Master Thesis
The Effect of M&A on Acquirer Performance in
the Newspaper Industry
J Liu s1591525
Email: j.liu.2@student.rug.nl
Faculty of Economics
University of Groningen
ABSTRACT
Newspaper industry has experienced a surge in merger and acquisition (M&A) during recent two decades. Based on the sample derived from Thomson, this paper focuses on two research questions: what is the effect of acquisition experience on post M&A performance? And what are the determinants of acquisition frequency? In order to investigate these two questions, I analyzed the effect of acquisition experience, transaction attitude, payment method, internationalization, percentage sought. Results show that, first, the effect of acquisition experience is negative but insignificant on acquirer performance. Second, the more the acquirers are engaged in cross-border M&A, the more they will undertake future acquisition. Finally, the effects of transaction attitude, percentage sought and payment method is inconclusive on acquisition frequency.
Key words: Merger and acquisition, newspaper industry, acquisition experience, count
CONTENT
1. INTRODUCTION……….4
2. THEORY AND HYPOTHESES………...7
2.1. M&A motivations and acquirer performance……….7
2.2. Acquisition experience and acquirer performance……….9
2.3. Determinants of acquisition frequency……….11
3. METHODOLOGY………..18
3.1. Data and sample………18
3.2. Measures………....19
3.2.1. Effect of acquisition experience on acquirer performance…….19
3.2.2. Determinants of acquisition frequency………...21
3.3. Statistical method………..22
4. RESULTS……….24
4.1. Effect of acquisition experience on acquirer performance ………24
4.2. Determinants of acquisition frequency………..27
5. DISCUSSION………..30
5.1. Effect of acquisition experience on acquirer performance …………...30
5.2. Determinants of acquisition frequency ……….31
6. CONCLUSION AND LIMITATION..………35
REFERENCE……….37
1. INTRODUCTION
Over the last two decades, mergers and acquisitions (M&A) has become one of the strategic initiatives of many firms that wish to survive from globalization and innovation. Firms expected to create greater economic efficiency as well as acquire complementary assets through M&A (Peltier, 2004). Like in other industries, newspaper industry confronted numerous changes, such as digitalization of content and widely used broadband network, much faster than people expected.
Generally, Newspapers Industry Profile: Global (2006) reported that global newspaper market grew slightly but steadily after a shock in 2001 and 2002. In 2005, revenues from copy sales were no longer dominant in market value; the leading role was taken by advertising, which generated $97.5 billion total revenue and was equivalent to 64.2 percent of overall market value. The market volumes of global newspaper shrank by 0.1 percent in the same year, with a volume of 374.8 million copies. With respect to geographical division, the US was responsible for 42.1 percent of the global market’s value, followed by Europe1 with 36.1 percent and Asia-Pacific2 with 21.8 percent, respectively. Looking forward, the growth in newspaper industry is predicted to slow down with an estimated compound annual growth rate of –0.5 percent for 2005-2010 periods. Although the future of newspaper industry is not optimistic, one cannot predict the death of newspaper. Cho, Martin and Lacy (2006) studied entry and exit of daily newspapers from 1987 to 2003, they found that among 305 newspapers that stopped their daily publication, 64 percent switched to serve the markets as weeklies, merged dailies or zoned editions; besides, while 111 dailies went out of business, 63 dailies started their publications.
Specifically, the newspaper market is facing two major problems now. One is a wide circulation of free newspapers. The free newspapers are often distributed in the streets, or from transport links to commuters. Their revenues are solely from the advertising. For instance, the world’s largest free-distributed newspaper “the Metro” was sent out in 93 major cities and 21 countries worldwide, which naturally accounted for the circulation reduction of traditional newspaper (Newspapers Industry Profile: Global,
1 Europe comprises Belgium, France, Germany, Italy, Netherlands, Spain and the UK in Global
Newspapers Industry Profile.
2006). The second problem is the emerging of other media. Newspapers in the developed countries gained declining profit from printed newspaper as people turned away to other media. According to the Newspaper Association of America (2006), in the US, visitors to newspaper websites take a proportion of one third (46 million) of overall Internet users. The number of visitors increased by 21 percent from the beginning to the end of 2005, and page views grew by 43 percent during the same period. As a result, many publishers made a range of efforts and initiatives in attempt to attract more general and targeted readers, especially the young readers. The efforts include modernizing the design of their newspapers, extending the distribution out of the original regions, establishing additional online products, etc. In the future, the newspapers are predicted to establish a wide range of products that combine their core business (the traditional newspaper) with all media (Cho, et al., 2006), which implies the growth of M&A in the newspaper industry.
Figure 1.
Number of M&As announced and completed between 1981-2000
0 20 40 60 80 100 120 140 160 180 200 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 Year N u m b e r number of mergers announced number of mergers completed
Source: Thomson Financial Securities Data’s Worldwide Mergers & Acquisition database.
acquisition and hence increase the number of successful acquisition experience. Therefore, to some extent, I expect the factors that affect acquirer performance will have impact on the number of future acquisitions (acquisition frequency) as well.
Summing up, this exploratory study aims to examining the acquirer performance after M&A. I will emphases on answering two research questions (RQ):
RQ1: What is the effect of acquisition experience on acquirer performance? RQ2: What are the determinants of acquisition frequency?
This study will contribute to the existing literatures on two aspects. First, I purely examine the effect of acquisition experience on acquirer performance in the post-M&A period. Second, I exploit determinants that used to test acquisition performance to study acquisition frequency. And count data model is applied along with variables to examine acquisition frequency. The implications suggest that further researches can be carried out when data for other factors come to be available or updated.
The remainder of the paper is processed as follows: in the following sections, I will describe the theory and estimate hypothesis. In section 3, I specify the samples and measurements for variables; particularly the count data model will be introduced. Section 4 provides empirical results with respects the effect of acquisition experience and the determinants of acquisition frequency. Section 5 presents the discussion of the findings related to hypothesis. In section 6, I conclude the findings of the study and give some limitations.
2. THEORY AND HYPOTHESES
(Vega-Céspedes & Hoshino, 2001). Thus, I think it is necessary to begin with the interpretation of M&A motivations.
2.1. M&A motivations and acquirer performance
There exist several theories that address M&A motivations and they can be classified into two broad categories: economic perspectives and managerial preference.
Economic perspectives are associated with efficiency stemming from synergy effect. Synergy “occurs when two operating units can be run more efficiently (i.e., with lower cost) and /or more effectively (i.e., with a more appropriate allocation of scarce resources, given environmental constraints) together than apart” (Lubatkin, 1983, p218). Generally, there are two sources of efficiency: economies of scale3 and economies of scope4 . They can be used to explain possible differences in acquisition outcomes. First,
economies of scale may help to achieve several potential sources of efficiency. Thus, firms are able to produce more quantity of products with same amount of inputs. For instance, firms could use common source of raw materials or components in the case of purchasing or inventory, thereby reduce average costs and may gain competitive advantage over rivals (Sharma & Ho, 2002). Second, economies of scope encourage firms to share complementary resources and knowledge beyond organizational and country boundaries. Gammelgaard, Husted and Michailova (2004) figured out that small firms with firm-specific assets could benefit from utilizing capital or distribution channel of larger and experienced acquirers. In addition to efficiency, market power5 plays a role in economic perspective as well. Firms exert market power through either imposing higher price on their buyers, or compelling their suppliers to offer at a lower price; thereby firms could increase their profit margin and become more productive (Sharma &Ho, 2002). Unlike economics of scale, market power mechanism does not provide efficiency from inputs utilization directly; it is derived from transferring income from less capable to more capable firms (Lubatkin, 1983).
3 Economies of scale refer to the reduction of per unit cost as the volume of output increase.
4 Economies of scope are present when the joint output of a firm is greater than the output if that could be
achieved by multiple firms.
Managerial preference is considered as the second motivation of M&A. This point of view argues that managers have different incentives with shareholders when they make decision on acquisition. If M&A is driven by managers’ personal benefits rather than maximizing shareholder gains, managers are more likely to overpay for the target. Particularly, Morck, Shleifer and Vishny (1990) assumed that managers engaged in acquisitions were eager to realize both personal benefit and market value of the firm. These managers might give up market value of the firm and overpay for investment that provide them with large personal benefit. As a result, profitability from acquisition with this purpose might be lower than transaction without this purpose (Jaggia & Thosar, 1993; Sudarsanam & Mahate, 2006). Another explanation for overpayment is that managers suffer from hubris and consequently overestimate their own ability to operate the targets (Conn, et al., 2005). Therefore, in order to avoid wealth abandonment by managers, shareholders are proposed to enhance monitoring and control mechanism over investment decisions, and/or add some management ownership shares to deter market value dissipation made by managers (Sharma & Ho, 2002). In addition to overpayment, managers are under the pressure of replacement especially when firms performed poorly, so managers are interested in expanding business in the new projects whose performance will be enhanced running by them than by other alternatives. After succeeding in acquisitions, the managers could get more promotion opportunities, increased their value to the shareholders and made themselves costly to be replaced (Morck et al., 1990).
In summary, the prediction of post-M&A performance based on M&A motivations imply that M&A with economic perspective is expected to improve acquirer performance, while M&A with managerial preference is more likely to violate acquirer value.
2.2. Acquisition experience and acquirer performance
understand the effect of prior acquisition experience on acquisition performance. This theory argued that both past and present environmental influences have impact on behavior. Several studies provided evidences on this argument. For instance, Thorndike (1898) demonstrated that rewarded behavior tended to strengthen and punished behavior tended to weaken (the effect of past condition on behavior). Looking at the present influences on behavior, the past condition would be generalized to the present condition when conditions were similar (Pinder, 1984); when conditions were dissimilar, generalizing past experience to present condition would lead to unfavorable results (Mazur, 1994). Applying the theory to firm acquisitions, prior experience was regarded as a source of competitive advantage. Good and/or similar previous performance was more likely to be repeated and poor and/or dissimilar performance more likely lead to behavioral changes (Haleblian & Finkelstein, 1999). The second mechanism is performance feedback approach, it examined the outcome of prior behaviors and their effect on future behaviors. This approach argued that firms modified their behavior based on the feedbacks from prior performance. For instance, firms regarded successful acquisition as a positive feedback, which in turn encouraged future acquisitions. Specifically, managers were more confident to repeat the behavior than alternatives that have limited experience, because managers had learnt related skills and knowledge from the success, and found less risk to repeat it (Levitt & March, 1988). By contrast, firms with poor acquisition performance regarded it as a negative feedback and less likely undertook future acquisitions. Managers would reevaluate the current strategies and search for new strategies that might improve firm performance (Haleblian, et al., 2006).
which demonstrated that the firm performance improved as the number of acquisition increase. In the second study, Hayward (2002) tested performance of 214 acquisitions made by 120 firms during 1990-1995. The hypothesis predicted an inversed U-shaped relationship between the similarity of prior acquisition and focal acquisition performances. However, the results showed a positive relationship between experience and performance when the prior experience was not very similar or dissimilar to the focal acquisition.
Hypothesis 1: acquisition experience has a positive effect on acquirer performance.
2.3. Determinants of acquisition frequency
According to the literatures presented above, the effect of factors that may influence the number of future acquisition (acquisition frequency) will be further examined. To my knowledge, there are little studies on the determinants of acquisition frequency. Therefore, I begin with explaining the relationship between firm performance and future acquisition so that I can employ factors determining acquirer performance to examine acquisition frequency.
Haleblian et al., (2006) argued that routine and performance feedback could be used to explain firm behavior. Routine that stems from experience is a central concept in behavioral learning theory (Levitt & March, 1988). It is a source of competitive advantages and plays an important role in firms’ strategy making process. March (1991) indicated that firms could learn specific capabilities and skills from accumulating experience through a certain routine. Furthermore, firms with specific routines were subject to inertial pressure and the likelihood of the routines being repeated was increased (Nelson & Winter, 1982; Szulanski, 1996). However, firms, which stick on a particular routine by which make them proficient, may overestimate the future success of this routine and neglect capability exploration (March, 1991). Study by Amburgey and Miner (1992) argued that decision maker might interpret poor performance as the consequence of unsuccessful execution and/or external factors, rather than the strategy itself. Hence, they concluded that acquirers were more likely to make the same type of acquisition if they had made a particular type of acquisition, even though the outcome of previous acquisition was not favorable. In other words, once a routine is established, it gains its life and acquirers own such routine are more likely to make future acquisitions regardless of the performance of prior acquisition.
on the effect of acquisition behavior. Haleblian et al., (2006) indicated that acquirers could learn to refine their existing routines from previous acquisition experiences, which in turn encouraged firms to exploit those routines in the future acquisitions and made the firms better acquirers. As reported in the study by Levitt & March (1988), the likelihood that a routine will be used is increased when the feedback from the prior acquisition experience is positive, which would encourage firms to engage in future acquisitions. By contrast, poor acquisition performance implied that the current routine was inappropriate and the likelihood to use the routine decreased, consequently, managers were suggested to make some effective modifications on the current routine or search for the better ones (Cyert & March, 1963). In sum, performance feedback complemented the effect of routine. They indicate that good performance improves the effect of routine-based persistence and firms with better performance are more likely to increase the number of future acquisitions.
managements requested excessive premium. The empirical results showed that the friendly and hostile transactions had very different effect on firm performance. For instance, Cosh and Guest (2001) examined the performance of listed firms in the UK from 1985-1996 and predicted greater post-acquisition improvement in hostile transactions. Their findings supported the hypothesis and argued that acquirer performance after a hostile transaction achieved a significant higher profitability arising from improved operating profit margins. As to friendly acquisitions, they found no evidence for the improvement of profit returns after acquisition, but significantly negative long-run share returns following the acquisitions. Sudarsanam and Mahate (2006) tested significance of difference in abnormal returns based on a sample of 519 UK acquisitions during 1983-1995. They predicted greater wealth gains from hostile acquisitions. Consistently, their findings indicated that hostile acquirers generated greater returns after acquisition than friendly acquirers, and hostile acquirers delivered significant higher shareholder value than friendly acquirers. Fowler and Schmidt (1989) predicted a positive relationship between hostile acquisition and acquirer performance. Contrary to their expectation, their got a surprising finding that hostile acquisitions had a significantly negative effect on performance that resulted from nonproductive managerial behavior, management defection, and etc. In summary, studies on the effect of transaction attitude through M&A suggest both positive and negative results, but poorer performance owing to friendly attitude domains (Loughran & Vijh, 1997; Cosh & Guest, 2001; Sudarsanam & Mahate, 2006). I expect the acquirers with poor performance are not willing to undertake future acquisition. Therefore, these arguments lead to the second hypothesis:
Hypothesis 2: The greater the number of friendly M&A, the less the number of future M&A.
positive expectation on future returns from bidders. On the other hand, cash and stock offers have different tax treatment. In stock offers, target shareholders’ are benefited until the stock is sold; and the depreciation basis of acquired assets does not change. In cash offers, target shareholders are required to pay tax, so acquirer firms have to pay a higher price since their payment are expected to offset the tax burden of target shareholders (Loughran & Vijh, 1997). As a result, cash offers do not create as much potential returns as expected. These explanations are in line with some empirically findings. Travlos (1987) selected a sample of 167 acquirer firms engaged in successful takeovers during 1972-1981 and examined the role of payment method in explaining daily abnormal common stock returns. The hypothesis predicted that common stock financing was expected to have a negative impact, while cash financing had a non-negative effect. The findings supported the hypothesis and showed that acquirer shareholders experienced significant losses in pure stock offers, but earned normal rates of returns in cash offers at the announcement; the difference in the abnormal returns between stock offers and cash-offers was statistically significant. Conn et al., (2005) examined the post-acquisition share returns of UK acquirers in over 4000 acquisitions during 1984-1998. They found that cash mergers had significantly higher long run returns than non-cash acquisitions. Likewise, Loughran and Vijh (1997) examined acquisition returns in 947 acquisitions during 1970-1989. They predicted that firms would prefer to pay cash if their stock is undervalued and shareholders gained from the cash payment. Their conclusion supported that, on average, acquirer stock returns were greater in the case that cash was used in payment. Therefore, although the level of acquirers’ wealth gains in a cash offer is ambiguous, it is still expected to have higher returns than stock offers (Travlos, 1987; Conn, et al., 2005). I further expect acquirers associated with greater benefit from cash payment are more likely to engage in future acquisitions. In view of these arguments, the following hypothesis was formulated:
Hypothesis 3: the greater the number of cash-paid M&A, the greater the number of future M&A.
to undertake future acquisition. Taken the above arguments together, I propose the following hypothesis:
Hypothesis 4: the greater the number of cross-border M&A, the less the number of future M&A.
Percentage sought refers to the proportion of shares acquirer obtained from target over M&A. It relates to two main purposes on M&A of acquirers, financial interest and corporate control. Financial interests focus on the acquirers’ interest in sharing profits generated by targets from their investments and operations. Corporate control emphasizes on the control and influence of acquirer on target’s decision-making, such as product selection (O’Brien & Salop, 2000). Furthermore, financial interest is accompanied by corporate control that is, the larger a shareholder’s financial interest over the target, the greater its control degree will be. For instance, in a board without majority shareholder, the minority shareholders who are better at forming voting coalitions could execute disproportionately control rights over the firm (O’Brien & Salop, 2000). In addition, legal framework provides guidelines to illustrate these viewpoints. For instance, Paragraph 17, Opinion No. 18 of the Accounting Principles Board of US indicated that an investor was presumed to have the ability to exercise a significant influence on an investee if the investor owned 20 percent or more stock of the investee. It implies a firm that acquires an important portion of the target firm is able to have more impact on effectiveness of target than that acquires fewer portions (American Institute of Certified Public Accountants, 1973). One relevant study was by Fowler and Schmidt (1989), they predicted that performance improved significantly as percentage of ownership increased. This hypothesis was supported by their empirical findings. They explained that, as the percentage acquired increases, a target was under more control and led to improvement of integration effectiveness. Thus, I expect that acquirers who obtain more stock after M&A are better performers and are more likely to engage in future acquisitions. Therefore, the last hypothesis is:
To sum up, transaction attitude, payment method, internationalization and percentage sought are commonly discussed factors that have impact on acquirer performance. Because of the association between acquirer performance and acquisition experience as discussed in behavioral learning theory and performance feedback approach, I expect that firms which benefited from stronger prior acquisition performance are more likely to undertake acquisitions, and consequently increase the number of future acquisitions. As a result, these factors may affect acquisition frequency as well.
The conceptual framework of this paper is as follows:
3. METHODOLOGY 3.1. Data and sample
The sample of this study is selected from Thomson Financial Securities Data’s Worldwide Mergers & Acquisitions database (henceforth Thomson). Thomson provides information of M&A transaction in which at least one firm involved was in international newspaper industry (SIC2711). The classification is based on firms’ recorded SIC codes that classify a firm’s main business with regard to its revenue. Furthermore, because financial data used for measuring acquirer performance are less accessible for private firms, I limited the sample in public listed firms who announced M&As during 1990-2000. As a result, it yields 215 firms in the initial sample.
RQ2 RQ1 M&A performance Past Transaction attitude Payment method Internationalization Percentage
sought Prior acquisition
experience
This paper addressed research questions investigating acquirer performance (RQ1) and acquisition frequency (RQ2), respectively. For RQ1, Thomson did not provide data for measuring performance of an individual firm, so I collected date from other commonly used databases. For instance, Amadeus provides financial data for most important European companies; DataStream contains worldwide information on stocks, equities, indices, macro-economic data and Financial data on companies; Zephyr includes worldwide M&A, initial public offerings, joint ventures and private equity deals; and finally, firms’ websites list their annual reports of recent years. Of initial sample, only Amadeus and annual reports provide relevant data, and then 143 firms were eliminated because their financial data could not be found in neither of data sources. Therefore, sample for RQ1 reduced to 78 firms for which at least one measurement of acquirer performance was available.
Turning to acquisition frequency (RQ2), the initial sample (215 firms) was split into two time periods, 1990-1995 and 1996-2000, so that I can assess acquisition frequency in 1996-2000 and investigate factors prior to acquisitions during 1990-1995. In the first time period (1990-1995) there were 139 firms making M&A announcements. In the second time period (1996-2000) the sample contained 148 firms announcing M&As. After matching firms in both time periods, the sample remained 72 firms that have relevant data. The data for RQ2 were derived directly from Thomson that identified M&A announcement and completion for each transaction.
3.2. Measures
3.2.1. Effect of acquisition experience on acquirer performance
However, researchers, who are in favor of other measurements, indicate some drawbacks of event study. Firstly, the event study usually covers a very short window around announcement, during which the market reaction to this activity is unclear (Fowler &Schmidt, 1989; Haleblian, et at., 2006). Secondly, event study assumes informational efficiency of stock markets, which leads to bias when applied to real world (Haleblian & Finkelstein, 1999). An alternative technique for measuring M&A effect is accounting returns. Accounting returns approach is a widely acceptable benchmark for firm performance. It is usually applied to studies measuring accounting rates of returns in the long-run (De Meuse, Bergmann, Vanderheiden & Roraff, 2004). However, accounting returns approach was also criticized because it could result in performance variations that due to economic and/or market fluctuations and different acceptance of accounting principles (Cochran, Wartick and Wood, 1984; Fowler & Schmidt, 1989). In general, compared to the weakness of the event study and considering the time framework involved in this paper, I decided to employ accounting returns to measure acquirer performance.
Independent variables. Acquirers make M&As partly rely on their acquisition experience. Good performance may lead to behavioral persistent and bad performance leads to behavioral changes. Consistent with measurements of Haleblian and Finkelstein (1999), acquisition experience is expressed as total number of completed mergers during 1990-2000. The data provided by Thomson was employed to indicate the status of each transaction (completed or abandoned). I aggregated these completed mergers identified by acquirer names and used it to measure acquisition experience.
Control variables. Because I focus on acquirer performance within a single industry, the industry-specific variances can be controlled for. Several control variables with potential to affect acquirer performance were incorporated, they were acquirer size, acquirer age and pre-transaction performance.
Firm size (acquirer size) indicates an organization’s access to complementary resources. It is a source of economies of scales and market power (Coombs & Bierly Ш, 2006). Larger firms are easier to access funds internally and externally; and they may benefit from the performance and can create entry barriers (Short & Keasey, 1999). However, larger firms may suffer from over-diversify and low degree of transparency in managerial behavior (Fama & French, 1992). Hence, considering both sides of effects, the acquirer size was included as a control variable. Furthermore, because a firm’s employees is a competitive advantage that cannot be easily imitated by its rivals (Barney, 1991), acquirer size was measured by taking the natural logarithm of its number of employees. The logarithms were taken in order to correct for size of very large firms. The data was derived from Amadeus and/or firms’ annual reports.
Mueller (1969) argued that older firms generated poorer internal rates of returns so that they made the greatest efforts towards outside investment for growth. Evans (1987) concluded that firm age was an important determinant and had significant effect on firm performance. Hence, acquirer age was controlled for and is measured by the number of years from incorporation year of its oldest part of the firm to the year 2001 or 2005. The data was derived from Amadeus and/or firms’ annual reports.
Finkelstein, 1999; Haleblian, et al., 2006). This variable was employed with same measurements for analyzing acquirer performance (ROA and ROE) and collected from the year when firms made their first M&A during 1990-2000. The data was derived from Amadeus and/or firms’ annual reports.
3.2.2. Determinants of acquisition frequency
Dependent variable. I now turn to the variables for the second research question. The measurement of acquisition frequency is similar to the measurement of acquisition experience in RQ1 that is, the number of completed mergers by acquirers, but the time period is narrowed to 1996-2000.
Independent variables. The time period for RQ2 switched to 1990-1995 for all independent variables. The transaction attitude is classified by the attitude or recommendation of target managers or board of directors toward M&As. If both board and managers approved the offer, the M&A was identified as friendly. This variable was measured as average number of friendly transactions that is, the number of friendly transactions divided by number of total mergers of an acquirer.
The payment of acquisitions can be in cash, stock or the combined. Thomson analysts classified cash payment with non-cash payment that included whole stock-financed transactions or combination of stock and cash. The payment method is measured by average number of cash-only transaction that is, the number of cash-paid transactions of an acquirer divided by its total mergers.
The internationalization is an indicator for whether a firm acquired targets within the same country (domestic M&A) or beyond its country boundaries (cross-border M&A). It is measured as average number of cross-border transactions that is, the number of cross-border mergers of an acquirer divided by its total mergers.
This empirical study used two different methodologies with regard to two research questions, respectively. To answer the first research question, I employed multiple regression technique to test the relationship between the acquisition experience and acquirer performance. For estimation in the second research question, the multiple regression models failed to explain efficiently the characteristics of dependent variable that is measured in counts. Hence, count data model based on Poisson distribution was employed. Count data model attracts increasing attentions in econometric researches in recently years (Jaggia & Thosar, 1993; Haynes, Thompson & Wright, 2000). Counts refer to the number of occurrences of the event and the number of events is assumed to be independently and identically distributed. In this paper, I used such count data model in the context of the number of completed M&As by acquirers.
While the multiple regression model is straightforward, the count data model in form of Poisson distribution is less utilized and is explained below.
The Poisson model. Many economic variables appear in form of Poisson
distribution is often used as benchmark for count data model. “… (it) is motivated by the usual considerations for regression analysis by also seeks to preserve and exploit as much as possible the nonnegative and inter-valued aspect of the outcome” (Cameron & Trivedi, 1998, p2). Poisson distribution is a nonlinear model and the response variable in the model is nonnegative integers. Let yi follows a Poisson distribution
P (yi) =e-µµyi / yi! , µ>0, yi=0, 1, 2, … (1)
where yi denotes the number of completed M&As by acquirers; µ denotes the parameter
of the distribution. Furthermore, to incorporate independent variables x, parameter µ is defined as
µ = E (yi│xi, β) =exp (x i’ β) (2)
where β is a parameter vector which measures the effect of x on µ. Given independent observations, the maximum log-likelihood estimator (MLE) of parameter β is derived from maximizing log-likelihood function:
L (β) =
∑
= n i 1
[yi x i’ β –exp (x i’ β) –ln yi!] (3)
mean and variance is one. This equality is referred to equidispersion, which is not frequently hold in real life cases. By contrast, overdispersion that implies variance exceeding mean is exhibited rather than underdispersion (variance is less than mean). There are two sources of overdispersion: the first one is neglecting assumption in Poisson distribution that successive events are independent; the second one is unobserved heterogeneity (Cameron & Trivedi, 1998). The consequence of using overdispersed data into Poisson distribution is the “estimated covariance matrix will be biased downwards, producing spuriously small estimated standard errors of the parameter estimates and overstated t-statistics” (Haybes, et al, 2000, p1211).
In order to test the overdispersion, a regression-based test is introduced, because there exits “population regression-type relation between the variance and the mean of an overdispersed random variable” (Cameron & Trivedi, 1990, p348). This test is based on the least-squares regression:
(yi-µi)2-yi= α g(µi) (4)
where g(µi) is a specified function and equals to 1, µi orµi2 ; α is an unknown parameter
and equals zero if Poisson restriction of equidispersion exists. In other words, overdispersion exists in Poisson distribution if the estimated coefficient is significantly positive.
One solution of overdispersion is to use a robust estimator, such as quasi-maximum likelihood (QML) estimators, to account for it. The objective function of QML estimator corresponding to Poisson family is:
-
∑
= n i 1 exp (xi,β)+∑
= n i 1 yixiβ (5)4. RESULTS6
4.1. Effect of acquisition experience on acquirer performance
I began with the first research question that investigated the relationship between acquisition experience and acquirer performance, and controlled for the acquirer size, acquirer age as well as acquirer’s pre-transaction performance. Table 1 in appendix reported the descriptive statistics and bivariate correlations of all the variables. The means of two acquirer performance measurements (ROA and ROE) indicated an increase in post-M&A period that is, both ROA and ROE are doubled in 2005 comparing to those in 2001. This finding may result from the overall global newspaper growth. Newspapers Industry Profile: Global (2006) reported that the market value of global newspaper increased from $145 billion (in 2001) to $151.8 billion (in 2005). I expect this industry-level growth to attribute to firm-industry-level development as well. However, when compare with pre-M&A period, the results are ambiguous. Acquirer performance in 2001 underperformed than those in its pre-M&A period, ROA and ROE shrank to half of its pre-M&A level; while acquirer performance in 2005 was recovered and was equivalent to its pre-M&A performance.
Furthermore, for the correlations between dependent variables and explanatory variables (independent variable and control variables), I found that the correlations between explanatory variables were relatively low, the highest correlation was between age and size (r=0.350) at 10% significance level. Regarding correlations between dependent variables and explanatory variables, they were not statistically significant at p<0.01 level, with the exception of correlation between pre-transaction measured as ROE and acquirer performance that were significant positive (p<0.01).
Table 2 summarized the test for multicollinearity and heteroskedasticity. The Variance Inflation Factors7 of independent variables are far below the ultimate threshold of 10, which tell that the variables are not highly related to each other, so I concluded no multicollinearity problem. Regarding heteroskedasticity, the White tests for
6 All statistical results related to count data model are performed by using EViews 5.0, otherwise, using
SPSS v. 12.1.
7 Variance Inflation Factor (VIF) measures the increase in the variances of the coefficients result from the
heteroskedasticity showed that the null of homoscedasticity was rejected at 1% significance level.
Table 2.
Results of Multicollinearity and White Heteroskedasticity Test of RQ1
Model 1 Model 2 Model 3 Model 4
Tolerance(Variance Inflation Factor in parentheses)
Experience 0,878 (1,139) 0,884 (1,131) 0,842 (1,188) 0,844 (1,184) Size 0,87 (1,149) 0,889 (1,125) 0,862 (1,160) 0,849 (1,178) Age 0,881 (1,135) 0,892 (1,122) 0,878 (1,139) 0,876 (1,141) Pre-transaction performance 0,985 (1,015) 0,996 (1,004) 0,99 (1,010) 0,977 (1,023) White Heteroskedasticity F-statistic (Prob. In parentheses) 3,489 (0,004) 2,982 (0,010) 4,553 (0,001) 3,848 (0,002) Obs*R-squared (Prob. In parentheses) 20,252 (0,009) 18,216 (0,020) 23,182 (0,003) 20,892 (0,007) Cases in analysis 50 48 48 46
Note: Dependent variables in each model refer to ROA in 2001, ROE in 2001, ROA in 2005 and ROE in 2005, respectively.
I ran separate regressions on each performance measurement with independent variable and control variables. The estimated results corresponding to multiple regression models for post-M&A acquirer performance were summarized in Table 3. Each of them was conducted through two steps. In the first step, the control variables was entered into the models (the first columns in the models, Table 3), and then in the second step, independent variable was added into the models (the second columns in the models, Table 3). In doing so, the adjusted R-square8 , which provides a better estimation for the population value, did not change much. In general, the adjusted R-square indicated that the best performance was in 2001 measured as ROE (model 2). The adjusted R-square in model 2 showed that the explanatory variables could explain 39.9% of the variance for acquirer performance. Furthermore, the results for acquisition experience, acquirer size and pre-transaction performance were straightforward. First, the coefficients for acquisition experiences were negative in all models, and the coefficients were statistically
8 Adjust R-square is useful for “comparing models with different numbers of explanatory variables
significant for measurements in 2001(model 1 and model 2, p<0.1) but insignificant in 2005. This result did not support the hypothesis 1 that stated the acquisition experience was positively related to acquirer performance. Second, for the acquirer size, the relationship with acquirer performance was positive and statistically significant (p<0.1), except for that in model 2 (positive but insignificant). It indicated that the larger acquirers had better performance than smaller acquirers. Finally, the coefficients for pre-transaction performance showed that it had a significantly (p<0.1) positive effect on acquirer performance, excepted model 3. This positive effect implied that performing well prior to M&A resulted in better performance after that. With regard to acquirer age, the messages from all the models showed that the effect was not precise. The coefficients were statistically insignificant and there was no consistent relationship with acquirer performance: the sign of coefficient was negative in model 1, model 3 and model 4, but positive in model 2.
Table 3.
Results of Multiple Regression Analysis Predicting Acquirer Performance
Model 1 Model 2 Model 3 Model 4
(Constant) (4.316) -7.865* -8.453(4.224) * (13.234) -16.143 (12.575) -18.177 (6.164) -9.857 -10.522 (6.274) (16.651) -18.914 (16.931) -20.795 Acquisition experience 1996-2000 -0.158* (0.087) -0.629 ** (0.258) (0.125) -0.086 (0.349) -0.260 Size (0.562) 1.086* 1.314(0.562) ** (1.715) 1.815 (1.665) 2.692 (0.780) 2.102** 2.255(0.815) *** (2.180) 5.144** 5.603(2.277) ** Age (0.029) -0.024 (0.029) -0.012 (0.088) 0.007 (0.085) 0.051 (0.039) -0.048 (0.041) -0.041 (0.110) -0.076 (0.114) -0.057 Pre-transaction performance 0.351 * (0.174) 0.340 * (0.170) 0.367 **** (0.074) 0.359 **** (0.070) (0.240) 0.384 (0.242) 0.383 0.200 *** (0.066) 0.199 *** (0.067) Adjusted R2 0.108 0.151 0.331 0.399 0.146 0.136 0.233 0.224 Cases in analysis 50 50 48 48 48 48 46 46
Note: 1. * p<0.1; ** p<0.05; *** p<0.01; **** p<0.001(standard errors in the parentheses)
2. Dependent variables are ROA in 2001(model 1), ROE in 2001(model 2), ROA in 2005(model 3) and ROE in 2005 (model 4).
4.2. Determinants of Acquisition frequency
I now turn to the second research question that is, what are the determinants of acquisition frequency? In doing so, I refer to the effect of transaction attitude, payment
method, internationalization and percentage sought. Descriptive statistics and
correlations between all the variables were reported in Table 4. The means of independent variables indicated some characteristics of the completed M&A. Of 72 firms who completed M&A during 1990-2000, their prevalent attitude towards M&A was friendly; they were not in favor of cash payment (36.1%); and they preferred the domestic targets (78.9%=1-21.1%). Besides, the average percentage of stake they sought in transaction was 71.69%, which implied their control powers in the targets were strong after M&A.
Table 4.
Descriptive Statistics and Bivariate Correlations for RQ2
Mean S.D. 1 2 3 4 5 1. Acquisition frequency 1996-2000 5,611 7,371 Pearson Correlation Sig. (2-tailed) N 72 Independent variables
2. Attitude 0,846 0,241 Pearson Correlation 0,049
Sig. (2-tailed) 0,681
N 72 72
3. Payment method 0,361 0,326 Pearson Correlation -0,040 -0,344**
Sig. (2-tailed) 0,740 0,003
N 72 72 72
4.
Internationalization 0,211 0,325 Pearson Correlation 0,292* 0,043 0,037
Sig. (2-tailed) 0,013 0,722 0,756
N 72 72 72 72
5. Percentage sought 71,691 26,726 Pearson Correlation 0,076 0,449** -0,315** -0,016
Sig. (2-tailed) 0,539 0,000 0,009 0,900
N 67 67 67 67 67
Furthermore, considering the correlations between independent variables (in Table 4), the highest correlation was between transaction attitude and percentage sought (r=0.449, p<0.01), therefore, I found no evidence of multicollinearity. Looking at the correlations with dependent variables (acquisition frequency), the positive correlation for internationalization was statistically significant (p<0.05). It implied that acquiring cross-border target successfully encouraged acquirers to engage in more M&A and obtained additional acquisition experience. For the other independent variables, the positive correlations for transaction attitude and percentage sought and negative correlations for payment method were not statistically significant.
The Poisson distribution requires the equality of conditional mean and conditional variance. Hence, before running the count data model, the existence of overdispersion was tested on the basis of regression-based test. The results was summarized in Table 5. The coefficient was positive and statistically significant (p<0.001), which suggested the presence of overdispersion.
Table 5.
Results of overdispersion for acquisition frequency
Regression-based test
Frequency_f^2 1,0234 ****(0,239)
Cases in analysis 67
Note: 1. Frequency_f refers to the forecasting of acquisition frequency and equals to the conditional mean of acquisition frequency.
2. Dependent variable is: (Acquisition experience-Frequency_f)2-Acquisition frequency
3. Method: Least Squares
4. **** p<0.001 (standard errors in the parentheses)
cross-border M&A on acquisition frequency. As to the other independent variables, the coefficient of transaction attitude was negative, but not statistically significant. This result stated the hypothesis 2, which predicted a negative impact of friendly M&A on acquisition frequency, can neither be supported nor be rejected. The coefficient of
payment method was insignificantly negative with acquisition frequency, which implied
the hypothesis 3 that predicted a positive effect of cash payment on acquisition frequency cannot be supported or rejected. Finally, the coefficient of percentage sought was positive but statistically insignificant, and the hypothesis 5, which predicted the greater percentage the acquirers sought in M&A the more likelihood they would engage in subsequent acquisition, could not be supported or rejected.
Table 6.
Results of Poisson model Predicting Acquisition Frequency
QML (Huber/White) QML (GLM) (Constant) 1.855**** (0.507) 1.855** (0.762) Transaction attitude -0.680 (0.511) -0.680 (0.855) Payment method -0.343 (0.498) -0.343 (0.480) Internationalization 1.052** (0.408) 1.052*** (0.372) Percentage sought 0.005 (0.004) 0.005 (0.007) Adjusted R2 0.079 0.079 cases in analysis 67 67
Note: 1. Dependent variable: acquisition frequency; 2. Method: Poisson Count
3.* p<0.1; ** p<0.05; *** p<0.01; **** p<0.001(standard errors in the parentheses)
5. DISCUSSION
5.1. The effect of acquisition experience on acquirer performance
up, the similarity and size of loss of acquisition experience may account for the inconclusive effect of acquisition experience on acquirer performance in this study. 5.2. Determinants of acquisition frequency
With respect to payment method, I found that whether or not the M&A is paid in cash or in non-cash, it appears not to affect the number of future acquisition. This result may be due to the inconclusive relationship between acquisition experience and acquirer performance in this study. Even though cash-paid M&A lead to better performance, it is uncertain whether this successful experience can be applied to the future acquisition. Taken the impact of similarity between experiences into consideration, if the subsequent acquisition is in a dissimilar industry, the current superior performance associated with cash payment cannot be properly utilized, and acquirers may decide not to engage in that acquisition. Furthermore, as mentioned before, the favor of cash payment was due to the fact that investors viewed it as a positive signal of future returns. However, Conn et al., (2005) argued that this argument was criticized in cross-border acquisitions, since the payment method would be influenced by other factors, such as imperfect information when acquired abroad. If acquirer shareholders recognize this, then the greater returns of cash payment connected to signaling effect would be neutralized. Thus, the mixed relationship may result from the interaction between cross-border (internationalization) effect and payment method that was not examined in the study. In a word, cash-paid M&A cannot ensure positive gains to acquirers, so that acquirers are uncertain if they will engage in future acquisitions.
contributed to the popularity of border M&A. The increasing number of cross-border M&A and their growing importance in global market enabled managers to become more experienced in dealing with misunderstanding arising from different cultures and economic structures (Shimizu, Hitt, Vaidyanath & Pisano, 2004). Conn et al., (2005) concluded that high-tech firms gained positive returns in cross-border acquisitions, while non-high-tech cross-border acquisitions reported negative long-run returns. Linked to the firms in newspaper industry, they were initially regarded as non-high-tech. However, the growth of increasingly technological innovation gave rise to the utilization of high technology associated with digitalization, printing and distribution. Consequently, newspaper firms were expected to be increasingly high-tech and could gain positive returns from cross-border acquisitions. In general, cross-border M&A may generate less potential hazards than expected and result in better performance than domestic M&A, so that acquirers can learn from prior cross-border experience and expected more future acquisitions
acquirers were inexpert in such capability, their performance would be poorer, and they were unwilling to make future M&A. Another interpretation for the negative effect might arise from the regulatory scrutiny. For instance, until 2004, firms were requested to notify European Commission about their combination no later than one week after a transaction agreement. Following the standardized procedure, the proposed combination could be approval, conditional approval and unacceptable (Aktas, Bodt & Roll, 2004). In the case that firms acquired great portion of stock from the target, they could increase the concentration level in that industry, and become more sensitive to the regulation changes (Aktas, Bodt & Roll, 2007). As a result, prior experiences with higher regulatory scrutiny might discourage firms to engage in future acquisitions.
6. CONCLUSION AND LIMITATION
much of factors that have impact on acquirer performance and acquisition frequency, thereby the results well supported this idea.
The findings of this paper gave rise to some implications of further researches. First, acquisition experience did not seem to be the most significant factors that have impact on acquirer performance. There were several studies highlighting the considerable effect of acquirer size (Evans, 1987; Short & Keasey, 1999; Coombs & Bierly Ш, 2006). The results in Table 3 confirmed that the acquirer size (as a control variable in this paper) had a significant positive effect on acquirer performance. As mentioned above, acquirer size had positive effect on performance that resulted from reduced financing constraints and economies of greater scale products (Short & Keasey, 1999). Second, different measurements can capture different perspectives of firms’ performance so that other indices of performance are suggested in further studies. For instance, Tobin’s Q is defined as the market value of the firm divided by replacement cost of its tangible assets (Short & Keasey, 1999). In other words, this measurement reflected the value of firm’s intangible assets, such as opportunities of future growth, management quality and monopoly power (Morck, et al., 1989).
REFERENCE
Aktas, N., Bodt, Eric de., & Roll, Richard. 2004. Market response to European regulation of business combinations. Journal of Financial and Quantitative Analysis, 39, 731-757.
Aktas, N., Bodt, Eric de., & Roll, Richard. 2007. Is European M&A regulation protectionist? The Economic Journal, 117, 1096-1121.
American Institute of Certified Public Accountants. APB Accounting Principles, 2, Opinion No. 18, Paragraph 17, Commerce Clearing House, Chicago, 30 June 1973. Amburgey, T. L., & Miner, A. S. 1992. Strategic momentum: The effects of repetitive
positional, and contextual momentum on merger activity. Strategic Management Journal, 13,335–348.
Aw, M. & R. Chatterjee. 2004. The Performance of UK Firms Acquiring Large Cross-Border and Domestic Takeover Targets. Applied Financial Economics, 14, 337–49. Barney, J. B. 1991. Firm resources and sustained competitive advantage. Journal of
Management, 17, 99-120.
Cameron, A. C, & Trivedi, P. K.1998. Regression analysis of count data. Cambridge University Press.
Cameron, A. C., & Trivedi, P. K. 1990. Regression- based tests for overdispersion in the Poisson model. Journal of Econometrics, 46, 347-364.
Cho, H., Martin, H., & Lacy, S. 2006. An industry in transition: entry and exit in daily newspaper markets, 1987-2003. Journalism & Mass Communication Quarterly, Vol. 83 Issue 2, p381-396
Cochran, P.l., Wartick, S.L., & Wood, R.A. 1984. The average age of boards and financial performance, revised. Quarterly Journal of Business and Economics, Autumn 1984, v. 23, iss. 4, 57-63.
Cochran, P.L, & Wood, R. A. 1984. Corporate social responsibility and financial performance. Academy of Management Journal, 1984, Vol. 27, Ni. 1, 42-56.
Conn, R. L., Cosh, A., Guest, P. M., & Hughes, A. 2005. The impact on UK acquirers of domestic, cross-borderer, public and private acquisitions. Journal of Business Finance & Accounting, 32(5) & (6), 0306-686X.
Cosh, A., & Guest, P. 2001. The long-run performance of hostile takeovers: UK evidence. ESRC Centre for Business Research, ESRC Centre for Business Research - Working Papers.
Cyert., R. M., & March, I. G. 1963. A behavioral theory of the firm. Englewood Cliffs, NJ: Prentice-Hall
De Meuse, K.P., Bergmann, T. J., Vanderheiden, P. A., & Roraff, C. E. 2004. New evidence regarding organizational downsizing and a firm’s financial performance: a long-term Analysis, Journal of Managerial Issues, Vol. XVI No.2, 155-17.
Evans, D.S. 1987. The relationship between firm growth, size, and age: estimates for 100 manufacturing industries. Journal of Industrial Economics, Jun87, Vol. 35 Issue 4, 567-581.
Fama, E. F., & French, K. R. 1992. The cross section of expected stock returns. Journal of Finance 47, 427-466.
Fowler, K.L., & Schmidt, D. R. 1989. Determinants of tender offer post-acquisition financial performance. Strategic Management Journal. Vol. 10, No. 4. pp339-350. Gammelgaard, J., Husted, K., & Michailova, S. 2004. Knowledge- sharing behaviour and
post-acquisition integration failure. CKG working paper No. 6/2004.
Gourieroux, C., Monfort, A., & Trognon, C. 1994. Pseudo-maximum likelihood methods: applications to Poisson models. Econometrica, 52,701-720.
Gregory, A. 1997. An examination of the long run performance of UK acquiring firms.
Journal of Business, Finance and Accounting, 24, pp971-1002.
Haleblian, J., & Finkelstein, S. 1999. The influence of organizational acquisition experience on acquisition performance: a behavioral learning perspective. Administrative Science Quarterly, Vol. 44, Issue 1, p29-56.
Haleblian, J., Kim., J., & Rajagopalan, N. 2006. The influence of acquisition experience and performcne on acquisition behaviour: evidence from the U. S. commercial banking industry. Academy of Management Journal. Vol. 49, 357-370.
Harrison, J. R., Hitt, M. A., Hoskisson, R.E., & Ireland, R.D. 1991. Synergies and post acquisition performance: differences versus similarities in resource allocation.
Journal of Management, 17: 173-190.
Haynes, M., Thompson, S., & Wright, M. 2000. The determinants of corporate divestment in UK. International Journal of Industrial Organization, 18 (2000) 1201-1222.
Hayward, M. L. 2002. When do firms learn from their acquisition experience? Evidence from 1990-1995. Strategic Management Journal, 23, 21-29.
Hitt, M. A., Harrison, J. S., Ireland, R. D., & Best, A. 1993. Lifting the veil of success in
mergers and acquisitions. Paper presented at the Annual Meeting of the Strategic
Management Society, Chicago.
Jaggia, S., & Thosar, S. 1993. Multiple bids as a consequence of target management resistance: a count data approach. Review of Quantitative Finance and Accounting, 3,447-457
Larsson, R., & Finkelstein, S. 1999. Integrating strategic, organizational, and human resource perspectives on mergers and acquisitions: a case synergy of synergy realization. Organization Science, 10, 1-26.
Levitt, B., & March, J. G. 1988. Organizational learning. Annual review of sociology, 14, 319-340.
Loughran, T., & Vijh, A.M. 1997. Do long-term shareholders benefit from corporate acquisitions? Journal of Finance, 52, 1765–1790.
Lubatkin, M. 1983. Mergers and the performance of the acquiring firm. Academy of
Management Review, 8 (2), 218–225.
March, J. G., 1991. Exploration and exploitation in organizational learning. Organization Science, 2: 71-87.
Mazur, J. 1994. Learning and behavior. Englewood Cliffs, NJ: Prentice-Hall.
Meschi, P. X., & Metais, E. 2006. International acquisition performance and experience: a resource-based view. Evidence from French acquisitions in the United States (1988-2004). Journal of International Management. 12, 430-448.
Moeller, S. B., & Schilingemann, F. P. 2005. global diversification and bidder gains: a comparison between cross-border and domestic acquisitions. Journal of Banking & Finance, 29, 533-564.
Morck, R. Shleifer, A. & Vishny, R.W. 1990. Do managerial objectives drive bad acquisitions? Journal of Finance, 45, 31-48.
Mueller, D.C. 1969. A theory of conglomerate mergers. Quarterly Journal of Economics. 83, 643-650
Nelson, R. R., & Winter, S. G. 1982. An evolutionary theory of economic change. Cambridge, MA: Harvard University Press.
Newspaper Association of America. 2006. The source: newspapers by the numbers. Newspaper Association of America.
Newspaper industry portfolio: Global. 2006. Newspapers Industry Profile: Global, August 2006.
Norusis, M. J. 2005. SPSS 13.0 statistical procedures companion. Upper Saddle River, N.J. : Prentice Hall.
O’Brien, D.P., & Salop, S. C. 2000. Competitive effects of partial ownership: financial interest and corporate control. Antitrust Law Journal, 67,559-614.
Pautler, P.A. 2003. Evidence on mergers and acquisitions. The Antitrust Bulletin/Spring. Peltier, S. 2004. Mergers and acquisitions in the media industries: were failures really
unforeseeable? Journal of Media Economics,17 (4), 261-278
Pinder, C.C . 1984. Work motivation: theory, issues and applications. Glenview, IL: Scott, Foresman.
Rumelt, R. P. 1982. Diversification strategy and profitability. Strategic Management Journal, 3(4), 359-369.
Sharma, D., & Ho, J. 2002. The impact of acquisitions on operating performance: some Australian evidence. Journal of Business Finance & Accounting, 29(1)& (2), Jan/Mar 2002, 0306-686X.
Shimizu, K., Hitt, M. A., Vaidyanath, D., & Pisano, V. 2004. Theoretical foundations of cross-border mergers and acquisitions: a review of current research and recommendations for the future. Journal of International Management, 10, 307-353. Short, H., & Keasey, K. 1999. Managerial ownership and the performance of firms:
evidence form the UK. Journal of Corporate Finance, 5, 79-101.
turnover in hostile and friendly acquirers? British Journal of Management, Vol. 17, S7-S30.
Szulanski, G. 1996. Exploring internal stickiness: impediments to the transfer of best practice with the firm. Strategic management Journal, 17, 27-42.
Thorndike, E. L. 1898. Animal intelligence: an experimental study of the associative processes in animals. Psychological Monographs, 2 (8).
Travlos, N. G. 1987. Corporate takeover bids, methods of payment, and bidding firms’ stock returns. The Journal of Finance. Vol. XLII, No. 4, 943-963.
Vega-Céspedes, C., & Hoshino, Y. 2001. Effects of ownership and internalization advantages of performance: the case of Japanese subsidiaries in the United States and Latin American. Review of Pacific Basin Financial Markets & Policies, Vol. 4 Issue 1, 69-95.
Vermeulen, F., & Barkema, H. 2001. Learning through acquisition. Academy of
APPENDIX
Table 1.
Descriptive Statistics and Bivariate Correlations for RQ1
Mean Deviation Std. ROA01 ROE01 ROA05 ROE05 Size Age01 Age05 Experience ROA ROE ROA01 2,211 11,780 Pearson Correlation
N 74
ROE01 7,7978 33,21715 Pearson Correlation ,754(**)
N 72 72
ROA05 6,1932 12,48145 Pearson Correlation ,511(**) 0,217
N 65 63 69
ROE05 17,696 35,38867 Pearson Correlation ,406(**) ,393(**) ,832(**) N 65 63 68 69 Explanatory variables Experience 7,91 13,623 Pearson Correlation -0,164 -,235(*) -0,016 0,019 ,243(*) ,252(*) ,252(*) N 74 72 69 69 70 77 77 78 Size 7,0257 2,65321 Pearson Correlation 0,084 -0,028 0,2 ,287(*) N 69 67 62 62 70 Age01 53,29 48,974 Pearson Correlation -0,071 0,007 -0,145 -0,03 ,350(**) N 73 71 69 69 69 77 Age05 57,29 48,974 Pearson Correlation -0,071 0,007 -0,145 -0,03 ,350(**) 1,000(**) N 73 71 69 69 69 77 77 ROA 7,5285 9,43241 Pearson Correlation 0,161 ,347(*) 0,239 0,182 -0,07 -0,037 -0,037 -0,049 N 53 52 50 49 52 54 54 55 55 ROE 16,755 75,94983 Pearson Correlation ,421(**) ,580(**) 0,121 ,422(**) 0,122 0,095 0,095 0,029 ,330(*)
N 52 51 49 48 51 53 53 54 53 54
Table 7.
Summary of studies regarding the effect of acquisition experience on performance
Study Sample Statistical
method Performance measurement Experience measurement Result Fowler & Schmidt (1989) 42 industrial manufacturing firms engaged in major tender offer during 1975-1979 Stepwise multiple regression analyses Return on common equity (ROCE) & total return to shareholders (RSH)
The number of acquisitions made before the year of tender offer positive Haleblian & Finkelstein (1999) 449 large completed majority acquisition during 1980-1992 Ordinary least squares (OLS) Cumulative abnormal returns (CAR) The number of previous acquisitions U-shaped Hayward (2002) 214 acquisitions by 120 firms in 6 industry during 1990-1995 OLS & ordered probit model9
CAR Sum of recent
acquisitions by acquirers positive Meschi & Metais (2006) 291 acquisitions completed by French acquirers in the US during 1988-2004 Multiple regression model Average abnormal returns (AR) & CAR
The number of
past acquisitions Inversed U-shaped
