, jossa arvostustasot ovat korkeammat. Eri maiden lainsäädännön merkitys puolestaan heijastuu siten, että korkeampi sijoittajansuoja lisää todennäköisyyttä käyttää omaa pääomaa, muta samalla korostuu haitallisen valikoitumisen kustannukset. Yrityskaupan jälkeisen toiminnan tehokkuus osoitetaan olevan parempi ylimääräisten kassavarantojen kasvaessa. Kassavarantojen taso osoitetaan olevan laskevia yrityskaupan jälkeen, jolloin ne lähestyvät normaalitasoa. AVAINSANAT Yrityskauppa, yritysosto, fuusio, haitallinen valikoituminen, tiedon epäsymmetria, pääomarakenne, kassavaranto, varovaisuuden periaate, maksutapa. I Table of Contents I. Introduction ............................................................................................................................... 1 1.1 Background ........................................................................................................................ 1 1.2. Contribution to existing literature ...................................................................................... 2 1.3 Research question and main findings ................................................................................ 3 1.4 Limitations of the study ..................................................................................................... 4 1.5 Structure of paper .............................................................................................................. 5 II. Theory and Previous Literature ................................................................................................ 5 2.1 Capital structure ................................................................................................................. 5 2.2 Adverse selection of cash reserve ...................................................................................... 7 2.2.1 Theoretical background for the adverse selection effect of corporate cash reserve ... 7 2.2.2 Other informational implications of cash ................................................................... 9 2.3 M&A financing decision ................................................................................................. 10 2.3.1 Determinants of payment method ............................................................................ 10 2.3.2 Effects of financing decision on post-acquisition performance ............................... 13 2.4 Announcement effects of acquisitions ............................................................................. 14 III. European context .................................................................................................................... 17 3.1 Distinct characteristics for the region .............................................................................. 17 3.2 Taxation differences ........................................................................................................ 19 3.3 Cross-border acquisitions in Europe................................................................................ 21 IV. Hypotheses .............................................................................................................................. 22 V. Data and Methodology ............................................................................................................ 28 5.1 Data description ............................................................................................................... 28 5.2 Methods ........................................................................................................................... 32 5.2.1 Logit regression ........................................................................................................ 32 5.2.2 Heckman two-stage test ........................................................................................... 33 II 5.2.3 Wilcoxon rank sum .................................................................................................. 34 5.2.4 Wald test ................................................................................................................... 35 5.2.5 Winsorising .............................................................................................................. 36 5.2.6 CAR calculation ....................................................................................................... 36 VI. Results ..................................................................................................................................... 37 6.1 Excess cash ratio .............................................................................................................. 37 6.2 Bidder share performance and CARs .............................................................................. 41 6.3 Probability of using cash as consideration ...................................................................... 53 6.4 Operating performance .................................................................................................... 60 VII. Robustness tests ...................................................................................................................... 64 7.1 Growth opportunities adjusted bidder announcement effects of excess cash ................. 64 7.2 Sample selection bias correction for probability of using equity financing .................... 67 7.3 Probability of using equity financing in cross-border M&A cases ................................. 69 7.4 Comparing post-transaction operating performance with a matched firm sample .......... 72 VIII. Summary and conclusions ............................................................................................... 75 7.1 Main findings ................................................................................................................... 76 7.2 Conclusions and suggestions for further research ........................................................... 79 REFERENCES ............................................................................................................................... 81 APPENDIX A: Mergers and acquisitions terminology ................................................................. 88 APPENDIX B: Variable definitions .............................................................................................. 89 APPENDIX C: Control variables correlation matrices .................................................................. 90 APPENDIX D: Survey results for cash holding rationale ............................................................. 91 APPENDIX E: Marginal effects of variables on probability of using stock ................................. 92 III List of Tables Table 1: M&A tax differences in major EU-15 countries .............................................................. 20 Table 2: Descriptive statistics of equity financed offer bidders ..................................................... 31 Table 3: Excess ratios across equity financed offer bidders .......................................................... 41 Table 4: The negative announcement effect of a bidder's excess cash ratio .................................. 46 Table 5: The announcement effect of excess cash reserve accounting for high tangible assets .... 52 Table 6: Logistic estimation of the probability of using stock ....................................................... 56 Table 7: The effect of excess cash reserve on a bidder's post-acquisition uses of funds ............... 62 Table 8: Effect of excess cash reserve on probability of using equity - Adj. for growth rates ...... 66 Table 9: Sampling bias corrected probability of using equity in M&A cases ............................... 68 Table 10: Cross-border M&A cases - Determinants for using equity ............................................ 71 Table 11: Matched firm analysis on post-acquisition uses of funds .............................................. 74 Table 12: Summary of results ........................................................................................................ 77 List of Figures Figure 1: Deal count and average deal value ................................................................................. 29 Figure 2: Distribution of excess cash ratios ................................................................................... 38 Figure 3: Bidders cumulative abnormal returns around deal announcement ................................. 43 Figure 4: Bidder CAR vs. excess cash ratio ................................................................................... 50 1 I. Introduction 1.1 Background Several mergers and acquisitions related research has focused on studying methods of payment and share price announcement effects, both for cross-border and domestic M&As. Findings vary significantly over time and across geographical focus, however a series of common fundamentals have been established: 1) equity financed transactions imply lower share price returns around the announcement date (Myers and Majluf, 1984); 2) full equity offers are more likely to be extended and accepted during equity market highs (Choe, Masulis, Nande, 1993); 3) method of payment does not affect the operating performance of the combined entity post-acquisition (Healy et al., 1999). Reflecting on theory and previous literature, the frames for this study are set relatively conscientiously: equity financed transactions are expected to have negative share price announcement effects, increase in probability at market highs, and create no operational value after the acquisition on average. The thesis builds on an understanding that firms having cash and marketable securities on their balance sheets, which exceed the required amount to finance daily operations, do not create value for shareholders. The return on cash is significantly lower than the required rate of return on equity. Therefore, it would be understandable to deploy these low return assets in ventures that provide higher returns. For some reason in situations where this cash could be deployed, firms haphazardly decide not to – instead they use more expensive sources of finance. The question is: Why do firms use equity financing if internal funds are also available? In the presence of capital market imperfections deriving from asymmetric information between managers and capital providers, liquidity may be considered to be in a strategic role. Financial flexibility in the form of excess cash reserve may convey 1) growth opportunities; 2) stability through economic downturns; 3) guard against the underinvestment1 problems caused by expensive external financing, which are seen valuable for shareholders. Firms that experience high growth prospects and need a high level of liquidity to ensure corporate flexibility, are 1 The underinvestment occurs because there are realized project payoffs that cover the investment expenditure at the time of the investment but not the investment expenditure and the promised payment on leverage at later dates (Myers 1977). 2 expected to act against established corporate finance theories. Cash reserves do, however, induce an adverse selection cost by sending a stronger signal of overvaluation when equity financing is used. It is this signal that my thesis addresses. This study is motivated by the study conducted by Gao (2011) and reflects results on his findings. In his study, Gao (2011) explores the effects of excess cash ratio of bidder announcement effects and the relation between excess cash reserves and equity financing. Based on a US sample, the author finds evidence that announcement effects and probability of using equity as a method of payment is negatively related to excess cash reserves. He also finds that post-acquisition performance improves along the cash richness of the bidder prior to the M&A case. 1.2. Contribution to existing literature The aim of the paper is to study EU-15 firms’ liquidity and its relation to M&A outcomes. Fundamentally I differentiate between firms with excess cash reserves and non-cash rich firms, in order to test theories such as adverse selection of cash reserve and asymmetries of information. Adding to the vast amount of literature on the topic, I provide results for the EU-15 countries and apply methods that capture market irrationality and two-sided asymmetries of information according to Rhodes-Kropf and Viswanathan (2004) and Shleifer and Vishny (2003). I find that the methodology Gao (2011) uses adds to the existing literature, with close reference to Harford (2005), by inspecting equity financed transactions. Gao (2011) uses a US based sample and inspects acquisitions during the period 1990-2005. Hence, I bring a more contemporary data set with a different geographical focus. In addition, I provide robustness tests that account for the growth opportunities that are prevalent in certain market conditions and industry cycles. A majority of previous literature has focused on the US and UK. By inspecting a European-wide sample of M&A, the thesis contributes to the existing literature, as it allows the evaluation of the impact of a wide range of institutional settings and legal and regulatory rules on the pattern of M&A activity. Having a sample consisting of several different countries enables me also to consider cross-border acquisitions from a very specific perspective. In comparison to the US and UK, Continental European firms are characterized by weaker investor protection and less developed capital markets (La Porta et al., 1998) and by more concentrated ownership structure 3 (Faccio and Lang, 2002). EU directives on M&A have harmonized regulation across the EU-15 countries, but there still exists differences in implementation. The same provisions applied in the UK, where shareholder rights are paramount, and in the rest of Europe, where corporate governance is dominated by stakeholders, are largely debated in the EU. The lack of experience by national authorities in evaluating complex transactions, the independence of regulatory agencies, and the degree of explicit or implicit government involvement are clear barriers to a homogeneous implementation of rules across countries. The thesis provides a unique insight on the differences between the two legislative areas, where both firm and investor behavior is confined to the two distinct areas. 1.3 Research question and main findings As stated, M&A is a well-studied area of finance, but certain aspects require more scrutiny, especially in the European context. Building on the existing theories and research results, I provide insights on three main concentration areas: 1) the announcement effect of fully stock financed acquisitions; 2) the probability of fully financing acquisitions with stock, 3) the post- transaction operating performance and uses of funds. Differentiating between different market conditions and different legislative areas, the thesis engages in answering the possible differences in results for different environments. In my study, I provide results that highlight the impact of excess cash holdings on firm and investor behavior. The results are achieved through OLS and Logit regression, Wald tests, Ranksum tests, and Heckman two-stage tests. First, I find that excess cash reserves are negatively related to the announcement effect of equity financed bidders at 1 percent significance. The impact is more pronounced in hot equity markets than in non-hot equity markets, which portray the overvaluation component of the market that is signaled through equity bids. The effect is also shown to be emphasized in legislative areas with higher investor protection, i.e. in areas where equity financing is used more likely. When the probability of using equity is tested, the excess cash reserve variable is highly significant and negative. Therefore probability of using equity is lower when internal, cheaper funds are available. The finding conveys rationality in firm behavior. Third, the operating performance improves along bidder cash richness, which implies that the higher synergies are expected following the weaker announcement effect. Overall, the 4 results presented in this study have both complementary and contradictory aspects regarding the theoretical frameworks that I construct in Section 2. 1.4 Limitations of the study There are three main limitations to this study that act as constraints when evaluating the applicability of the results presented in this study. First, the availability of M&A data and firm related metrics are constrained to the data sources. Securities Data Company (SDC) Mergers and Acquisitions database limits the geographical diversity of the sample used in this study. For an unknown reason, the amount of M&A activity is unproportionally high for the United Kingdom as observations for smaller countries are lacking. Further, the number of observations for smaller countries is not consistent with the distribution of M&A activity. Therefore, the sample is slightly biased towards countries where transactions are recorded with higher precision. Similarly, I use ThomsonReuters database to retrieve firm financials, where observations are sensitive to the accuracy and correctness of firm coding. Again, for some smaller countries there exists a relatively lower ‘hit ratio’ in finding sufficient information. This consequently makes country level comparisons difficult to perform. However, I find that the sample used is sufficiently large and a fair representation of firm behavior in the EU. Second, even though I use a sample starting from the year 1999 (the year the Euro was introduced to the world financial markets as an accounting currency) there is heterogeneity between country legislation developments during the past 12 years. The impact of these changes in country legislation is impossible to fully account for. I perform tests that include country and year differences, which are seen to be sufficient measures to incorporate this effect. Further, I differentiate between Code and Common Law countries to test for differences across various legislations. Third, an important variable in explaining decisions on method of payment in M&A cases is excluded from this study. Corporate control (elaborated on in Section 2.3.1) is excluded due to the hand picking nature of the variable. Including over 3,000 acquisitions in the specific test where management ownership would be used as an explanatory variable, makes hand picking a data collection method that is not seen feasible.