1
The Long-Term Post-Acquisition Stock Performance of European Acquiring Firms
Using the Calendar-Time Abnormal Return (CTAR) Methodology
Master Thesis Finance
Author: Rutgur Boersma Student Number: 1752898
August 31, 2010
University of Groningen Faculty of Economics & Business
MSc Business Administration, Specialization Finance
Abstract
This paper examines the long-term post-acquisition stock performance of European acquirers with the Calendar-Time Abnormal Return (CTAR) method for the period 1997-2006. The long-term performance of European acquirers is examined for different types of mergers and acquisitions (M&A) namely; all, domestic, cross-border, intercontinental and intracontinental M&A. Overall, I find no clear significant evidence of long-term abnormal performance four each type of M&A. Moreover, I do not find significant evidence that domestic M&A outperform cross-border M&A and that intracontinental M&A outperform intercontinental M&A. I also examine whether the way in which a deal is financed affects long-term performance of European acquirers and find that on the long-term cash acquisitions perform better than share acquisitions, as expected. However, cash acquisitions do not significantly outperform share acquisitions in all subsamples. Finally, I examine whether the legal system of European countries affects the long-term post-acquisition abnormal performance by distinguishing between civil law (continental Europe) and common law (U.K. and Ireland) European countries. Acquirers form common law countries perform better than acquirers from civil law countries on the long-term, although the former does not significantly outperform the latter.
JEL classification: G12, G14, G34
2 1. Introduction
Researchers have devoted considerable time to answer the question whether mergers and acquisitions (M&A) are wealth creating or wealth reducing events for shareholders. The bulk of research examines the short-term stock performance of both acquiring and target firms surrounding announcement dates of M&A. Strong evidence exists for value creation from the perspective of target firms and mixed results from the perspective of acquiring firms (Agrawal and Jaffe, 2000). Although the short-term stock performance of acquiring firms surrounding announcement dates of M&A gives mixed results, the conclusion of Roll (1986), that the null hypotheses of zero abnormal return should not be rejected, seems to hold (Agrawal and Jaffe, 2000).
The long-term post-acquisition stock performance of acquiring firms has received considerable less attention of researchers and gives mixed results. Moreover, the results vary by the type of M&A (e.g. domestic and cross-border M&A). For example, Gregory (1997), Loughran and Vijh (1997) and André et al. (2004) find significant negative long-term post-acquisition abnormal returns for domestic M&A while Mitchell and Stafford (2000) and Dutta and Jog (2009) find no evidence of significant long-term post-acquisition abnormal returns for domestic M&A. The empirical evidence of long-term post-post-acquisition stock performance for cross-border M&A also gives mixed results. For example, André et al. (2004) and Black et al. (2001) find significant negative post-acquisition abnormal returns for cross-border M&A while Conn et al. (2005) and Gregory and McCorriston (2001) find insignificant negative long-term post-acquisition abnormal returns for cross-border M&A.
3
I refer to the long-term as a three-year event window starting from the completion date of mergers and acquisitions. Long-term event studies differ from short-term event studies because the former is sensitive to the model applied for computing normal returns, which may partially explain the mixed findings in past research (Rau and Vermaelen, 1998). I follow the view of Mitchell and Stafford (2000) and Fama (1998) that the calendar-time abnormal return (CTAR) methodology is superior to the buy-and-hold abnormal return methodology in detecting long-term abnormal returns as explained in the methodology section.
The main objective of this thesis is to examine the long-term post-acquisition stock performance of European acquirers for domestic, cross-border, intracontinental and intercontinental M&A in the period 1997-2006. More specific, I examine whether the long-term post-acquisition performance between domestic and cross-border M&A significantly differs because cross-border M&A are more complex than domestic M&A due to differences in political and economic environments, corporate organizations, cultures, traditions, tax rules, and law and accounting rules between the acquirer and target country (Sudarsana, 2003). I also examine whether intercontinental and intracontinental M&A long-term post-acquisition abnormal performance significantly differs because the establishment of the Economic and Monetary Union (EMU) and the European Union (EU) has brought a single market with standardized laws applying in all the member states, ensuring the freedom of movement of people, goods, services, and capital. These developments are likely to make intracontinental M&A less complex than intercontinental M&A with respect to the differences in political and economic environments, corporate organizations, cultures, traditions, tax rules, and law and accounting rules between the acquirer and target country.
I also test if the method of payment (cash or shares) affects long-term post-acquisition abnormal performance for all types (all, domestic, cross-border, intracontinental and intercontinental) of M&A. When the acquirer uses shares as the method of payment, investors know that the shares of the acquirer are overvalued, hence the share price of the acquirer will ceteris paribus fall and vice versa (Myers and Majluf, 1984). If the form of payment affects long-term post-acquisition abnormal returns, then the market does not react efficiently to the news conveyed by the form of the payment.
4
protection and more concentrated ownership structures compared to the U.S. and the U.K. (La Porta et al.,1998). To my knowledge, no study has examined if long-term post-acquisition abnormal performance differs between civil law countries (continental Europe) and common law countries (U.K. and Ireland). More concentrated ownership structures by outsiders, decrease agency costs as block holders are more able to actively monitor the firm compared to dispersed shareholders (denis and McConnel, 2003). However, block holders can also expropriate wealth from minority shareholders if their control gives rise to private benefits that are not available to other shareholders (denis and McConnel, 2003). Lower shareholder protection and the uncertain effect of concentrated ownership structures imply higher agency costs for acquirers from civil law countries then for acquirers from common law countries. If the market underestimates the additional costs of higher agency costs, then the acquiring firm’s stock performance will probably be worse than expected. Furthermore, if the market slowly adjusts the acquiring firm’s stock price after experiencing that the acquiring firm’s performance is worse than expected, one would find negative long-term post-acquisition abnormal performance. Therefore, I will test if long-term post-acquisition abnormal performance significantly differs between European acquirers from civil law and common law countries.
The remainder of the paper proceeds as follows; the next section reviews previous work with respect to long-term post-acquisition performance. The third section discusses how I collected the data and gives the summary statistics of the data. Section 4 discusses the applied methodology and explains why I prefer the CTAR methodology for measuring long-term performance. The main results are discussed in section 5 and section 6 concludes.
2. Literature review 2.1 Theories
5
and strong. This study assumes that prices reflect all publicity available information, which is equivalent to the semi-strong form of efficiency.
Short-term event studies differ from long-term event studies in that the former examines abnormal returns within a short event window (e.g. a few days) around a cleanly dated event, whereas the latter examines abnormal returns within a long event window (e.g. three years) after a cleanly dated event. The assumption of the EMH is that any lag in the response of prices to an event is short-lived (Fama, 1998). The behavioural finance literature questions this assumption, arguing that the market slowly adjust stock prices to new information. Therefore, one must examine abnormal returns over a long horizon to test for market inefficiency.
Evidence of significant long-term post-acquisition abnormal returns imply that the market slowly adjusts stock prices of the acquiring firm to information on mergers and acquisition. Thus, significant negative long-term post-acquisition abnormal returns imply a slow correction on an initial overreaction at the time of the M&A announcement and vice versa. The representativeness and the conservatism bias are behavioural explanations for over and under reactions respectively to certain information for which the market slowly adjusts (Ross et al., 2008).
The representativeness bias implies overweighting the results of small samples leading to an overreaction in stock returns (Ross et al., 2008). Stock market bubbles might be an example of overreactions to good news. During the late 1990s internet companies showed high levels of revenue growth. Since many investors believed that these high levels of revenue growth would last indefinitely, stock prices rose. After many investors realized that the probability of high levels of revenue growth would last indefinitely were overstated, stock prices plummeted.
The conservatism bias implies that the market is slow in adjusting stock prices to new information leading to an under reaction in stock prices. For example, stock prices rise slowly following announcements of positive earnings surprises and vice versa (Kolasinski and Li, 2005).
2.2 Long-term studies
6
on the calendar months in which an event takes place and the latter does not. Another important issue affecting measurement of long-term abnormal returns reflects weighting procedures. Ordinary Least Square (OLS) and Weighted Least Square (WLS) regressions as well as Value Weighted (VW) and Equally Weighted (EW) event firm portfolios affect long-term abnormal returns (André et al., 2004). The methodology section discusses the two methods for measuring long-term abnormal returns and the impact of weighting procedures more extensively.
2.2.1 Domestic mergers and acquisitions
Agrawal and Jaffe (2000), which reviewed 22 studies concerning long-term post-acquisition performance during 1974-1998, report that except for Frank et al. (1991) each paper shows more or less evidence of significant negative long-term post-acquisition stock performance. However, several more recent studies did not found significant long-term post-acquisition abnormal returns (e.g. Mitchell and Stafford, 2000, Dutta and Jog, 2009, and André et al., 2004). In addition, these more recent studies applied more sophisticated techniques (e.g. CTAR and BHAR) for measuring long-term abnormal returns. Because the more recent studies apply more sophisticate techniques, which can affect measures of long-term abnormal returns in a severe way, we put more weight on the findings of these later studies. Table 1.0 gives an overview of the studies that examine the long-term post-acquisition performance for domestic mergers and acquisitions.
7
Other studies that also performed more sophisticated methodologies are Gregory (1997), Loughran and Vijh (1997), and Rau and Vermaelen (1998) all of which found significant long-term abnormal performance. Gregory (1997) examined the U.K. market in the period 1984 to 1992 by applying six methodologies including the CTAR method (-0.8% per month) reporting significant negative long-term abnormal returns for all methodologies. Loughran and Vijh (1997) examined long-term abnormal performance for U.S. domestic M&A by applying the BHAR methodology. They find significant long-term abnormal returns of -15.9% although they do not control for the dependence problem discussed by Mitchell and Stafford (2000). Rau and Vermaelen (1998) find significant negative long-term abnormal returns for U.S. domestic mergers and acquisitions using the BHAR method. After controlling for the dependence problem described by Mitchell and Stafford (2000) these authors find BHARs of -4% for a three year post event period. In sum, the evidence of long-term post-acquisition performance gives mixed results. The most recent studies report some evidence of no long-term abnormal performance while the studies performed in the 1990s report significant negative long-term abnormal returns.
The empirical literature shows mixed results with respect to the method of payment hypothesis for domestic mergers and acquisitions. Loughran and Vijh (1997), find that stock acquisitions earn significant negative long-term post-acquisition abnormal returns (-24%) and cash acquisitions earn significant positive long-term post-acquisition abnormal returns (30.5%) over a five year period. Andre et al. (2004) and Gregory (1997) both find insignificant negative long-term post-acquisition abnormal returns following cash acquisitions and significant negative long-term post-acquisition abnormal returns following stock acquisitions. Furthermore, Andre et al. (2004) find that acquisitions financed with cash significantly outperform acquisitions financed with stock. Frank et al. (1991) and Dutta and Jog (2009) find negative and positive signs for equity and cash acquisitions respectively although insignificant. These authors also find that acquisitions financed with cash do not significantly outperform stock acquisitions.
2.2.2 Cross-border mergers and acquisitions
8
Table 1.0; an overview of studies examining the long-term post-acquisition performance for domestic mergers and acquisitions.
André et al. (2004) and Black et al. (2001) find significant negative long-term post-acquisition abnormal returns for cross-border M&A while Conn et al. (2005) and Gregory and McCorriston (2001) did not. Of these studies, André et al. (2004) and Conn et al. (2005) both performed the calendar-time abnormal return (CTAR) method for Canadian and U.S acquirers respectively. The results of both authors differ as André et al. (2004) found significant negative long-term abnormal returns for Canadian cross-border acquirers while Conn et al. (2005) found insignificant negative long-term abnormal returns. The Buy-and Hold Abnormal Return (BHAR) method also gives mixed results. Black et al. (2001) and Conn et al. (2005) find almost identical significant negative long-term abnormal returns of -13.2% and -13.4% respectively. However, Gregory and McCorriston (2001) find insignificant negative long-term post-acquisition abnormal returns for U.K. acquirers of -9.3%. André et al. (2004) also examined whether cross-border M&A significantly outperform domestic M&A and conclude that cross-border M&A significantly perform worse than domestic M&A.
9
Table 1.1; an overview of studies examining the long-term post-acquisition performance for cross-border mergers and acquisitions.
2.3 Hypotheses 2.3.1 Type of M&A
10
Cross-border M&A have also disadvantages compared to domestic mergers and acquisitions. Geographic diversification may also lower firm value through shareholders of diversified firms have more difficulty to monitor managerial actions compared to purely domestic firms (Bodnar et al., 1997). These authors point out that managers might adopt or maintain value decreasing diversification strategies that decrease shareholder value as long as managerial private benefits exceed their private costs. According to Sudarsana (2003), cross-border M&A are more complex than domestic M&A due to differences in political and economic environments, corporate organizations, cultures, traditions, tax rules, and law and accounting rules between the acquirer and target country. André et al., (2004) finds that cross-border M&A significantly perform worse than domestic mergers and acquisitions. Therefore, I expect that cross-border M&A significantly underperform compared to domestic M&A on the long term.
Hypothesis 1A: Long-term post-acquisition abnormal performance of cross-border M&A significantly differ from domestic M&A.
Conn et al. (2005) examine cross-border M&A for U.K. acquirers and report that long-term post-acquisitions abnormal returns are lower when cultural differences are greater between the U.K. and target countries. The establishment of the Economic and Monetary Union (EMU) and the European Union (EU) has brought a single market with standardized laws applying in all the member states, ensuring the freedom of movement of people, goods, services, and capital. To the extent that these developments within Europe make intracontinental cross-border M&A less complex with respect to differences in political and economic environments, corporate organizations, cultures, traditions, tax rules, and law and accounting rules between the acquirer and target country (Sudarsana, 2003), I expect that intracontinental M&A significantly outperform intercontinental M&A on the long-term. However, I will only find long-term post-acquisition abnormal returns if the market is not able to efficiently price the acquirer shares to the information of the type of the acquisition on the short term and slowly corrects for an initial under or overreaction on the long-term.
Hypothesis 1B: Long-term post-acquisition abnormal performance of intracontinental M&A significantly differ from intercontinental M&A
2.3.2 Method of payment hypothesis
11
market or if the valuation of the target shares is uncertain. When the acquirer uses shares as method of payment investors know that the shares of the acquirer are overvalued, hence the share price of the acquirer will ceteris paribus fall (e.g. Myers and Majluf, (1984), Frank, Harris and Titman, (1991), Loughran and Vijh, (1997), Rau and Vermaelen (1998) and Grinblatt and Titman, (2004)). Acquisitions financed with cash have a similar, but opposite effect. Thus, the method of payment hypothesis predicts that on average long-term post-acquisition abnormal returns are positive following cash acquisitions and negative following share acquisitions. However, the method of payment hypothesis will only hold if the market is not able to efficiently price the acquirer shares to the information how the acquisition is financed on the short term and slowly correct for an initial under or overreaction on the long-term. Significant negative (positive) long-term post-acquisition abnormal returns for cash and stock acquisitions imply an initial over (under) reaction.
According to Moeller and Schlingeman (2005), acquirers prefer to finance cross-border mergers and acquisitions with shares as it becomes more difficult for the acquirer to evaluate the foreign target. However, target firms often do not accept shares as the method of payment, which forces the acquirer to finance the cross-border M&A with cash (Gaughan, 2002). The positive signal from cash acquisitions might therefore diminish or non-exist for cross-border mergers and acquisitions (Moeller and Schlingeman, 2005).
Hypothesis 2A: Long-term post-acquisition abnormal returns for domestic (cross-border) M&A financed with cash significantly differ from domestic (cross-border) M&A financed with stock. I expect that the positive signalling effect for cash acquisitions diminish or non-exists to be higher for intercontinental M&A compared to intracontinental M&A because the acquirer is more difficult to evaluate due to higher information asymmetries.
Hypothesis 2B: Long-term post-acquisition abnormal returns for intracontinental (intercontinental) M&A financed with cash significantly differ from intracontinental (intercontinental) financed with
stock. 2.3.3 Legal system
12
al. (2003) are time dependent and find no evidence that investors underestimate the implications of agency costs or poor corporate governance. Finally, Johnson et al. (2009) find no significant relationship between corporate governance and long-term stock returns using the same sample as Gompers et al. (2003). I will test if higher agency costs have an effect on long-term post-acquisition returns by distinguishing between civil law and common law European countries. In Europe, the U.K. and Ireland are the only common law countries and all other countries within Europe are civil law countries (continental Europe).
Hypothesis 3: Long-term post-acquisition abnormal returns for continental European acquirers differs significantly from U.K. and Ireland acquirers
Continental Europe has less developed financial markets, weaker shareholder protection and more concentrated ownership structures compared to the U.S. and the U.K. (La Porta et al.,1998). More concentrated ownership structures by outsiders, decrease agency costs as block holders are more able to actively monitor the firm compared to dispersed shareholders (Milgrom and Roberts, 1992). However, block holders can also expropriate wealth from minority shareholders if their control gives rise to private benefits that are not available to other shareholders (denis and McConnel, 2003). Lower shareholder protection and the uncertain effect of concentrated ownership structures might imply higher agency costs for continental European countries compared to the U.K. and Ireland. If the market underestimates the additional costs of higher agency costs, then the acquiring firm’s stock performance will probably be worse than expected and vice versa. Furthermore, if the market slowly adjusts the acquiring firm’s stock price after experiencing that the acquiring firm’s performance is better or worse than expected, one would find long-term post-acquisition abnormal performance.
3. Data
This study uses the Zephyr database of Bureau van Dijk Electronic Publishing (BvDEP) to obtain event dates of mergers and acquisitions by European acquirers during the period 1997-2006. Following Mitchell and Stafford (2000), I use completion dates rather than announcement dates of mergers and acquisitions by European acquirers as a starting point to measure long-term post-acquisition abnormal returns for a three-year event window because this assures that the M&A are not withdrawn and rules out short-term market reactions surrounding the announcement date of mergers and acquisitions.
13
acquirer firm is listed on a major Euro stock exchange. Table 3.1 shows the European acquirer countries and the major stock exchanges that are examined in this study. Third, following André et al (2004), mergers and acquisitions have a deal value greater than ten million Euros. Fourth, firms with multiple M&A during the period 1997-2006 are included. Fifth, acquisitions are financed with cash or shares. Sixth, acquiring firms have an acquired stake in the target company greater than fifty percent. Seventh, the deal type is either a merger or an acquisition. Finally, monthly stock returns (for a three-year period beginning at the completion date) of acquiring firms are available in Thomson Reuters Datastream database.
Table 3.1; European acquirer countries and the major stock exchanges examined
County Stock Exchange County Stock Exchange County Stock Exchange
Austria Vienna Stock Exchange Finland Nasdaq OMX Helsinki Italy Italian Stock Exchange
Belgium Euronext Brussels France Euronext Paris Luxembourg Luxembourg Stock Exchange
Germany Frankfurt Stock exchange UK Londen Stock Exchange Netherlands Euronext Amsterdam
Denmark Nasdaq OMX Copenhagen Greece Athens Stock Exchange Portugal Euronext Lisabon
Spain Madrid Stock Exchange Ireland Irish Stock Exchange Sweden Nasdaq OMX Stockholm
14
3.1 Descriptive statistics for the total sample of M&A by European acquirers
Panel A of table 3.2 reports the distribution of the total number of M&A by European acquirers per acquirer country. The final sample consists of 1208 M&A of European acquirer firms during the research period 1997-2006. U.K. acquirers comprise 45.0 percent of the total number of acquisitions. Goergen and Renneboog, (2002) find a similar distribution in that U.K. acquirers comprise 40 percent of all European mergers and acquisitions. Next to the UK, France, Germany, Italy, Sweden, The Netherlands, and Spain acquired most firms with respect to the 15 European countries examined (EU 15), accounting together for 42.1 percent of the total number of M&A.
Panel B of table 3.2 reports the distribution of the total number of M&A by European acquirers per year for the whole research period. Throughout the research period, the total number of European M&A shows relative small increases and decreases except for the years 1997-1998, 1999-2000, and 2002-2003. A dramatic increase of the total number of M&A of 68.1 percent and 84.5 percent in 1997-1998 and 1999-2000 respectively followed by a severe decrease of -34.3 percent in 2002-2003.
3.2 Descriptive statistics by type of M&A
Panel A through D of table 3.3 report the number of M&A, the average deal value in millions of Euros, and the method of payment by year during the 1997-2006 period for domestic, cross-border, intercontinental, and intracontinental M&A respectively. The total sample consists of 1208 M&A of which 481 domestic M&A (39.8 percent of total sample) and 727 cross-border M&A (60.2 percent of total sample). Panel C and D subdivide the 727 cross-border M&A in 384 intercontinental (31.8 percent of total sample) and 343 intracontinental (28.4 percent of total sample). Intercontinental M&A consist of cross-border M&A of which the target firm is located outside the European continent while intracontinental M&A consist of cross-border M&A of which the target firm is located within the European continent.
15
for cross-border deals. The number of domestic M&A financed with a mixture of cash and shares equals 97 (20.2 percent) compared to 87 (12.0 percent) for cross-border M&A.
Panel C and D subdivide the 727 cross-border M&A in 384 intercontinental (panel C) and 343 intracontinental (panel D) M&A respectively. The results with respect to the method of payment show that intercontinental M&A are more often financed with cash or shares and less often with a mixture of cash and shares, although the differences are small and insignificant.
Table 3.2 Distribution of M&A by European Acquirers
Distribution of all mergers and acquisitions by acquirer country (Panel A) and by year (Panel B) during the 1997-2006 period. The M&A are performed by European acquirers that are a member of the European Union (EU) as of 1995.
Panel A. Distribution by country Acquirer country Total number of M&A per
country
Percentage of total number of M&A per country UK 543 45,0% France 163 13,5% Germany 75 6,2% Italy 74 6,1% Sweden 71 5,9% Netherlands 67 5,5% Spain 59 4,9% Belgium 31 2,6% Finland 30 2,5% Denmark 27 2,2% Ireland 27 2,2% Greece 23 1,9% Austria 14 1,2% Luxembourg 2 0,2% Portugal 2 0,2% Total 1208 100%
Panel B. Distribution by year
Year Total number of M&A per year
16 Table 3.3 Descriptive statistics by type of M&A
The number of M&A, the average deal value in millions of Euros, and the method of payment by year during the 1997-2006 period. The M&A are performed by European acquirers firms that are a member of the European Union (EU) as of 1995.
Panel A. Domestic M&A by European acquirers
Method of payment
Year
Total number of M&A per
year
Average deal value in millions Euros
Number of M&A financed with cash
Number of M&A financed with
shares
Number of M&A financed
with cash and shares 1997 19 83 8 4 7 1998 37 473 28 1 8 1999 41 713 19 11 11 2000 56 653 31 12 13 2001 61 333 32 11 18 2002 72 216 47 12 13 2003 50 479 35 10 5 2004 45 217 30 8 7 2005 47 122 36 5 6 2006 53 196 38 6 9 Total 481 3.485 304 80 97
Panel B. Cross-border M&A by European acquirers
Method of payment
Year
Total number of M&A per
year
Average deal value in millions Euros
Number of M&A financed with cash
Number of M&A financed with
shares
Number of M&A financed
17 Table 3.3 Descriptive statistics by type of M&A (continued)
The number of M&A, the average deal value in millions of Euros, and the method of payment by year during the 1997-2006 period. The M&A are performed by European acquirers that are a member of the European Union (EU) as of 1995.
Panel C. Intercontinental M&A by European acquirers
Method of payment
Year
Total number of M&A per year
Average deal value in millions Euros
Number of M&A financed with cash
Number of M&A financed with
shares
Number of M&A financed with cash and shares
1997 13 207 9 3 1 1998 24 717 22 2 0 1999 31 563 28 1 2 2000 72 599 49 11 12 2001 64 283 45 7 12 2002 38 545 25 6 7 2003 32 501 29 3 0 2004 45 252 38 2 5 2005 29 91 27 2 0 2006 36 407 32 2 2 Total 384 4.163 304 39 41
Panel D. Intracontinental M&A by European acquirers
Method of payment
Year
Total number of M&A per year
Average deal value in millions Euros
Number of M&A financed with cash
Number of M&A financed with
shares
Number of M&A financed with cash and shares
1997 15 307 15 0 0 1998 18 619 15 1 2 1999 25 1.064 20 3 2 2000 51 982 24 12 15 2001 51 284 34 6 11 2002 56 501 47 3 6 2003 27 140 23 2 2 2004 34 193 30 2 2 2005 41 380 36 2 3 2006 25 121 22 0 3 Total 343 4.591 266 31 46 4. Methodology
4.1 Methods to measure long-term abnormal performance
18
findings of long-term post-acquisition performance in past research (Rau and Vermaelen, 1998). The most advanced and widely used methods for measuring long-term abnormal performance are the Buy and Hold Abnormal Return (BHAR) and the Calendar-Time Abnormal Return (CTAR) approach (Khotari and Warner, 2006). I will discuss the event time approach (BHAR) and the calendar-time approach (CTAR) and explain why I prefer the CTAR method for detecting long-term abnormal returns.
The Buy and Hold Abnormal Return (BHAR) method measures long-term abnormal returns as the average multiyear return difference from investing in a portfolio that consists of firms that complete an event and investing in a portfolio that consist of firms with similar characteristics as the sample firms but did not complete the event (Mitchell and Stafford, 2000). The Calendar-Time Abnormal Return (CTAR) approach measures long-term abnormal performance through computing monthly portfolio returns consisting of firms that experienced an event within previous T months for the entire sample period. Next, the monthly portfolio returns are regressed on the Fama and French three-factor model where the intercept captures long-term monthly abnormal performance.
Barber and Lyon (1997) and Lyon et al. (1999) prefer the BHAR method for calculating long-term abnormal returns because it “precisely measures investor experience”. However, the BHAR methodology has several limitations in measuring long-term abnormal returns (Mitchell and Stafford, 2000). Long-term buy-and-hold returns tend to be right skewed (e.g. Khotari and Warner, 2006 and Mitchell and Stafford, 2000). The right skewed buy-and-hold returns arise because the lower bound of returns is -100 percent and the returns are unbounded on the upside leading to a skewness bias for long-term abnormal performance test statistics (Khotari and Warner, 2006). According to Lyon et al. (1999), the skewness bias is mitigated by applying a bootstrapping procedure introduced by Ikenberry and Lakonishok (1995). This bootstrapping procedure simulates an empirical null distribution of the estimator, relaxing the assumption of normality. However, Mitchell and Stafford (2000), report that the bootstrapping procedure assumes independent event-firm abnormal returns.
19
The Calendar-Time Abnormal Return (CTAR) methodology is an important improvement over the BHAR methodology (which assumes independence of individual firm abnormal returns) because the former constructs monthly event portfolios, which automatically accounts for cross-sectional correlations of individual event firm abnormal returns in the portfolio variance (Mitchell and Stafford, 2000). In addition to the dependence problem Fama, (1998) argues that the BHAR method is also subject to the “bad model problem”. The bad model problem implies that systematic errors arise with imperfect expected return proxies, which become more severe with long-horizon measures of returns. The CTAR approach calculates monthly returns, which are less sensitive to the bad model problem (Fama, 1998).
A potential problem with the CTAR methodology is that it assumes constant factor loadings over time. Constant factor loadings are unlikely because each month the event portfolio is refreshed as new firms may enter the event portfolio and firms that have been part of the event portfolio for a pre specified period drop out which may lead to biased estimates (Mitchell and Stafford, 2000). These authors address the problem of constant factor loadings by calculating Mean Calendar-Time Abnormal Returns (MCTAR), which mitigates the restriction of constant factor loadings. The MCTAR is similar to the CTAR methodology in that the event portfolio abnormal returns are calculated in calendar-time. The estimates of long-term abnormal returns of the MCTAR approach are similar to the CTAR approach, suggesting that the regression intercept for the CTAR methodology is not biased by assuming constant factor loadings (Mitchell and Stafford, 2000).
20
heteroskedasticity problem using an OLS regression. These authors also test to what extent the CTAR methodology has difficulty in detecting long-term abnormal returns because event activity is averaged over months as reported by Louhran and Ritter (2000). They conclude that the CTAR approach has more power to detect long-term abnormal returns than the BHAR method controlling for cross-sectional dependence of individual firm BHARs.
I follow the view of Mitchell and Stafford (2000) and Fama (1998) that the calendar-time abnormal return methodology is superior to the buy-and-hold abnormal return methodology in detecting long-term abnormal returns. The CTAR approach is preferred over the BHAR approach because the monthly constructed portfolios automatically account for cross correlations of individual sample firm returns (Mitchell and Stafford, 2000). In addition, the CTAR approach is less subject to the “bad model problem” and allows better statistical inferences (Fama, 1998). Following Mitchell and Stafford (2000) and André et al. (2004), I will use equal and value weighted monthly acquisition portfolios. In equally weighted monthly acquisition portfolios, each firm in the acquisition portfolio is equally weighted. Value weighted monthly acquisition portfolios weight each firm by their market capitalization. Following André et al. (2004) equally weighted and value weighted monthly acquisition portfolios are regressed on the Fama and French three-factor model using a WLS regression where the weights are proportional to the square root of the number of firms present in each monthly acquisition portfolio. I also regress equally weighted and value weighted monthly acquisition portfolios on the Fama and French three factor model using an OLS regression requiring a minimum of ten firms in the monthly event portfolios to mitigate the heteroskedasticity problem as advocated by Mitchell and Stafford (2000). Section 4.2 discusses how I constructed the monthly acquisition portfolios as well as the construction of the three zero investment portfolios for the Fama and French three-factor model.
4.2 Calendar-Time Abnormal Return (CTAR) approach
21
book to market ratios and excess market returns. The intercept of the Fama and French three-factor model measures long-term abnormal returns and is significant if the intercept significantly differs from zero. Equation 1 shows the Fama and French three-factor model in formula form.
t t t t ∑ (1)
The dependent variable is the monthly excess return of the equal or value weighted return for calendar month t for the portfolio of event firms that experienced the event within the previous T months denoted as RPt over the risk-free rate denoted as RFt. For the risk-free rate, I use one-month German Euribor rates because all acquirer firms in the acquisition portfolio are European firms and equal or value weighted monthly acquisition portfolio returns are measured in Euros. The one-month German Euribor rates are obtained from the Thomson Reuters Datastream database for each month through the research period. The independent variables consist of the market excess return Rmt over the risk-free rate RFt, and two zero investment portfolios (HML and SMB) to mimic the risk factors common to all securities. The two zero investment portfolios are the difference between the return on a value weighted portfolio of high and low book to market stocks denoted as HMLt and the difference between the return on a value weighted portfolio of small and big stocks denoted as SMBt. β0, β1, and β2 are the factor loadings for the three zero investment portfolios. The intercept α measures average monthly long-term abnormal returns. The efficient market hypothesis holds if the intercept does not significantly differ from zero indicating no long-term abnormal performance for firms that engaged in mergers and acquisitions.
22
In order to measure long-term abnormal performance a benchmark return is needed. This benchmark return is constructed through the Fama and French three-factor model, which controls for the risk factors excess market return, size (SMB), and book to market ratios (HML), common to all securities. The estimations of the factor loadings for size (SMB) and book to market ratios (HML) become more reliable as the securities that are included become more representative for the market that is examined. In order to get the best possible risk premiums for excess market return, size and book to market ratios all securities trading on major European indices of all European countries that are a member of the European Union (EU) as of 1995, are included. Table 4.1 shows the major European indices per European country used for the construction of the Fama and French three-factor model, which I discuss next.
Table 4.1; major European indices per European country used for the construction of the Fama and French three-factor model
County Stock Exchange County Stock Exchange County Stock Exchange
Austria Vienna Stock Exchange Finland Nasdaq OMX Helsinki Italy Italian Stock Exchange
Belgium Euronext Brussels France Euronext Paris Luxembourg Luxembourg Stock Exchange
Germany Frankfurt Stock exchange UK Londen Stock Exchange Netherlands Euronext Amsterdam
Denmark Nasdaq OMX Copenhagen Greece Athens Stock Exchange Portugal Euronext Lisabon
Spain Madrid Stock Exchange Ireland Irish Stock Exchange Sweden Nasdaq OMX Stockholm
I follow the procedure as described by Fama and French (1993) for the construction of the two zero investment portfolios HML and SMB and the calculation of the market return RMt. For all securities trading on the major European indices, I collected monthly stock returns, market capitalization data, and yearly book to market ratios for the entire research period using Datastream. I will first discuss how I constructed the two zero investment portfolios HML and SMB followed by the calculation of the market return RMt.
23
capitalization and book to market ratios; BH, BM, BL and SH, SM, SL. The BH, BM, and BL portfolios consist of those securities with the highest market capitalization (B) with the top 30% (High), middle 40% (Medium) and bottom 30% (Low) book to market ratios respectively. For example, the BH portfolio consists of firms within the big group (B) based on size and the high group (High) based on book to market ratios. The SH, SM, and SL portfolios consist of those securities with the lowest market capitalization (S) with the top 30% (High), middle 40% (Medium) and bottom 30% (Low) respectively. For example, the SL portfolio consists of firms within the small group (S) based on size and the low group (L) based on book to market ratios. For each of the six portfolios monthly value weighted returns are calculated for each year in the research period. Finally, the monthly SMB portfolios are constructed through subtracting the average return of the portfolios within the small group (SH, SM, SL) from the average return of the portfolios within the big group (BH, BM, BL) for each year in the research period. The monthly HML portfolios are constructed through subtracting the average return of the portfolios within the high group (SH, BH) form the average return of the portfolios within the low group (SL, BL) for each month throughout the research period.
The monthly excess market returns (RMt – RFt) are calculated by subtracting the monthly value weighted returns of the stocks in the six portfolios (i.e. BH, BM, BL, SH, SM, and SL) from the monthly risk-free rates for which I used one-month German Euribor rates.
5. Results
This section discusses the results of the regressions by applying the Calendar-Time Abnormal Return (CTAR) methodology. I perform an Ordinary least Square (OLS) as well as a Weighted Least Square (WLS) with equally weighted (EW) and value weighted (VW) acquisition portfolios. Section 5.1 discusses the results regarding different types of mergers and acquisitions. Section 5.2 examines whether the method of payment affects long-term post-acquisition abnormal returns. Finally, section 5.3 examines the results of long-term post-acquisition performance for European acquirers by distinguishing between common law and civil law European countries.
24
months with relatively high event activity are handled the same as months with relatively low event activity. If abnormal performance for months with relatively high event activity differs from months with relatively low event activity, it becomes more difficult to detect abnormal performance because event activity is averaged over months. The composition of the acquisition portfolios change over time, which may lead to heteroskedasticity as the variances of the residuals are related to the amount of firms in the acquisition portfolio. The heteroskedasticity problem may result in an inefficient estimator by applying the Ordinary Least Square (OLS) regression (Mitchell and Stafford, 2000). Following André et al. (2004) and Dutta and Jog (2009), the square root of the number of firms in each monthly acquisition portfolio is used for the weights in the weighted least square (WLS) regression.
For the OLS regression, I require that each monthly acquisition portfolio consist of a minimum of ten firms to mitigate the heteroskedasticity problem as advocated by Mitchell and Stafford (2000). These authors also test to what extent the CTAR methodology using an OLS regression has difficulty in detecting long-term abnormal returns because event activity is averaged over months as reported by Louhran and Ritter (2000). They conclude that the CTAR approach using an OLS regression has sufficient power in detecting long-term abnormal returns.
Both the OLS and WLS regressions are applied using equally weighted (EW) and value weighted (VW) monthly acquisition portfolios because past research has shown that using different weighting schemas for the monthly acquisition portfolios tend to influence long-term abnormal returns. In addition, most researchers distinguish between EW and VW monthly acquisition portfolios (e.g. Dutta and Jog, 2007; André et al., 2004; Mitchell and Stafford, 2000) which makes comparison between my results and those of other researchers possible. In equally weighted monthly acquisition portfolios, the return of each firm is equally weighted. Value weighted monthly acquisition portfolios weight the return of each firm by their market capitalization.
5.1 Type of Mergers and acquisitions
25
acquisition portfolios give positive long-term abnormal returns for both the OLS and the WLS regression. Loughran and Ritter (2000) and Fama (1998) report that abnormal performance reduces and even disappears using value weighted portfolios, which is consistent with my results. The results indicate that small firms underperform compared to large firms as long-term abnormal returns are for equally weighted acquisition portfolios and positive for value weighted acquisition portfolios.
The two types of regressions (OLS and WLS) show almost similar results regarding the sign of the factor loadings for the risk premiums for all types of M&A with EW and VW acquisition portfolios. With EW acquisition portfolios, all types of M&A show highly significant positive risk premiums for size and the market return for both types of regressions. For VW acquisition portfolios, only the factor loadings for the market return show highly significant positive risk premiums for all subsamples except for domestic M&A, which also shows a highly significant positive risk premium for size. However, the WLS regressions show more often significant underperformance for EW acquisition portfolios compared to the OLS regressions, which is consistent with the argument of André et al. (2004). If abnormal performance for months with relatively high event activity differs from months with relatively low event activity it becomes more difficult to detect abnormal performance because event activity is averaged over months (Loughran and Ritter, 2000). Using WLS regression solves this problem because the WLS regression allows to weight months in which event activity is high more heavily and therefore has less difficulty in detecting long-term abnormal performance (André et al., 2004).
The first subsample includes all mergers and acquisition with a total of 1208 events. Equally weighted acquisition portfolios for all mergers and acquisitions show marginal significant negative long-term post-acquisition abnormal returns of -0.470 and -0.559 percent per month using an OLS and WLS regression respectively. However, value weighted acquisition portfolios for all M&A show positive insignificant long-term post-acquisition abnormal returns of 0.345 and 0.278 percent per month using an OLS and WLS regression respectively.
26
abnormal returns of 0.322 percent per month with (VW) acquisition portfolios applying a WLS regression.
The third subsample (crossborder M&A), show insignificant negative longterm abnormal returns of -0.478 (OLS) and significant negative long-term abnormal returns of -0.675 (WLS) percent per month with equally weighted acquisition portfolios. With VW acquisition portfolios long-term abnormal returns for cross-border M&A become positive and insignificant for both types of regressions. The results are consistent with those of Conn et al. (2005) who report insignificant negative long-term abnormal returns for cross-border M&A performed by U.K. firms of -0.3 percent per month with EW acquisition portfolios using an OLS regression although the results show larger underperformance. Splitting the cross-border M&A into intercontinental and intracontinental M&A the results show that both intercontinental and intracontinental M&A have negative long-term abnormal returns for EW acquisition portfolios, as the intercepts are negative. However, only intercontinental M&A give significant negative long-term abnormal returns of -0.606 (OLS) and -0.89 (WLS) per month. An open question remains whether the long-term abnormal performance of different subsamples significantly differs, which I address next.
27
portfolios I find that the long-term abnormal returns for these subsamples are insignificant for both types of regressions and for EW and VW acquisition portfolios.
5.2 Method of payment
This sub section discusses the results for the method of payment for all types of M&A. I examine cash and share acquisition using a WLS regression with EW and VW acquisition portfolios. I did not perform an OLS regression because the requirement of a minimum of ten firms in each calendar month is often not satisfied for share acquisitions as the majority of the sample consists of cash acquisitions. Table 5.3 shows the results of the method of payment by type of mergers and acquisitions using an OLS regression with equally weighted (EW) and value weighted (VW) monthly acquisition portfolios. With equally weighted acquisition portfolios, all intercepts are negative for both methods of payments and more negative for share acquisitions in all subsamples. The intercepts for value weighted acquisition portfolios are all positive and higher for cash acquisitions. With equally weighted and value weighted acquisition portfolios cash acquisitions perform better compared to share acquisitions. I examined whether cash and share acquisitions significantly differ for all five subsamples. Except for the total sample (all M&A VW) none of the subsample shows evidence in favour of the method of payment hypothesis as cash acquisitions do not significantly outperform share acquisitions for each subsample.
5.3 Legal system
28
Table 5.1; Long-term post-acquisition abnormal returns using the Calendar-Time Abnormal Return (CTAR) methodology on a OLS regression
This table estimates long-term post-acquisition abnormal returns for a three-year period after the completion date of mergers and acquisitions (M&A) from January 1997 through December 2007 for European acquirers. All M&A are performed by European acquirers that are a member of the European Union (EU) as of 1995. I estimate the long-term post-acquisition abnormal returns for five subsamples; all, domestic, cross-border, intercontinental, and intracontinental M&A of which the sample size equals 1208, 481, 727, 384, and 343 respectively. I examine both equally weighted (EW) and value weighted (VW) monthly acquisition portfolio returns. The excess monthly portfolio acquisition returns are regressed on the Fama and French three-factor model using a OLS regression where each monthly acquisition portfolio consist of a minimum of ten event firms and an:
where Rpt-RFt is the monthly excess return of the equal or value weighted return for calendar month t for the portfolio of event firms that experienced the event within the
previous T months denoted as RPt over the risk-free rate denoted as RFt. For the risk-free rate, I use one-month German Euribor rates. β0, β1, and β2 are the factor loadings
of the portfolio on each risk factor; HML (book to market ratio), SMB (size), and the market. The intercept α measures average monthly long-term abnormal returns. The t-statistics for each parameter are shown in parentheses. To test for differences in the returns of the portfolios, I regress the differences in the returns of two subsample portfolios on the Fama-French (1993) three-factor model. *** Significance at the 0,01 level; ** Significance at the 0,05 level; * Significance at the 0,10 level.
Equally weighted acquisition portfolios (OLS) Value weighted acquisition portfolios (OLS)
29
Table 5.2; Long-term post-acquisition abnormal returns using the Calendar-Time Abnormal Return (CTAR) methodology on a WLS regression
This table estimates long-term post-acquisition abnormal returns for a three-year period after the completion date of mergers and acquisitions (M&A) from January 1997 through December 2007 for European acquirers. All M&A are performed by European acquirers that are a member of the European Union (EU) as of 1995. I estimate the long-term post-acquisition abnormal returns for five subsamples; all, domestic, cross-border, intercontinental, and intracontinental M&A of which the sample size equals 1208, 481, 727, 384, and 343 respectively. I examine both equally weighted (EW) and value weighted (VW) monthly acquisition portfolio returns. The excess monthly portfolio acquisition returns are regressed on the Fama and French three-factor model using an WLS regression where the square root of the number of firms in each monthly acquisition portfolio is used for the weights:
where Rpt-RFt is the monthly excess return of the equal or value weighted return for calendar month t for the portfolio of event firms that experienced the event within the
previous T months denoted as RPt over the risk-free rate denoted as RFt. For the risk-free rate, I use one-month German Euribor rates. β0, β1, and β2 are the factor loadings
of the portfolio on each risk factor; HML (book to market ratio), SMB (size), and the market. The intercept α measures average monthly long-term abnormal returns. The t-statistics for each parameter are shown in parentheses. *** Significance at the 0,01 level; ** Significance at the 0,05 level; * Significance at the 0,10 level.
Equally weighted acquisition portfolios (WLS) Value weighted acquisition portfolios (WLS)
30
Table 5.3; Long-term post-acquisition abnormal returns by method of payment using the Calendar-Time Abnormal Return (CTAR) methodology on a WLS regression This table shows the long-term post-acquisition abnormal returns by method of payment (cash and shares) for all, domestic, cross-border, intercontinental, and intracontinental M&A of which the total sample size equals 1024 (cash 874 and stock 150) M&A. Please see table 5.2 for a description of the variables included in this table.
Equally weighted acquisition portfolios (WLS) Value weighted acquisition portfolios (WLS)
31
Table 5.4; Long-term post-acquisition abnormal returns by legal system using the Calendar-Time Abnormal Return (CTAR) methodology on a WLS (panel A) and OLS (panel B) regression
This table shows the long-term post-acquisition abnormal returns for different legal systems by distinguishing between civil law (continental Europe) and common law (U.K. and Ireland) European acquirers of which the sample size equals 638 and 570 respectively. Please see table 5.1 (OLS) and 5.2 (WLS) for a description of the variables included in this table.
Panel A. Long-term post-acquisition abnormal returns by legal system on a WLS regression
Equally weighted acquisition portfolios Value weighted acquisition portfolios
Type of M&A α % β0 β1 β2 Adjusted
R2 (1)-(2) α % α % β0 β1 β2 Adjusted R2 (1)-(2) α % -0,468 -0,192 0,345 1,096 0,455 -0,544 -0,061 0,938 Civil law (1) (-1,700)* (-1,436) (4,557)*** (17,798)*** 0,783 -0,113 (1,108) (-2,724)** (-0,543) (10,214)*** 0,622 -0,212 (-0,304) (-0,714) -0,547 0,145 0,367 1,053 0,715 -0,206 -0,338 0,618 Common law (2) (-1,620) (0,922) (3,982)*** (13,567)*** 0,654 (2,364)** (-1,465) (-4,098)*** (8,891)*** 0,617
Panel B. Long-term post-acquisition abnormal returns by legal system on a OLS regression
Equally weighted acquisition portfolios Value weighted acquisition portfolios
Type of M&A α % β0 β1 β2 Adjusted
32 6 Conclusion
This thesis examines the long-term post-acquisition performance of European acquiring firms for the period 1997-2006 using the Calendar-Time Abnormal Return (CTAR) methodology. Moreover, I examined the long-term performance for different types of mergers and acquisition (M&A) namely; all, domestic, cross-border, intercontinental and intracontinental M&A. I also examined whether or not the way in which a deal is financed affects long-term post-acquisition abnormal performance for European acquirers. Finally, I examined whether differences in the legal system of European acquirers affects long-term post-acquisition abnormal performance by distinguishing between civil law (continental Europe) and common law (U.K. and Ireland) European countries.
The Efficient Market Hypothesis (EMH) predicts that no long-term post acquisition stock performance should be found as the market incorporates new information correctly into stock prices within a short period of time. However, the behavioural finance literature argues that the assumptions regarding the efficient market hypothesis are not likely to hold in the real world implying that the market may not be efficient.
I perform an Ordinary least Square (OLS) as well as a Weighted Least Square (WLS) with equally weighted (EW) and value weighted (VW) acquisition portfolios to measure long-term abnormal performance. Overall, the results indicate no long-term post-acquisition abnormal performance for European acquirers for the period 1997-2006. Regarding the different types of mergers and acquisitions I find evidence of long-term post-acquisition abnormal performance in some cases with equally weighted acquisition portfolios for both types of regressions although the WLS regression captures evidence of long-term abnormal performance more often. However, when using value weighted acquisition portfolios no evidence is found for long-term-abnormal performance except for domestic M&A, which is only significant at the 0.10 percent level. This result is consistent with the findings of Loughran and Ritter (2000) and Fama (1998) which report that abnormal performance reduces and even disappears using value weighted acquisition portfolios. Furthermore, the sign of the intercept changes from negative for EW acquisition portfolios to positive for value weighted acquisition portfolios indicating that small firms underperform compared to larger firms.
33
found that cross-border mergers and acquisition underperform compared to domestic M&A. Intercontinental mergers and acquisition underperform compared to intracontinental M&A for the OLS and WLS regression and with EW acquisition portfolios. However, the difference between the long-term performance of intercontinental and intracontinental is not significant.
Regarding the way in which a deal is financed, I find that cash acquisitions perform better compared to share acquisitions for both equally weighted and value weighted acquisitions portfolios. Except for the total sample (all M&A) none of the subsample shows evidence in favour of the method of payment hypothesis as cash acquisitions do not significantly outperform share acquisitions.
The results with respect to differences in the legal system indicate that common law European acquirers outperform civil law European acquirers. However, for both regression types and using equally and value weighted acquisition portfolios, the results show that the long-term abnormal performance between common and civil law European acquirers does not significantly differ.
34 References
Agrawal, A., Jaffe, J., 2000. “The post-merger performance puzzle, Advances in Mergers and Acquisitions,” Cary Cooper and Alan Gregory, Eds., Stamford, CT, JAI Press
André, P., Kooli, M., and L’Her, J.F., 2004. “The long-run Performance of Mergers and Acquisitions : Evidence from the Canadian Stock Market,” Financial Management, vol. 33, pp. 27-43
Aw M.S.B., Chatterjee R.A., 2004 “The performance of UK firms acquiring large cross-border and domestic takeover targets,” Applied Financial Economics Vol. 14, pp. 337-349
Barber, B.M, and Lyon J. D., 1997. “Detecting long run abnormal returns : The empirical power and specification of test statistics, “ Journal of Financial Economics, Vol. 43, pp. 341-372
Black, E.L., Cames, T.A., and Jandik, T., 2001. “The long-term success of cross-border mergers and acquisitions,” Available at SSRN: http://ssrn.com/abstract=270288 or doi:10.2139/ssrn.270288 Bodnar, G.M., Tang, C., and Weintrop J., 1998. “Both sides of corporate diversification: The value
impacts of geographical and industrial diversification,” NBER Working Paper No. W6224 Brav, A., 2000. Inference in long-horizon event studies: A Bayesian approach with application
to initial public offerings,” Journal of Finance, Vol. 55 pp. 1979-2016
Bruner, R. F., 2002. “Does M&A Pay ? A Survey of Evidence for the Decision Maker,” Journal of Applied Finance, Vol. 12, pp. 48-68.
Caves, R.E., 1971. “International corporations: The industrial economics of foreign investment,” The Industrial Economics of Foreign Investment, Vol. 38, pp. 1-27
Conn, R., Cosh , A., Guest, P., and Hughes, A., 2005. “The impact on U.K. acquirers of domestic, cross- border, public and private acquisition” ESRC Centre for Business Research, University of
Cambridge Working Paper No 276
Core, J. E., Guay, W. R., and Rusticus T. O., 2006. “Does weak governance cause weak stock returns: An examination of firm operating performance and investors’ expectations,” Journal of Finance, Vol. 61, pp. 655–87
35
Denis D. K., and McConnell J. J., 2003. “International corporate governance,” The Journal of Financial and Quantitative Analysis, Vol. 38, pp. 1-36
Dube S, Glascock J.L., 2006. “Effects of the method of payment and the mode of acquisition on performance and risk metrics,” International Journal of Managerial Finance, vol. 2, pp.176-195
Dutta S., Jog V., 2009. “The Long Term Performance of Acquiring Firms: A Re-examination of an Anomaly”, Journal of banking and Finance, Vol. 33, pp. 1400-1412
Fama E.F., 1998. “Market efficiency, long-term returns, and behavioural finance,” Journal of Financial Economics, vol. 49, pp. 283-306
Fama E.F., French K.R., 1993. “Common risk factors in the returns on stocks and bonds,” Journal of Financial Economics, vol. 33, pp. 3-65
Franks J., Harris R., Titman S., 1991. “The post merger share-price performance of acquiring firms,” Journal of Financial Economics, vol. 29, pp. 81-96
Gaughan, P.A., 2002. “Mergers, acquisitions, and corporate restructurings, third edition John Wiley and Sons, New York
Goergen, M. and Renneboog, L., 2002. “Shareholder wealth effects of European domestic and cross- border takeover bids,” Available at SSRN: http://ssrn.com/abstract=372440
Gompers, P.A., Ishii, J.L., and Metrick, A., 2003. “Corporate governance and equity prices,” The Quarterly Journal of Ecomics, Vol. 118, pp.107-155
Gregory A., 1997 “An examination of the long run performance of UK acquirers,” Journal of Business Finance & Accounting, Vol. 24, pp. 971-1002
Grinblatt M., and Titman S., 2004. “Financial Markets and Corporate Strategy,” 2nd ed. McGraw-Hill, 1221 Avenue of the Americas, New York
Gregory, A., and McCorriston, S., 2002. “Foreign acquisitions by UK limited companies: long-run performance in the US, continental Europe and the rest of the world,” University of Exeter Hymer, S.H., 1997. “The international operations of national firms: A study of direct investment,”
36
Ikenberry, D., Lakonishok, J. and Vermaelen, T., 1995. “The underreaction to open market share repurchases,” Journal of financial economics, Vol. 39, pp. 181-208
Johnson, S.A., Moorman, T.C., and Sorescu, S., 2009. “A re-examination of corporate governance and equity prices,” Oxford University Press, Vol.22, pp. 4753-4786
Khotari, S.P., and Warner, J.B., 2006. “Econometrics of event studies,” Handbook of corporate finance: Empirical Corporate Finance, Volume A Elsevier
Kogut, B., 1983. “Foreign direct investment as a sequential process", Charles P. Kindelberger and David (Eds.), The Multinational Corporations in the 1980’s, MIT Press, Cambridge, MA. Kolasinski, A., and Li, X., 2005. “Do managers detect mispricing? Evidence from insider trading and
post-earnings-announcement drift,” Massachites Institute of Technology: unpublished paper La Porta, R., Lopez-de-Silanes, F., and Shleifer, A., 1998. “Corporate ownership around the world,”
The Journal of Finance, Vol. 54, pp. 471-518
Loughran, T., and Ritter, J.R., 2000. “Uniformly least powerful test of market efficiency”, Journal of Financial Economics, Vol. 55, pp. 361-389
Loughran T., Vijh A.M., 1997 Do Long-term shareholders benefit from corporate acquisitions?,” Journal of Finance, Vol. 52, pp. 1765-90.
Lyon, J.D., Barber, B.M., and Tsai, C., 1999. “Improved methods for tests of long run abnormal stock returns,” Journal of finance, Vol. 54 pp. 165-201
Mitchell M.L., Stafford E., 2000 “Managerial decisions and long-term stock price performance,” Journal of Business, Vol. 73 pp. 287-329.
Moeller S., Schlingemann F.P., Stulz R., 2003. “Do shareholders of acquiring firms gain from acquisitions,” Research Technology Management, Vol. 46, pp.62-62
Moeller, S., and Schlingemann, F.P., 2005. “Global diversification and bidder gains: A comparison between cross-border and domestic acquisitions,” Journal of Banking & Finance, Vol. 29, pp. 533-564
37
Rau P., Vermaelen T., 1998 “Glamour, value and the post-acquisition performance of acquiring firms,” Journal of Financial Economics Vol. 49, pp. 223-53
Roll R., 1986 “The hubris hypothesis of corporate takeovers,” Journal of Business, Vol. 59, pp. 197- 216.
Ross, S.A., Westerfield, R.W., and Jordan, B.D., 2007. “Corporate finance fundamentals,” McGraw-Hill Education
