Are Investor Reactions to Mergers and Acquisitions Dependent upon the Economic Cycle?

Transcription

1 Are Investor Reactons to Mergers and Acqustons Dependent upon the Economc Cycle? Chrst Wann Unversty of Tennessee at Chattanooga Na H. Lamb Unversty of Tennessee at Chattanooga Ths paper utlzes an event study methodology to nvestgate the possble dfferences n market reactons to same- and cross-ndustry merger and acquston actvty durng dfferent economc cycles. Target frms from same- and cross-ndustry mergers experence larger postve cumulatve abnormal returns durng recessons than n non-recessons. Ths suggests that good news n bad tmes s worth more than good news n good tmes. The study also fnds evdence that same-ndustry acqurers experence small but sgnfcantly hgher CARs than cross-ndustry acqurers durng non-recessons. Ths result may ndcate the market s preference for synergstc same-ndustry mergers over dversfyng mergers. INTRODUCTION Behavoral fnance scholars have found that nvestors do not make ther decsons n a vacuum (Schjven & Htt, 2012). Studes have shown that nvestors are nfluenced by ther perceptons on whether a strategc acton wll create value (March & Smon, 1958). Drawng from the behavoral theory of the frm (Cyert & March, 1963), we argue that nvestors do not react to merger and acquston (M&A) announcements based on ratonal or effcent calculatons. Rather, nvestors are nfluenced by ther own perceptons and bounded ratonalty; that s, nvestors wll react to M&A announcement dfferently n tmes of recessons and non-recessons. We thus predct asymmetrc stock market reactons to smlar M&A news for frms n recessons and non-recessons. Pror lterature suggests that bad s stronger than good (Baumester, Bratslavsky, Fnkenauer, and Vohs, 2001; Kahneman and Tversky, 1979) and could be related to evolutonary psychology. Humans that were more vglant towards bad thngs were more lkely to survve threats and would ncrease ther chances of producng more of ther genes. Durng prmtve tmes, only one nstance of gnorng a potental bad outcome could result n death (McDermott, Fowler, and Smrnov, 2008). However, gnorng a postve outcome only resulted n a mssed chance for pleasure or advancement (Baumester, Bratslavsky, Fnkenauer, and Vohs, 2001). In modern tmes, the extensve coverage of negatve economc news durng economc contractons further aggravates the perceptons of economc uncertanty (Bloom 2009, 2014). However, the same attenton s not gven to postve economc phases (Soroka 2006). Therefore, humans tend towards nconsstent and context-dependent reactons to negatve and postve events. McQueen and Roley (1993), for nstance, show that when the economy s strong, the stock market responds negatvely to news about Journal of Accountng and Fnance Vol. 16(6)

2 hgher real economc actvty whle the same surprse n a weak economy s assocated wth hgher stock returns. Ths result suggests that good news durng bad tmes s worth more than good news durng good tmes. Mergers and acqustons (M&As) are an mportant method of expanson or dversfcaton (Htt, Freeman, & Harrson, 2001). Much of the pror lterature examnes whether M&As nfluence stock prces wthout the nformatve context provded by economc cycles. Typcally, studes of M&A returns fnd large postve target returns (Chevaler, 2004; Xuan, 2014; Wang and Xe, 2009; Bhagat, Dong, Hrshlefer, and Noah, 2005) and small negatve acqurer returns (Andrade, Mtchell, and Stafford, 2001; Moeller, Schlngemann, and Stulz, 2005; Kng, Dalton, Daly, and Covn, 2004). Therefore, ths paper seeks to contrbute to the extant lterature that examnes the nfluence of M&A announcements on stock prce by ncorporatng the economc cycle. Ths study nvestgates the M&A announcement day stock returns across dfferent phases of the busness cycle as defned by the Natonal Bureau of Economc Research (NBER). Specfcally, we use a tradtonal Fama-French three-factor event study approach (Fama & French, 1993) to examne announcement day abnormal returns assocated wth frms (both acqurer and target frms) n recessons and non-recessons. Under ths framework, we examne the followng questons: 1) Are same-ndustry target returns dfferent from cross-ndustry target returns? 2) Are same-ndustry acqurer returns dfferent from cross-ndustry acqurer returns? 3) Do economc cycles affect same-ndustry target returns and cross-ndustry target returns equally? 4) Do economc cycles affect same-ndustry acqurer returns and cross-ndustry acqurer returns equally? In the followng secton, we revew some of the relevant lterature and layout our theoretcal framework. We then descrbe the sample data and methodology used. Next, we emprcally test our hypotheses and dscuss the results. Fnally, we conclude wth mplcatons, lmtatons, and drectons for future research. LITERATURE REVIEW M&As and Economc Condtons Many studes have been conducted to examne the nfluence of M&As on stock prce (Bruner, 2002; McQueen & Roley, 1993; Shlefer & Vshny, 2003). Tradtonal M&A studes, however, generally overlooked the lnk between M&A announcements and the underlyng economc condtons that are present at the announcement. There are good reasons to expect M&A announcements to be perceved dfferently n dfferent phases of the busness cycle. To the authors knowledge, there s only one study that consders the current economc cycle. In ths study, target frm cumulatve abnormal returns (CARs) are 3.53% to 8.12% hgher durng recessons than non-recessons (Wann and Lamb, 2016). Ths result suggests that good news durng bad tmes s more valuable than good news n good tmes. Fama and French (1989) fnd that expected prces for stocks and bonds are hgher when economc condtons are weak (e.g., durng recesson) and vce versa. When economc condtons are weak as n a recesson, more frms are more lkely to experence performance problems. Jensen (1991) argues that M&As can be an effectve tool to help frms n fnancal dstress. Other scholars have found that dstressed frms are more lkely to be sold durng recessons (Bard & Rasmussen, 2003). Even though M&As can be a tool for frms to cope wth recessons, few studes have yet examned the mpact of M&As on stock prce n recessons. Though scholars have argued that M&A actvtes tend to occur less often n recessons (Cools, Gell, Kengelbach, & Roos, 2007), others argue that M&As n recessons often result n better deals (Rhodes & Stelter, 2009). The clams may be vald. For nstance, frm values tend to drop n recessons. Thus, the acqurer frm may be able to buy the target frm at a dscount (from ts real market value). However, f ths s the only reason, then we should see smlar nfluence for the acqurer frms and target frms, meanng, both acqurer frms and target frms should experence stock prce ncreases after a recesson. Scholars have not reached an agreement whether M&A announcement has the same nfluence for both the acqurer frms and target frms n recessons and non-recessons. For example, Goergen and 62 Journal of Accountng and Fnance Vol. 16(6) 2016

3 Renneboog (2004) fnd that target frms accrue postve returns whle acqurer frms experence nsgnfcant return durng non-recessons. Other scholars, however, fnd contradctory results for acqurer frm s stock prce (Moeller, Schlngemann, & Stulz, 2005). Dng and Rahaman (2010) fnd that recessons tend to reveal the true rsks of frms that engaged n acqustons durng non-recessons. These frms tend to become acqured durng the next recesson. On the other hand, frms that acqure durng recessons tend to perform well and not be acqured n subsequent recessons. Therefore, a recesson provdes market partcpants wth nformaton about the true fnancal strength of acqurng frms. In summary, scholars have yet reached an agreement whether M&A announcement has a postve mpact on a frm s stock prce n recessons or non-recessons. We argue that nvestors react dfferently to smlar M&A announcements n dfferent states of economc condtons due to changng preferences and expectatons. Tradtonal Effcent-Markets Hypothess Fama (1970) argues that a frm s stock prce should reflect all avalable nformaton on the markets. In ths reasonng, every nvestor s nformed of any market constrants and can make effcent and ratonal nvestment decsons (thus reflected n the stock prce). For nstance, Klen (2001) suggests that f there s an M&A, ths means t wll generate postve returns for both frms, ncreasng shareholder wealth, and thus resultng n hgher stock prce. Effcent-markets hypothess has three assumptons. Frst, nvestors are ratonal. Second, rratonal decsons are random and rare, and fnally, the mpact of rratonal decsons wll eventually be corrected by the market (Schjven & Htt, 2012). Based on the effcent-markets hypothess, M&A announcement should have the same mpact on stock prce regardless whether the economc condtons are weak or strong. Because nvestors are ratonal, they should be able to make the best decsons based on frms M&A announcements. Despte ts popularty, scholars have found several drawbacks. Frst, effcent-markets hypothess downplays nformatonal asymmetry that s well-studed n the management feld (Aharon, Thany, & Connelly, 2011; Connelly, Certo, Ireland, & Reutzel, 2011). Informaton asymmetry often exsts between managers and nvestors because managers are the ones that oversee the day-to-day operatons of the frm (Myers & Majluf, 1984). Investors often rely on publc nformaton whereas the managers often hold the keys to prvate nformaton. Second, effcent-markets hypothess assumes that the markets are effcent but fals to explan how and why. It treats the market as a black box where thngs just naturally happen (Zajac & Westphal, 2004). If markets were to be as effcent as the effcent-markets hypothess assumes, there would be no opportunty for arbtrage. Many studes have shown that markets are not as effcent as what s beleved (Cordng, Chrstmann, & Wegelt, 2010; Hunter & Coggn, 1988). Target frms have been found to experence postve abnormal returns of 20%, 21.52%, and the range of 17.96% to 44.78% n studes by Ish and Xuan (2014), Wang and Xe (2009), and Bhagat, Dong, Hrshlefer, and Noah (2005), respectvely. Acqurer frms have been found to experence negatve returns (Moeller, Schlngemann, and Stulz, 2005; Kng, Dalton, Daly, and Covn, 2004). Gven the complexty of M&As, the nformaton asymmetry between the managers and nvestors wll be hgh. Addng the complexty of dfferent economc condtons such as recessons, t s plausble that nvestors wll react dfferently. Thus, M&A announcements are lkely to nfluence stock prce n dfferent ways n recessons and non-recessons. We thus relax the assumptons of the effcent-markets hypothess and turn to the behavoral theory of nvestors. Behavoral Theory of Investors Cyert and March (1963) propose that people make decsons based on dfferent factors, such as routnes and learnng. In essence, people are bounded ratonal. Dfferent scholars have found support for the behavoral theory by examnng the perceptons of rsks and uncertanty and how they affect a frm s strateges (Argote & Greve, 2007; Gavett & Rvkn, 2007). For nstance, Cuypers and Martn (2010) examne the perceptons of frm rsk wth respect to real optons on nternatonal jont ventures, and fnd exogenous uncertanty strongly nfluences nvestment. Others fnd that famly frms often perceve rsks Journal of Accountng and Fnance Vol. 16(6)

4 dfferently than non-famly frms; thus famly frms dsplay dfferent rsk tolerance behavors (Gomez- Meja, Haynes, Nunez-Nckel, Jacobson, & Moyano-Fuentes, 2007). Behavor theory can explan the nvestors behavors because t takes nto consderaton nformaton asymmetry and people s reactons to t. Scholars have shown that when a target frm s a prvate frm, t wll be harder for nvestors to value t than a publc target frm (Offcer, Poulsen, & Stegemoller, 2009). In ths ven, behavoral theory suggests that the nvestors reactons on the stock market are the results of problematc search (Greve, 2003). Ths means that the nvestors problem s the nformaton asymmetry and thus they seek to fnd the nformaton needed to make the best nvestment decsons. Thus, an M&A announcement wll trgger an nvestor to search for nformaton n order to maxmze ther wealth (Kock, 2005). For nstance, Verones (1999) fnds that shareholders overreact to bad news n good tmes. There has been evdence that M&As can be affected by a recesson (Aguar & Gopnath, 2005) and Gaughan (2011) fnds that the recesson of 2008 decreased most frms number of M&A actvtes. Buldng on the arguments above, an M&A announcement wll nfluence an nvestor dfferently n tmes of recesson vs. non-recesson. Durng a recesson, an M&A announcement shows that a target frm s worthy of beng acqured. Because the nvestors have no prvate nformaton avalable to them, they have to rely on publc nformaton. Ths can be thought of as good news n bad tmes. Investors are lkely to react postvely to the M&A announcement for the target frm. On the other hand, n a non-recesson, the nvestors may not value the good news n good tmes as much as good news n bad tmes. DATA Sample The sample ncludes frms wth annual data on the Thompson One database as well as research ndustral fle from January 1, 1971 to December 31, The sample excludes ncomplete M&As or those that were repurchases. Repurchases occur when frms buy back ts own shares. The ntal search gave us a total of 64,989 M&A announcements, and they ncluded deal-level nformaton (e.g. target country and M&A sze). The ntal sample contaned a total of 73,277 unque frms. Stock return data are obtaned from the Center for Research and Securty Prces (CRSP) daly stock event fle. After matchng the orgnal sample to the avalable CRSP stock return data, the sample sze decreased to 31,815 frms. One purpose of the study s to analyze event perod cumulatve abnormal returns for same- and crossndustry M&A actvty. There are twelve ndustry categores: consumer products and servces, consumer staples, energy and power, fnancals, healthcare, hgh technology, ndustrals, materals, meda and entertanment, real estate, retal, and telecommuncatons. Table 1 descrbes the sample n terms of percentages of acqurer and target frms by ndustry. The largest percentage of acqurer and target frms represent the technology, ndustral, fnancals, and health care ndustres. Table 2 reports descrptve statstcs for merger announcements by busness cycle phase and merger type. Panel A reveals that there were a total of 3,492 total frms nvolved n mergers and acqustons durng recessons versus a total of 28,323 frms durng non-recessons. The total percentage of frms nvolved n M&A actvty durng recessons and non-recessons s 11% and 89%, respectvely. As expected, merger actvty s reduced durng economc downturns (Cools, Gell, Kengelbach, & Roos, 2007; Aguar & Gopnath, 2005). The target frm sample conssts of 363 (9.6%) frms durng recessons and 3,402 (90.4%) frms durng non-recessons. Smlarly, the acqurer frms sample conssts of 3,192 (11.2%) frms durng recessons and 24,921 (88.8%) frms durng non-recessons. The total number of target frms that had returns avalable n CRSP s 3,765 whle the number of acqurer frms s much larger at 28,050. The lower avalablty of target stock return data s due to the lack of publcly traded shares for a majorty of these frms. 64 Journal of Accountng and Fnance Vol. 16(6) 2016

5 TABLE 1 INDUSTRY COMPOSITION OF ACQUIRER AND TARGET FIRMS Industry Acqurer Target Consumer Products and Servces 6.7% 6.5% Consumer Staples 5.4% 3.9% Energy and Power 6.7% 7.2% Fnancals 13.4% 20.3% Healthcare 10.9% 13.4% Hgh Technology 19.0% 18.3% Industrals 15.3% 9.6% Materals 7.4% 5.2% Meda and Entertanment 4.3% 3.7% Real Estate 3.9% 3.7% Retal 3.5% 3.5% Telecommuncatons 3.5% 4.8% Total 100% 100% TABLE 2 DESCRIPTIVE STATISTICS ON MERGER ANNOUNCEMENTS DURING RECESSIONS AND NON-RECESSIONS AND FOR SAME- AND CROSS-INDUSTRY MERGER ANNOUNCEMENTS Panel A Target % Acqurer % Total % Recessons % 3, % 3, % Non-recessons 3, % 24, % 28, % Total 3, % 28, % 31, % % 11.8% 88.2% 100.0% Panel B Target % Acqurer % Total % Same-Industry 2, % 19, % 22, % Cross-Industry 1, % 8, % 9, % Total 3, % 28, % 31, % % 11.8% 88.2% 100.0% We also provde data for same-ndustry and cross ndustry announcements n Table 2 Panel B. Panel B reveals that same-ndustry mergers occur more often than cross-ndustry mergers. Same-Industry mergers accounted for 69.7% (22,178) of the total announcements, leavng 30.3% (9,637) attrbuted to cross-ndustry announcements. Same-ndustry target frms make up 72.4% of the sample whle crossndustry target frms represent 27.6% of the sample. Ths s lkely due to the percepton of an easer assmlaton n acqurng and managng a frm wthn a frm s own ndustry compared to a dversfyng acquston. Table 3 provdes more detaled data for the types of mergers that occur durng recessons and nonrecessons. Durng recessons, there are 2,428 (69.5%) same-ndustry mergers whle there are only 1,064 (30.5%) cross-ndustry mergers. Therefore, a majorty of mergers durng recessons nvolve frms wthn Journal of Accountng and Fnance Vol. 16(6)

6 the same ndustry. Smlar fndngs occur durng non-recessons. In non-recessons, 69.7% of M&A actvty nvolves frms wthn the acqurer s same ndustry and 30.3% occurs outsde of the frm s ndustry. TABLE 3 DESCRIPTIVE STATISTICS ON MERGER ANNOUNCEMENTS BY ECONOMIC PHASE AND TYPE OF MERGER Economc Phase Type of Merger Target % Acqurer % Total % Recessons Same-Industry % 2, % 2, % Cross-Industry % % 1, % Total % 3, % 3, % % 10.4% 89.6% 100.0% Non-recessons Same-Industry 2, % 17, % 19, % Cross-Industry % 7, % 8, % Total 3, % 24, % 28, % % 12.0% 88.0% 100.0% METHODOLOGY Abnormal Performance Measurement The Fama-French (Fama & French, 1993) three factor model s used to estmate abnormal returns usng the followng cross-sectonal equaton: R, t Rft = α β ( Rmt Rft ) γ SMBt δ HMLt + ε, t where t = -100,., -11. Also, R,t s stock s return, Rf t s a rsk-free rate, and (Rm t Rf t, SMB t, HML t ) represents the market rsk premum, frm sze, and book-to-market factors n perod t, respectvely. Rm t Rf t s the value-weghted return on all NYSE, AMEX, and NASDAQ stocks mnus the one-month T- bll rate. The hstorc excess return of small captalzaton stocks over large captalzaton stocks s measured by SMB (.e. small mnus bg). The hstorc excess return of hgh book-to-market equty (value stocks) over low book-to-market equty s measured by HML (.e. hgh mnus low). Fama and French (1992) observed that small cap stocks and stocks wth a hgh book-to-market rato tend to have superor hstorc performance relatve to the market as a whole. These two factors proxy for addtonal rsk factors beyond that of the tradtonal CAPM beta. Fama and French (1992) showed that these three stock market factors (Rm t Rf t, SMB and HML) capture a larger and more statstcally sgnfcant fracton of the varaton n stock returns. The resduals produced by the three factor model solate the frm-specfc component of returns better than the resduals of the tradtonal CAPM model (Fama & French, 1993). Abnormal returns are calculated usng the estmated coeffcents wth the followng equaton for t= - 10,., 5: AR t t t = R, Rf α β ( R Rf ) γ SMB δ HML (2) mt t Once the excess abnormal returns (ARs) are calculated, excess CARs are calculated for each frm over the event perod (t = -10,, 0,.5): t t (1) 66 Journal of Accountng and Fnance Vol. 16(6) 2016

7 5 CAR t AR, t t= 5 EMPIRICAL RESULTS = (3) We study the possble asymmetrc market reacton to target and acqurng frms durng recessonary tme perods as opposed to non-recessonary tme perods. We examne several peces of ths ssue n the followng way. We dstngush between mergers nvolvng target frms n ndustres smlar to the acqurng frm versus mergers nvolvng target frms n ndustres dfferent from the acqurng frm. We expect that same-ndustry mergers should result n sgnfcantly dfferent cumulatve abnormal returns than those n cross-ndustry mergers. Ths s expectaton arses due to the synerges that naturally arse from combnng two frms n the same ndustry. Frst, we smply examne whether there are dfferences n returns for same-ndustry and cross-ndustry cumulatve abnormal returns. Then, we expand our analyss to study the effect of the busness cycle on same-ndustry and cross-ndustry mergers. Are Same-Industry Target Returns Dfferent from Cross-Industry Target Returns? To test ths hypothess, the sample of mergers s segregated nto groups based upon two basc types of mergers. We group frms nto same-ndustry mergers and cross-ndustry mergers. The results n Panel A of Table 4 show the cumulatve abnormal returns from the day before the announcement to 5 days afterwards for these two categores of target and acqurer frms. Panel A ndcates that there s no sgnfcant dfference between the CARs of same-ndustry and cross-ndustry frms. Ths fndng s contrary to our prevously stated expectaton of hgher same-ndustry target returns. However, the study has not yet ncorporated the possble effects of the current phase of the busness cycle. TABLE 4 CUMULATIVE ABNORMAL EVENT PERIOD RETURNS FOR SAME-INDUSTRY VERSUS CROSS-INDUSTRY MERGERS Days Relatve to AD A. Target Frms I. Same-Industry (A1) CAR (%) (-5, t) 2.99% 16.15% 20.86% 20.64% p-value (n=2,724) II. Cross-Industry (A2) CAR (%) (-5, t) 3.13% 16.68% 20.63% 23.39% p-value (n=1041) Dfference n CARs (A1) - (A2) -0.14% -0.53% 0.23% -2.75% p-value B. Acqurng Frms I. Same-Industry (B1) CAR (%) (-5, t) -0.09% 0.10% 0.24% 0.52% p-value (n=8,779) II. Cross-Industry (B2) CAR (%) (-5, t) 0.09% 0.06% 0.20% 0.16% p-value (n=3,810) Dfference n CARs (B1) - (B2) -0.18% 0.03% 0.04% 0.36% p-value Journal of Accountng and Fnance Vol. 16(6)

8 Are Same-Industry Acqurer Returns Dfferent from Cross-Industry Acqurer Returns? Panel B of Table 4 reveals a small, but statstcally sgnfcant dfference n cumulatve abnormal returns for acqurer frms on day 5 after the announcement. Same-ndustry acqurers experence 0.36% hgher CARs than cross-ndustry acqurers (p=.0163). Pror research also provdes evdence of nsgnfcant short-term acqurer returns (Goergen and Renneboog, 2004). However, ths fndng may ndcate that the market vews same-ndustry M&A actvty more favorably than cross-ndustry M&A actvty. The fndngs n ths table ndcate that the market values acqurer frms that engage n relatvely safer same-ndustry mergers. Ths seems to mply that the market values acqustons that nvolve samendustry expertse over dversfyng, cross-ndustry mergers. Ultmately, ths fndng appears to be related to agency theory expectatons rather than nformaton asymmetry or behavor theory. Next, we analyze the nsght ganed by examnng the cumulatve abnormal returns n same-ndustry and cross-ndustry mergers across busness cycles. For reportng purposes, we analyze target frms n recessons and non-recessons separately from acqurng frms. For example, Table 5 only reports the results for target frms and Table 6 only reports the results for acqurer frms. Do Economc Cycles Affect Same-Industry Target Returns and Cross-Industry Target Returns Equally? There are many dfferent comparsons that can be studed n Panel A of Table 5. Frst, we examne whether same-ndustry target frms experence larger postve abnormal returns durng recessons than same-ndustry target frms durng non-recessons. As mentoned before, Wann and Lamb (2016) fnd that target frm CARs are 3.53% to 8.12% hgher durng recessons than non-recessons. Same-ndustry target frms durng non-recessons experence sgnfcant CARs n the range of 15.80% to 20.21% after the merger announcement. Durng recessons, same-ndustry target frms durng recessons experence sgnfcant CARs n the range of 19.49% to 27.45%. Therefore, target frms from same ndustry mergers experence 3.69% to 7.53% larger postve CARs durng recessons than n nonrecessons over tme. The dfference between the recesson and non-recesson CARs are statstcally sgnfcant from day 0 to 5. The trend n the dfferences also ncreases monotoncally. Thus, t appears that good news n bad tmes s stronger than good news n good tmes. Second, we examne whether cross-ndustry target frms experence larger postve abnormal returns durng recessons than cross-ndustry target frms durng non-recessons. Durng non-recessons, crossndustry target frms durng recessons experence sgnfcant CARs n the range of 16.20% to 23.51%. Cross-ndustry target frms durng recessons experence sgnfcant CARs n the range of 21.00% to 23.83% over the tme perod studed. Even though the target CARs durng recessons are larger than those n non-recessons, the dfferences between the CARs are not statstcally sgnfcant at the 5% level. However, at the 10% level, cross-ndustry target returns are 4.80% hgher durng recessons. Ths evdence s also consstent wth good news beng worth more n bad tmes. Fnally, there are no statstcal dfferences between CARs for same-ndustry and cross-ndustry target frms durng non-recessons (Panel C) and recessons (Panel D). Do Economc Cycles Affect Same-Industry Acqurer Returns and Cross-Industry Acqurer Returns Equally? We examne whether same-ndustry acqurer frms experence larger postve abnormal returns durng recessons than same-ndustry acqurer frms durng non-recessons. The results are presented n Table 6. Pror lterature suggests that we should not fnd statstcal dfferences (Wann and Lamb, 2016; Goergen and Renneboog, 2004). Panel A of Table 6 reveals that same-ndustry acqurer frms durng recessons experence sgnfcant CARs of 0.28% to 0.54% on days 1 and 5 durng non-recessons. Durng recessons, same-ndustry acqurer frm CARs are not statstcally sgnfcant. The dfference between the recesson and nonrecesson CARs are not statstcally sgnfcant durng event perod. The busness cycle phase does not appear to affect CARs for same-ndustry acqurer frms. 68 Journal of Accountng and Fnance Vol. 16(6) 2016

9 TABLE 5 CUMULATIVE ABNORMAL RETURNS FOR SAME-INDUSTRY VERSUS CROSS- INDUSTRY TARGET FIRMS DURING RECESSIONS AND NON-RECESSIONS Days Relatve to AD A. Same-Industry Mergers I. Non-Recesson (A1) CAR (%) (-5, t) 2.98% 15.80% 20.21% 19.92% p-value (n=2,466) II. Recesson (A2) CAR (%) (-5, t) 3.04% 19.49% 27.07% 27.45% p-value (n=258) Dfference n CARs (A1) - (A2) -0.06% -3.69% -6.86% -7.53% p-value B. Cross-Industry Mergers I. Non-Recesson (B1) CAR (%) (-5, t) 2.92% 16.20% 20.27% 23.51% p-value (n=936) II. Recesson (B2) CAR (%) (-5, t) 4.88% 21.00% 23.83% 22.34% p-value (n=105) Dfference n CARs (B1) - (B2) -1.96% -4.80% -3.56% 1.17% p-value C. Dfference n CARs (A1) - (B1) 0.06% -0.40% -0.06% -3.59% p-value D. Dfference n CARs (A2) - (B2) -1.84% -1.51% 3.24% 5.11% p-value Next, we examne whether cross-ndustry acqurer frms experence dfferent abnormal returns durng non-recessons and recessons. In Panel B of Table 6, we fnd that cross-ndustry acqurer frm CARs are not statstcally sgnfcant durng the event perod, durng both phases of the economc cycle. Further, at the 10% level, there are no sgnfcant dfferences between abnormal returns for acqurers n recessons and non-recessons. Panel C reports the dfferences n same-ndustry and cross-ndustry acqurer returns durng nonrecessons. Fve days after the announcements, same-ndustry acqurers earn 0.34% hgher CARs than cross-ndustry acqurers durng non-recessons (p=0.0336). Therefore, the results found n Table 4 seem to orgnate from non-recessonary tmes. Ths fndng of a postve result for acqurers could ndcate that market partcpants vew non-recessons as safer tmes to engage n M&A actvty than durng recessons. Fnally, there are no statstcal dfferences between CARs for same-ndustry and cross-ndustry acqurer frms durng recessons (Panel D). Ths s somewhat surprsng n the sense that recessons should be vewed more negatvely and reflected as more negatve CARs for acqurers. Journal of Accountng and Fnance Vol. 16(6)

10 TABLE 6 ABNORMAL ANNOUNCEMENT DAY RETURNS FOR SAME-INDUSTRY VERSUS CROSS- INDUSTRY ACQUIRER FIRMS DURING RECESSIONS AND NON-RECESSIONS Days Relatve to AD A. Same-Industry Mergers I. Non-Recesson (A1) CAR (%) (-5, t) 0.10% 0.11% 0.28% 0.54% p-value (n=17,284) II. Recesson (A2) CAR (%) (-5, t) -0.14% -0.05% 0.00% 0.37% p-value (n=2,170) Dfference n CARs (A1) - (A2) 0.23% 0.16% 0.28% 0.17% p-value B. Cross-Industry Mergers I. Non-Recesson (B1) CAR (%) (-5, t) -0.09% 0.11% 0.27% 0.20% p-value (n=7,637) II. Recesson (B2) CAR (%) (-5, t) -0.01% -0.41% -0.48% -0.13% p-value (n=959) Dfference n CARs (B1) - (B2) -0.08% 0.52% 0.76% 0.32% p-value C. Dfference n CARs (A1) - (B1) 0.19% 0.00% 0.00% 0.34% p-value D. Dfference n CARs (A2) - (B2) -0.12% 0.36% 0.48% 0.50% p-value CONCLUSION Ths paper utlzes an event study methodology to nvestgate the possble dfferences n market reactons to same- and cross-ndustry merger and acquston actvty durng recessons and nonrecessons. Frst, the paper reports the frequency of same- and cross-ndustry mergers and the number of frms nvolved n M&A actvty durng dfferent phases of the economc cycle from 1971 to Second, the paper nvestgates whether the market regards the announcement of same-ndustry and crossndustry mergers and acqustons equally. Thrd, the paper studes the mpact of the busness cycle on announcement perod returns for same- and cross-ndustry mergers and acqustons. The study fnds that the largest percentage of acqurer and target frms represent the technology, ndustral, fnancals, and health care ndustres. Most M&A actvty occurs wthn the frm s same ndustry (70%) as opposed to dfferent a dfferent ndustry (30%). Durng tmes of recessons, M&A actvty occurs much less frequently than durng non-recessons. Roughly 10% of the sample M&As occur durng recessons compared to 90% that occur durng non-recessons. The frst research queston explores whether same-ndustry target returns are dfferent from crossndustry target returns. The reported results ndcate no dfference between the target returns of same- 70 Journal of Accountng and Fnance Vol. 16(6) 2016

11 ndustry and cross-ndustry mergers. Ths fndng seems surprsng due to the expectaton that samendustry mergers would be nterpreted more postvely by the market than cross-ndustry mergers. The second research queston s: Are same-ndustry acqurer returns dfferent from cross-ndustry acqurer returns? Smlar to the frst research queston, the expected answer s yes. The results ndcate that same-ndustry acqurers experence small but sgnfcantly hgher CARs than cross-ndustry acqurers 5 days after the announcement. Therefore, there s some evdence that market partcpants vew samendustry mergers more postvely than cross-ndustry mergers. Ths s lkely due to the perceved synerges of acqurng a frm n an ndustry that s famlar rather than a frm n a completely dfferent ndustry. The thrd research queston nvestgates whether economc cycles have an equal effect on same- and cross-ndustry target returns. Ths paper shows sgnfcant dfferences n returns to target frms across the busness cycle. Target frms from same-ndustry mergers experence 3.69% to 7.53% larger postve CARs durng recessons than n non-recessons over the event perod. Further, target frms from crossndustry mergers experence 4.80% hgher returns on the announcement day durng recessons. These results mply that good news n bad tmes s worth more than good news n good tmes. In other words, market partcpants exhbt an asymmetrc response to target frms based upon the economc cycle. Ths result ndcates the presence of a behavoral bas whch warrants further nvestgaton of the mpact of economc cycles on research results n much of the exstng fnance lterature. The fourth and fnal research queston s: Do economc cycles affect same-ndustry acqurer returns and cross-ndustry acqurer returns equally? The answer s yes for most of the results. However, as reported before, the second research queston reports hgher same-ndustry acqurer returns 5 days after the merger announcement. Ultmately, ths specfc result can be attrbuted to non-recessonary tme perods where same-ndustry acqurers earn 0.34% hgher CARs than cross-ndustry acqurers fve days after the announcement. Ths study s not wthout lmtatons. Ths research only examnes short run market perceptons regardng M&A actvty and provdes no nformaton as to the long term performance of the newly formed company. Further, a survey of nvestors may reveal dfferent opnons regardng reactons to M&A actvty for same- and cross-ndustry target and acqurers. Survey results may also reveal that the economc cycle s not deemed as mportant by nvestors when analyzng M&A deals. However, survey results can be based and ths paper does provde a helpful understandng of what actually occurs wth respect to average short-run stock market returns. Future research could further segregate the results found n ths study by these ndustres to see f there s n ndustry effect. Ths may prove useful snce about 60% of the sample mergers occur n the technology, ndustral, fnancals, and health care ndustres. Smlarly, ndustry effects could be studed wthn the context of the current economc cycle. Other future research could study whether there are dfferences between target and acqurer returns n cross-border and same-country M&A actvty based upon the phase of the economc cycle. REFERENCES Aguar, M., & Gopnath, G. (2005). Fre-sale foregn drect nvestment and lqudty crses. Revew of Economcs and Statstcs, 87, (3), Aharon, Y., Thany, L., & Connelly, B. L. (2011). Manageral decson-makng n nternatonal busness: A forty-fve-year retrospectve. Journal of World Busness, 46, (2), Andrade, G., Mtchell, M., & Stafford, E. (2001). New evdence and perspectves on mergers. Journal of Economc Perspectves, 15, (2), Argote, L., & Greve, H. R. (2007). A behavoral theory of the frm-40 years and countng: Introducton and mpact. Organzaton Scence, 18, (3), Bard, D. G., & Rasmussen, R. K. (2003). Chapter 11 at twlght. Stanford Law Revew, 56, Baumester, R.F., Bratslavsky, E., Fnkenauer, C., & Vohs, K. D. (2001). Bad s stronger than good. Revew of General Psychology, 5, Journal of Accountng and Fnance Vol. 16(6)

12 Bhagat, S., Dong, M., Hrshlefer, D., & Noah, R. (2005). Do tender offers create value? New methods and evdence. Journal of Fnancal Economcs, 76, Bloom, N., (2009). The mpact of uncertanty shocks. Econometrca, 77, (3), Bloom, N., (2014). Fluctuatons n uncertanty. Journal of Economc Perspectves, 28, (2), Bruner, R. F. (2002). Does M&A pay? A survey of evdence for the decson-maker. Journal of Appled Fnance, 12, (1), Chevaler J., (2004). What do we know about cross-subsdzaton? Evdence from mergng frms. The B.E. Journal of Economc Analyss & Polcy, 4, (1), Connelly, B. L., Certo, S. T., Ireland, R. D., & Reutzel, C. R. (2011). Sgnalng theory: A revew and assessment. Journal of Management, 37, (1), Cools, K., Gell, J., Kengelbach, J., & Roos, A. (2007). The brave new world of M&A: How to create value from mergers and acqustons, Boston: Boston Consultng Group. Cordng, M., Chrstmann, P., & Wegelt, C. (2010). Measurng theoretcally complex constructs: The case of acquston performance. Strategc Organzaton, 8, (1), Cuypers, I. R., & Martn, X. (2010). What makes and what does not make a real opton? A study of equty shares n nternatonal jont ventures. Journal of Internatonal Busness Studes, 41, (1), Cyert, R. M., & March, J. G. (1963). A behavoral theory of the frm, Englewood Clffs, NJ: Prentce Hall. Dng, D. & Rahaman, M. M., (June 26, 2010). Booms, busts, and frm ext: Evdence from M&A actvtes across busness cycles, avalable at SSRN: or Fama, E. F. (1970). Effcent captal markets: A revew of theory and emprcal work. Journal of Fnance, 25, (2), Fama, E. F., & French, K. R. (1989). Busness condtons and expected returns on stocks and bonds. Journal of Fnancal Economcs, 25, (1), Fama, E. F., & French, K. R. (1992). The cross secton of expected stock returns. The Journal of Fnance, 47, (2), Fama, E. F., & French, K. R. (1993). Common rsk factors n the returns on stocks and bonds. Journal of Fnancal Economcs, 33, (1), Gaughan, P. A. (2011). Mergers, acqustons, and corporate restructurngs, New Jersey: John Wley and Sons, Inc. Gavett, G., & Rvkn, J. W. (2007). On the orgn of strategy: Acton and cognton over tme. Organzaton Scence, 18, (3), Goergen, M., & Renneboog, L. (2004). Shareholder wealth effects of European domestc and cross border takeover bds. European Fnancal Management, 10, (1), Gomez-Meja, L. R., Haynes, K. T., Nunez-Nckel, M., Jacobson, K. J., & Moyano-Fuentes, J. (2007). Socoemotonal wealth and busness rsks n famly-controlled frms: Evdence from Spansh olve ol mlls. Admnstratve Scence Quarterly, 52, (1), Greve, H. R. (2003). Organzatonal learnng from performance feedback: A behavoral perspectve on nnovaton and change, Cambrdge, UK: Cambrdge Unversty Press. Htt, M. A., Freeman, R. E., & Harrson, J. S. (2001). The Blackwell handbook of strategc management, Oxford: Blackwell Busness. Hunter, J. E., & Coggn, T. D. (1988). Analyst judgment: The effcent market hypothess versus a psychologcal theory of human judgment. Organzatonal Behavor and Human Decson Processes, 42, (3), Ish, J., & Xuan, Y., (2014). Acqurer-target socal tes and merger outcomes. Journal of Fnancal Economcs, 112, (3), Jensen, M. C. (1991). Corporate control and the poltcs of fnance. Journal of Appled Corporate Fnance, 4, (2), Kahneman, D. & Tversky, A., (1979). Prospect theory: An analyss of decson under rsk. Econometrca, 47, (2), Journal of Accountng and Fnance Vol. 16(6) 2016

13 Kng, D. R., Dalton, D. R., Daly, C. M., and Covn, J. G. (2004). Meta analyses of post acquston performance: Indcatons of undentfed moderators. Strategc Management Journal, 25, (2), Klen, P. G. (2001). Were the acqustve conglomerates neffcent? The Rand Journal of Economcs, 32, (4), Kock, C. J. (2005). When the market msleads: Stock prces, frm behavor, and ndustry evoluton. Organzaton Scence, 16, (6), Lee, D. R., & Verbrugge, J. A. (1996). The effcent market theory thrves on crtcsm. Journal of Appled Corporate Fnance, 9, (1), March, J. G., & Smon, H. A. (1958). Organzatons, New York: Wley. McDermott, R. Fowler, J. H. & Smrnov, O., (2008). On the evolutonary orgn of prospect theory preferences. Journal of Poltcs, Vol. 70, (2), McQueen, G., & Roley, V. V. (1993). Stock prces, news, and busness condtons. The Revew of Fnancal Studes, 6, (3), Moeller, S. B., Schlngemann, F. P., & Stulz, R. M. (2005). Wealth destructon on a massve scale? A study of acqurng frm returns n the recent merger wave. The Journal of Fnance, 60, (2), Myers, S. C., & Majluf, N. S. (1984). Corporate fnancng and nvestment decsons when frms have nformaton that nvestors do not have. Journal of Fnancal Economcs, 13, (2), Offcer, M. S., Poulsen, A. B., & Stegemoller, M. (2009). Target-frm nformaton asymmetry and acqurer returns. Revew of Fnance, 13, (3), Rhodes, D., & Stelter, D. (2009). Seze advantage n a downturn. Harvard Busness Revew, 87, (2), Schjven, M., & Htt, M. A. (2012). The vcarous wsdom of crowds: Toward a behavoral perspectve on nvestor reactons to acquston announcements. Strategc Management Journal, 33, (11), Shlefer, A., & Vshny, R. W. (2003). Stock market drven acqustons. Journal of Fnancal Economcs, 70, (3), Soroka, S N (2006). Good news and bad news: Asymmetrc responses to economc nformaton. Journal of Poltcs, 68, (2), Verones, P. (1999). Stock market overreactons to bad news n good tmes: A ratonal expectatons equlbrum model. Revew of Fnancal Studes, 12, (5), Wang, C. & Xe, F., (2009). Corporate governance transfer and synergstc gans from mergers and acqustons. Revew of Fnancal Studes, 22, (2), Wann, C. R., & Lamb, N. H., (2016). Merger and acquston returns: Does the busness cycle matter? Workng Paper, Unversty of Tennessee at Chattanooga. Zajac, E. J., & Westphal, J. D. (2004). The socal constructon of market value: Insttutonalzaton and learnng perspectves on stock market reactons. Amercan Socologcal Revew, 69, (3), Zuckerman, E. W. (2004). Structural ncoherence and stock market actvty. Amercan Socologcal Revew, 69, (3), Journal of Accountng and Fnance Vol. 16(6)