The effect of executive equity-based compensation on the
payment method in acquisitions: a special case for cash
acquisitions.
Amsterdam Business School
Name Catriona de Goede
Number 10280375
MSc in Business Economics Specialization Finance
Supervisor Vladimir Vladimirov Completion 15 August 2016
Abstract
This paper investigates the effect of equity-based compensation of bidder executives on the choice of payment method in corporate acquisitions. The payment method could be cash, stock or a mixture of cash and stock. Additionally, this study also examines whether equity-based compensation also influences the chosen source of funding in case of only cash financed acquisitions. These types of funding could be divided into debt, non-debt and corporate funds. The equity-based compensation is used because it is a good sensitivity measure, it measures the alignment between bidders executives and shareholders. The results show that equity-based compensation is significantly positive related to stock financing. This means that higher EBC makes stock financing more likely. There is no significant relation found between the EBC and the source of funding of cash acquisitions. This means that the chosen source of funding is not influenced by the EBC.
2
Statement of Originality
This document is written by Catriona de Goede 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
Table of Content
1. Introduction 4
2. Literature Review and Hypotheses 6
2.1 Equity-based executive compensation 6
2.2 Financial Strength and Cash Source of Funds 7
2.3 Executive ownership 9
2.4 Investment opportunities 9
2.5 Information asymmetry 10
2.6 Unlisted & Subsidiary targets 11
2.7 Deal Mode 12
2.8 Hypotheses 12
3. Data and descriptive statistics 14
3.1 Data 14
3.2 Descriptive statistics 16
4. Methodology 20
4.1 Ordered probit regression model 20
4.2 Probit regression model 22
5. Results 23
5.1 Ordered probit regression results 23
5.2 Probit regression results 27
6. Conclusion and discussion 29
Appendix A – Variable definitions 31
References 32
4
1. Introduction
Corporate takeovers are large external investments, they bring some major changes to the whole organization and if completed acquisitions turned out to be successful, this will increase economic activity. Higher economic activity and successful acquisitions will create more value for the society. It is therefore important to make the right payment method choice and compensation structure for bidder executives to make the completed acquisition
successful. Previous studies show that the executive compensation structure in acquisitions can influence the post-acquisition performance of the acquiring firm. An optimal
compensation structure presumes that a strong pay-performance sensitivity helps to promote better acquisition decisions (Murphy (1999) and Aggarwal (2008)). A good sensitivity measure could be the equity-based compensation (EBC) part of bidder executives introduces by Datta, Iskandar-Datta, and Raman (2001), because this part of the executive compensation is positively influenced by firm performance. Most acquisitions were paid by stock financing, cash financing or a mixture of cash and stock. Prior literature shows that there are several aspects that influence the choice of payment method of these acquisition, for example, the investment opportunities, asymmetric information, capital structure and control position, and behavioral arguments (Betton, Eckbo and Thornburn, 2008). In the special case of cash acquisitions, the source of funding also depends on bidder’s characteristics. The source of funding can be divided in debt-like consideration, non-debt-like considerations and internal corporate funds. For example, Schlingemann (2004) reports different relations between bidder gains and the source of funds in cash acquisitions. While prior studies focus on the main determinants of payment method and the effect of executive compensation on the firm performance after the acquisition, this article examines the effect of the sensitivity measure, the EBC, on the payment method choice. Does the equity-based compensation of bidder executives influence the payment method choice? And more specific, does this equity-based compensation also influence the source of funding in cash acquisitions?
Both stock and cash financing have advantages and disadvantages. From the bidder’s perspective, the financing decision can depend strongly on its debt capacity and existing leverage and their maintenance of the existing governance structure like the ownership of the main shareholders. For example, Martin (1996) finds that bidders with higher investment opportunities use most of the time pure stock as payment method. Amihud, Lev, and Travlos (1990) and Faccio and Masulis (2005) both show that when the managerial ownership is at
5 intermediate level, the probability of using cash as final consideration is higher. Bidder
executives can protect their ownership by using cash as their main financing method. Looking more closely to the cash acquisitions is especially interesting, because there are several possibilities in how to fund these acquisitions and it turns out that the source of funding of cash acquisition influence bidder gains (Schlingemann, 2004). Vladimirov (2015) reports that if bidders can finance their cash bids in a competitive market for capital, they will issue debt. This is the cheapest form of financing from a bidder’s perspective. It turns out that bidders will choose only equity financing if they have difficulties with raising external
financing or when they are cash constrained (Vladimirov, 2015). On the other hand, Schlingemann (2004) finds higher bidder gains if they used internal funds to finance their cash acquisitions. He did not find any effect on debt financing on bidder gains. The three different sources of funding can be classified as either external or internal funding. Debt and external equity issuance are considered as external funding as the use of internal corporate funds is identified as internal funding. External funding usually creates more firm value than the use of internal funding. Therefore, it is expected that high EBC bidders are more likely to use external financing rather than internal financing.
Several recent papers find evidence for the link between executive compensation, and both completing the acquisition and the post-acquisition firm performance. Bliss and Rosen (2001) and Harford and Li (2007) find a positive relation between executive compensation and completing an acquisition. Harford and Li (2007) report that even in acquisitions where bidding shareholders are worse off, the executives are better off three quarters of the time. Even if the bidder is performing poorly after the completion of the acquisition, executives’ compensation increases. This could imply that executives can use acquisitions to manipulate their compensation. Higher compensation could give executives incentives to complete acquisitions even if this will have a deteriorating effect on bidding shareholders’ wealth. Lehn and Zhao (2006) conclude that the corporate governance and the external market for corporate control generally work well in disciplining managers who pursue acquisitions to the detriment of their stockholders. Huson, Parrino, and Starks (2001) also suggest that the effectiveness of internal monitoring mechanisms is not dependent on the intensity of the takeover market. Taking these articles into account, it would be expected that high EBC bidders would use stock financing to show that completing the acquisition will create value for bidder shareholders.
The results of this study indicate that stock financing in a corporate takeover becomes more likely when the equity-based compensation of bidder executives increases. This results
6 control for several factors, which are included as additional explanatory variables, and
appear to influence the payment method decision for acquisitions. Two of these explanatory variables, for example, are the financial leverage and Tobin’s q. Financial leverage measures bidder’s financial strength, and Tobin’s q is a measure of bidder’s investment opportunities. Both are measured prior to the acquisition announcement. An ordered probit regression analysis is used to order the payment method as dependent variable. For testing the influence of equity-based compensation on the source of funding in cash acquisitions, there are no significant results found for the main independent variable, the EBC. This means that there is no relation between the equity-based compensation and the source of funding choice in cash acquisitions.
In the next section earlier research concerning the payment method determinants, the sources of cash financing and the equity-based compensation will be reviewed. This section also covers the hypotheses that will be tested. Section three describe the sample selection and descriptive statistics. Section four and five describe the methodology and results,
respectively. The final section provides a short summary of the empirical findings and concludes.
2. Literature Review and Hypotheses
This section will discuss existing relevant literature on the executive compensation and the choice of payment method in acquisitions. First, the equity-based compensation as part of bidder executives’ compensation is discussed. Secondly, the existing literature about the determinants of the choice of payment method in acquisitions is addressed. Finally, the results of previous literature are translated into two different hypotheses to answer the main research question and the sub-question about cash acquisitions.
2.1 Equity-based executive compensation
Executive compensation often consists of a fixed and a variable component, in which the variable share is usually equity-based. The equity-based compensation (EBC), following the paper of Datta et al. (2001), is the sum of the value of stock options granted to the executives as percentage of total compensation paid to them one year prior to the announcement date. Datta et al. (2001) find evidence that the EBC of executives influences the target selection. For example, they report that high equity-based executives tend to seek targets with high growth opportunities and low equity-based executives tend to acquire targets with low growth
7 opportunities. Additionally, their findings indicate that the stock price reaction to the
announcements for high EBC bidders is significantly positive and significantly negative for low EBC bidders. The reason behind this stock price reaction to the announcement is found by Guay (1999). His paper finds that stock options significantly increase the sensitivity of executives’ wealth to equity risk. Other shareholders will interpret the acquisition
announcement of high EBC bidder as a signal that this external investment will create value for them.
Both Guay (1999) and Datta et al. (2001) found that the sensitivity of the EBC incentivizes executives to invest in risky and more valuable projects when the potential loss from underinvestment in valuable risk-increasing projects is greatest. On the other hand, Bliss and Rosen (2001) find that CEOs in the banking industry with more EBC are less likely to make an acquisition, suggesting that the executives are motivated by the form of their compensation contracts. This could mean that executives in the banking industry are less likely to make an acquisition because of the risk they will face if the acquisition turned out to deteriorate shareholders’ wealth.
Even though Lehn and Zhao (2006) conclude that corporate governance and corporate control generally work well in disciplining managers, it is proven by Harford and Li (2007) that even if the bidder performs worse after the deal is completed, the executive
compensation increases. They show that bidding executives are rewarded with new stock and options grants following the acquisitions. These grants are high enough to compensate the loss the executives face if the firm’s post-acquisition performance is poorly. Given this, it could be that executives, given that their EBC is high, see completing an acquisition as an incentive to increase their compensation, even if this could deteriorate bidder shareholders’ wealth.
2.2 Financial Strength and Cash Source of Funds
The financial strength and cash accessibility of the bidder are important determinants of the payment method decision in acquisitions. Hovakimian, Opler and Titman (2001) find a strong positive relation between tangible assets as a percentage of total assets and cash accessibility. This means that if the bidder has a higher percentage of tangible assets, cash financing is more attractive due to the easier accessibility to debt issuing. Faccio and Masulis (2005) support their findings by showing that using cash is more likely if collateral is higher. In addition, Jensen (1986) implies with the free cash flow theory, that bidder executives with unused borrowing power and large free cash flows are more likely to undertake low-benefit
8 or even value-destroying acquisitions. As a consequence, the theory predicts they will
generate lower total gains. To show that high equity-based compensated bidder executives do not invest in value-destroying acquisitions, they could signal this using stock as payment method. This idea is also supported by Faccio and Masulis (2005), who find that higher financial leverage could constrain bidders in their ability to issue debt and make stock financing more likely.
However, the pecking-order financing hierarchy states that internal financing and debt financing are more attractive than external equity financing (Myers, 1984). This suggests that using stock is less favorable. This pecking-order theory supports the findings of Guay (1999), who finds that bidders will use equity only in case of highly profitable investments, because stock issuance can generate highly adverse selection costs.
The sources of funding cash acquisitions can be classified as debt, non-debt and internal funds, in which non-debt could be considered as external equity funding. Schlingemann (2004) examined the effect of the source of funds in cash acquisitions on bidder gains. In this paper, he reports that, holding the form of payment constant, prior financing decisions are a significant factor in explaining the cross section bidder returns. The results show a significantly positive relation between equity issuance in the fiscal year preceding the announcement and bidder gains, a significantly negative relation between internally generated free cash flow and bidder gains, and no significant relation between the amount of cash raised by debt issuance and bidder gains.
There is found an opposite effect in the paper of Vladimirov (2015). His paper shows empirically that abnormal announcement returns are lower when they finance their cash bids with previously issued equity, compared to using debt to finance their cash bids. The market reacts significantly worse to cash acquisitions that are not financed with debt. Moreover, Vladimirov (2015) reports that smaller firms (i.e. firms in less-developed capital markets and private firms) have limited access to debt financing, which means they are constrained in using internal funds or external equity. On the other hand, when cash constraints are completely absent, the bidders might start wasting money on unprofitable investments following the free cash flow theory of Jensen (1986). Combining this with the findings of Harford and Li (2007), this excessive use of internal cash flows could suggest that bidder executives can use these cash flows easily to complete acquisitions and to increase their total compensation no matter the effect on their equity-based compensation part. All things considered, if high EBC bidders use cash to finance their acquisitions, they will use debt or external equity issuance to signal bidder shareholders the acquisition will increase value.
9
2.3 Executive ownership
Executive ownership plays a major role in determining the payment method for completing an acquisition. Following Datta et al. (2001), the executive ownership is measured by the sum of numbers of shares excluding stock options, shares acquiring after option exercise and the number of restricted stock holdings known at the fiscal year-end prior to the acquisitions announcement. Amihud et al. (1990) report that the larger the executive ownership fraction of the bidder, the more likely the use of cash financing in corporate acquisitions. The reason behind this is that executives can secure their ownership and control position by using cash as a payment method. If they use stock, they risk to dilute this ownership and control position. In addition, Martin (1996) finds a nonlinear relationship between executive ownership and stock-financed deals. This nonlinear relationship means that only over the intermediate ownership range, the ownership has a negative significant effect on stock financing. Bidders with a relatively high or low ownership percentage are less concerned about the dilution of their control position. This nonlinear relation is supported by the findings of Jung, Kim and Stulz (1996) and Ghosh and Ruland (1998).
Furthermore, Amihud et al. (1990) find that negative bidders’ abnormal returns are associated with stock financing and bidders with low executive ownership. Given the greater potential cost to executives of losing control under stock financing, the fact that they still apply this method of payment may signal to investors that the acquisition is not value decreasing. This is the same for bidders that are known as high EBC bidders (Guay, 1999).
2.4 Investment opportunities
Bidder’s investment opportunities play an important role in the payment method decision for acquisitions. Combining the agency costs of debt with the asymmetric information model of Myer and Majluf (1984), Jung et al. (1996) argue that executives with growth objectives prefer to raise capital with equity, because it gives them more discretion over the funds raised compared to debt financing. They also find a positive stock price reaction after the
acquisition announcement if bidders use stock as payment method for bidders with valuable investment opportunities. This idea is supported by Martin (1996). In his paper he measures bidder investment opportunities using three different variables: Tobin’s q, average annually compounded sales growth rate, and the preceding cumulative stock returns. This research will only take the Tobin’s q and the preceding cumulative stock return (pre-CAR) into account. Only these two variables are included because both seem to be equal between the three payment method categories. Tobin’s q-ratio is the sum of bidder’s market value of equity,
10 long-term debt, short-term debt, and preferred stock dividend, divided by the book value of bidder’s assets, calculated as of the fiscal year-end preceding the announcement. Pre-CAR is the abnormal return cumulated over the 120 days preceding event day – 5, where event day zero is the day the acquisition is announced. For both variables, an increase in Tobin’s q and pre-CAR will be indication of growing investment opportunities.
Faccio and Masulis (2005) use the market-to-book ratio to measure bidder’s
investment opportunities, in which the MTB-ratio is calculated by the market value of equity plus the book value of debt, divided by the book value of equity plus the book value of debt. In addition, the results of Faccio and Masulis (2005) show that an increase in investment opportunities make stock financing more likely. They argue that a higher MTB-ratio is correlated with high levels of tax-deductible R&D expenses, along with low current earnings and cash dividend. This attributes to a lower bidder’s need for additional debt tax savings, which makes cash financing less attractive. Because Tobin’s Q and the MTB-ratio are more or less the same, only Tobin’s Q will be used in this research.
As mentioned before, Datta et al. (2001) have found that high EBC bidders like to acquire target’s with high growth opportunities. This means that executives with high equity-based compensation are willing to take on riskier and more valuable acquisitions to growth and create value for the firm. Acquiring those targets with growth opportunities will increase bidders’ investment opportunities and makes stock financing more likely.
2.5 Information asymmetry
Before the acquisition will take place, there exists asymmetric information about the true value of both the target and the bidder. This asymmetric information plays an important role in determining the payment method of acquisitions. The payment method that is used could already reveal some relevant information about the true value of the bidder and target. Hansen (1987) predicts a higher likelihood of using stock financing when the asymmetric information, with respect to the target’s true value, is high. If the bidder is not sure about the true value of the target, it will use stock as payment method to force the target to share the risk the bidder is facing by paying too much. With stock financing, the target will also share in any post-acquisition revaluation effect. This type of asymmetric information about the true value of the target is measured by the relative deal size, measured as the ratio of the deal value (excluding assumed liabilities), divided by the deal value (excluding assumed liabilities) plus bidder’s market value of equity 20 days prior to the announcement date (Martin, 1996).
11 Investors often don’t know the true value of the bidder, because there also exists asymmetric information on the true value of the bidder. The payment method in acquisition could solve this asymmetric information, because it has been shown by Myers and Majluf (1984) and Hansen (1987) that bidders use stock if their stock value is overpriced and use cash if their stock is underpriced. The stock price run-up can be an indication about valuation of bidder’s stock value. Stock price run-up is calculated from a bidder’s buy and hold
cumulative stock return over the year preceding the announcement month (Faccio and Masulis, 2005).
Using cash if the bidder stock is undervalued is also implied by the market-driven acquisition hypothesis of Shleifer and Vishny (2003). They argue that synergies must create value for the bidder in cash acquisitions to act in bidder shareholders’ interest. This means that poor performance following stock financing is consistent with bidder value-maximizing behavior, poor performance after cash financed acquisitions is not. Moreover, Travlos (1987) finds larger negative announcement effects following stock financing, this is part of bidder value-maximizing behavior and thus in the favor of bidder shareholders’ wealth and high equity-based compensated executives. This confirms that high EBC bidders use stock financing more likely if they are overvalued.
To prevent asymmetric information in acquisitions where the target and the bidder operate in the same industry, bidders will offer directly stock financing. Sellers are reluctant to accept the stock as payment more often when the bidder is active in the same industry. For high EBC bidders an acquisition in the same industry should indicate bidders have already more information about the target, because they are well known with the industry risks and prospects (Faccio and Masulis, 2005). This should support the idea the high EBC bidders make a value increasing decision, which is also in favor of bidder shareholders’ wealth.
2.6 Unlisted & Subsidiary targets
The legal form and public status of targets are also factors that could influence the payment method choice in acquisitions. Generally the ownership structure of an unlisted target is highly concentrated. Therefore, the takeover of an unlisted target could dilute the executive’s ownership of the bidder. As already mentioned before in section 2.3, it has been shown by Martin (1996) and Faccio and Masulis (2005) that bidder executives want to avoid this
dilution by using cash as final payment method. If the target is a subsidiary, the reason behind the sell could be that the selling corporation needs cash to restructure their core competency and goals. In contrast, Fuller, Netter and Stegemoller (2002) report that bidder shareholders
12 gain when buying a private or subsidiary target, but lose when they buy a public target. They also find a higher return if the target is large and the bidder offers stock instead of cash to finance the acquisition. This is a good example of the conflict of interest. Bidder executives prefer cash if the target is unlisted or a subsidiary and bidder shareholders prefer stock because of the higher stock returns. If executives are highly equity-based compensated, this should mean that they will also gain by using stock. When using stock they will risk their control position but they will get a higher total compensation anyway (Harford and Li, 2007).
2.7 Deal Mode
Deals can have different modes, that are also important determinants for making the payment method choice. For example, a deal could be characterized as a tender offer, a friendly takeover or a hostile takeover. A tender offer is an offer made by the bidder directly to the target shareholder to purchase target shares (Betton et al., 2008). Because there is no approval needed from the target’s management, this speeds up the completion process. If the bidder makes an offer to buy the target with the use of equity, this must be made in accordance with the Securities Act of 1933, which can lead to a substantial delay since a registration statement must be reviewed by the SEC (Martin, 1996). This is not in the interest of the initial bidder, as other bidders might come along with competing offers. So if the bidder wants to complete the deal as soon as possible, they should use cash as main payment method.
A take-over could also be characterized as hostile. A hostile takeover occurs when the target’s board of directors fights against the takeover attempt. If the bidder wants to succeed, it must gather enough shares to replace the board of directors, but the speed of the process could be crucial for the success of a hostile takeover offer (Huang and Walkling, 1987). This means that cash acquisitions have a higher probability of success than stock acquisitions, because cash financing has a shorter time period to complete the acquisitions. This research only incorporates completed acquisitions, so will not look into the relation between the payment choice and the success of an acquisition. However, it seems that the deal mode might be of any influence in the choice for a payment method and will therefore be included as a control variable in the analysis of this study.
2.8 Hypotheses
To answer the main question about the influence of the share of equity-based compensation of executives on the final payment method in acquisitions, two hypotheses will be tested. The first hypothesis will be tested with the use of an ordinal probit regression analysis, which will be discussed in detailed later on in section four.
13 H1: “A higher EBC for executives will increase the likelihood of pure stock financing”
Hypothesis 1 is based on prior literature about the link between executive compensation and completing an acquisition. If shareholders know about the high equity-based compensation of executives, they will see an acquisition announcement as a positive signal of value creation in the firm (Datta et al., 2001). This indicates that no matter the payment method, high EBC bidders will always create a positive stock price reaction when they announce an acquisition. Guay (1999) finds that this equity-based compensation is positively related to firms’
investment opportunities, which makes stock financing more likely, given the findings of both Martin (1996), and Faccio and Masulis (2005). Even the findings of Harford and Li (2007) suggest that executives can use acquisition to manipulate their compensation. Corporate governance could be used to control this manipulation, since it generally works well in disciplining executives (Lehn and Zhao, 2006).
To answer the sub-question about the influence of equity-based compensation for bidder executives on the source of funding in cash acquisitions, the second hypothesis will be tested with a normal probit regression analysis:
H2: “Higher EBC for bidder executives will increase the use of external funding.”
Hypothesis 2 is based on the findings of Schlingemann (2004) and Vladimirov (2015), that are discussed in the previous sections. They describe different sources of funds to finance bidders’ cash acquisitions. Both distinguish between debt, non-debt and corporate funds, which in turn could be either external or internal financing. Even though the pecking-order financing hierarchy places both debt and external equity financing below internal funds, Schlingemann (2004) and Vladimirov (2015) find empirical evidence that bidder gains will be higher after using equity issuance from the year preceding the announcement and after issuing debt, respectively. Using internal funds when bidder’s cash constraints are completely absent, would give signals that bidders executives use acquisitions to manipulate their total compensation. This kind of acquisitions are often value-destroying investments. Therefore, high EBC bidders should use the source of funding that will signal that the bidder is investing in a value-increasing acquisition, that is in favor of both, the bidder executives and
14
3. Data and descriptive statistics
This section first discusses the data sources and preparation of the final samples used to test the two hypotheses. It discusses also some general information behind the sample. Secondly, the descriptive statistics of the dataset is discussed. This gives some more insights into the mean values of the variables that are used to test the two hypotheses.
3.1 Data
The initial sample includes all announced and completed deals in the United States between January 1993 and December 2015. The sample is collected from Thomson Financial
Securities Data’s SDC, Worldwide Mergers & Acquisition Database.
The sample is created by using the following criteria: (1) bidders and targets are both located in the United States; (2) bidders are public firms and listed on a stock market in the U.S.; (3) there are no restrictions on the public status of the target; (4) bidders should acquire at least 5% ownership of the target. After this initial screening, 24,549 deals are collected. Information about the bidder’s and target’s identity and industry (4-digit SIC code),
announcement date, the effective date, deal value, method of payment, the source of funding and the type of deal are also collected from the SDC database.
The bidder’s accounting characteristics and executive compensation at fiscal year-end prior to the announcement are collected from Compustat and Execucomp respectively. CRSP is used to collect daily stock prices and holding returns one year prior to the announcement. After merging these three databases with the initial sample from the SDC database, 4,062 deals and 986 bidders remain in the sample. This merge is based on the bidder’s unique 6-digit CUSIP-code from the SDC database.
Only deals of which the final payment method is known, to be only cash, stock or a mix of cash and stock, are included. After controlling for the available accounting
information, executive compensation and holding stock returns one year prior to the
announcements, 794 different bidders and 1,593 deals are left in the final sample. As shown in Table I, most of the completed deals in the final sample are cash financed. The sample includes 984 (62%) cash deals, 351 (22%) stock deals and 258 (16%) deals that are financed with a mixture of stock and cash. It is interesting that mixed deals contain a higher
15 proportion of stock (53.1%) than cash (33.9%) on average. This is in contrast with the
findings of Faccio and Masulis (2005), who find a higher proportion of cash (56.9%) than stock (43.1%) on average. Panel A shows the number of deals per year based on the method
Table I
General information about the final sample
Final sample includes 794 different bidders and 1,593 completed deals. Panel A shows the number of completed deals per year based on the method of payment. Panel B shows the number of completed deals per industry based on the 2-digit SIC-code based on the method of payment. Panel C shows how many bidders made only one, two, three, four or more than five completed deals in this final sample.
Panel A: Frequency deals per year
Stock
Only Mixed Payment
Cash Only Total Percent 22% 16% 62% 100% Total 351 258 984 1,593 1993 11 2 5 18 1994 21 5 21 47 1995 32 12 32 76 1996 39 17 37 93 1997 35 18 40 93 1998 35 16 39 90 1999 41 14 52 107 2000 29 17 48 94 2001 18 20 38 76 2002 11 15 37 63 2003 6 8 49 63 2004 12 12 64 88 2005 9 10 51 70 2006 9 16 54 79 2007 3 10 65 78 2008 7 8 40 55 2009 3 9 30 42 2010 8 10 49 67 2011 5 7 47 59 2012 8 10 57 75 2013 4 9 40 53 2014 3 4 41 48 2015 2 9 48 59
Panel B: Frequency deals per Industry
Agriculture, Forestry, Fishing - - 2 2
Mining 14 20 26 60
Construction 1 8 10 19
Manufacturing 122 98 563 783
Transportation & Public Utilities 15 24 38 77
Wholesale Trade 7 1 28 36
Retail Trade 14 12 46 72
Finance, Insurance, Real Estate 127 63 101 291
16 Table I - Continued
Panel C: Frequency deals per Company
Maximum Deals Frequency Percent
1 395 49.75 2 198 24.94 3 102 12.85 4 49 6.17 > 5 50 6.30 Total 794 100%
of payment. The sample clearly shows the global and strategic merger wave in the late 1990s. On average 75 deals are completed in one year and only two stock deals are completed in 2015. Reason for this low number of completed deals could be the longer duration of getting a stock deal completed. Stock deals have an average duration of four months, which is twice as high as the duration of cash deals; cash deals have an average duration of two months.
Panel B shows the number of deals completed per industry, based on the 2-digit SIC-code. The manufacturing, finance, insurance and real estate and services industry have the most completed deals. Panel B is also divided over the method of payment, which shows clearly that only in the finance, insurance and real estate industry most deals were paid with only stock, while in other industries the payment method is mostly cash only. Panel C gives an overview of how many deals a company has completed in the time period from 1993 till 2015. In this sample International Business Machines Corp. and Pfizer Inc. completed most acquisitions; they completed 10 deals in a time period of 23 years.
In the second part of this research, only cash-financed acquisitions are left in the sample. Because the information about the source of funding is not always available in the SDC database, there are only cash-financed acquisitions taken into account from which we know what the source of funding is. This results in only 423 deals from 331 different bidders.
3.2 Descriptive statistics
Table II Panel A presents the descriptive statistics of bidder, target and deal-specific variables. Some of these variables show large differences across the different payment methods, which can give a clear overview of the influence on the payment method. The growth opportunities indicators Tobin’s Q, the MTB-ratio and pre-CAR, are the highest for stock-financed deals. This is in line with the expectation that bidders with higher investment opportunities will prefer stock financing. The pre-CAR of cash-financed deals is even
17 negative, which indicates that cash financing might be the most preferable payment method for bidders in case of low investment opportunities.
Remarkable is the highest collateral percentage for mixed payments. The expectation would be that the higher the percentage of collateral, the more the bidder is able to finance their cash bids with debt and will use cash only to finance the acquisition. Stock financed deals are characterized by the highest average values for financial leverage (61.9%). This
Table II
Descriptive Statistics by Payment Method
'Stock Only' includes deals that were financed only with stock. 'Mixed Payment' includes deals that were paid with a combination of cash and stock. 'Cash Only' includes deals that were financed with only cash, which really includes actual cash, liabilities and newly issued notes. Panel A includes the averages of some continuous variables and binary variables sorted by the method of payment. Panel B shows the averages of executive compensation variables of fiscal year-end prior to the announcement sorted by payment method. These variables are given in thousands $. Equity-based compensation (EBC) is calculated by the sum of all stock options granted to executives divided by total compensation of those executives, all based on the fiscal year-end prior to the announcement. A detailed variable description is given in Appendix A.
Panel A: Bidder, Target and Deal characteristics
Variables Stock Only
Mixed
Payment Cash Only Total
Tobin's q 2.035 1.362 1.700 1.719 MTB-ratio 3.297 2.056 2.675 2.712 Pre-CAR 0.040 0.006 -0.021 -0.003 Collateral 0.314 0.474 0.412 0.401 Financial Leverage 0.619 0.588 0.518 0.552 EBIT/Sales 0.162 0.153 0.143 0.148 EBITDA/Total assets 0.113 0.121 0.147 0.135
Cash flow (mil. $) 389 340 727 590
Total Assets (mil. $) 35,047 18,910 18,078 21,925
Ownership 0.035 0.034 0.028 0.031
Market value (mil. $) 13,000 10,360 15,795 14,297
Deal value (mil. $) 1,432 1,788 545 942
Relative value 0.109 0.162 0.082 0.101 Run-up 0.288 0.214 0.172 0.205 Market Run-up 0.122 0.088 0.096 0.100 Tender offer 0.014 0.039 0.185 0.124 Hostile takeover 0.009 0.016 0.027 0.021 Bankrupt 0.000 0.000 0.008 0.005 Unlisted target 0.339 0.469 0.600 0.521 Subsidiary 0.037 0.008 0.037 0.032 Intra-industry 0.860 0.837 0.780 0.807
18 Table II - Continued
Panel B: Bidder's Executive Compensation
Salary 2,263 2,318 2,612 2,488
Bonus 2,348 1,921 1,733 1,899
Other annual (short-term) 153 133 115 126
Restricted Stock granted 827 768 848 831
Stock options granted 5,871 2,939 3,615 4,002
Long-term incentive plan pay-outs 515 928 432 530
All other (long-term) 165 42 132 125
Total compensation 12,142 9,050 9,487 10,001
Equity-based compensation (%) 31.09% 23.03% 20.00% 22.93%
Panel C: Percentage Source of Funds
Debt 75% Non-Debt 5%
Borrowings 24.82% Common stock issue 3.78%
Line of credit 32.39% Rights issue 0.24%
Debt issue 9.69% Preferred stock issue 0.95%
Bridge loan 5.44% Mezzanine financing 0.00%
Foreign lenders 2.36%
Junk bond issue 0.00% Corporate funds 56%
convinces the idea that financial constrained bidders prefer stock as financial instrument in financing the deal. As predicted, bidders that completed only cash-financed deals have on average higher cash flows in the year preceding the acquisition announcement, than bidders that completed only stock deals. The same holds for the ratio of EBITDA over total assets and is in line with the findings of Jensen (1986). Bidders of stock financed acquisitions have on average a higher value of total assets than bidders which complete cash or mixed deals. This is remarkable because the greater the size of the bidder, the more cash as payment method is expected, which is also assumed by Faccio and Masulis (2005).
An intermediate percentage of ownership should lead to more cash financed deals, because the risk of losing control will be the highest if they use stock as payment method in that case (Martin, 1996). In this sample the ownership is based on the total shares the
executives own over the total shares outstanding, which shows low percentages of ownership. The lowest percentage ownership is found in only cash financed deals (2.8%), while the highest percentage is for pure stock financed deals (3.5%). The intermediate percentage is found for mixed deals (3.4%).
The average deal value for stock financed deals is almost three times larger than that for cash deals, although the highest average deal value is found for mixed deals. This contradicts with findings of earlier studies that classify takeover premiums according to the
19 payment method. The evidence shows that takeover premiums, and thus higher deal values, are indeed significantly greater in only cash deals than in pure stock financed deals (Betton et al., 2008). On the contrary, the highest relative value is found for mixed deals (16.2%). The lowest is for cash deals (8.2%) and the intermediate percentage is for pure stock deals (10.9%). The relative value gives a good indication for the risk the bidder takes with this takeover. If the relative value is higher, the higher the risk the bidder takes, which means stock-financing would be more preferable. The stock price run-up in the year prior to the announcement is the highest for stock financed deals (28.8%), the lowest for pure cash deals (17.2%) and the intermediate for mixed deals (21.4%).
Looking at the mode of acquisitions, 18.5% of cash deals is characterized as a tender offer. This is much higher than the number of tender offers in mixed deals (3.9%) and in stock financed deals (1.4%). Because stock financing will take a longer period of time to complete, this indicates that cash financing would be more preferable in tender offers. On average, in only a half per cent of all acquisitions the target is bankrupt. The zero percentage for stock and mixed deals indicates that bidders use only cash if their target is bankrupt. Due to the highly concentrated ownership structure of private targets, 60.0% of the cash deals are unlisted targets. Only in 33.9% of the stock deals the target was unlisted, which is nearly half of the percentage in cash deals. Although most of all deals were completed in the same industry, the highest percentage is found for stock financed deals. 86.0% of all stock deals were completed in the same industry.
Panel B presents the differences over the payment methods and executives
compensation attributes at the fiscal year-end prior to the announcement date. The variable part of the compensation, which are the restricted stocks and stock options granted at the fiscal year-end prior to the announcement, is the highest for pure cash financed deals. This could imply that the executives choose cash as most preferable payment method because it is proven that cash deals do not have that negative stock price effect after the deal is announced (Travlos, 1986). On the other hand, the equity-based compensation (EBC) of bidder
executives is the highest for pure stock financed deals, which could indicate that they only use stock if they are sure the deal will add value to the bidder in the future. This would also be in favor of bidder shareholders’ wealth.
Table II, Panel C shows the percentages of the different sources of funds that were used in the cash acquisitions sample. Bidders combine sometimes two till four different types of funds, which explain why the percentages do not add up to 100%. This final sample includes a small percentage of cash acquisitions that used non-debt sources of funds to
20 finance their cash bids, which is in contrast with the 21% of Vladimirov (2015). Also
surprising is the percentage of credit line, which is the most used financing source in cash acquisitions instead of borrowings. Taking debt and non-debt funds together as external funds and corporate funds as internal funds, only 24.4% of all cash acquisitions are financed by only internal funds. 75.6% of all cash acquisitions are financed with external and/or internal funds.
4. Methodology
To test the two hypotheses, two different regression analyses are used; the ordered probit regression analysis, and the normal probit regression analysis. First, the ordered probit regression analysis and model is discussed that is used to test the first hypothesis. Thereafter, the normal probit regression analysis and model that is used to test the second hypothesis is discussed.
4.1 Ordered probit regression model
First, to look at the effect of equity-based compensation on the financing decision, the payment method is the dependent variable. The sample includes only acquisitions paid with pure cash, stock or a mixture of stock and cash. Because bidders do not always know the exact percentage of cash in their mixed payments, it might be more attractive to order the different payment methods and use the ordered probit regression analysis to test the first hypothesis. In this case, the different payment methods are ordered with cash in the highest order. This means that zero indicates the acquisition is pure stock financed, one if the acquisition is paid with a mix of stock and cash, and two if the acquisition is pure cash financed. The general form of the ordered probit model is:
(1) 𝑦𝑖∗ = 𝑥𝑖′𝛽 + 𝑢𝑖
with
(2) 𝑦𝑖 = 𝑗 𝑖𝑓 𝛼𝑗−1 ≤ 𝑦𝑖∗ < 𝛼𝑗
where i is the acquisition, x’ the vector of independent variables, and β is the vector of regression coefficients. The dependent variable can be seen as a single latent variable, which is not observable only when it crosses a threshold. In this case the j is the ordered value of the
21 dependent variable, with the underlying latent variable that must lie between the boundaries of two thresholds, 𝛼𝑗−1 and 𝛼𝑗. The probability that acquisition i will select ordered value j is:
(3) 𝑝𝑖𝑗 = 𝑝(𝑦𝑖 = 𝑗) = 𝑝(𝛼𝑗−1 ≤ 𝑦𝑖∗< 𝛼𝑗) = 𝐹(𝛼𝑗− 𝑥𝑖′𝛽) − 𝐹(𝛼𝑗−1− 𝑥𝑖′𝛽)
where F for the ordered probit regression model is the standard normal cumulative
distribution function. The coefficients of an ordered probit regression model cannot be seen as the regular OLS coefficient. To interpret the ordered probit coefficients, marginal effects need to be calculated. These marginal effects of an one unit increase in regressor x, on the probability of selecting ordered value j is:
(4) 𝜕𝑝𝑖𝑗
𝜕𝑥𝑜𝑖= {𝐹
′(𝛼
𝑗−1− 𝑥𝑖′𝛽) − 𝐹′(𝛼𝑗− 𝑥𝑖′𝛽)}𝛽𝑜
The ordered probit regression is usually estimated using the maximum likelihood. Therefore, these marginal effects give a prediction of the probability of getting in the higher or lower threshold.
Given the general form of an ordered probit regression model, the ordered probit regression model that is used to test this first hypothesis looks like:
(5) 𝑃𝑖(0 = 𝑆𝑡𝑜𝑐𝑘; 1 = 𝑀𝑖𝑥; 2 = 𝐶𝑎𝑠ℎ) = 𝛽1𝐸𝐵𝐶𝑖+ 𝛽2𝐶𝑜𝑙𝑙𝑎𝑡𝑒𝑟𝑎𝑙𝑖+
𝛽3𝐶𝑎𝑠ℎ 𝑓𝑙𝑜𝑤𝑖 + 𝛽4𝐹𝑖𝑛′𝐿𝑒𝑣𝑒𝑟𝑎𝑔𝑒𝑖 + 𝛽5𝐿𝑜𝑔(𝑇𝑜𝑡𝑎𝑙 𝐴𝑠𝑠𝑒𝑡𝑠)𝑖 + 𝛽6𝑂𝑤𝑛𝑒𝑟𝑠ℎ𝑖𝑝𝑖 + 𝛽7𝑄𝑖 + 𝛽8𝑃𝑟𝑒𝐶𝐴𝑅𝑖 + 𝛽9𝑅𝑉𝑎𝑙𝑢𝑒𝑖 + 𝛽10𝑅𝑈𝑁𝑈𝑃𝑖 + 𝛽11𝐼𝑛𝑡𝑟𝑎 𝐼𝑛𝑑𝑢𝑠𝑡𝑟𝑦𝑖 +
𝛽12𝑈𝑛𝑙𝑖𝑠𝑡𝑒𝑑 𝑇𝑎𝑟𝑔𝑒𝑡𝑖 + 𝛽13𝑆𝑢𝑏𝑠𝑖𝑑𝑖𝑎𝑟𝑦𝑖 + 𝛽14𝑇𝑒𝑛𝑑𝑒𝑟𝑖 + 𝛽15𝑀𝑘𝑡 𝑅𝑈𝑁𝑈𝑃𝑖 + 𝑢𝑖
The main independent variable is the equity-based compensation (EBC). A negative coefficient of this main independent variable would make cash as payment method less likely. The opposite, a significant negative coefficient means that a higher equity-based compensation would make stock financing more likely. Other independent variables are included to control their influence on the payment method based on earlier studies of Martin (1996) and Faccio and Masulis (2005). The included control variables: collateral, cash flow, financial leverage, and the logarithm of total assets are included to control for the capital structure and financial strength of the bidder; ownership is included to control for bidder executives’ control effect; q and pre-CAR are included to control for the investment and
22 growth opportunities; relative value, run-up, intra-industry, and market run-up control for the existing asymmetric information; and dummy variables unlisted target, subsidiary, and tender are included to control for the legal form of the target and the mode of the acquisition respectively.
4.2 Probit regression model
To test the influence of equity-based compensation on the source of funding in cash
acquisitions (hypothesis 2), a normal probit regression analysis is used. For this second part, the sample used, includes only acquisition financed with cash. The dependent variable for this probit regression model is the source of funding of these cash acquisitions, which equals one if these acquisitions are funded with external financing, and equals zero otherwise. This means that if the dependent variable equals one, the cash acquisitions is financed by debt and/or non-debt funds. Debt-like considerations include borrowings, credit lines, debt issuance, bridge loans or foreign lenders. Non-debt considerations include common stock issue, rights issue, preferred stock issue and mezzanine financing. The dependent variable equals zero if the cash acquisition is financed by internal corporate funds. The probit regression model used to test the second hypothesis is:
(6) 𝑆𝐹𝑖(1 = 𝑒𝑥𝑡𝑒𝑟𝑛𝑎𝑙 𝑓𝑢𝑛𝑑𝑠; 0 = 𝑖𝑛𝑡𝑒𝑟𝑛𝑎𝑙 𝑓𝑢𝑛𝑑𝑠) = 𝛽1𝐸𝐵𝐶𝑖+ 𝛽2𝐶𝑜𝑙𝑙𝑎𝑡𝑒𝑟𝑎𝑙𝑖+ 𝛽3𝐶𝑎𝑠ℎ 𝑓𝑙𝑜𝑤𝑖 + 𝛽4𝐹𝑖𝑛′𝐿𝑒𝑣𝑒𝑟𝑎𝑔𝑒𝑖 + 𝛽5𝐸𝐵𝐼𝑇/𝑆𝑎𝑙𝑒𝑠𝑖 + 𝛽6𝐸𝐵𝐼𝑇𝐷𝐴/𝑇𝐴𝑖 + 𝛽7𝐿𝑜𝑔(𝑇𝑜𝑡𝑎𝑙 𝐴𝑠𝑠𝑒𝑡𝑠)𝑖 + 𝛽8𝑄𝑖+ 𝛽9𝑃𝑟𝑒𝐶𝐴𝑅𝑖 + 𝑢𝑖
The main independent variable in this probit regression is similar to the one used in the ordered probit regression model, which is the equity-based compensation (EBC). A positive coefficient of the EBC indicate a positive relation between the external financing and the equity-based compensation. This means that higher equity-based compensation increases the probability of using external funds to finance cash acquisitions. Control variables are
included to control for their influence on the source of funding. These control variables are based on the variables used by Schlingemann (2004) and Vladimirov (2015). The ratio of earnings before interest and taxes (EBIT) over sales, and the ratio of earnings before interest, tax, depreciation and amortization (EBITDA) over total assets, both measured from the preceding fiscal year-end before the announcement, are included. These two ratio’s together with Collateral, Cash flow, Financial leverage, and the logarithm of total assets are used to
23 proxy, for example, bidder’s accessibility to borrowing debt on the capital market (Hadlock and Pierce, 2010), which could increase the use of external financing.
5. Results
This section discusses the results of the ordered probit regression and the normal probit regression analysis. First, the results of the ordered probit regression analysis are related to the hypothesis 1, that tests the relation between equity-based compensation and the payment method choice. Secondly, the results of the normal probit regression analysis are related to the second hypothesis about the relation between the source of funding in cash acquisitions and the equity-based compensation of bidder executives.
5.1 Ordered probit regression results
Table III shows the ordered probit regression results of six different regression models. The first two regressions are comparable to the model used in the paper of Faccio and Masulis (2005). In these two regressions the control variable for the mode of the acquisition is not included. The third and fourth regression models are based on the paper of Martin (1996). These two regression models excluded the legal form of the target and included the control variable for the mode of the acquisitions. The final two regressions include all control
variables from Faccio and Masulis (2005) and Martin (1996) combined. All control variables are included which shows that some variables changed in significance after adding some new control variables about the target. For example, the dummy variable that indicates if the target is a subsidiary is equal to one, is significant in the first two regressions, but lost significance in the fifth and sixth regressions. Regressions two, four and six include the main independent variable, equity-based compensation. The odd regressions are excluding the EBC, which clearly show the effect of this main independent variable on the other control variables.
Looking at the three odd regressions including the EBC, it turns out that in all three regressions this variable is significantly negative. This means that higher EBC makes cash financing less likely. Given the mean values of all included independent variables of the odd regressions, the marginal effects shows that a 10 per cent increase in equity-based
compensation makes it around 2% more likely to use stock as payment method. Using cash as payment method is around 3% less likely if the equity-based compensation increased with 10%. Even though the negative coefficient on the equity-based compensation is significant at
24 Table III
Ordered Probit Regressions Explaining the choice of Payment Method
In all models the dependent variable is the payment method choice, which is zero for pure stock acquisitions, one for mixed acquisitions and two for pure cash acquisitions. The p-values in the parentheses below are based on the Huber/White heteroscedasticity - consistent standard errors. Variable explanation can be found in Appendix A.
(1) (2) (3) (4) (5) (6) EBC -0.772*** -0.969*** -0.846*** (0.000) (0.000) (0.000) Collateral 0.235*** 0.242*** 0.171* 0.180** 0.186** 0.193** (0.008) (0.006) (0.050) (0.040) (0.034) (0.030) Cash flow 0.000*** 0.000*** 0.000*** 0.000*** 0.000*** 0.000*** (0.001) (0.002) (0.001) (0.002) (0.001) (0.001) Financial Leverage -1.317*** -1.316*** -1.179*** -1.176*** -1.180*** -1.183*** (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) Log(Total assets) 0.033 0.020 -0.048* -0.060** 0.034 0.020 (0.218) (0.459) (0.062) (0.021) (0.209) (0.462) Ownership -1.104** -1.325** -0.839 -1.130** -0.918* -1.168** (0.041) (0.017) (0.112) (0.039) (0.095) (0.038) Tobin's Q -0.114*** -0.086*** -0.132*** -0.096*** -0.120*** -0.089*** (0.000) (0.001) (0.000) (0.000) (0.000) (0.000) Pre-CAR -0.285 -0.282 -0.357* -0.342* -0.287 -0.277 (0.121) (0.115) (0.065) (0.068) (0.133) (0.137) Relative Value -1.002*** -1.162*** -1.715*** -1.838*** -0.766*** -0.933*** (0.000) (0.000) (0.000) (0.000) (0.004) (0.000) Run-up -0.323*** -0.296** -0.297** -0.270** -0.303** -0.280** (0.009) (0.014) (0.019) (0.028) (0.016) (0.023) Intra-industry -0.126 -0.131 -0.165* -0.161* -0.081 -0.079 (0.147) (0.131) (0.060) (0.067) (0.378) (0.383) Unlisted target 0.512*** 0.463*** 0.825*** 0.778*** (0.000) (0.000) (0.000) (0.000) Subsidiary 0.475** 0.467** 0.252 0.243 (0.039) (0.049) (0.341) (0.383) Tender 1.249*** 1.319*** 1.610*** 1.648*** (0.000) (0.000) (0.000) (0.000) Market Run-up -0.472* -0.534** -0.473* -0.546** -0.449* -0.516** (0.052) (0.031) (0.055) (0.031) (0.073) (0.044) Intercept 1 -1.444*** -1.746*** -2.348*** -2.636*** -1.040*** -1.362*** (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) Intercept 2 -0.908*** -1.201*** -1.800*** -2.072*** -0.458* -0.767*** (0.000) (0.000) (0.000) (0.000) (0.084) (0.005) Observations 1,558 1,558 1,558 1,558 1,558 1,558 Pseudo R^2 0.0836 0.0956 0.1019 0.1207 0.1403 0.1541 Log Likelihood -1,321.59 -1,304.20 -1,295.15 -1,268.07 -1,239.83 -1,219.93 Robust pval in parentheses. *** p<0.01, ** p<0.05, * p<0.1
25 1% level for all three models which included this main independent variable, the marginal effect is not really high. Given that a higher equity-based compensation makes stock financing more likely is in line with the first hypothesis. So this could indicate that bidder executives will use the stock payment method to give signal to the bidder shareholders they acquire a target company, which will create value to the firm. This increasing value is in favor of both the bidders executives and shareholders’ wealth, but also good for the economic activity. If more acquisitions will be seen as value-maximizing investments, this will have a positive effect on the whole economy and society.
Specifically looking at the control variable for the financial strength and cash accessibility for the bidders, most of them stay highly significant over all six regressions. Only the logarithm of total assets are significant in the third and fourth model, but in these two regressions, the coefficients are both negative. This negative sign indicate that the higher the total book value of assets, the more likely stock financing is used. This contradicts with the findings of Faccio and Masulis (2005), who find that higher total assets makes cash financing more likely. For both cash flow and the percentage of collateral all coefficients are positive and highly significant. The coefficients for the financial leverage are negative and highly significant for all six regressions. These results are all in line with the expectation that if cash accessibility is higher, cash financing is more likely, and if bidders are cash
constrained, stock financing is more likely as final payment method.
Interesting are results of the coefficients of ownership. Prior studies show a negative relation between the percentage ownership and stock financing. The results shown in table III, suggest the opposite. These ownership coefficients are negative and significant, except for the third regression. This regression shows a non-significant negative effect of ownership on the likelihood of using cash as payment method. Comparing this with regression four, which included the main independent variable, the equity-based compensation, this
ownership coefficient is significant again. This suggests that equity-based compensation is influenced by the percentage ownership, but that these two variables are not highly
correlated. Following the argumentation of Martin (1996), who finds a nonlinear effect on stock financing, the negative results are in line with his findings. If the percentage ownership is higher, stock as payment method becomes more likely.
The control variables for bidder’s investment opportunities, are both negative and significant in the third and fourth regression. Tobin’s q stays negative and significant over all six regressions. This negative result indicates that bidders with higher investment
26 opportunities will more likely use stock as final payment method. This is in line with the results of both Martin (1996) and Faccio and Masulis (2005).
Asymmetric information plays an important role in the payment method decision, because the final choice of payment method can reveal much information about the true value of the bidder as well as the target. Both relative value and stock price run-up stay negative and highly significant over all six regressions. The higher the relative value, the more likely stock as final payment method is used. The higher the stock price run-up preceding the
announcement, the more likely bidder’s stock price is overvalued and will use stock financing as final payment method. If investors will see the stock payment method as a signal of
overvaluation, this has a negative stock price effect after the announcement. This negative stock price reaction will be restored over time if it turns out that high EBC bidders used stock financing for a value-maximizing investment. Even the intra-industry dummy is only
significant in the third and fourth regression, the negative coefficient indicate also that when the bidder acquires a target in the same industry, stock financing is more likely. The negative significant coefficient of market run-up indicates that if it turns out that the market is
overvalued, stock financing is more likely to be used as final payment method.
Thereafter, controlling for the legal form of the target, the results of the first two regressions show significantly positive coefficients for unlisted and subsidiary targets. This means that if the target is unlisted and/or a subsidiary, cash financing as payment method is more likely. Controlling for the mode of the acquisition, the results show that if the
acquisition is characterized as a tender offer, cash financing is significantly more likely than stock or mixed financing. If both control variables for the mode of acquisitions and the legal form of the target are included, the results show that the dummy variable for subsidiary targets change to non-significant positive coefficients in the last two regressions. This suggest that if the acquisition is characterized as a tender offer, the legal form of the target does not influence the final payment method. The same holds for the intra-industry dummy, which turns into a non-significant negative coefficient in the last two regressions by including the tender offer dummy variable.
Overall, given the mean values of all independent values, the average probability that the acquisition is financed with stock is 22%, 16% probability that the acquisition is financed with a mixture of stock and cash, and 62% probability that the acquisition is paid with cash.
27
5.2 Probit regression results
Table IV shows the probit regression result for four different regression models. For these last regressions, only cash acquisitions are included in the sample. These four regressions test the second hypothesis about the relation between the equity-based compensation and the source of funding in cash acquisitions. Due to the merge between Compustat, Execucomp and CRSP with all cash acquisitions of the Thomson One SDC database, the total number of observation is relatively low to the whole sample of acquisition.
The results show non-significant negative coefficients for the second and fourth regression including the main independent variable equity-based compensation. This suggests that the equity-based compensation does not influence the choice of the source of funding in cash acquisitions. The second hypothesis expected a positive relation between equity-based compensation and external funding. The negative coefficients show the opposite effect. So even though the results are not significant, the coefficients show a negative relation between equity-based compensation and external funding. This negative effect could indicate that if high EBC bidders use cash to finance the acquisition, they use cash because completing acquisitions is a good way to increase executive’s total compensation despite the post-acquisition performance.
Looking at the sign and significance of the control variables, the results in the first two regressions show significant effects for the percentage of collateral, cash flow and financial leverage. The positive significant effect of collateral on the probability of external funding is in line with the expectation. If a bidder owns more intangible assets, it is easier to issue debt and use this debt to finance acquisitions. The same holds for the positive
significant effect of EBITDA over total assets in the last two regressions. The negative significant effect of cash flow on the probability of using external funding, is also in line with the expectation. If bidders experience high cash flows, the probability of using internal funds is higher.
By including control variables Tobin’s q and pre-CAR, which are good indicators of bidder’s investment opportunities, the coefficients of Tobin’s q show a negative significant effect on the probability of using external funding. Bidders with high investment
opportunities can be seen as risky companies, and this risk can increase the cost of debt (Myers, 1977), which makes internal funding more attractive. This means that the negative significant effect of Tobin’s q on the probability of using external funding can be clarified by the increase of the cost of debt if that kind of bidders take more risk.
28 Table IV
Probit Regression explaining financing choice in Cash Acquisitions
The dependent variable is the source of funding in cash financed acquisitions. It will take value one if the source of funding is debt, and will be zero otherwise. The p-values in the parentheses are based on the Huber/White heteroscedasticity - consistent standard errors. Variable explanation can be found in Appendix A. (1) (2) (3) (4) EBC -0.382 -0.277 (0.237) (0.390) Collateral 0.564** 0.553** 0.419* 0.419* (0.011) (0.011) (0.058) (0.057) Cash flow -0.000** -0.000* -0.000** -0.000** (0.049) (0.054) (0.045) (0.050) Financial leverage 0.887** 0.946** 0.649 0.696 (0.046) (0.036) (0.141) (0.118) EBIT/Sales -0.452 -0.537 -0.430 -0.487 (0.590) (0.522) (0.604) (0.560) EBITDA/Total assets 1.337 1.473 2.695** 2.714** (0.174) (0.144) (0.018) (0.017) Log(Total assets) -0.020 -0.019 -0.012 -0.012 (0.757) (0.768) (0.846) (0.852) Tobin's q -0.169** -0.162** (0.011) (0.015) Pre-CAR 0.314 0.298 (0.400) (0.424) Intercept 0.073 0.102 0.279 0.294 (0.879) (0.833) (0.564) (0.545) Observations 414 414 414 414 Pseudo R^2 0.0525 0.0556 0.0659 0.0675
Log pseudo likelihood -215.78 -215.07 -212.73 -212.36 Robust pval in parentheses. *** p<0.01, ** p<0.05, * p<0.1
Overall, the results show no significant effect of the equity-based compensation on the probability of using external financing, which is not in line with the second hypothesis. These insignificant results could be a consequence of the small sample size. Besides this, a
combination of internal and external funding is used to finance cash acquisition, which is not taken into account in this probit regression model. Also a combination of debt and non-debt sources of funding are used to finance these cash acquisitions. Both debt and non-debt are taken into account as external financing, which could also be a limitation in testing the second hypothesis.
29
6. Conclusion and discussion
This study examined the relation between equity-based compensation of bidder executives and the final payment method in acquisitions, with as special case the relation between equity-based compensation and the source of funding in cash acquisitions. While corporate takeovers bring some major changes in the whole organization, successful acquisitions increase economic activity and create value for the whole society. This means it is important to make the right payment method decision and executive compensation structure to make the corporate acquisition successful. Prior studies focusses more on the main determinants that influence the method of payment in corporate acquisitions and the effect executive
compensation on the post-acquisitions firm performance, whereas this research examines if there is a relation between these two important determinants and the post-acquisition firm performance.
To answer the main and sub-question, two different hypotheses are tested using the ordered probit regression analysis and the normal probit regression analysis respectively. The first hypothesis about the relation between EBC and the payment method is tested by using the ordered probit regression with the payment method as ordered dependent variable. The main independent variable is the EBC, controlled by variables which measure the bidder’s financial strength and cash accessibility, investment opportunities, asymmetric information, the mode of the acquisition and the public and legal form of the target. The results of the EBC are significantly negative for all three regressions, which means that higher EBC for bidder executives make stock as final payment method more likely. This is in line with the first hypothesis. Using stock as payment method could give a good signal to bidder
shareholders that the acquisition will create value for the firm and thus increases shareholders’ wealth. Taking all mean values of the included explanatory variables, the probability the acquisition is financed with stock is 22%, 16% the acquisition is financed with a mixture of cash and stock, and 62% that the acquisition is financed with cash.
The second hypothesis about the relation between the EBC of bidder executives and the source of funding in cash acquisitions is tested by using the normal probit regression model with external funding for cash acquisitions as dependent variable. The main
independent variable is the same as the one for the ordered probit regression model, which is the EBC of bidder executives. In the probit regression model the EBC is controlled by
variables indicating the financial strength and cash accessibility of the bidder, and by variable indicating bidder’s investment opportunities. The results of this probit regression are not
30 significant for the main independent variable. This is in contrast with the second hypothesis, which expected a positive relation between the equity-based compensation and external funding in cash acquisitions. These insignificant results can be a consequence of the small sample of cash acquisition from which the source of funding is known. Even though Vladimirov (2015) found thousands of cash acquisition of which the source of funding is known, this study lost a lot of cash acquisition due to the merge with Execucomp, Compustat and CRSP. Not all information about executives compensation is available for the whole sample which excluded some cash acquisitions from which the source of funding is known. Another reason for insignificant results might be the limitation of the dependent variable. The dependent variable external funding equals one if the source of funding is debt or non-debt, but some acquisitions use a combination of debt, non-debt and internal corporate funds. These combinations are not taken into account in this probit regression analysis. A solution to test the effect of equity-based compensation on the use of external, internal or a combination of external and internal funding, an ordered probit regressions analysis can be used.
Concluding, there is a positive relation between the equity-based compensation of bidder executives and using stock as final payment method. Stock financing is more likely if the equity-based compensation is higher. There is no relation between the equity-based compensation of bidder executives and the source of funding in cash acquisitions. This means that the magnitude of the equity-based compensation does not indicate which source of funding is used to finance these cash acquisitions. The relation between the equity-based compensation and the sources of funding can be expanded by looking more detailed at the influence of the EBC on the specific debt and non-debt sources. There might be an additional positive relation between the EBC and external equity issuance prior to the announcements.
