University of Amsterdam BSc Economics & Business
Specialization Finance and Organization
Impact of the Valuation Multiple on the Method
of Payment in M&A Transactions
Evidence from US-bidders in 1990-2015
Author: J.M. Fritz
Student number: 11030763 Thesis supervisor: E. Zhivotova Finish date: January 2018
Abstract
This thesis examines whether the target’s valuation multiple has an effect on the method of payment on announcements made between 1990 and 2015. I expect that a higher value of the target increases the use of cash as the method of payment in a transaction. The univariate and regression analyses show that this expectation is true. In addition, consistent with previous studies, I find that several bidder, target and deal characteristics affect the method of payment. My results are statistically significant and robust to firm- and deal-specific characteristics, and they are not sensitive to the method used to measure the likelihood of the payment choice. The extension to current literature makes it interesting to test other valuation methods in future research.
Statement of Originality:
This document is written by Joey Mitchell Fritz (11030763), who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document are 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.
Table of Contents
1. Introduction
4
1.1 Contributions 5
1.2 Thesis Structure 5
2. Literature Review
6
2.1 Literature on Method of Payment 6
2.2 Literature on Firm Valuation 7
2.3 Formulation of Hypothesis 8
3. Methodology and Data Collection
9
3.1 Research Method 9 3.2 Model 10 3.2.1 Method of Payment 10 3.2.2 Valuation Multiple 10 3.2.3 Bidder Considerations 10 3.2.4 Target Considerations 11 3.2.5 Deal Characteristics 12 3.2.6 Market Conditions 12 3.3 Collection of Data 13
4. Empirical Tests and Results
14
4.1 Univariate Results 14
4.2 Regression Results 17
4.3 Robustness of Results 19
5. Conclusion
22
5.1 Theoretical Contributions 22
5.2 Limitations and Future Research 22
References
23
Appendices
25
Appendix A: Description of Variables 25
Appendix B: Announcements over the Sample Period 26
1. Introduction
Baker McKenzie (2017) predicted that the total value of global M&A transactions will rise to US$3.2 trillion in 2018, catching up to the US$4 trillion peak before the global economic crisis. This important and growing field in corporate finance attracts the attention of financial academics, since there is yet much to discover. Most M&A research has traditionally been focused on market reactions, while the payment decision has received relatively less attention.
The method of payment is important, since it shows how firms prefer to finance major investments (Amihud, Lev, & Travlos, 1990). When structuring the acquisition, the bidder is faced with the choice between stock and cash as the main method of payment. Since the bidder is likely to have insufficient internal funds to finance a cash bid, it is also a choice between debt and equity financing. This can have an impact on more general corporate finance issues, such as the capital structure, corporate governance and agency costs.
Faccio and Masulis (2005) found that the payment method can be strongly influenced by the bidder’s debt capacity and leverage. In addition, it can also be strongly influenced by the desire of the bidder’s shareholders to maintain the existing corporate governance structure because issuing stock can dilute the voting power of the existing shareholders (Martin, 1996). Furthermore, the method of payment can have significant effects on the affected parties’ shareholder wealth. Several studies (Travlos, 1987; Servaes, 1991; Brown & Ryngaert, 1991) found significantly negative average announcements returns to acquirers when the method of payment is stock rather than cash.
In other words, the method of payment remains a continuous point of discussion within bidding firms. Despite numerous studies examining the consequences of choosing a payment method and the determinants of the method of payment, there are still determinants that have not received attention. One important element of an acquisition is the valuation of the target, because it ultimately determines the final purchase price to be paid by the bidding firm. The price, in this way, determines the magnitude of the effects on the issues involved in a transaction.
Valuing a target can be accomplished by using different valuation techniques, of which the use of valuation multiples is most common (Asquith, Mikhail & Au, 2005). By combining the results of previous studies on firm valuation (Giammarino & Heinkel, 1986; Fishman, 1989; Hirshleifer & Png, 1989) and the method of payment (Travlos, 1987; Moeller, 2005), it can be expected that a relation exists between both variables. Therefore, I empirically study the relation between the valuation of the target and the method of payment. The research question of this thesis is as follows:
“What is the effect of target valuation on the method of payment used in M&A transactions?”
1.1 Contributions
In order to gain a deeper understanding of the method of payment, I extend existing literature by adding a new explanatory variable. In general, bidding firms can use both debt and equity to finance an acquisition. However, debt providers will have a limit on the amount of debt that can be provided. On the other hand, existing shareholders do not want their share to be diluted. Since the valuation of the target firm determines what price will be paid, the amount paid also determines the magnitude of these issues. Therefore, it is important that valuation of the target is captured by a new variable.
Besides finding an answer to the research question, this thesis also makes other contributions to current literature. The valuation multiple, which is the most commonly used valuation method, is used as a proxy for target valuation, while there are also other valuation methods. Adding a new explanatory variable could result in testing the relationship of the method of payment and other valuation methods.
In addition, since there has been a massive increase in M&A activity in the recent
decade, most of the studies of the payment choice are based on samples from before this rise. I study a longer period, which allows me to observe whether the effects on the method of
payment examined in earlier studies hold for this extensive sample.
Finally, studies on the method of payment use different proxies and types of regression models to come to their results. Even in recent studies, no clear proxy and model is established. In the main empirical analysis, I use univariate tests and simple logistic regressions which is also used by previous studies (Martin, 1996; Gosh & Ruland, 1998; Boone, Lie, & Lui, 2014). In section 4.3, I compare these results with the results obtained from an ordered logistic
regression and a fractional logistic regression.
1.2 Thesis Structure
In the following section, the theoretical background and hypothesis are provided and developed. The third section of this study discusses the research method and the data
collection method. This is followed by a presentation of the empirical findings and a discussion of the results. The paper ends with a discussion of the contribution to current literature and a description of possible future research areas.
2. Literature Review
In this section, existing scientific literature that guides the empirical analysis is discussed. To begin, several studies on the method of payment are discussed. Second, literature on the valuation of target firms by the acquiring firms is examined. This section ends with the development the hypothesis based on existing literature.
2.1 Literature on Method of Payment
The method of payment generally consists of two major categories: cash payment and stock payment. In a cash payment, the acquirer purchases a certain amount of assets or stocks from the target company by paying cash. It is likely that the acquirer has insufficient cash or cash equivalents to finance the entire purchase price, however, and therefore, it is necessary to attract external funds (Berk & DeMarzo, 2014). This can be accomplished by issuing corporate bonds or increasing the amount of debt, which is constrained by several factors (e.g., sufficient collateral, qualified requirements).
In a stock payment, the acquirer issues new stocks to buy the stocks or assets of the target companies. The most popular form is stock exchange, in which shares of the acquirer are exchanged for stocks or assets of the target company. Issuing new shares could lead to a loss of control for existing shareholders and potential conflicts with new shareholders (Moeller, 2005). This shows that the method of payment is ultimately a tradeoff between debt and equity financing.
After a period of strong interest by financial researchers, the method of payment has historically received considerably less attention relative to other studies on M&As. Over the last few years, the number of publications on the method of payment has increased once again. Most of the early empirical literature on this topic concentrates on the market reactions after the announcement (Travlos, 1987; Servaes, 1991; Brown & Ryngaert, 1991), with the determinants of these financing decisions generally given limited attention.
A prominent empirical study on determinants of the payment method was conducted by Martin (1996). He found that the higher the acquirer's growth opportunities, the more likely the acquirer is to use stock to finance an acquisition. In addition, acquirer managerial ownership is not related to the probability of stock financing over small and large ranges of ownership, but is negatively related over a middle range because this is where they are in a vulnerable control-position. In addition, he also found that stock-financing decreases with an acquirer's higher cash availability. Faccio and Masulis (2005) found results similar to Martin’s (1996) when examining European transactions. In addition, they included three measures of bidder financial strength, which were found to be significant. They also observed that stock financing is less likely for non-public targets, which supports bidder aversion to creating a new blockholder.
There are some studies that examine the effect of the bidder’s valuation on the method of payment. Hansen (1987), for instance, predicted that bidders offer stock when the target has superior information regarding its own value. In case of double-sided information asymmetry, bidders are more likely to offer stock when they are overvalued and cash when they are undervalued. In line with this, Brown and Ryngaert (1991) predicted that bidders who believe their own stock is overvalued will choose stock payment. Chemmanur, Paeglis and Simonyan (2009) found that acquirers choosing a stock offer are overvalued, while those choosing a cash offer are correctly valued. They concluded that private information held by both bidders and targets together determine the medium of exchange.
When looking at recent studies that examine the determinants of the method of
payment, we still did not find any studies addressing target valuation. We see that Boone et al. (2014) studied determinants related to adverse selection, taxation and contracting costs. Huang, Officer and Powell (2016), on the other hand, focused on cross-border deals and found that targets in countries with high governance risk are more likely to receive stock. In addition, Ismail and Krause (2010) found that the fraction of cash offered in a transaction is lower when the premium is higher.
2.2 Literature on Firm Valuation
Before making the announcement to acquire another firm, extensive research is conducted. One of the most important aspects in this process is valuing the target firm. In general, this can be accomplished by using valuation models, which can be categorized into three main categories: accrual-based models, hybrid valuation models and multi-period valuation models. The accrual-based valuation models contain widely used multiples, such as price-earnings ratio and enterprise value-earnings multiple. Hybrid valuation models, on the other hand, are models based on return on equity while multi-period valuation models contain the discounted cash flow model and the Gordon's growth model (Berk & DeMarzo, 2014).
The main message from literature on valuation model choice in practice is that analysts prefer accrual-based models over other valuation models (Barker, 1999; Block, 1999;
Demirakos, Strong & Walker, 2004). Asquith et al. (2005) found that almost 99% of the valuation reports mention that analysts use some sort of accrual-based models. They also found that these multiples are consistent in accuracy. In contrast, only 25% of the reports use hybrid valuation models and 12.8% use multi-period models.
The advantage of using accrual-based models is that such is based on actual prices of real firms, rather than what may be unrealistic forecasts of future cash flows (Berk & DeMarzo, 2014). Another advantage of the use of multiples is that one relies on the market’s assessment of the value of other firms with similar prospects. Block (1999) explains the popularity of
multiples by showing that it is difficult to make multi-period forecasts in an uncertain corporate environment and to estimate the appropriate discount rate for multi-period models.
2.3 Formulation of Hypothesis
Travlos (1987) noted that cash payment transactions are associated with higher acquisition premiums than transactions using other methods of payment. This finding is also supported by Moeller (2005). Several studies (Giammarino & Heinkel, 1986; Fishman, 1989; Hirshleifer & Png, 1989) studied the relation between the bidder’s valuation of the target and the initial acquisition premium offered. They found that an acquirer is willing to pay a premium if he believes that the target is worth it. So, a high bidder valuation is signaled by offering a high initial acquisition premium.
If both cash payment transactions and high target valuations are associated with high acquisition premiums, there might exist a relationship between the method of payment and the valuation of target firms. To be more specific, there might be an increase in the use of cash relative to other methods of payment when the target valuation is higher.
The target valuation is measured using a valuation multiple. This is because most analysts prefer this type of valuation model (see section 2.2). In addition, these multiples are easier to obtain and are less prone to errors when setting expectations for the future (Berk & DeMarzo, 2014). In general, a company with a high multiple is worth more to investors. Therefore, I expect that a higher value of the multiple results in an increase in the use of cash as the method of payment. The following hypothesis is tested:
Hypothesis: A higher valuation of the target has a positive effect on the use of cash as the method of payment.
3. Methodology and Data Collection
To begin, this section describes the chosen research method. Then, all the variables are described and the regression model is introduced. This is followed by a description of the data collection method.
3.1 Research Method
The main objective of this study is to determine whether the valuation multiple has a significant effect on the payment method used in M&A transactions. Since no previous studies on this relationship exist, there is no clear methodology to be used. First, it should be decided which payment methods are included in the study. Hansen (1987), Fishman (1989), Eckbo, Giammarino and Heinkel (1990) and Huang et al. (2016) primarily focused on the choice of stock versus cash. Boone et al. (2014), however, argued that mixed payments differ from cash and stock payments. Because they come with their own diversity, they should be treated as a separate payment category. This view is shared by Amihud et al. (1990), Martin (1996), Faccio and Masulis (2005) which include mixed payments in their studies.
Then, I observe that recent studies on the method of payment (Chemmanur et al., 2009; Huang et al., 2016; Alexandridis, Antypas & Travlos, 2017) first conduct univariate tests before using different types of regression models and proxies for their dependent variable. García-Feijóo, Madura and Ngo (2015) used the proportion of stock used in the acquisition. In contrast, Faccio and Masulis (2005) used the fraction of cash as the dependent variable. They noted that these detailed percentages are often either not specified or suffer from inconsistencies. This is captured in their study by using an ordered logit regression, which allowed them to focus on the qualitative decision to finance with cash, stock or a mixture.
Martin (1996) ran three separate logistic regressions for various subsamples consisting of only cash deals, only stock deals and mixed deals. In each of the three regressions, he used only two subsamples. This allowed him to focus on the broad categories of the method of payment choice, and therefore yields significantly larger samples. Gosh and Ruland (1998) and Boone et al. (2014) effectively used the same method by running multinomial logit regressions on each payment type.
In the empirical analysis, I follow the trend observed in recent studies by first conducting univariate tests on the main variables of interest in section 4.1. Because I am interested in whether a higher valuation multiple increases the use of cash as the method of payment, two regression models are used in section 4.2 to focus on the qualitative decision to finance with either cash instead of stock, or cash instead of a mixture between cash and stock. The first regression has a binary dependent variable, which takes the value of 1 when the considered transaction is only-cash and 0 if it is only-stock. Announcements with a mixed payment are left out of this regression. The second regression includes cash-financed and mixed
announcements. The dependent variable takes the value of 1 when the transaction involves solely cash and 0 if a mixed payment is considered. When testing the robustness of the results in section 4.3, I follow Martin (1996) by using an ordered logistic regression and Faccio and Masulis (2005) by using the fraction of cash as the dependent variable.
3.2 Model
3.2.1 Method of Payment
Following Martin (1996), Faccio and Masulis (2005), cash payments are defined to include cash, non-contingent liabilities and newly issued notes. Stock is defined as shares with full voting or inferior voting rights. Mixed payments are defined as a mixture between cash and stock payments, which fall in the definitions of cash and stock in this study.
3.2.2 Valuation Multiple
(EVEBITDA). The main independent variable of interest is the valuation of the target firm. As noted earlier in section 2.2, the accrual-based models are most frequently used when determining firm value. The most common valuation multiple is the price-earnings ratio, but due to its simplicity, theory predicts that this ratio has some limitations. One of them is that this ratio does not factor in the amount of debt that a company has. Since studies like Kemsley and Nissim (2002) show that having debt creates a tax shield, and therefore influences firm valuation, it is important to incorporate this.
To capture the effect of firms with different levels of leverage, valuation multiples based on the firm’s enterprise value are often used (Berk & DeMarzo, 2014). The enterprise value includes the value of the firm’s assets, debt and equity, and therefore represents the total value of the firm’s underlying business rather than just the value of equity. A commonly used multiple when examining enterprise value is the enterprise multiple which is the firm’s enterprise value divided by the earnings before interest, taxes, depreciation and amortization. The enterprise multiple, therefore, is the proxy that this study uses to capture target valuation.
3.2.3 Bidder Considerations
(COLLATERAL). According to Faccio and Masulis (2005), the method of payment can be constrained by several factors. They found that when bidders have more tangible assets, they are more likely to use cash. In addition, Myers (1977) argued that the cost of debt is usually lower for firms with a high value of tangible assets, because their debtholders are less subject to moral hazard. Since cash-financing is more attractive when the costs of debt are lower, one would expect that a higher value of collateral increases the use of cash.
Karampatsas, Petmezas and Travlos (2014) also included this variable into their study and found that there was an increase in cash-financing, but this was not statistically significant.
(FINLEV). Given that most bidders have limited cash and liquid assets, cash offers generally require debt financing (Faccio & Masulis, 2005). So, a bidder’s currency decision can be strongly influenced by its existing leverage. The predicted sign of this variable is ambiguous as Faccio and Masulis (2005) reported a negative relation between leverage and the use of cash, while Harford, Klasa & Walcott (2009) and Burch, Nanda & Silveri (2012) found a positive relation.
(LNASSETS). Bidder size is likely to influence its financing choices (Faccio & Masulis, 2005). Larger firms are often more diversified, and thus have proportionally lower expected bankruptcy costs. In addition, they are likely to have better access to public debt markets, making them having a higher borrowing capacity. Thus, it is expected that larger firms are more likely to use cash when acquiring another firm. Following Chemmanur et al. (2009), Faccio, Masulis (2005) and Huang et al. (2016), I take the natural logarithm of this variable.
(RELCASH). According to the pecking-order theory, Myers (1984) found that firms prefer to finance investments by internal funds over external borrowing and equity issuance. In line with Martin (1996), the value of the variable is calculated by dividing the amount of cash held by the acquirer before the announcement by the total deal value. It is expected that when the relative cash holdings are higher, this will increase the likelihood of using cash to pay for the transaction.
(MKTBOOK). Jung, Yong-Cheol & Stulz (1996) showed that when the bidder has a higher book ratio, the deal is more likely to be stock-financed. When the market-to-book value is high, investors think that the company has high investment opportunities, so they prefer stock to cash. According to Faccio and Masulis (2005), high market-to-book firms also often have high levels of tax-deductible R&D expenditures and low current earnings. This reduces the need for creating a tax shield by taking on debt. So, high market-to-book companies have less desire to use cash as the method of payment. I expect that a higher market-to-book value increases the chance of using stock as the method of payment.
3.2.4 Target Considerations
(PUBLIC). When the target is not publicly traded, the target’s shareholder’s liquidity needs are also important to consider. Shareholders of unlisted companies might prefer to liquidate their investment by accepting cash as the method of payment (Faccio & Masulis, 2005). In addition, Alexandridis et al. (2017) found that acquirers use significantly less cash-financing in public acquisitions. From following literature, I expect that there is a greater likelihood of cash-financing when the target is not publicly listed.
(RELATED). Shareholders of the target are assumed to be better informed with industry risks and prospects of firms in the same industry than firms from another industry (Faccio & Masulis, 2005; Eckbo et al., 2016). So, uncertainty about bidder equity value and future
bidder’s industry. To proxy this, a dummy variable is introduced that indicates whether both parties are operating in related industries. Since the information asymmetry is reduced when the parties are operating in the same industry, I expect that shareholders of a target that operates in the same industry is more likely to accept stock as the method of payment.
(FOREIGN). Selling stock to foreign investors can entail several problems that could affect the method of payment. First, investors can have a home bias (Chan, Covrig, & Ng, 2004) that makes them prefer cash instead of foreign stock when their company is acquired. Second, a cross-border share exchange exposes the investor to exchange risk and greater trading costs (Huang et al., 2016). This makes them again prefer cash. To proxy this, a dummy variable is introduced that indicates whether both parties are incorporated in the same country. I expect that shareholders of a target that is incorporated in another country are more likely to accept cash as the method of payment.
3.2.5 Deal Characteristics
(RELSIZE). Hansen (1987) found that bidders have greater incentives to use stock as the method of payment when the asymmetric information about the target assets is high. This information asymmetry is likely to rise as target assets rise in value relative to those of a bidder. In line with Martin (1996), Faccio and Masulis (2005), a variable that measures the relative deal value is introduced to capture this effect. According to the previous literature, I expect that a higher value of this variable increases the likelihood of stock financing.
(HOSTILE). In hostile acquisitions, the bidder might want to complete the deal relatively quickly in order to deter competition (Fishman, 1989; Berkovitch & Narayan, 1990). This is because stock payments can lead to substantial delays in the United States, while cash payments are more rapid. This makes cash-financed acquisitions preferable in hostile situations. In addition, Schwerth (2000) found that hostile bidders are less likely to offer only stock. To capture these effects, a dummy variable indicates whether the target company’s management or board of directors perceived the bid as hostile. I expect that an announcement that is perceived as hostile increases the likelihood of using cash as the method of payment.
3.2.6 Market Conditions
(WAVE). The history of M&As has witnessed six merger waves so far, and the relevant merger waves for this study are 1993-1999 (WAVE90) and 2003-2007 (WAVE00). Two theories explain why merger waves occur. First, the neoclassical theory states that merger waves occur when firms in specific industries react to economic shocks (Alexandridis, Mavrovitis & Travlos, 2012). This could explain why merger activity is often clustered by industry (Andrade & Stafford, 2004).
The overvaluation hypothesis, however, states that mergers are the result of temporary misevaluations. When the firm valuations deviate from their fundamentals, managers use overvalued stock to buy assets of undervalued firms (Shleifer & Vishny, 2003; Rhodes-Kropf &
Viswanathan, 2004). Therefore, most of the deals during a merger wave are stock-financed (Harford et al., 2009). In contrast to this finding, García-Feijóo et al. (2015) found that the proportion of stock used to finance decreases during merger waves when they isolate each merger wave. So, it is ambiguous which effect merger waves have on the method of payment.
To account for merger clustering by industry, Alexandridis et al. (2017) are followed by including industry fixed effect based on the Fama and French-12 in all specifications. In addition, there is an alternative specification where fixed year effects are included and replace the two variables that capture the relevant merger waves for this sample.
All of the introduced variables results in the following regression model:
-./0 = %"+ %&343567890+ %:;.<<973=9<0+ %>?6@<340+ %A<@9BB37B0+ %C=3<;9B!0
+ %D-E75..E0+ %F/G5<6;0+ %H=3<97380+ %I?.=36J@0+ %&"=3<B6K30 + %&&!.B76<30+ %&:L943900+ %&>L943000+ N0
3.3 Collection of Data
The sample of M&As was retrieved from the Thomson One Database. This database covers changes in economic ownership of businesses worldwide, from minority stake purchases through 100% takeovers. In order to collect a recent sample of announcements, I decided to use the data collection method employed by Alexandridis et al. (2017) as the main guidance. The initial sample includes completed and withdrawn deals announced between 1990 and 2015 by bidders from the United States with a deal value of at least 5 million in dollar terms.
Following most of the literature (Alexandridis et al., 2017; Huang et al., 2016), minority stake purchases, acquisitions of remaining interest, privatizations, leveraged buyouts, spin-offs, recapitalizations, self-tenders, exchange offers and repurchases are excluded. The restrictions utilized in relation to firm status and the country of incorporation is that acquiring firms are listed in the United States. Regarding the target’s location and status, the only imposed restriction is that both are known. Until this stage, there are 24.595 announcements that satisfy these criteria. This is in line with Alexandridis et al. (2017).
In addition, it is required for each announcement to have full information on the method of payment. This means that only deals to be financed with cash, stock or a mixture between cash and stock are considered. For each deal to be included in the sample, there must also be full information on the main explanatory variable and all other variables. The accounting data was retrieved from Compustat. These additional restrictions result in 2.028 deals satisfying all criteria.
Following Huang et al. (2016) and Alexandridis et al. (2017), both tails of the distribution are winsorized at the 1% level to control for possible outliers in the continuous firm values. This applies for three variables: EVEBITDA, RELCASH and MKTBOOK.
4. Empirical Tests and Results
In this section, I test the hypothesis presented earlier. First, I discuss the summary statistics and conduct univariate tests to determine if there are any differences across the variables between the different methods of payment. Second, I run logit regressions to test the significance of all the introduced variables. Finally, the results are tested for robustness.
4.1 Univariate Results
Table 1 presents the summary statistics for the total sample, the mean for all variables and the median values for the continuous variables under each payment method and the differentials between these methods. The difference in means is tested by using simple t-tests. To test the difference in medians, non-parametric Wilcoxon signed rank tests are used.
On average, a target has a valuation multiple of 15.77. If we categorize this into the method of payment, it is found that the value of the EV/EBITDA is, on average, significantly higher for only-cash announcements. If we look at the median values of the EV/EBITDA under each payment category, we find similar results. This supports the suggestion that there is a relationship between the method of payment and the valuation multiple. To be more specific, it confirms the expectation that in cash-financed announcements, the target has a higher
valuation compared to other payment methods.
If we look at the bidder’s characteristics, we first see that, on average, 30% of the bidder’s value of total assets consists of tangible assets. The average and median value of COLLATERAL is significantly higher for cash-financed announcements compared to stock-financing, which is in line with Myers (1997), Faccio and Masulis (2005). In line with
Karampatsas et al. (2014) is the insignificant difference between cash-financing and a mixture between both methods.
Second, it is found that bidders in cash-financed announcements have a significantly higher percentage of total financial debt. This supports the findings of Harford et al. (2009) and Burch et al. (2012).
Third, it is observed in Table 1 that the average (median) book value of the bidder’s assets is $1.4 (1.6) billion. The value of the assets of bidders that use cash-financing is significantly larger compared to bidders that use stock-financing or a mix between cash and stock. This finding is in support of Faccio and Masulis (2005).
If we look at the cash holdings of the bidders before the announcement, we see that acquirers have, on average, 74% of the total deal value in cash. The median value is lower because it corrects for some announcements in the sample where the acquirer is many times larger than the target, and therefore has more cash than the total deal value. By looking at both mean and medians, it is found, in line with Myers (1984) and Martin (1996), that bidders with
Table 1
Descriptive statistics and univariate tests for the sample of 2028 announcements of acquisitions and mergers between 1990 and 2015.
Total Only Cash Only Stock Mix Only Cash – Only Stock Only Cash – Mix
Variable Mean Median SD Min Max Mean Median Mean Median Mean Median Mean Median Mean Median
EVEBITDA 15.77 11.56 12.12 4.43 53.69 17.09 12.61 14.06 10.86 15.31 11.39 3.03*** (0.698) 1.75*** (0.005) 1.78*** (0.590) 1.22*** (0.009) COLLATERAL 0.30 0.22 0.24 0.00 0.95 0.32 0.27 0.29 0.19 0.30 0.22 0.04*** (0.013) 0.08*** (0.000) 0.02 (0.013) 0.04** (0.039) FINLEV 0.44 0.44 0.17 0.03 1.29 0.46 0.46 0.42 0.41 0.44 0.44 0.04*** (0.009) 0.04*** (0.004) 0.02** (0.009) 0.02** (0.046) LNASSETS 7.29 7.35 1.73 1.65 12.06 7.57 7.67 6.86 6.96 7.26 7.44 0.70*** (0.089) 0.71*** (0.000) 0.30*** (0.094) 0.23*** (0.006) RELCASH 0.74 0.21 1.25 0.01 4.86 0.90 0.39 0.54 0.16 0.69 0.19 0.36*** (0.066) 0.23*** (0.000) 0.21*** (0.058) 0.19*** (0.000) MKTBOOK 3.67 2.74 2.73 0.98 11.43 3.26 2.50 4.39 3.38 3.62 2.60 -1.13*** (0.154) -0.89*** (0.000) -0.35*** (0.138) -0.10** (0.021) PUBLIC 0.86 0.35 0 1 0.83 0.94 0.82 -0.11*** (0.017) 0.01 (0.020) RELATED 0.31 0.46 0 1 0.28 0.38 0.31 -0.10*** (0.024) -0.03 (0.026) FOREIGN 0.13 0.34 0 1 0.21 0.07 0.07 0.15*** (0.017) 0.15*** (0.017) RELSIZE 0.50 0.26 0.78 0.01 4.72 0.39 0.20 0.69 0.58 0.48 0.38 -0.30*** (0.039) -0.39*** (0.000) -0.08* (0.045) -0.18*** (0.008) HOSTILE 0.08 0.26 0 1 0.11 0.04 0.05 0.07*** (0.013) 0.06*** (0.014) WAVE90 0.43 0.49 0 1 0.31 0.61 0.43 -0.31*** (0.025) -0.12*** (0.026) WAVE00 0.19 0.40 0 1 0.27 0.09 0.18 0.17*** (0.019) 0.09*** (0.022) N 2028 927 569 532
Note. Detailed variables descriptions are provided in Appendix A. Only the median values for the continuous variables are provided. The difference in means is tested by t-tests and the standard errors are given in parentheses. The difference in medians are tested by non-parametric Wilcoxon signed-rank tests of which the p-value is given in parentheses. ***, ** and * indicate significance at the 1, 5, and 10% levels, respectively.
Finally, we see that bidders have, on average, a market-to-book ratio of 3.67. This means that the market value of the bidder is 3.67 times higher than the book value. The median and average ratios are highest for only-stock deals. The results of the univariate tests on this variable are significant and negative. This is consistent with Jung et al. (1996), who state that when the bidder has a higher market-to-book ratio, the deal is more likely to be stock-financed.
By looking at the target’s considerations in Table 1, we see that in 86% of all the announcements, the target is publicly listed. If we compare this percentage under the three payment methods, we see that the target is significantly more often listed publicly in stock-financed announcements compared to cash-stock-financed announcements. This in line with Faccio, Masulis (2005) and Alexandridis et al. (2017).
Next, it is observed that in 31% of announcements, both acquirer and target belonged to the same 4-digit SIC code industry. This percentage is significantly higher for stock-financed deals relative to cash. This is because shareholders of the target are assumed to be better informed with industry risks and prospects of firms in the same industry (Faccio & Masulis, 2005; Eckbo et al., 2016). There is no significant difference between this percentage and the use of cash or a mixture between cash and stock.
If we look at the target’s country of incorporation, we see that 13% of all targets are incorporated in a country outside the U.S. If this is specified by payment method, we see that this is 7% for stock-financed and mixed-deals. Cash-financed deals, on the other hand, have a significantly higher percentage of foreign targets. This is in line with the home bias found in previous studies such as Chan et al. (2004).
If we look at the deal characteristics, it is seen than the average (median) relative deal value is 0.50 (0.26), indicating that bidders are, on average, two (to four) times larger than their targets. Bidders using cash-financing are, on average, relatively bigger than the target
compared to the other two payment methods. This is consistent with Martin (1996), Faccio and Masulis (2005).
Moreover, in 8% of the announcements, the target’s management or board of directors perceived the bid as hostile. As expected, this percentage is significantly higher for cash-financed deals relative to mixed and stock-cash-financed deals.
Consistent with Martynova, Renneboog (2008) and García-Feijóo et al. (2015), Appendix B supports the occurrence of two merger waves in the sample period. By examining Table 1, we find that 43% off all the announcements made were between 1993 and 1999. The dominant method of payment in this merger wave was stock. We see that cash-financing was significantly less preferred. In contrast, we see that between 2003 and 2007, cash was the dominant method of payment. This was also observed by García-Feijóo et al. (2015), which found that the
percentage of stock used in a transaction is positively related to WAVE90, and negatively related to WAVE00.
4.2 Regression Results
Running regressions allows me to include fixed industry and year effects compared to the univariate tests. It also provides a more economical interpretation, since you can now mention by how much the chance increases to use a specific payment method compared to another. The results obtained in this sub-section are comparable to those obtained in the previous sub-section.
The main variable of interest is the valuation multiple. It was expected that an increase in the valuation multiple would increase the use of cash-financing. The expectations are confirmed in Table 2 where the coefficients on EVEBITDA in regressions (1) and (3) are highly significant and positive. This result indicates that for target firms with a higher valuation multiple, acquiring firms is more likely to use cash as the method of payment relative to stock-financing. In line with this finding is the sign and significance of the coefficient on LNEBITDA in specification (2). A 1% increase in the EV/EBITDA increases the likelihood of using cash-financing by 1.63%.
The expectations are also confirmed in specifications (4), (5) and (6), where all the coefficients are positive and highly significant. If we control for fixed year effects in specification (6), it is observed that an increase of one of the valuation multiple increases the likelihood of using cash-financing by 1.19% instead of a mixture between cash and stock.
If we look at the bidder’s considerations, we see that, in line with Myers (1997), Faccio and Masulis (2005), COLLATERAL is positive and significant in the first three regressions in Table 2. This means that when bidders have more tangible assets relative to their total assets, they are more likely to use cash instead of stock. In line with Karampatsas et al. (2014), the coefficient in the remaining three specifications is positive, but insignificant which was also found in section 4.1.
According to literature, the financial leverage of the bidder has an ambiguous effect on the method of payment. Faccio and Masulis (2005) found a negative relation between leverage and cash-financing. In contrast, and in line with Hartford et al. (2009) and Burch et al. (2012), a significant and positive relation is found in all specifications. This means that bidders with higher levels of leverage are more likely to use cash relative stock or a mixture between cash and stock. It is seen that the coefficients in the first three specifications are almost twice as large as in the other three specifications.
In addition, the coefficients on LNASSETS are positive and highly significant in all
specifications. This supports the finding of current literature that larger firms have, on average, a higher borrowing capacity, which enables them to use cash-financing more frequently (Faccio & Masulis (2005). For specification (1), this means that a 1% increase in the total assets of an acquirer increases the likelihood of using cash-financing relative to stock-financing by 1.31%.
The positive and highly significant coefficients on RELCASH in all specifications confirm the findings of Myers (1984) and Martin (1996), which supports the pecking-order theory,
Table 2
Logistic regression results on the method of payment.
Stock = 0 Cash = 1 Mixed = 0 Cash = 1 Variable (1) (2) (3) (4) (5) (6) EVEBITDA 0.121*** (0.049) 0.149*** (0.052) 0.146*** (0.055) 0.166*** (0.059) LNEVEBITDA 0.488*** (0.203) 0.532*** (0.216) COLLATERAL 0.638** (0.324) 0.661** (0.326) 0.558* (0.336) 0.440 (0.305) 0.447 (0.307) 0.399 (0.323) FINLEV 1.503*** (0.404) 1.491*** (0.404) 1.436*** (0.427) 0.809*** (0.413) 0.804** (0.413) 0.726* (0.422) LNASSETS 0.267*** (0.043) 0.273*** (0.043) 0.211*** (0.046) 0.130*** (0.042) 0.130*** (0.042) 0.128*** (0.044) RELCASH 0.172*** (0.058) 0.170*** (0.058) 0.190*** (0.061) 0.486*** (0.087) 0.487*** (0.087) 0.498*** (0.093) MKTBOOK -0.158*** (0.023) -0.155*** (0.024) -0.167*** (0.026) -0.067** (0.027) -0.066** (0.027) -0.076*** (0.028) PUBLIC -0.938*** (0.204) -0.944*** (0.204) -0.867*** (0.224) -0.326* (0.174) -0.332* (0.174) -0.305* (0.184) RELATED -0.628*** (0.147) -0.632*** (0.147) -0.664*** (0.155) -0.174 (0.151) -0.166 (0.151) -0.122 (0.157) FOREIGN 0.948*** (0.207) 0.944*** (0.207) 0.862*** (0.214) 0.817*** (0.215) 0.818*** (0.215) 0.820*** (0.219) RELSIZE -0.274** (0.107) -0.274** (0.107) -0.312** (0.144) -0.287** (0.134) -0.287** (0.133) -0.299** (0.135) HOSTILE 1.464*** (0.293) 1.472*** (0.294) 1.543*** (0.300) 1.223*** (0.254) 1.217*** (0.254) 1.183*** (0.261) WAVE90 -0.699*** (0.143) -0.694*** (0.143) -0.164 (0.148) -0.164 (0.148) WAVE00 0.717*** (0.198) 0.722*** (0.198) 0.399** (0.175) 0.402** (0.175)
Fixed Year Effects NO NO YES NO NO YES
Fixed Industry Effects YES YES YES YES YES YES
N 1496 1496 1496 1459 1459 1459
Pseudo R2 21.73% 21.74% 26.45% 16.81% 16.78% 19.16%
Note. Detailed variables descriptions are provided in Appendix A. In specification (1), (2) and (3) MOP equals 0 for stock-financing and 1 for cash-financing. In specification (4), (5) and (6) MOP equals 0 for a mixture between cash- and stock-financing and 1 for cash-financing. Robust standard errors are in parentheses. EVEBITDA and LNEBITDA are tested one-sidedly; all other coefficients are tested two-sidedly. ***, ** and * indicate significance at the 1, 5, and 10% levels, respectively. In all regression specifications fixed industry effects are included.
2, it is seen that most of the coefficients on the market-to-book ratios are negative and significant at the 1% level. This is in line with Jung et al. (1996), Faccio and Masulis (2005).
When looking at the target’s considerations, it is observed that when the target is not publicly listed, their shareholders might be more willing to accept cash instead of stock as the method of payment since this liquidates their investment. This argumentation is supported in Table 2, where the coefficients on PUBLIC are negative and significant at the 1% level. This finding is also in support of Faccio, Masulis (2005) and Alexandridis et al. (2017). The effect is only significant at the 10% level for the coefficients in the last three specifications. The
liquidation might be sufficient for them to invest in other opportunities, while a payment in only stock does not allow for this.
In support of existing literature (Faccio & Masulis, 2005; Eckbo et al., 1990), shareholders of a target that operates in the same industry are significantly more likely to receive a stock-financed offer. This is because the target’s shareholders have less information asymmetry than otherwise. In regressions (4), (5) and (6), there is a preference for mixed-financed deals, but this effect is insignificant. This insignificance was also found in the previous section.
Again in support of previous literature and the findings in section 4.1, when a target is incorporated in another country, it significantly increases the likelihood of being cash-financed. As stipulated in studies (Chan et al., 2004), investors might have a home bias and there might be difficulties with a cross-border share exchange. Therefore, they have a strong preference for only-cash deals relative to deals where some stock is involved.
RELSIZE is negative and significant in all specifications. If the target is a relatively larger addition to the value of the acquirer, the probability of choosing stock or a portion of stock is significantly higher. This is in line with the empirical findings of Martin (1996), Faccio and Masulis (2005). Further, this finding supports the information asymmetry prediction by Hansen (1987), stating that a relatively bigger target motivates the need for stock financing to mitigate the problems of information asymmetry on target valuation.
In addition, the variable HOSTILE is positive and highly significant in all specifications. This means that an announcement that is perceived as hostile increases the chance of using cash-only as the method of payment. This is because stock generally causes delays, and cash therefore allows the parties to consummate the deal quickly (Fishman, 1989; Berkovitch & Narayan, 1990). For specification (3), for example, a deal perceived as hostile increases the likelihood of cash-financing with 4.69% relative to stock-financing.
It is ambiguous which effect merger waves had on the method of payment. From
regressions (1) and (2), it is seen that during the merger wave from 1993 until 1999, there was a significantly higher probability of stock-financing. During the second relevant merger wave (2003–2008), however, it was more likely to use cash. Both findings are consistent with García-Feijóo et al. (2015) and section 4.1. We see that when including fixed year effects in
specifications (3) and (6), most of the coefficients increase in size. Despite the increase in robust standard errors, this increases the significance of many coefficients.
4.3 Robustness of Results
In the previous analysis, evidence indicates that the valuation multiple has a significant effect on the method of payment. More specifically, a higher EVEBITDA increases the likelihood of using cash as the method of payment. In this section, several robustness tests are conducted to ensure that the results are not sensitive to sample selection and model specification issues.
The robustness is assessed by using two other regression techniques, considering only
completed deals, introducing the PE-multiple as an alternative proxy for the enterprise multiple, and introducing additional variables. The results reported in Appendix C show that most of the variables remain statistically significant, indicating that the findings are robust.
Following Martin (1996), an ordered logistic regression is used in specification (1), where the dependent variable is 0 if the acquisition is stock-financed, 1 if mixed-financed and 2 if cash-financed. In contrast to the simple logistic regressions, this allows me to use all payment
categories in one specification. Compared to the results of specification (1) in Table 2, most of the coefficients follow the same direction and have the same significance levels. The only exception is that the coefficient on COLLATERAL has a lower significance level. This is in line with the findings in section 4.2, where it was found that there is only a significant increase in the likelihood of using cash-financing relative to stock-financing when the value of collateral is high.
In the previous sections, broad methods of payment categories were used as opposed to the more detailed percentages used in other empirical studies on the method of payment (Faccio & Masulis, 2005; Harford et al., 2009; Ismail & Krause, 2010; Eckbo et al., 2016). The dependent variable of specification 2 in Appendix C represents the fraction of cash used in the transaction. Because this variable must be in the interval [0, 100], a fractional logistic regression is used. Again, this allows me to use the entire sample instead of only two subsamples. When comparing the results to the base specification, some variables differ in size, but there are no notable differences in sign and coefficients.
Following Huang, Officer and Powell (2016), in specification (3), only completed deals are considered. Again, this does not result in any significant changes compared to the base specification.
In specification (4), the price-earnings multiple is used as an alternative proxy for the enterprise multiple. Since it is only possible to obtain the share price for listed and some
subsidiary targets, the number of observations decreased to 931. In line with results in the base specification, the coefficient on PE is highly significant and positive. This means that for target firms with a higher PE-ratio, acquiring firms are more likely to use cash as the method of payment relative to stock-financing. All other variables have the same sign and level of significance compared to the base specification except the coefficient on PUBLIC. This is because there were only five non-public firms in this sample due to the collection of the PE-ratio.
As mentioned earlier, the overvaluation theory states that more acquisitions will occur in periods of bubbles. When firm valuations deviate from fundamentals, acquirers use their
overvalued stock to buy assets of less overvalued firms (Shleifer & Vishny, 2003; Rhodes-Kropf & Viswanathan, 2004; Chemmanur, Paeglis, & Simonyan, 2009). Since merger activity is correlated with stock market performance, the adjusted closing prices of the S&P500 at the announcement date are used to capture this effect. In specification (5), I observe that the
coefficient on SP500 is negative and highly significant, which confirms the overvaluation theory. Again, there are no notable differences for the other variables compared to the base
specification.
In line with Ismail and Krause (2015), the variable RATE is introduced. The aim is to control for times of high interest rates that might increase the costs of using debt to finance acquisitions, and therefore influence the method of payment. The variable is the 10-year United States government bond-yield at the announcement date. I observe that this variable is negative and highly significant. This means that a higher interest rate increases the likelihood of stock-financing. This finding is consistent with Ismail and Krause (2015). Again, the introduction of this variable does not result in any significant changes compared to the base specification.
5. Conclusion
5.1 Theoretical Contributions
This study seeks to answer the following question: “What is the effect of target valuation on the method of payment used in M&A transactions?” To this end, an empirical study was conducted that examined take-over announcements made by United States bidders between 1990 and 2015. From the literature, it was expected that a higher valuation of the target increases the use of cash as the method of payment. To test the expectations, univariate tests and regression techniques were used.
From the first set of tests, it was concluded that the value of the EV/EBITDA is significantly higher for only-cash transactions compared to other payment categories. The regressions results also confirmed that a higher valuation of the target increases the likelihood of using cash-financing instead of stock-financing or a mixture. This means that the null
hypothesis is rejected by two different econometric techniques. Hence, the valuation of a target has an impact on the method of payment as predicted by literature.
In addition, after an increase in M&A activity in the recent decade, it was uncertain if these effects would still hold. This study provides evidence that all examined bidder, target and deal characteristics influence the method of payment. This study also showed that two merger waves occurred during the sample period.
Finally, in studies on the method of payment, there is no clear proxy and model established. I have showed by using univariate tests and three different types of regression models that the obtained results are robust. Moreover, by using the PE-multiple as a proxy for the target valuation, it was again shown that there is a higher probability of using cash-financing when the valuation of the target is higher.
5.2 Limitations and Future Research
At first, this study relied solely on announcements made by United States bidders. It is unclear whether the obtained results are generalizable and would remain the same with bidders from other countries. For example, in Europe, the valuation of targets might be conducted in another way. Also, access to debt markets may vary between the United States and Europe.
As there were no previous studies on the effect of the target’s valuation on the method of payment, it would be interesting to examine other proxies for the valuation of a target. It could be that there also exists a connection between hybrid or multi-period valuation models, for example. In addition, the enterprise multiple could be specified in another way. The multiples used in this study relied on past earnings, while most investors are looking for attractive opportunities in the future.
References
Alexandridis, G., Antypas, N., & Travlos, N. G. (2017). Value creation from M&As: New evidence. The Journal
of Corporate Finance, 45, 632–650.
Alexandridis, G., Mavrovitis, C. F., & Travlos, N. G. (2012). How have M&As changed? Evidence from the sixth merger wave. The European Journal of Finance, 18(8), 663–688.
Amihud, Y., Lev, B., & Travlos, N. G. (1990). Corporate Control and the Choice of Investment Financing: The Case of Corporate Acquisitions. The Journal of Finance, 45(2), 603–616.
Andrade, G., & Stafford, E. (2004). Investigating the economic role of mergers. Journal of Corporate Finance,
10(1), 1–36.
Asquith, P., Mikhail, M., & Au, A. (2005). Information content of equity analyst reports. Journal of Financial
Economics, 75(2), 245–270.
Baker McKenzie (2017). Global Transactions Forecast 2018: Deal Appetite Rising. Retrieved from: http://insight.bakermckenzie.com/forecast-overview
Barker, R. G. (1999). The role of dividends in valuation models used by analysts and fund managers. The
European Accounting Review, 8(2), 195–218.
Berk, J. B. & DeMarzo, P. M. (2014). Valuing Projects and Firms. In D. Battista (Ed.), Corporate Finance (pp. 206–300). Essex: Pearson Education.
Berkovitch, E., & Narayanan, M. P. (1990). Competition and the Medium of Exchange in Takeovers. The
Review of Financial Studies, 3(2), 153–174.
Block, S. B. (1999). A study of financial analysts: Practice and theory. Financial Analysts Journal, 55(4), 86–95. Boone, A. L., Lie, E., & Liu, Y. (2014). Time trends and determinants of the method of payment in M&As. The
Journal of Corporate Finance, 27, 296–304.
Brown, D. T., & Ryngaert, M. D. The Mode of Acquisition in Takeovers: Taxes and Asymmetric Information.
The Journal of Finance (46)2, 653–669.
Burch, T. R., Nanda, V., & Silveri, S. (2012). Taking stock or cashing in? Shareholder style preferences, premiums and the method of payment. Journal of Empirical Finance, 19(4), 558–582.
Chan, K., Covrig, V., & Ng, L. (2005). What Determines the Domestic Bias and Foreign Bias? Evidence from Mutual Fund Equity Allocations Worldwide. The Journal of Finance, 60(3), 1495–1534.
Chemmanur, T. J., Paeglis, I., & Simonyan, K. (2009). The medium of exchange in acquisitions: Does the private information of both acquirer and target matter? The Journal of Corporate Finance, 15(5), 523–542.
Demirakos, E. G., Strong, N. C., & Walker, M. (2004). What valuation models do analysts use? Accounting
Horizons, 18(4), 221–240.
Eckbo, B. E., Giammarino, R. M., & Heinkel, R. L. (1990). Asymmetric Information and the Medium of Exchange in Takeovers: Theory and Tests. The Review of Financial Studies, 3(4), 651–675. Faccio, M., & Masulis, R. W. (2005). The Choice of Payment in European Mergers and Acquisitions. The
Journal of Finance, 60(3), 1345–1388.
Fishman, M. J. (1989). Preemptive Bidding and the Role of the Medium of Exchange in Acquisitions. The
Journal of Finance, (44)1, 41–57.
García-Feijóo, L., Madura, J., & Ngo, T. (2012). Impact of industry characteristics on the method of payment in mergers. The Journal of Economics and Business, 64, 261–274.
Ghosh, A., & Ruland, W. (1998). Managerial Ownership, the Method of Payment for Acquisitions, and Executive Job Retention. The Journal of Finance, 53(2), 785–798.
Giammarino, R. M., & Heinkel, R. L. (1986). A Model of Dynamic Takeover Behavior. The Journal of Finance,
41(2), 465–480.
Hansen, R. G. (1987). A theory for the choice of exchange medium in mergers and acquisitions. The Journal of
Business, (60)1, 75–95.
Harford, J., Klasa, S., & Walcott, N. (2009). Do Firms have Leverage Targets? Evidence from Acquisitions.
Journal of Financial Economics, 93(1), 1–14.
Hirshleifer, D. A., & Png, I. P. (1989). Facilitation of Competing Bids and the Price of a Takeover Target. The
Review of Financial Studies, 2(4), 587–606.
Hovakimian, A., Opler, T., & Titman, S. (2001). The Debt-Equity Choice. Journal of Financial and Quantitative
Analysis, (36)1, 1–24.
Huang, P., Officer, M. S., & Powell, R. (2016). Method of payment and risk mitigation in cross-border mergers and acquisitions. The Journal of Corporate Finance, 40, 216–234.
Ismail, A., & Krause, A. (2010). Determinants of the method of payment in mergers and acquisitions. The
Quarterly Review of Economics and Finance 50(4), 471–484.
Jung, K., Yong-Cheol, K., & Stulz, R. (1996). Timing, investment opportunities, managerial discretion, and the security issue decision. Journal of Financial Economics (42)2, 159–185.
Karampatsas, N., Petmezas, D., & Travlos, N. G. (2014). Credit ratings and the choice of payment method in mergers and acquisitions. The Journal of Corporate Finance, 25, 474–493.
Kemsley, D., & Nissim, D. (2002). Valuation of the Debt Tax Shield. The Journal of Finance, 57(5), 2045–2073. Martin, K. J. (1996). The Method of Payment in Corporate Acquisitions, Investment Opportunities and
Management Ownership. The Journal of Finance, 51(4), 1227–1246.
Martynova, M., & Renneboog, L. (2008). A century of corporate takeovers: What have we learned and where do we stand? Journal of Banking & Finance, 32(10), 2148–2177.
Moeller, T. (2005). Let’s make a deal! How shareholder control impacts merger payoffs. Journal of Financial
Economics, 76(1), 167–190.
Moeller, S. B., Schlingemann, F. P., & Stulz, R. M. (2004). Firm size and the gains from acquisitions. The
Journal of Financial Economics, 73(2), 201–228.
Myers, S. C. (1977). Determinants of Corporate Borrowing. Journal of Financial Economics, 5(2), 147–175. Myers, S. C. (1984). The Capital Structure Puzzle. The Journal of Finance, 39(3), 574–592.
Officer, M. S. (2013). Termination fees in mergers and acquisitions. The Journal of Financial Economics, 69(3), 431–467.
Rappaport, A., & Sirower, M. L. Stock or Cash? Harvard Business Review, 77(6), 147–158.
Rhodes-Kropf, M., & Viswanathan, S. (2004). Market Valuation and Merger Waves. The Journal of Finance,
59(6), 2685–2718.
Schwert, G. W. (2000). Hostility in Takeovers: In the Eyes of the Beholder? The Journal of Finance, 55(6), 2599–2640.
Shleifer, A., & Vishny, R. W. (2003). Stock market driven acquisitions. Journal of Financial Economics, 70(3), 295–311.
Servaes, H. (1991). Tobin’s Q and the Gains from Takeovers. The Journal of Finance, 46(1), 409–419. Travlos, N. G. (1987). Corporate Takeover Bids, Method of Payment, and Bidding Firms’ Stock Returns. The
Appendices
Appendix A: Description of Variables
Variable Definition
ASSETS Sum of the bidder’s total current assets, long-term receivables, investment in
unconsolidated subsidiaries, other investments, net property plant and equipment, and other assets, prior to deal announcement. Source: Compustat.
RELCASH Ratio of the amount of cash held by the bidder before the announcement of the total
deal value. Source: Thomson One & Compustat.
COLLATERAL Ratio of the bidder’s property, plant and equipment (PPE) to the book value of total assets in the fiscal year prior to the acquisition announcement. Source: Compustat.
EVEBITDA Ratio of the target’s enterprise value to the EBITDA. The enterprise value is
calculated by multiplying the number of actual target shares outstanding from its most recent balance sheet by the offer price and then adding the cost to acquire
convertible securities, plus short-term debt, straight debt and preferred equity minus cash and marketable securities. The EBITDA is equal to the earnings before interest, taxes, depreciation and amortization for the most recent fiscal year prior to the announcement of the transaction. Source: Thomson One & Compustat.
FOREIGN Indicator variable that equals one in deals in which the target firm is incorporated in
the United States. Source: Thomson One.
FINLEV Ratio of the bidder’s total financial debt (long-term debt plus debt in current liabilities)
to the book value of total assets in the fiscal year prior to the acquisition announcement. Source: Compustat.
HOSTILE Indicator variable that equals one for deals defined as “hostile” or “unsolicited”, and
which equals 0 otherwise. Source: Thomson One.
MKTBOOK Ratio of the bidder’s market value of equity plus book value of debt over the sum of
book value of equity plus book value of debt, prior to deal announcement. Source: Thomson One & Compustat.
PUBLIC Indicator variable that equals one in deals in which the target firm is a publicly traded
firm. Source: Thomson One.
RATE The yield of 10-year US government bonds on the announcement date. Source:
CRSP.
RELATED Indicator variable that equals one if the bidder’s and target’s primary 4-digit SIC code
coincides, and which equals 0 otherwise. Source: Thomson One.
RELSIZE Ratio of the price offered in the announcement (excluding assumed liabilities) over
the bidder’s market capitalization as of four weeks prior to deal announcement. Source: Thomson One & Compustat.
SP500 The adjusted closing price of the S&P-500 on the announcement date. Source:
CRSP.
WAVE90 Indicator variable that equals one for deals in the period from 1993 to 1999.
Appendix B: Announcements over the Sample Period
Number of announcements by year and the method of payment, 1990–2015.
0 25 50 75 100 125 150 175 200 225 Nu mb er o f An n o u n ce me n ts
Appendix C: Robustness Tests
Logistic regression results on the method of payment.
(1) (2) (3) (4) (5) (6) EVEBITDA 0.116*** (0.040) 0.112*** (0.037) 0.135*** (0.052) 0.132*** (0.051) 0.129*** (0.050) PE 0.479*** (0.031) COLLATERAL 0.382* (0.219) 0.597*** (0.223) 0.718** (0.347) 0.691* (0.408) 0.651** (0.327) 0.661** (0.330) FINLEV 1.163*** (0.296) 0.652** (0.305) 1.539*** (0.440) 2.150*** (0.501) 1.478*** (0.410) 1.406*** (0.407) LNASSETS 0.171*** (0.028) 0.121*** (0.029) 0.241*** (0.047) 0.212*** (0.055) 0.228*** (0.044) 0.238*** (0.045) RELCASH 0.259*** (0.052) 0.241*** (0.058) 0.153*** (0.060) 0.295*** (0.086) 0.178*** (0.059) 0.160*** (0.059) MKTBOOK -0.123*** (0.017) -0.109*** (0.019) -0.151*** (0.025) -0.206*** (0.030) -0.167*** (0.024) -0.150*** (0.023) PUBLIC -0.613*** (0.126) -0.534*** (0.134) -0.845*** (0.212) -0.423 (1.323) -0.933*** (0.204) -0.867*** (0.203) RELATED -0.183* (0.098) -0.198** (0.100) -0.658*** (0.159) -0.601*** (0.191) -0.624*** (0.148) -0.634*** (0.148) FOREIGN 0.863*** (0.164) 0.864*** (0.144) 0.995*** (0.231) 0.737*** (0.284) 0.871*** (0.210) 0.900*** (0.208) RELSIZE -0.151** (0.062) -0.345*** (0.085) -0.263** (0.111) -0.277** (0.126) -0.306** (0.125) -0.295** (0.117) HOSTILE 1.165*** (0.210) 1.450*** (0.200) 1.876*** (0.591) 1.239*** (0.325) 1.526*** (0.292) 1.488*** (0.293) WAVE90 -0.475*** (0.105) -0.618*** (0.107) -0.802*** (0.154) -0.535*** (0.184) -0.261 (0.167) -0.313* (0.169) WAVE00 0.546*** (0.130) 0.560*** (0.141) 0.791*** (0.217) 0.695*** (0.244) 0.791*** (0.197) 0.788*** (0.200) SP500 -0.001*** (0.000) RATE -0.227*** (0.055)
Fixed Industry Effects YES YES YES YES YES YES
N 2028 1921 1290 931 1496 1496
Pseudo R2 15.25% 16.37% 21.65% 21.66% 23.20% 22.60%
Note. Detailed variable descriptions are provided in Appendix A. In specification (1) MOP equals 0 for stock-financing, 1 for a mixture between cash- and stock-stock-financing, 2 for cash-financing. In specification (2) MOP equals the fraction of cash in consideration relative to the sum of all payment methods. In specification (3), (4), (5) and (6) MOP equals 0 for stock-financing and 1 for cash-financing. Robust standard errors are in parentheses. EVEBITDA and PE are tested one-sidedly; all other coefficients are tested two-sidedly. ***, ** and * indicate significance at the 1, 5 and 10% levels, respectively.
