1 UNIVERSITEIT VAN AMSTERDAM
The difference in
post-acquisition performance in
horizontal and non-horizontal
acquisitions
Abstract
Horizontal acquisitions affect firm performance differently compared to non-horizontal acquisitions. In a sample of 121 acquisitions performed by firms of the S&P500 in the years 2009, 2010 and 2011, an analysis on how horizontal acquisitions affect three year post-acquisition cumulative abnormal returns compared to non-horizontal acquisitions has been done. Although the relative target to the acquirer size of the acquisitions is negatively related to firm performance in the full sample, horizontal acquisitions benefit from larger relative target size compared to non-horizontal acquisitions. As expected, the relative change in market share due to acquisitions was larger for horizontal acquisitions compared to non-horizontal acquisitions. Still, this increase did not go along with an increase in firm performance, as the relative change in market share was negatively related to firm performance for both types of acquisitions.
Bachelor Thesis
Author: Thomas Neijenhuis
Email: thomasneijenhuis@hotmail.com Student number: 10270604
FEB – Faculty of Economics and Business Programme: Finance and Organization Supervisor: I. Sakalauskaite
2
Statement of Originality
This document is written by Student Thomas Neijenhuis who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
3
1. Introduction
Economic literature has looked at different aspects of mergers and acquisitions and their impact on firm performance. The aspects investigated are for instance why mergers and acquisitions are done, how mergers and acquisitions create value for a firm or how mergers and acquisitions influence their competitors and the market itself. How a firm acquires a target could be in different ways. Firms can acquire other firms in a hostile takeover or in a tender offer. There are also differences in financing an acquisition or the method of payment. Firms can pay with cash, common stock or a mixed offering. These examples are just a fraction of the aspects regarding mergers and acquisitions. Finally, there are also differences in the type of acquisition. Firms that acquire a target in the same industry perform a horizontal acquisition. Firms that acquire a target that is upward or downward on the production chain perform a vertical acquisition. Conglomerate acquisitions are acquisitions in which the target and acquirer are not related in their daily businesses. Each type of acquisition has different ways of creating value for a firm. This is why motivation for each type is also different.
Existing research on this topic has conflicting evidence. Although underperformance after acquisitions is common in the majority of studies, when controlling for different types of acquisitions the evidence is not one-sided. This is why in this analysis, the expectation for differences in firm performance for horizontal acquisitions and non-horizontal acquisitions will be zero. On the other hand, market power is assumed to be a major motivator for horizontal acquisitions. When looking at the relative change in market share due to an acquisition, the expectation will be in favour of horizontal acquisitions. Also, due to efficiency reasons, larger relative target to acquirer size is expected to benefit horizontal acquisitions more than non-horizontal acquisitions. Especially for non-horizontal acquisitions this is the case, since direct competitors are acquired. When evaluating this relation in the full sample compared to only horizontal acquisitions, the expectation will be this relation has larger impact on horizontal acquisitions.
There has not been a lot of research in this topic as of late. Also, when examining existing literature, differences in types of acquisitions are reviewed but there is a lack of evidence in how these differences are created. When looking at motivators for types of acquisitions and using these motivators as a tool to distinguish differences, an explanation could be given for how these differences are created.
This paper focusses on whether and how horizontal acquisitions have a different effect on firm performance relative to non-horizontal acquisitions (i.e. vertical and conglomerate). For this, a sample of firms from the S&P500 that acquired targets during 2009, 2010 and 2011
4 has been obtained. Acquisitions have been classified as horizontal or non-horizontal using the four-digit SIC code. To answer the research question, an event study comparing cumulative abnormal returns of horizontal- and non-horizontal acquisitions three years after an acquisition will be implemented. The focus will lie on change in relative market share and relative target to acquirer size.
In chapter two the related literature about horizontal- and non-horizontal acquisitions and post-acquisition firm performance will be reviewed. The data and methodology used in this research will be discussed in chapter three, empirical results will be elaborated in chapter four and the conclusion will be found in chapter five.
2. Literature Review
2.1 Incentives for acquisitions
Acquisitions are done because it is believed a target can create value for an acquirer. Since an acquisition is an investment, the net present value of this investment decision should be higher than zero. The net present value represents the expected dollar gain to shareholders of the acquiring firm. (Halpern 1983). There are several factors which create value through an acquisition. These factors represent incentives for management to acquire.
An incentive for managers to acquirer another company is the creation of synergies. Synergy is achieved when the combined value of the two firms is more than the value of the two separate firms in total. Especially for horizontal acquisitions, economies of scale and the creation of market power are incentives (Healy et al. 1992). Due to the fact that firms in the same industry sell the same product, a firm increases its production through a horizontal acquisition. At this point economies of scale could be realized. Also, when a competitor leaves the market and its assets become available to the acquirer, an increase in market share could be realized. Tremblay & Tremblay (1992) however, did not find support for the fact these were motivations for horizontal acquisitions, but did not exclude the fact they create value for the firm. Economies of scope is a motivational aspect that arises in acquisitions in general, since it thrives on diversification.
Financial motivation is another incentive for acquisitions. Halpern (1983) states that through an acquisition, a firm can redeploy their excess cash. Also through diversification, the probability of default can be reduced which causes lower expected bankruptcy costs and increases the debt capacity of the new firm. Halpern (1983) also elaborates the exploitation of asymmetric information. This means the acquirer has more information about the target than its competitors. This is information concerning share prices of the target firm. The share price
5 could be underpriced, so the target firm has a higher real value then the market shows. The acquirer could exploit this by taking over the target firm and receiving the firm at a discount.
While the incentives outlined above tend to increase the value of a firm, acquisitions can also be done which leads to a decrease in firm-value. These incentives are often of a managerial kind. Examples of these incentives are empire building or overconfidence by CEO’s. Empire building arises when a CEO tends to maximize growth in sales and assets, without taking into consideration whether this growth actually benefits the firm (Halpern 1983). Ferris et al. (2005) find evidence that CEO overconfidence is related to a number of critical aspects in international merger activity. They find that when CEO’s are overconfident, they tend to underestimate the risk associated with the merger and overestimate the possible synergy gains following the acquisition.
2.2 Acquisitions and firm performance
Extensive research has been done on the effect of acquisitions on firm performance. One way to measure this effect is to implement an event study, which measures cumulative abnormal returns of a company during the period surrounding the announcement date of the acquisition. Moeller et al. (2004) find an abnormal announcement return of 1.1%, which can be seen as the economic benefit of the acquisition to the acquiring-firm. Still this does not imply this measures the increases of wealth to these shareholders, since they also find that the average dollar return is negative. Announcement date returns are also used by James & Wier (1987). They found a positive relation between the number of potential targets for the acquirer and the gains to an acquirer and a negative relation between potential bidders and gains for an acquirer.
Another performance measure focuses on the returns of the acquirer in the years after an acquisition. Loderer & Martin (1992) find negative cumulative abnormal returns during the first three years post-acquisition but not for the years four and five post-acquisition. Agrawal et al. (1992) also find acquiring firms underperform. They find a statistically significant loss of 10% to shareholders during the five years post-merger, using cumulative abnormal returns. Other studies relate to these findings, since Barnes (1984) as well as Raghavendra Rau & Vermaelen (1998) find three year post-acquisition negative cumulative abnormal returns of respectively 4.7% and 15.23%. Raghavendra & Vermaelen (1998) find that the negative return is mainly caused by a group of firms with low book to market ratio, which tend to make relatively poor acquisition decisions.
6 Healy et al. (1992) use operating cash flow as a measure of postmerger performance. They find significant increases in the postmerger operating cash flow due to asset productivity relative to their industry company peers. They also found that mergers were not at the expense of a company’s long term sustainability, since relative to their industry the capital expenditure as well as the R&D rates were maintained.
2.3 Horizontal vs. non-horizontal acquisitions
As has been mentioned in the first part of this section, acquisitions can create value for a firm. Different types of acquisitions create value in different ways. The findings in relevant literature about this topic are divided. Montgomery & Singh (1987) state that related
acquisitions create value through economies of scale, economies of scope and market power. Unrelated acquisitions on the other hand, create value through more general aspects like reduced finance cost, increased administrative efficiencies or superior human capital. Still, these general aspects also count for related acquisitions, which implies that ceteris paribus related acquisitions create more value.
In a sample of 105 acquisitions of market value greater than 100 million dollars, Singh & Montgomery (1987) found cumulative abnormal returns were significantly higher for the acquired firm in related acquisitions than unrelated acquisitions (35.9% against 26.9%). This was not the case for the acquirer. This is consistent with other studies. As stated earlier, Barnes (1984) found negative post-acquisition cumulative abnormal returns. There were no differences in return for different types of mergers. Loderer & Martin (1992) also tested their sample for differences in acquisition form but the results did not differ for different forms of acquisitions.
From these findings, contrary to the assumption made by Singh & Montgomery (1987) that related acquisitions create more value ceteris paribus, it cannot be concluded that
horizontal acquisitions create more value than non-horizontal acquisitions on the basis of post-acquisition cumulative abnormal returns to the acquirer. Still there is evidence stating there are differences in returns in types of mergers or acquisitions. Healy et al. (1992) find support for mergers in overlapping businesses to perform better. They add dummies to
separate their sample in groups of low, medium and highly overlapping business acquisitions. They find a 5.1% higher post-merger return for the group of highly overlapping businesses. Agrawal et al. (1992) on the other hand, find post-merger underperformance for the total sample, but find non-conglomerate mergers perform significantly worse than their conglomerate counterpart for every but one of the five time periods investigated.
7
2.4 Determinants of post-acquisition performance
When looking at how value is created through acquisitions, market power and the size of the target relative to its acquirer are returning subjects. In evaluating an impact of an acquisition on firm performance, the question remains whether this impact is caused by the acquisition or other factors that influenced firm performance during that time period. Agrawal et al. (1992) and Loderer & Martin (1992) point out, when post-acquisition underperformance reflects the impact of an acquisition, then a relatively large target acquisition will have greater underperformance than a relatively smaller target acquisition. Agrawal et al. (1992) divide their sample in portfolios of relative target size. They find that the portfolios of largest relative target acquisitions experience largest underperformance. This may imply acquisitions with larger relative target size had more impact on the post-merger cumulative abnormal returns. Statistically however, they find no significance regarding relative target size affecting cumulative abnormal returns. A resemblance can be seen from a study of James & Wier (1987), who found little evidence for relative size of target and post-merger performance. Loderer & Martin (1992) on the other hand, did not find negative cumulative abnormal for the lowest quintile of relative target size while this was the case for the full sample. This could be interpreted as an impact of the acquisition on cumulative abnormal returns for a company. While relative size of a target comparing to its acquirer may not have a direct relation with firm performance, it could still be a useful tool for the impact of an acquisition on firm performance comparing to other random factors.
For acquisitions, and in particular horizontal acquisitions, market power is an incentive. As through a horizontal acquisition a competitor leaves the market and its assets become available to the acquirer, one might assume an increase in market share. The evidence of existing literature on the gains from an increase in market share is divided. James & Wier (1987) test their market power hypothesis on a sample of banks. There was a lack of a
significant relation between acquirer returns and the change in market share for banks. These findings confirmed earlier studies by Jensen & Ruback (1983), who found gains in corporate acquisitions did not come from the creation of market power. Still there is evidence
acquisitions did create value through creating market share. Barton & Sherman (1984) saw an increase in profit gain after large acquisitions, because of a higher market share which caused firms to increase their prices. Mueller (1985) on the other hand found results contradictory to earlier statements. He compared market shares of companies that had mergers to a control group of companies that did not have mergers in that period. He compared conglomerate
8 mergers to horizontal mergers. It was found that both groups of mergers had experienced substantial losses in market shares compared to their control group.
3. Empirical analysis 3.1 Hypotheses
According to earlier findings, acquisitions create market share, especially horizontal acquisitions. Although evidence is divided on whether an increase in market share increases gains for a company, market power is expected to be a motivation for horizontal acquisitions. This is why horizontal acquisitions are expected to affect firm performance more through an increase in relative market share compared to non-horizontal acquisitions. This will be called the market power hypothesis.
Relative target size to the acquirer could be a measure of how large the impact of an acquisition in general is. Economies of scale and scope are realized through an increase in firm size. These factors are more beneficial for horizontal acquisitions than for non-horizontal. That is why the second hypothesis will state horizontal acquisitions create value through larger relative target size compared to non-horizontal acquisitions. This is called the relative target
size hypothesis.
3.2 Methodology
In this research, the difference between the three year post-acquisition firm performance of a horizontal acquisitions compared to a non-horizontal acquisition is measured. The focus lies especially on whether this difference is caused by increase in relative firm size and increase in market share. To estimate this, the regression will be as follows:
CARjt = β0 + β1Horizontaljt + β2Relativefirmsize + β3 RelativeΔmarketshare + β4Horizontal*Relativefirmsize + β5Horizontal*RelativeΔmarketshare + β6Xjt +εjt
To measure the difference between the types of acquisitions, the variable Horizontal is added. We look at β2 to see the effect of the relative firm size increase on the cumulative abnormal returns, and at β3 for the effect of the relative change of market share by the acquisition. The interaction variables β4 and β5 are added to see whether these effects are bigger for horizontal acquisitions compared to non-horizontal acquisitions. The final variable is a term that includes control variables.
9
3.3 Data description
This research uses the databases Zephyr, Compustat and CRSP. A list of acquisitions is obtained from Zephyr that satisfied the following conditions: the acquisition is confirmed and completed, the acquirer is a firm in the S&P 500 and the acquisition took place in the year 2009, 2010 or 2011. A list of 363 acquisitions was obtained which satisfied these conditions. All acquisitions with a total target value or estimated total target value lower than $500.000 were eliminated, which resulted in 165 acquisitions. For companies that had more than one acquisition, the acquisition with the largest target value was used to avoid biased results due to double counting (Loderer & Martin 1992). This resulted in 133 companies and an equal amount of acquisitions. When obtaining the variables described in the remaining part of this section, another twelve companies of the sample were lost due to lack of data. An overview of the variables described can be found in Table 1.
To distinguish a horizontal acquisition from a non-horizontal acquisition, the four-digit SIC codes were used. If the acquirer and the target had the same SIC codes, the acquisition is defined as a horizontal acquisition.
Daily returns of the 121 companies and the S&P 500 during the 3 years after the completion date of the acquisition were obtained from CRSP. By calculating the covariance of the daily returns with the S&P500 returns of the 3 years after the completion date and dividing this by the variance of the post-acquisition 3 year daily returns of the company, the beta of the company was obtained. This means returns are compared after removing the part that can be explained by fluctuations in the market. The cumulative abnormal returns for these 3 years were calculated. The cumulative abnormal returns (CAR) will be the performance measure in this research.
To measure the effect of the increase in firm size through an acquisition on firm performance, data of the relative acquisition size is gathered by dividing the total target value or estimated target value by the acquirer’s market capitalization 20 days before the announcement date of the acquisition. The target value was obtained by Zephyr and the market capitalization was calculated by multiplying the price of the stock by total shares outstanding 20 days prior to the announcement date.
For the measure of relative change in market share for acquiring firms, the market share one calendar quarter before the completion date of the acquisition was subtracted from the market share three calendar quarters after the acquisition. This change was divided by the market share before the acquisition. The market share was obtained through Compustat by dividing the company’s total quarterly sales/turnover by the sum of all quarterly sales/turnovers
10
Table 1
This table described the variables used in this research.
Variables Description Data source
Performance measure
CAR Cumulative Abnormal Return CRSP
Measure
of value creation
Relative size acquisition
Total target value over market capitalization 20 days before the announcement date of acquisition
Zehpyr/CRSP Relative change in
market share Difference between market share three calendar quarters after- and one quarter before acquisition completion date over market share before acquisition Compustat Type of acquisition Horizontal acquisition
Equals 1 if the acquisition was made within a four-digit SIC industry, 0 if not
Zephyr
Control Variables
Ln assets The logarithm of total assets a quarter before the acquisition Compustat Leverage
Tobin’s Q
Market share before acquisition
Total debt over total assets a quarter before the acquisition Market value over book value of total assets
Sales/turnover of the firm divided by total sales/turnover in the total four-digit SIC code industry, a quarter before the acquisition
Compustat Compustat Compustat
from the particular quarter in the company’s SIC-four digit industry (James & Wier 1987). To account for other factors influencing cumulative abnormal returns, the analysis controls for firm size, leverage and investment opportunities for the acquirer firm. To control for firm size, the logarithm of total assets of the acquirer before the acquisition has been obtained. Leverage equals total debt of the calendar quarter prior to the acquisitions, divided by total assets during the same quarter. Tobin’s Q is the total market-to-book ratio of assets pre-acquisition. The logarithm of target value at the completion date was obtained from Zephyr to control for the size of the target.
3.4 Summary statistics
Table 2 represents a summary of the obtained data. The average 3 year cumulative abnormal return for the 124 companies was 4.39%. In the sample companies acquired on average a target that had a value of 19.26% of their market capitalization 20 days before the announcement date. An average relative gain in market share of 36.21% has been realized post-acquisition. Of all acquisitions in this sample, 61.16% are of a horizontal nature. The average total assets of the sample is $124476.80 in thousands of dollars. Debt was on average 18.52% of total assets. The market-to-book ratio is on average 1.16.
11
Table 2
Summary of variables for total sample
Variable #N observations Mean St.dev Min Max
Dependant variables
CAR 121 0.0439 0.4305 -1.7426 1.3153
Value creating variables
Relative target size 121 0.1926 0.3571 0.0079 3.1420 Relative change in market
share 121 0.3621 0.9952 -0.5188 7.1125 Separating variables Horizontal 121 0.6116 0.4894 0 1 Control Variables Assets 121 $124476.80 335115.90 $1370.57 $2,014,019.- Leverage 121 0.1852 0.1327 0 0.7135 Tobin’s Q 121 1.1566 0.9131 0.0388 5.5836 Market share before 121 0.2557 0.2788 0.0013 1
In Table 3 and 4, the sample has been divided in firms that had horizontal acquisitions and firms that had non-horizontal acquisitions. With 1.71%, horizontal acquisitions had lower cumulative abnormal returns than non-horizontal acquisitions, which had average CARs of 8.61%. Firms that acquired targets in the same industry had relative target sizes of 16.83%, compared to 23.10% for non-horizontal acquisitions. As expected, horizontal acquisitions increased their relative market share more than non-horizontal acquisitions, with 42.19% to 26.79% respectively.
Table 3
Summary of variables for horizontal acquisitions
Variable #N observations Mean St.dev Min Max
Dependant variables
CAR 74 0.0171 0.4725 -1.7426 1.3153
Value creating variables
Relative target size 74 0.1683 0.2267 0.0079 1.4719 Relative change in market
share 74 0.4219 1.1013 -0.5188 7.1125 Control Variables Assets 74 $168660.40 413291.1 $1370.574 $2014019 Leverage 74 0.1940 0.1142 0 0.5575 Tobin’s Q 74 1.1549 0.9658 0.0389 5.5837 Market share before 74 0.2166 0.2562 0.0013 1
12
Table 4
Summary of variables for non-horizontal acquisitions
Variable #N observations Mean St.dev Min Max
Dependant variables
CAR 47 0.0861 0.3553 -0.9059 0.9898
Value creating variables
Relative target size 47 0.2310 0.4987 0.0095 3.1420 Relative change in market
share 47 0.2679 0.8026 -0.2667 5.1536 Control Variables Assets 47 $54911.14 117520.7 $2633.1 $751216 Leverage 47 0.1714 0.1580 0.0011 0.7135 Tobin’s Q 47 1.1592 0.8334 0.0954 4.505671 Market share before 47 0.3174 0.3037 0.0040 1
Table 5
Correlation
Variables CAR Relative target size
Relative Δmarket share
Horizontal Lnassets Leverage Tobin’s Q Market share before acquisitio n CAR 1 -0.0947 -0.1799 -0.0785 -0.1445 -0.0112 0.1743 0.0037 Relative target size -0.0947 1 0.2512 -0.0860 -0.2222 0.4294 -0.1890 -0.2202 Relative Δmarket share -0.1799 0.2512 1 0.0757 -0.0390 0.0285 -0.0809 -0.2521 Horizontal -0.0785 -0.0860 0.0757 1 0.1725 0.0833 -0.0023 -0.1771 Lnassets -0.1445 -0.2222 -0.0390 0.1725 1 -0.1135 -0.4670 -0.0072 Leverage -0.0112 0.4294 0.0285 0.0833 -0.1135 1 -0.1499 -0.0275 Tobin’s Q 0.1743 -0.1890 -0.0809 -0.0023 -0.4670 -0.1499 1 0.1746 Market share before acquisition 0.0037 -0.2202 -0.2521 -0.1771 -0.0072 -0.0275 0.1746 1
Table 5 represents the correlation between variables. Relative target size and cumulative abnormal returns are negatively correlated. This implies that cumulative abnormal returns are lower when relative target size is higher and vice versa. In contrast to our market power
13 hypothesis, the relative change in market share is negatively correlated with cumulative abnormal returns. This means when the relative change in market share increases, cumulative abnormal returns decrease. Tobin’s Q on the other hand is positively correlated with CARs. This implies a higher market-to-book ratio benefits the cumulative abnormal returns. Horizontal acquisitions have smaller returns than nonhorizontal acquisitions, since a correlation of -0.0785 means when the horizontal acquisition dummy equals one, cumulative abnormal returns decrease compared to non-horizontal acquisitions.
4. Results
4.1 Evaluating first findings
With a three year post-acquisition cumulative abnormal return average of 4.39% the firms in this sample performed relatively well. Since the acquisitions done in this sample had a target value of 19.26% relative to their acquirer, this could imply the cumulative abnormal returns are affected by these events. This raises the question whether this value is created through the variables of interest and especially whether they affect horizontal acquisitions more than non-horizontal acquisitions.
4.2 Value creating variables on firm performance
In Table 6, the results of the basic regressions can be found. In the first regression, the variables of interest will be tested. In the second regression, control variables will be added. In Table 7 interaction variables will be added and will, like the previous table, contain two regressions. The first will be without control variables, the second will be with control variables.
Looking at Table 6, a first insight is given on how the value creating variables relate to cumulative abnormal returns. Although not all variables in regression one and two are significant, their signs can be interpreted. From regression one, it follows that horizontal acquisitions have lower cumulative abnormal returns than non-horizontal acquisitions since the dummy variable has a negative coefficient. The relative target to acquirer size has a negative impact on cumulative abnormal returns. This implies when the target is larger compared to the acquirer, cumulative abnormal returns are lower and vice versa. The relative market share change is significant at the 1% significance level. This implies there is a strong relation between the relative market share change and cumulative abnormal returns. In contrast to Barton & Sherman (1984), who found significant gains in profit due to the increase of market share, CARs decrease when the relative market share change is larger. This is contrary to the
14 expectation, as market power and the relative size of the target to the acquirer are assumed to be value creating variables.
Table 6
This table represents a regression of the basic variables of interest on firm performance for the entire sample with and without control variables.
This table contains the relation between the three year post-acquisition cumulative abnormal returns and separating-, value creating- and control variables. The dependant variable is the cumulative abnormal return, measured over three years post-acquisition. The horizontal dummy equals 1 for a horizontal acquisition, 0 for a non-horizontal acquisition. The relative target size equals the total target value divided by the market value of the acquirer 20 days prior to the announcement date. The relative change in market share is the market share in the quarter before the acquisition subtracted from the market share three calendar quarters after the acquisition, divided by market share of the acquirer before the acquisition. Lnassets is the natural logarithm of assets a quarter before the acquisition. Leverage is the debt a quarter to the acquisition over total assets a quarter to the acquisition. Tobin’s Q is the total market value a quarter before the acquisition over the book value of total assets. Brackets contain the t-values and superscripts ***, ** and * denote significance at 10%, 5% and 1% respectively.
CAR (1) CAR (2) Horizontal dummy
Relative target size Relative change in market share Lnassets Leverage Tobin’s Q -0.0631 (-0.84) -0.0734 (-1.02) -0.0688* (-2.95) -0.0682 (-0.79) -0.1202 (-1.24) -0.0729* (-2.81) -0.0286 (-0.83) 0.1513 (0.55) 0.0559 (1.16) Market share before -0.1461
(-0.71) Constant 0.1215** (2.22) 0.3742 (0.97) # observations R2 0.04 121 0.0801 121
In the second regression, control variables are added. The natural logarithm of assets is added to control for firm size, since acquisitions could have different impact on small firms compared to large firms. To control for the initial position in market share the market share before the acquisition has been added. Leverage and Tobin’s Q are variables that control for other factors that could influence cumulative abnormal returns. Although signs of the variables of interest do not change, there is an increase in the coefficient of the variables. The relative target size to acquirer has a negative influence on cumulative abnormal returns even when adding control variables. A reason for the negative coefficient of relative target size to acquirer might be the fact that larger targets take more time to implement into the existing firm and
15 diseconomies of scale and scope. The relative market share change remains significant at the 1% significance level when controlling for factors mentioned above.
4.3 Value creating variables and horizontal acquisitions
In Table 7, interaction terms are added to find how a relative change in market share and relative target to acquirer size relate to horizontal acquisitions compared to non-horizontal acquisitions. In regression 3, no control variables are added. Both value creating variables are significant at the 1% significance level and have a negative coefficient. When looking at the interaction term for relative target size, a change in sign occurs. This implies horizontal acquisitions benefit from increases in relative target size compared to non-horizontal acquisitions. These benefits could have been achieved through economies of scale, since economies of scale do not quickly arise in non-horizontal acquisitions. The change in sign also occurs at the interaction term for relative market share change. Still, it is very close to zero and the interaction term is insignificant, which implies there is no significant difference between the relative change in market share for horizontal acquisitions compared to non-horizontal acquisitions. This is conflicting with earlier findings, regarding market power as major incentive for horizontal acquisitions, and the market power hypothesis.
Adding control variables has an amplifying effect on relative target size. As can be found in Table 7, regression 4, the effect increases. The variable itself is significant at the 5% significance level, implying an increase in target size relative to the acquirer decreases cumulative abnormal returns. When interacting it with the horizontal dummy, the sign is positive. Even when controlling for firm size, leverage and other factors influencing cumulative abnormal returns, horizontal acquisitions benefit from larger relative targets. Still, since the interaction variable is not significantly different from zero, the relative target size hypothesis cannot be rejected. The relative change in market share however, remains negative when control variables are added. Since it is insignificant and just like regression 3 close to zero, it can be concluded no relation is found between relative change in market share and horizontal acquisitions compared to non-horizontal acquisitions. The market power hypothesis can be rejected, since it is not significantly different from zero.
16
Table 7
This table represents a regression of the variables of interest on firm performance for the entire sample with and without control variables including interaction variables. This table contains the relation between the three year post-acquisition cumulative abnormal returns and separating-, value creating- and control variables. This table also contains interaction variables. The dependant variable is the cumulative abnormal return, measured over three years post-acquisition. The horizontal dummy equals 1 for a horizontal acquisition, 0 for a non-horizontal acquisition. The relative target size equals the total target value divided by the market value of the acquirer 20 days prior to the announcement date. The relative change in market share is the market share in the quarter before the acquisition subtracted from the market share three calendar quarters after the acquisition, divided by the market share prior to the acquisition. Lnassets is the natural logarithm of assets a quarter before the acquisition. Leverage is the debt a quarter to the acquisition over total assets a quarter to the acquisition. Tobin’s Q is the total market value a quarter before the acquisition over the book value of total assets. Brackets contain the t-values and superscripts ***, ** and * denote significance at 10%, 5% and 1% respectively.
CAR (3) CAR (4) Horizontal dummy
Relative target size Relative change in market share -0.1119 (-1.27) -0.1337* (-3.19) -0.0783* (-4.01) -0.1242 (-1.23) -0.1912** (-2.55) -0.0725* (-2.96) Horizontal*Relative target size 0.2663 (1.38) 0.3185 (1.55) Horizontal*relative change in market share 0.0038 (0.11) -0.099 (-0.27) Lnassets `-0.0255 (-0.74) Leverage 0.1681 TobinsQ
Market share before
(0.59) 0.0652 (1.37) -0.1475 (-0.72) Constant # observations R 0.138** (2.42) 121 0.049 0.3464 (0.9) 121 0.0919 5. Conclusion
The effect of acquisitions on firm performance has been studied extensively. As acquisitions differ in size, type and other factors, they affect firm performance in different ways. When evaluating two variables that create value for a firm through an acquisition, and especially in a horizontal acquisition, differences for horizontal and non-horizontal acquisitions have been analysed. As economies of scale and the creation of market power are large incentives for horizontal acquisitions, relative target to acquirer size and relative change in
17 market share are the main variables of interest. In contrast to expectation, these variables had significant negative relations with cumulative abnormal returns. When interacting them with a dummy that distinguishes a horizontal acquisitions from a non-horizontal acquisition, the effect changed. Still, both interaction terms were insignificant, so the hypotheses relative target to acquirer size and relative market share change create value for horizontal acquisitions compared to non-horizontal acquisitions cannot be assumed to be true. To examine this topic further, other measures of market power could be implemented. Also, it would be interesting to examine the effect of why larger relative acquisitions could have negative influence on cumulative abnormal returns, as seen in this research.
18
References
Agrawal, A., Jaffe, J. F., & Mandelker, G. N. (1992). The Post-Merger Performance of Acquiring Firms: A Re-Examination of an Anomaly. The Journal of Finance, 1605-1621.
Barnes, P. (1984). The Effect of a Merger on the Share Price of the Attacker, Revisited.
Accounting & Business Research, 45-49.
Barton, D. M., & Sherman, R. (1984). The Price and Profit Effects of Horizontal Merger: A Case Study. The Journal of Industrial Economics, 165-177.
Bhattacharyya, S., & Amrita, N. (2011). Horizontal acquisitions and buying power: A product market analysis. Journal of Financial Economics, 97-115.
Ferris, S. P., Jayaraman, N., & Sabherwal, S. (2013). CEO Overconfidence and International Merger and Acquisition Activity. Journal of Financial and Quantitative Analysis, 137-164.
Halpern, P. (1983). Corporate Acquisitions: A Theory of Special Cases? A Review of Event Studies Applied to Acquisitions. The Journal of Finance, 297-317.
Healy, P. M., Palepu, K. G., & Ruback, R. S. (1992). Does corporate performance improve after mergers? Journal of Financial Economics, 135-175.
James, C. M., & Wier, P. (1987). Returns to Acquirers and Competition in the Acquisition Market: The Case of Banking. Journal of Political Economy, 355-370.
Loderer, C., & Martin, K. (1992). Postacquisition Performance of Acquiring Firms. Financial
Management, 69-79.
Moeller, S. B., Schlingemann, F. P., & Stulz, R. M. (2004). Firm size and the gains from acquisitions. Journal of Financial Economics, 201-228.
Mueller, D. C. (1985). Mergers and Market Share. Review of Economics Statistics, 259-267. Raghavendra Rau, P., & Vermaelen, T. (1998). Glamour, value and the post-acquisition
performance of acquiring firms. Journal of Financial Economics, 223-253.
Singh, H., & Montgomery, C. A. (1987). Corporate acquisition strategies and economic performance. Strategic management journal, 377-386.
Tremblay, V. J., & Tremblay, C. H. (1988). The Determinants of Horizontal Acquisitions: Evidence from the US Brewing Industry. The Journal of Industrial Economics, 21-45.
