Price differences in M and A for direct and indirect equity payments

Amsterdam Business School MSc Business Economics, Finance Track

Master Thesis

Price differences in M&A for direct and indirect equity payments

Bunnik, Mark

May 2015

Thesis supervisor: Dr. V.N. Vladimirov

1 Statement of Originality

This document is written by Student Mark Bunnik 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.

2

Content

Section 1. Introduction 3

Section 2. Literature review 4

2.1. Method of payment 4

2.1.1. Equity 4

2.1.2. Cash 5

2.1.3. Other related findings 6

2.2. Method of financing 6

2.2.1. Internally generated cash flows 6

2.2.2. Debt 7

2.2.3. Equity 8

2.3. Pecking order theory 8

2.4. Managerial discretion and information sensitivity 9

2.5. Contribution 9 2.6. Hypotheses 9 Section 3. Methodology 10 3.1. Data 10 3.1.1. Firm characteristics 11 3.1.2. Deal characteristics 11 3.2. The model 11

Section 4. Data and descriptive statistics 14

Section 5. Results 19

Section 6. Robustness checks 22

Section 7. Conclusion 24

3

1. Introduction

The number and size of mergers and acquisitions is huge, every year 4-8% of all publicly traded companies is taken over (Berk et al., 2009). The reason for this is that companies hope to achieve large synergies, earnings growth, diversification or an entry into a new market in order to generate more value for their shareholders. However firms in the United States lose a lot of shareholder value when acquiring other companies (Moeller & Schlingemann, 2005). The shareholders of the acquirer lost up to 12 cents per dollar spent on the acquisition in the period 1998 through 2000, this sums up to a combined loss of $240 billion in that period in the United States alone.

Therefore it would be very useful for managers to know how they could make less costly acquisitions, hence lose less shareholder value with acquisitions. Or more favorable profitable takeovers. Some factors that could improve the return of takeovers are better research of the target, more emphasis on the implementation of the strategy or simply offer a lower price for the target’s shares. The latter is the focus of this paper, however if acquirers could simply offer less for the targets they will do so. Takeover prices is a extensively analyzed topic in the financial economics, an overview of this is given by Betton, Eckbo and Thorburn (2008). It is clear that method of payment is of great influence for the prices. Next to that the method of financing also plays an important role in the determination of the offer price (Schlingemann, 2004; Vladimirov, 2014).

In this paper the focus is on the effect of the method of payment combined with the method of financing on the price for cash-constrained companies. The methods of payments that are compared in this paper are cash financed by selling equity and direct equity payments. Next to that direct equity payments will be compared with cash payments with other financing methods. This paper will contribute to the literature by combining the method of payment and method of financing in order to make a bridge in the current literature. The main research question is: Is selling equity to bid in cash the same as bidding in equity?

This paper answers this question with the use of data on mergers and acquisitions from the Thomson One database between 1980 and 2014 from all over the world and where both the target and the acquirer are public companies. Another important quality of the data is that the source of financing for the cash payments needs to be known. With this dataset regressions are ran to compare takeover transactions that are paid for fully in equity or in cash of which the financing method is known. The analyses will be made to find the difference between equity payments and cash payment with the different financing methods. After controlling for other factors that could influence the offer price the result is that, ceteris

4 paribus, cash offers financed by selling equity leads to 2.2% lower takeover premium than paying directly in equity. The other financing methods have do not have a significant different takeover price than direct equity payments. In other words, the results show that there is no price difference for mergers and acquisitions between equity payments and cash payments financed by debt of internal financing. Given the previous results for the difference between direct and indirect equity payments it is expected that the otherwise financed cash payments have a higher takeover price than the equity payments in order to come to the on average higher takeover price of cash payments compared to equity payments (Betton, Eckbo & Thorburn, 2008). The last finding is that if the method of payment is cash equity financing leads to the lowest takeover prices followed by debt financing and internal financing leads to the highest takeover prices.

The paper continues as follows, section 2 provides the literature review of all the related literature and theories for this study. Section 3 explains the methodology that is used in this paper. Followed by the data and descriptive statistics in section 4. Section 5 discusses the results of the analysis. The robustness checks are in section 6 and section 7 concludes the paper.

2. Literature review

In this section an overview is given of the related literature for this study. The different methods of payments are discussed in section 2.1, followed by the methods of financing in section 2.2. Thereafter the pecking order theory is discussed in section 2.3 and Managerial discretion and information sensitivity in section 2.4. Section 2.5 shows the contribution of the paper and section 2.6 provides the hypotheses.

2.1. Method of payment

Takeover premiums is an extensively analyzed topic in the financial economics, an overview of this is given by Betton, Eckbo and Thorburn (2008). From all those analyses it is clear that the method of payment is of great influence for the premiums. There are two main types of payment methods; equity payments and cash payments. Mixes of equity and cash payments are also possible and often used.

2.1.1. Equity

Equity payments are payments done with stock or with stock like securities. Bidders typically use this method of payment when they have low cash balances or when their stock is

5 overvalued (Shleifer and Vishny, 2003). Overvalued firms cannot just issue equity and invest in bonds, since the shareholders will then know that the firm is overvalued. This is a problem due to asymmetric information (Myers & Majluf, 1984). The shareholders do not know whether a firm issues new equity for the reason that they are overvalued or that there is a good investment opportunity (Schlingemann, 2004). As reaction negative stock price reaction for acquirer for offering equity to pay for acquisitions (Martin, 1996). As a result there are high costs for adverse selection and financing associated with using equity as method of payment (Schlingemann, 2004). When a bidder is overvalued the chances are high that the market as a total is overvalued, including the target. When this is the case it is only profitable for the bidder to bid in equity when the bidder is overvalued more than the target. In contrast to the bidder, the seller will only agree to an equity payment if he will undervalue the bidder’s shares in to order obtain a larger claim to the profit of the newly-formed firm (Vladimirov, 2014). This suggests a higher takeover price.

Another reason for using equity payments is that the seller has tax benefits of receiving equity over cash (Faccio and Masulis, 2005). If the seller is paid in cash he will immediately have to pay capital gain taxes over his return, however when he is paid in equity the capital gain taxes are deferred until the received shares are sold. Next to that the bidder shares the risk of overpayment with the target (Hansen, 1987), therefore this is often used with mergers (Betton, Eckbo & Thorburn, 2008). Lastly, higher management ownership results in a lower chance of stock payments, for the reason that this could lead to dilution of control of the management. In particular when the targets shareholders are highly concentrated (Martin, 1996). This is not the case when there are supermajority voting rights (Faccio and Masulis, 2005) because they will hardly notice a difference in who holds the common shares.

2.1.2. Cash

When bidder choose cash as method of payment they face lower constraints such as bank requirements for loans or providing information to shareholders before issuing new equity. This provides the managers with more flexibility. In addition, due to regulatory reasons bidding in cash is faster than other methods of payment. Therefore it is often used as tender offer to preempt the competition (Martin, 1996).

Hansen (1987) and Fishman (1989) argue that targets will only accept the offer if the cash offered has more value than the target company, therefore underpayment is very unlikely. Bidder should only pay in cash if they are undervalued, the more they are

6 undervalued the higher the cost of a stock offer, hence better to offer in cash. This results in higher bidder returns when offering in cash than with equity offers (Martin, 1996). Even though on average the takeover premiums are higher than bids in equity (Betton, Eckbo & Thorburn, 2008). This contradicts the suggested higher takeover price for equity payments of the previous section 2.1.1..

Another advantage of a cash offer is that it mitigates the risk for the seller’s shareholders that they become a minority shareholder in the newly-formed firm facing a new majority shareholder with the possibility of moral hazard problems (Faccio and Masulis, 2005).

2.1.3. Other related findings

Other related literature has been written by Boateng and Bi (2013), they found that there is a difference in the performance in the pre-bid period between acquisitions financed by cash and those financed with equity. They did not mention price differences between the methods of payments. De la Bruslerie (2013) showed that the higher the percentage of cash in the payment the higher the tender offer. However his data set consisted out of acquisitions with payment mixes and not 100% cash or equity payments.

2.2. Method of financing

When the method of payment is cash then the method of financing is also of importance (Schlingemann, 2004; Vladimirov, 2014), however this part of the deal structure is less comprehensively researched than the method of payment. There are three methods of financing for cash payments; Internal generated cash, debt and issuing new equity. It is also possible for a firm to finance itself with a mix of these methods.

2.2.1. Internally generated cash flows

The first method of financing for cash bids is using the internally generated cash flows. Firms need sufficient free cash flows in order to bring up enough cash to offer a winning bid. Usually this is only possible with large, well established companies. However current cash stockpiles in U.S. at record highs at $2 trillion in 2013 (Pinkowitz, Sturgess & Williamson, 2013). Jensen (1986) argues that managers with easy access to money could lead to agency problems such as investing in projects with a negative NPV or self-dealing. This is in line with findings of Moeller and Schlingemann (2005) of large wealth destruction in M&A transactions. However Pinkowitz, Sturgess and Williamson (2013) contradict this theory and

7 they find that cash-rich firms 23% less likely than non-cash-rich firms to make a offer in cash.

In contrary to the large cash rich companies, most bidder are cash constrained and do not have enough cash and liquid assets to pay cash for acquisitions (Faccio and Masulis, 2005). Therefore it is not possible for them to use internally generated cash flow as financing method and are forced to use one of the other financing methods.

2.2.2. Debt

The amount a firm is able to borrow is strongly influenced by its debt capacity and current debt level (Faccio and Masulis, 2005). More tangible assets leads to easier borrowing as it can be used as collateral. Next to that larger firms are usually more diversified, therefore are less volatile and have smaller probabilities on bankruptcy. Hence large firms are able to borrow more and easier than smaller firms (Faccio and Masulis, 2005). When the debt increases so do the tax benefits but also the bankruptcy costs. Therefore companies are always searching for the optimal debt level, this is the trade-off theory.

The cost of debt financing depends greatly on the competitiveness of the capital market, the lower the competitiveness the higher the financier can set the financing costs which results in less borrowing and lower takeover prices. If the bidder has more bargaining power, the financing cost will be lower and the payments they make are more often financed with debt (Vladimirov, 2014). Cheaper financing results in higher and more aggressive bids than when the bidder is not cash constraint, hence more overbidding when debt financing is cheap. As a consequence the takeover premium in five to eight percent higher when financed with debt (Vladimirov, 2014).

Debt also has some other effects, first of all monitoring by institutions that provide the loans, most commonly banks. Since the banks have an interest in that the company runs well and creates enough cash flows to make its interest payments, the bank will monitor the performance of the borrower. Due to this monitoring the management discretion is reduced and the managers are less likely to show moral hazard problems. Another possible effect is debt overhang, this is the case when a company has too much debt. If that company has a positive NPV investment opportunity but all the value will go to the debt holders instead of the shareholders the investment in forgone. To mitigate this problem companies should not issue too much debt.

8 2.2.3. Equity

The last method of financing is issuing new equity to generate cash for investments. This is only used when the bidder has no access to competitive (debt) financing. Financier try to make a profit of the bidders cash constraint, bidders pass on the high costs and bid less than if they were not cash constraint. This result in lower take over premiums than debt financed bids (Vladimirov, 2014). If value depends much on growth than equity financing is preferred over debt financing (Martin, 1996). Small firms that do not have access to competitive financing have to issue equity in order to generate cash for investments, this could be both public and private equity. This is also the case for countries with less developed capital markets. As mentioned before in section 2.1.1. companies should only issue equity when they are overvalued, however sometimes they do not have another option and are forced to issue new equity.

Equity financing is also more expensive than debt financing and therefore also leads to lower take over prices (Vladimirov, 2014). Using this method of financing the bidder’s gains are the highest of the different financing methods for paying in cash (Schlingemann, 2004).

2.3. Pecking order theory

All the previously discussed advantages and disadvantages of the different methods of financed leads companies to have certain preferences for certain methods of financing. This is called the theory is the pecking-order theory which was firstly noticed by Myers (1984). He argues that there are three types of funding for firms; retained earnings (internally generated cash), debt and equity. For outside investors there is risk involved with both debt and equity, where the risk with equity is larger than the risk of debt. This risk is primarily because of adverse selection problems. The outside investors need to be compensated for the risk with a premium, where the premium for equity is larger. Due to the extra costs firms prefer to financing with retained earnings, then financing with debt and lastly financing with equity. However Frank and Goyal (2003) show that in reality equity financing is not dominated by debt financing in magnitude. For this paper the focus is on cash constrained companies, therefore the most favorable method of financing, retained earnings, will not be used. Only the least favorable method of financing, equity financing, is used in this paper. Both selling equity and bidding directly in equity are in the last category.

9 2.4. Managerial discretion and information sensitivity

There also is a trade-off between obtaining financing and managerial discretion which depends on information sensitivity. When a firm cannot generate internal funding it has to give information to outside investors in order to obtain external financing. How sensitive this information is depends on the firms size (Vladimirov, 2014), the smaller the firm the more sensitive information they have to provide. The information is needed by the investors to solve the adverse selection problem, but also enables them to monitor the firm and helps them to prevent moral hazard problems as overinvestment by managers (Stulz, 1990). However the higher the level of management discretion the higher the likelihood of investing in negative NPV projects given a poor investment opportunity set. And the more likely managers are to forgo positive NPV projects given a good investment opportunity set.

2.5. Contribution

In contrast to all the studies mentioned above which look at the payment methods or financing methods of cash, this study will focus on the differences between offering directly with equity or selling the equity to bid in cash. By doing so it will make a bridge in the current literature between methods of payments and methods of financing. Next to that the findings in both the method of payment and method of financing are not all in line with each other, therefore this subject needs to be more researched.

2.6. Hypotheses

The literature and theory’s discussed above find that the price premium for an acquisition is higher when the method of payment is cash instead of equity. However the method of financing the cash offers is also of importance for the acquisition price. Myers’ pecking order theory (1984) and Vladimirov’s findings (2014) show that when the cash offer is financed by selling equity that the price is lower than with the other financing methods. Therefore the question arises whether cash offers are still higher than equity offers even if they are financed by selling equity. This question also answers the main research question and is tested by the following hypothesis:

Hypothesis 1: Bidding in equity leads to lower acquisition prices than bidding in cash generated from selling equity.

10 From the literature it also follows that cash offers financed by debt have an even higher premium. Therefore the difference between the premiums when offering in equity of offering cash financed by debt should be even higher. This is tested by the following hypothesis:

Hypothesis 2: Bidding in equity leads to lower acquisition prices than bidding in cash financed by debt.

Finally the effect of offering cash that is internally financed on the premium is analyzed with the use of the following hypothesis:

Hypothesis 3: Bidding in equity leads to lower acquisition prices than bidding in internally financed cash.

3. Methodology

This section will describe the data and methodology used to analyze the price difference in takeovers with different methods of payment and financing. In section 3.1. the type of data needed for this analysis and their expected effect are provided. Thereafter the models to test the hypotheses are given in section 3.2.

3.1. Data

The data sample build form the Thomson One database and consists of merger and acquisition transactions between 1980 and 2014. The mergers and acquisitions in the dataset have to satisfy the following conditions. (1) The takeover is fully completed. (2) Both the target and the acquirer are publicly listed. (3) The method of payment is fully in equity or the method is partially or fully in cash. (4) When the method of payment is cash the financing method is known. (5) After the transaction the acquirer owns at least 50% of the target company. (6) And the financial information of both the target and the acquirer is available. Conditions four allows partial cash payments in order to be left with a large enough sample. The financing method could also be mixed for the cash payments. However when the cash payments is financed by selling equity it is assumed that a large proportion of the total offer is financed in this way for the reason it is the most expensive financing method. Therefore companies will not use this method for a small cash requirement. The final data sample contains 3,500 deals of which 1,976 deals are paid for with equity and 1,524 are paid for with cash.

11 3.1.1. Firm characteristics

The size of the target is retrieved from Thomson One, this is expected to have a negative relationship with the offer price for the reason that it is more expensive to pay a premium for a large company than for a small company. The acquirer’s size is also retrieved from Thomson One and is expected the have a positive relationship with the offer price since a larger company is able to obtain more financing than smaller companies and are therefore able to offer a higher price. Whether or not the acquirer has toehold in the target is found in Thomson One and is expected to have a positive effect on the offer price because a part of the premium will come back to the acquirer through the already owned shares.

3.1.2. Deal characteristics

Whether or not the target tried to stop the acquisition using a poison pill is found on Thomson One and the relationship with the price is expected to be positive for the reason that the acquirer will most likely have to offer a higher price in order to complete a successful takeover. The relative value is the deal value divided by the size of the acquirer. The larger the acquirer is relative to the deal the easier it is for the acquirer to finance a premium, therefore a negative relationship is expected. Both the deal value as well as the size of the acquirer are retrieved from Thomson One. Lastly the runup return is the cumulative return of the target's stock in the 41 days prior to the bid, the stock prices are gathered from CRSP. In line with the existing literature (an overview is given by Betton, Eckbo & Thorburn, 2008), this is expected to have a positive relationship with the offer price.

3.2. The model

At first the premium paid by the bidder is determined as the natural logarithm of the price the acquirer paid per share divided by the target share price 42 days prior to the bid. The premium is determined in this way in order to be able to compare takeovers of different size and prices. The price 42 days prior to the bid is used instead of the price 1 day prior to the bid in order to rule out any influence of insider trading or leaked information that will possibly affect the share price. When the premiums are determined they are compared between the mergers and acquisition paid in cash and those paid in equity. The premium is determined as follows:

12 This results in the following simple regression model:

𝑃𝑟𝑒𝑚𝑖𝑢𝑚𝑖 = 𝛽1𝑖 𝐸𝑞𝑢𝑖𝑡𝑦 + 𝛽2𝑖𝐸𝑞𝑢𝑖𝑡𝑦𝐹𝑖𝑛𝑎𝑛𝑐𝑖𝑛𝑔 + 𝛽3𝑖𝐷𝑒𝑏𝑡𝐹𝑖𝑛𝑎𝑛𝑐𝑖𝑛𝑔 (2)

+ 𝛽4𝑖𝐼𝑛𝑡𝑒𝑟𝑛𝑎𝑙𝐹𝑖𝑛𝑎𝑛𝑐𝑖𝑛𝑔 + 𝜀

Where Premium is the natural logarithm of the difference between the price paid and the share price 42 days before the bid. PricePaid is the price paid per share, Price42Prior is the share price 42 days prior to the bid. Equity is a dummy that is 1 when the target is paid in equity. EquityFinancing is a dummy that equals 1 when the method of payment is cash and the method of financing is selling equity. DebtFinancing is a dummy that is 1 when the method of payment is cash and the method of financing is debt. InternalFinancing is a dummy that equals 1 when the method of payment is cash and this is internally financed. And lastly the error term ε.

The simple model will probably be affected by endogeneity problems and omitted variable biases, therefore the model is expanded with control variables such as characteristics of the bidder and the target that could influence the price. The control variables that are used are the control variables that were tested significant by Schlingeman (2004) or Vladimirov (2014). The control variables are:

 Toehold; The acquirer has a toehold in the target.

 Poison pill; The target company adopts a poison pill in order to stop the takeover.

 Runup return; The cumulative return of the target's stock in the 41 days prior to the bid.

 Relative value; The deal value divided by the size of the acquirer.

 Size target; The natural logarithm of the total assets of the target.

 Size acquirer; The natural logarithm of the total assets of the acquirer. This results in the following regression model:

𝑃𝑟𝑒𝑚𝑖𝑢𝑚_𝑖 = 𝛽_1𝑖 𝐸𝑞𝑢𝑖𝑡𝑦 + 𝛽_2𝑖𝐸𝑞𝑢𝑖𝑡𝑦𝐹𝑖𝑛𝑎𝑛𝑐𝑖𝑛𝑔 + 𝛽_3𝑖𝐷𝑒𝑏𝑡𝐹𝑖𝑛𝑎𝑛𝑐𝑖𝑛𝑔 (3) + 𝛽4𝑖𝐼𝑛𝑡𝑒𝑟𝑛𝑎𝑙𝐹𝑖𝑛𝑎𝑛𝑐𝑖𝑛𝑔 + 𝛾1𝑖𝑇𝑜𝑒ℎ𝑜𝑙𝑑 + 𝛾2𝑖𝑃𝑜𝑖𝑠𝑜𝑛𝑃𝑖𝑙𝑙 + 𝛾3𝑖𝑅𝑢𝑛𝑢𝑝

+ 𝛾4𝑖𝑅𝑒𝑙𝑎𝑡𝑖𝑣𝑒𝑉𝑎𝑙𝑢𝑒 + 𝛾5𝑖𝑆𝑖𝑧𝑒𝑇𝑎𝑟𝑔𝑒𝑡 + 𝛾6𝑖𝑆𝑖𝑧𝑒𝐴𝑐𝑞𝑢𝑖𝑟𝑒𝑟 + 𝜀𝑖

13 The model will also be tested for country, industry and year fixed effects. The analysis will use all three models (simple model as in equation 2, the extended model as in equation 3 and the latter with fixed effects) in order to see the effect of the additions in the regression model. After the regression is run the difference in the last model between the equity payment and the cash payments with the different method of financing is tested with the use the following Wald test: 2 𝐹_{1,𝑁−𝐾−1} = (𝛽1 − 𝛽2) √𝑆_𝛽2₁ + 𝑆_𝛽2_{2 − 2𝑆} 𝛽1,𝛽2 (4)

This test will show whether or not the coefficients are significantly different of one another. If they are different this proves that there is a difference in the takeover premium between selling equity to bid in cash and bidding in equity directly. In order to support all the hypotheses the coefficient for equity should be lower than for the cash payments. If this is the case it shows that equity payments have a lower takeover premium than cash payments financed with the different financing methods. With the results of the test it is also possible to find differences in takeover prices between cash payments with the different financing methods.

14

4. Data and descriptive statistics

All the variables used in the analyses are listed in table 1 with their definition and source of origin.

Table 1. Variable definitions

The table describes the variables used in this analysis in alphabetical order and their sources.

Variable Definition Source

Cash Dummy that equals 1 if the method of payment is cash financed by selling equity.

Thomson One

Debt Financing Dummy that equals 1 when the method of payment is cash and the method of financing is debt.

Thomson One

Equity Dummy that equals 1 if the method of payment is equity. Thomson One Equity Financing Dummy that equals 1 when the method of payment is cash

and the method of financing is selling equity.

Thomson One

Internal Financing Dummy that equals 1 when the method of payment is cash and this is internally financed.

Thomson One

Poison pill Dummy that equals 1 if the target adopted a poison pill. Thomson One Premium The natural logarithm of the difference between the price

paid per share and the share price 42 days prior to the bid, as provided in equation 1.

Thomson One /CRSP

Price42Prior The target's share price 42 days prior to the bid, this share price is used to rule out influence of insider trading or leaked information.

CRSP

Price Paid The price the acquirer paid for the target. Thomson One Relative Value The value of the deal divided by the total assets of the

acquirer.

Thomson One Runup The cumulative return of the target's stock in the 41 days

prior to the bid.

CRSP

Size acquirer The acquirer's total assets at time of the bid. Thomson One Size target The target's total assets at time of the bid. Thomson One Toehold Dummy that equals 1 if the acquirer owned shares of the

target before the announcement date.

Thomson One

Table 2 contains the annual distribution of takeover dates of 3,500 mergers and acquisitions over the period 1980-2014. The highest number of takeovers are at the end of the 1990’s and at the start of 2000’s, this is a period where the economy was strong and the market was leading towards the internet bubble.

15 Table 2. Annual distribution of takeover dates.

All the data is collected from the Thomson One (formerly SDC Platinum) mergers and acquisitions database. The data consists of M&A transactions between 1980 and 2014 where both the target and the acquirer were publicly listed.

Year n Percent 1980 3 0.09 1981 11 0.31 1982 9 0.26 1983 22 0.63 1984 24 0.69 1985 70 2.00 1986 90 2.57 1987 95 2.71 1988 111 3.17 1989 125 3.57 1990 68 1.94 1991 69 1.97 1992 73 2.09 1993 92 2.63 1994 153 4.37 1995 202 5.77 1996 183 5.23 1997 283 8.09 1998 301 8.60 1999 323 9.23 2000 235 6.71 2001 147 4.20 2002 81 2.31 2003 72 2.06 2004 67 1.91 2005 61 1.74 2006 81 2.31 2007 82 2.34 2008 52 1.49 2009 44 1.26 2010 43 1.23 2011 40 1.14 2012 58 1.66 2013 49 1.40 2014 81 2.31 Total 3,500 100

16 Table 3 presents the sample summary statistics. It contains the number of observations, mean, median, standard deviation, minimum and maximum of all variables used in this paper. A summary on the methods of payments used in the sample is provided in Panel A, 56.46% of the sample uses equity payments and 43.54% are cash payments.

Table 3. Sample summary statistics

All the data on firm and deal characteristics is collected from the Thomson One (formerly SDC Platinum) mergers and acquisitions database, the data on share prices and returns are from CRSP. The data consists of M&A transactions between 1980 and 2014 where both the target and the acquirer were publicly listed. The method of payment is fully in equity or in cash. Panel A provides a summary on the methods of payments used in the sample. The method of financing is provided in panel B. Panel C shows the summary statistics on the firm and deal characteristics before winsorizing. The summary statistics on the firm and deal characteristics after winsorizing at 90% are in panel D. The definitions of all the variables are given in table 1.

Panel A: Method of payment % n

Equity 56.46 1,976

Cash 43.54 1,524

Panel B: Method of financing % n

Selling equity 12.47 190

Debt 73.36 1,118

Internal financing 55.71 849

Panel C: firm and deal characteristics

before winsorizing n Mean Median Std. Dev. Min Max

Takeover premium 3140 0.31 0.32 0.40 -3.30 2.71

Acquirer has toehold 3500 0.12

Poison pill 3500 0.01 Runup 3413 0.27 0.20 0.35 -0.80 3.73 Relative Value 3162 0.74 0.15 3.27 0.00 85.07 ln(Size target) 3408 5.70 5.62 1.95 -0.61 13.78 ln(Size acquirer) 3176 7.60 7.69 2.19 0.47 14.36

Panel D: firm and deal characteristics

after winsorizing n Mean Median Std. Dev. Min Max

Takeover premium 3140 0.33 0.32 0.24 -0.16 0.80

Acquirer has toehold 3500 0.12

Poison pill 3500 0.01

Runup 3413 0.26 0.20 0.27 -0.15 0.88

Relative Value 3162 0.45 0.15 0.66 0.00 2.48

ln(Size target) 3408 5.68 5.62 1.74 2.74 8.99

17 Panel B provides the method of financing for the cash payments. It is possible that firms use multiple financing methods for one cash payment. Therefore the sum of the percentages of the financing methods is higher than 100%. Most of the cash offers are at least partially financed with internal cash flows (73.36%), more than half (55.71%) of the offers is financed with debt. And only 12.47% of the cash offers is financed by selling equity. This is in line with the pecking order theory, as discussed in section 2.3, that companies will only finance their operations with equity if there is no other financing method available.

Panel C shows the summary statistics on the firm and deal characteristics. The table shows that the average takeover premium in the sample is 31%. 12% of the acquirers has toehold in the target company before the bid and that only 1% of the targets tried to stop the takeover with the use of a poison pill. This number is probably this low in the sample for the reason that only successfully completed takeovers are included. When non completed takeovers are added this could be higher for the reason that a poison pill makes it more expensive for the acquirer to take over the target, this may lead to a withdrawal of the acquirer. The median runup return is 0.20% and median relative value is 0.15, this indicates that the acquirers on average takeover a company that is 1/7 of their own size. This is also supported by the fact that the average size of the target is smaller (around 1/7) than the average size of the acquirer.

The minimum and maximum observations in panel C show that there are large outliers in the sample. For example, the maximum relative value is 85.07. This means that the deal value was over 85 times higher than the size of the target. This is over 25 times the standard deviation from the average. In order to reduce the effect of the outliers the variables are winsorized at 90% level. The lowest and highest 5% of the observations are changed into the same level as the 5th percentile and the 95th percentile respectively. The summary statistics on the firm and deal characteristics after the winsorization are provided in panel D. The means of the variables did not change that much in comparison to panel C, except for the relative value since that variable had the biggest outliers. Since only the lowest and the highest 5% of the observation are adjusted the 50th percentile did not change and therefore did the median not change. The main effect of winsorizing the variables is that the outliers are removed and therefore the standard deviation is clearly smaller than before winsorizing. The lower standard deviation could lead to more significant results in the empirical analysis.

Table 4 presents the correlation of the variables. There is very little if any correlation between most of the variables except between the size of the target and the size of the acquirer. The reason why these variables are moderately correlated is probably because it is

18 easier for larger companies to acquirer large targets than it is for smaller companies. Therefore small companies acquire more small targets and larger companies acquirer the larger targets, hence the correlation. The size of the acquirer and the relative value of the deal are also moderately correlated. This is because the size of the acquirer is used to calculate the relative value. Lastly the equity payment is negative and moderately to highly correlated with the financing method. This is for the reason that they are all dummy variables and if equity payment equals zero than the method of payment is cash and one or more financing methods is used, turning one or more financing dummies to one. Because the rest of the variables are little correlated and especially with the main independent variable, the equity payment, the probability on a multicollinearity problem is small.

Equity payment 1

Equity financing -0.273 1

Debt financing -0.788 0.169 1

Internal financing -0.637 0.049 0.328 1

Acquirer has toehold -0.103 0.051 0.081 0.067 1

Poison pill -0.096 0.114 0.085 0.039 0.150 1

Runup -0.165 0.017 0.101 0.144 -0.045 0.015 1

Relative Value 0.024 0.091 0.071 -0.13 -0.099 0.042 -0.042 1

ln(Size target) -0.035 0.065 0.106 -0.022 0.055 0.044 -0.122 -0.099 1

ln(Size acquirer) -0.065 -0.038 -0.009 0.139 0.035 -0.014 0.026 -0.513 0.0668 1

Table 4. Correlation of variables

The cells denote the correlation coefficients between the variables.

ln(Size acquirer) Equity payment Acquirer has toehold Poison pill Runup Relative Value ln(Size target) Equity financing Debt financing Internal financing

19

5. Results

The results of the regressions of the determinants on the premium that the acquirer paid for the target when the method of financing for the cash bids is known is shown in table 5. The first regression uses the simple regression model as stated in equation 2 (see section 3.2) and provides no preliminary supportive evidence for the hypotheses (see section 2.6). The coefficient for equity payments is higher than all of the financing methods for the cash payments, which indicates that paying directly in equity for a merger of acquisition leads to a higher takeover premium than cash payments regardless of the method of financing.

Table 5.

The table shows the results of the regressions of the determinants on the premium that the acquirer paid for the target. The definitions of all the variables are given in table 1. The first regression uses the simple model provided in equation 2. The second model includes the control variables (equation 3) and model 3 includes the control variables and fixed effects. The significance level is based on robust standard errors (between brackets) in parentheses at the 1%, 5% and 10% level is indicated by ***, ** and * respectively.

Dependent variable: Premium

Model (1) (2) (3) Equity payment 0.306*** 0.114*** 0.012*** (0.006) (0.012) (0.003) Equity financing 0.102*** 0.026* -0.010 (0.022) (0.014) (0.008) Debt financing 0.232*** 0.071*** 0.010* (0.010) (0.010) (0.005) Internal financing 0.231*** 0.056*** 0.019*** (0.011) (0.009) (0.006)

Acquirer has toehold 0.011 0.031***

(0.012) (0.003) Poison pill 0.082** 0.034*** (0.037) (0.007) Runup 0.645*** 0.623*** (0.012) (0.003) Relative Value 0.038*** 0.031*** (0.006) (0.001) ln(Size target) -0.017*** -0.016*** (0.003) (0.001) ln(Size acquirer) 0.017*** 0.004*** (0.002) (0.001) Observations 3,140 2,803 2,802 R-squared 0.610 0.833 0.879 Country FE NO NO YES Industry FE NO NO YES Year FE NO NO YES

20 The difference in coefficients indicates that using equity as method of payment leads to 20.4% higher takeover price than cash payments financed by selling equity. It also shows a 7.4% and 7.5% higher takeover price for equity payments than cash payments financed with debt or internal financing respectively. This contradicts all three hypotheses.

When the control variables are added in the second model like in equation 3 (see section 3.2) the magnitude of the coefficients changes but the coefficient for equity payments still indicates that it has the highest takeover price in comparison to the different financing methods of the cash payments. Due to the change of magnitude of the coefficients the difference in takeover price between equity and cash payments is in this model 8.8% more expensive takeover with direct equity payments instead of indirect equity payments. Cash payments financed with debt or internal financing leads to respectively 4.3% and 5.8% lower takeover prices than equity payments. These results also contradicts the hypotheses.

In the third model the fixed effects are added to the complete model with the control variables. In this model is the significance of the coefficient of the equity financed and debt financed cash payments lower than in the other models. This in combination with the coefficients being much lower, the effect is less than 2% on the takeover premium, makes it more difficult to draw conclusions from this table alone. There is a more detailed analysis on the differences in takeover premium between the equity payments and the differently financed cash payments provided in table 6. The coefficients of the control variables in the regressions are in line with other findings in the literature (an overview is given by Betton, Eckbo & Thorburn, 2008) and the expectations mentioned in section 3.1.

The differences in takeover premium caused by the different methods of payment and different methods of financing in case of cash payments are provided in table 6. The differences are calculated by subtracting the first column variables from the top row variables. The differences are than tested with the use of the Wald test (see equation 4 in section 3.2). With the use of this table the hypotheses can be tested.

Hypothesis 1: Bidding in equity leads to lower acquisition prices than bidding in cash generated from selling equity.

The difference between direct and indirect equity payments is significant, however the difference is positive (see table 6). This means that direct equity payments have a 2.2% higher takeover price than cash payments financed by selling equity. This result contradicts the first hypothesis. In other words, hypothesis 1 is rejected, there is no empirical evidence that bidding in equity leads to lower acquisition prices than bidding in cash generated from

21 selling equity. In contrary, there is empirical evidence that bidding in equity leads to higher acquisition prices than bidding in cash generated from selling equity. This finding also provides an answer to the main research question, selling equity to bid in cash is not the same as bidding in equity.

Table 6.

The table provides the differences in takeover premium due to the different methods of payment and in case of a cash payment the different methods of financing. The difference is measured as the top row variables minus the first column variables. Between brackets is the p-value of the difference as a result of the Wald test as given in equation 4. The significance at the 1%, 5% and 10% level is indicated by ***, ** and * respectively.

_Equity payment Equity financing Debt financing Internal financing Equity payment - Equity financing 0.022*** - (0.002) Debt financing 0.011 -0.020 - (0.790) (0.128) Internal financing -0.007 -0.029*** -0.009 - (0.113) (0.000) (0.431)

Hypothesis 2: Bidding in equity leads to lower acquisition prices than bidding in cash financed by debt.

Table 6 shows that equity payments have a 1.1% higher premium than cash payments financed with debt, but the difference is not significant. Therefore there is no empirical evidence that there is a difference in the takeover premium between equity payments and cash payments financed with debt. This results in the rejection of hypothesis 2.

Hypothesis 3: Bidding in equity leads to lower acquisition prices than bidding in internally financed cash.

Table 6 provides preliminary supportive evidence for hypotheses 3. It shows that equity payments have on average a 0.7% lower takeover premium than internally financed cash payments. However the difference is very small and not significant. Therefore hypothesis 3 need to be rejected. There is no empirical evidence that bidding in equity leads to lower acquisition prices than bidding in cash that is internally financed.

22 Other findings of the analysis are the difference in takeover prices between the different financing methods for the cash payments. The table provides evidence that equity financing leads to the lowest takeover prices compared with debt financing and internal financing. The differences are respectively 2.0% and 2.9%, but only the difference with internal financing is significant. This finding, that equity financing leads to the lowest takeover prices, is in line with other empirical findings (e.g. Vladimirov, 2014. See section 2.2.3.). Debt financed cash payments have on average a 0.9% lower takeover premium than internally financed cash offers. However the difference is very small and not significant.

6. Robustness checks

The cash payments in the dataset are possibly diluted with other methods of payments. Some of the observations flagged as cash payments could have been partially paid with other methods. Therefore the analysis could be comparing full equity payments with mostly equity payments and a small cash payment. Table 7 provides the results of the same regressions as in the prior section, but here the payments are either fully in equity or fully in cash financed by selling equity. This solves the dilution of method of payment problem, however with these limitations there are only 48 observations left for cash payments financed by selling equity. The sample used for this analysis is not corrected for outliers with the use of winsorization, this could be of influence for the significance of this analysis.

Due to the few observations it is difficult to draw conclusions from the analysis. With more observations the almost 1% lower takeover premium for equity payments in model 3 could have been significant, or even completely different.

23

Table 7.

The table shows the results of the regressions of the determinants on the premium that the acquirer paid for the target when the method of payment is either fully in equity or fully in cash financed by selling equity. The definitions of all the variables are given in table 1. The first regression uses the simple model provided in equation 2. The second model includes the control variables (equation 3) and model 3 includes the control variables and fixed effects. The significance level is based on robust standard errors (between bracets) in parentheses at the 1%, 5% and 10% level is indicated by ***, ** and * respectively.

Dependent variable: Premium

Model (1) (2) (3)

Equity payment -0.089** 0.003 -0.009

(0.035) (0.027) (0.008)

Acquirer has toehold -0.093** 0.011

(0.043) (0.024) Poison pill 0.103 -0.020** (0.095) (0.009) Runup 0.682*** 0.670*** (0.026) (0.013) Relative Value 0.001 0.001** (0.002) (0.000) ln(Size target) -0.015*** -0.002** (0.005) (0.001) ln(Size acquirer) 0.008 0.000 (0.005) (0.001) Constant 0.385*** 0.164*** -1.891*** (0.034) (0.042) (0.029) Observations 1,628 1,609 1,609 R-squared 0.001 0.410 0.687 Country FE NO NO YES Industry FE NO NO YES Year FE NO NO YES

A problem that arises often with social sciences is the endogeneity problem. The problem is that one or more variables are correlated with the error term. As a consequence there is a bias in the estimates, which may cause the hypothesis to be falsely rejected (type 1 error) or not rejected the hypothesis when it is actually false (type 2 error). Since the results of the analyses in this paper reject two hypotheses and not rejects one hypothesis both type of errors could be the case in this analysis. Endogeneity can result from three reasons, the first reason could be measurement errors. However since all the data used in this paper is retrieved from databases and is not newly created data and it is corrected for outliers, it is not likely to a possible cause for an endogeneity problem. Unless there are imperfections in the databases.

24 Another circumstance that could inflict endogeneity is simultaneity, which means that both the dependent and independent variable have a simultaneous causal effect on each other. This also seems to be unlikely in this situation because there is no obvious explanation why the offer price should affect the method of payment.

The last and most common circumstance that can cause endogeneity are omitted variables. The variables that are omitted from the analysis are included in the error term causing it to correlate with the explanatory variables. In order to prevent the endogeneity problem from emerging in this analysis control variables are added to the regression model. The control variables are the factors that the existing literature has pointed out as consistent and significant control variables. Next to the control variables the model is controlled for country, industry and year fixed effects. Even after all preventive measures it is still possible that the model suffers from a endogeneity problem.

One aspect of the analysis that could inflict an endogeneity problem are the factors that affect both the method of payment and the method of financing. For example, if a company is not able to issue equity in order to pay in equity for the target it has to use cash as method of payment. As method of financing for this cash payment that company is still unable to issue equity, therefore is that financing method impossible. Or if a company is cash constrained and does not have access to debt financing the only payment options are direct and indirect equity payments. Hence, companies that pay either direct or indirect with equity probably have common characteristics. This could also affect the takeover price. A possible solution in this case is adding the amount of cash an acquirer has before the takeover as a control variable or even as an instrument for internally generated cash payments in an IV-regression. However Pinkowitz, Sturgess and Williamson (2013) show that cash-rich companies are less likely to make cash payments, which implies that the amount of cash is a bad instrument. The solution to the problem possibly lies within IV-regressions, however finding good instruments prove to be a hard task. This might be a good topic for future research.

7. Conclusion

Literature findings have shown that the method of payment has a great influence on the takeover price of mergers and acquisitions. In addition, few publications described cases when the method of payment was cash that than the method of financing also plays an important role as determinant for the takeover price. This paper combines the method of

25 payment with the method of financing in order to answer the following question: Is selling equity to bid in cash the same as bidding in equity? Next to that the difference in takeover premium between equity payments and cash payments with other financing methods are also analyzed in this paper. Lastly the difference in takeover prices between the different financing methods for cash payments is analyzed.

The results of the empirical analyses show that equity payments do not have a lower takeover price than cash payments financed by selling equity. In contrary, direct equity payments result in a 2.2% higher takeover price than indirect equity payments. Other literature argue that equity payments lead to lower takeover prices than cash payments (Betton, Eckbo & Thorburn, 2008). However they do not take into account how that cash is financed. When the effect of the method of finance is analyzed it is clear that equity financed cash payments have lower takeover prices than the other financing methods (both in this paper and by Vladimirov, 2014). From these results it can be concluded that the on average higher takeover price of cash payments lies with the debt and internally financed cash payments and not with the equity financed offers. With this in mind it can be stated that, ceteris paribus, cash payments financed by selling equity leads to the lowest takeover prices. Hence the answer to the main research question, Is selling equity to bid in cash the same as bidding in equity? Is no, selling equity to bid in cash is not the same as bidding in equity. Selling equity to bid in cash leads to lower takeover prices.

When the method of financing for the cash offer is debt, the takeover premium is 1.1% higher than equity payments. For internally financed cash payments the takeover premium is 0.7% lower than equity offers. However both of these differences are not significant. This is not consistent with the other literature. Given the previous results for the difference between direct and indirect equity payments it is expected that the otherwise financed cash payments have a higher takeover price than the equity payments in order to come to the on average higher takeover price of cash payments compared to equity payments (Betton, Eckbo & Thorburn, 2008). Since this is not the case the results suggest that there is no difference in takeover price between equity and cash payments or even that equity payments lead to higher takeover price on average.

The last results of the analysis are on the differences in takeover prices between cash payments with the different financing methods. It shows that financing cash payments by selling equity leads to the lowest takeover prices followed by debt financing, internal financing results into the highest takeover prices for cash bids. However the difference between debt and internal financing is very small and not significant. That equity financing

26 leads to the lowest takeover price of the different cash payments is in line with other literature (e.g. Vladimirov, 2014).

The differences between this paper’s findings and those of other literature could be due to the data. Due to a limited dataset with too few observations when only payments fully in cash with only one single financing method are used (see robustness check in section 6), the cash payments are possibly diluted with other methods of payments. Some of the observations flagged as cash payments could have been partially paid with other methods. Therefore the analysis could be comparing full equity payments with mostly equity payments and a small cash payment. This could explain the lack of differences between the cash payments with debt or internal financing and equity payments. For the reason that some cash payments are financed with more than one source this could also explain that this analysis did not find a difference in takeover price between all of the different financing methods.

Another limitation of this paper is that it only uses the takeover premium as the dependent variable in the analyses. The takeover premium is in the interest of the target’s shareholders, it is the premium they receive over the current value of their shares. While this premium is paid for with the money of the acquirer’s shareholders it is not necessary something negative for them. As long as their profit of the merger or acquisition is larger than the cost of the takeover they are satisfied. Therefore another dependent variable that could have been used is the bidder’s abnormal announcement return, this would be in line with the papers of Schlingemann (2004) and Vladimirov (2014).

Due to the availability of data, only mergers and acquisitions where both the target and the acquirer are public are used in the sample. Therefore it is unclear what the implications of the findings are for private companies.

According to the findings of this paper, companies that pay for mergers and acquisitions with cash financed by selling equity pay the lowest takeover prices. However this does not mean that they make the best deals. Selling equity is the most expensive financing method (see section 2.2.3.) and therefore it is unclear what the return of the acquirer is. There are a lot of other factors that play an import in determining the optimal method of payment and financing method. For instance, the current debt level and cash constraints. If the acquirer is cash constraint and unable to obtain debt financing, the conclusion that can be drawn from this paper is that it is more profitable to use an indirect equity payment than a direct equity payment (assuming that the costs of the additional equity issuance is equal in both cases).

27 Lastly, there are some directions for future research. The first direction is in line with the test in the robustness check. It is the same test but only with fully cash payments with only one method of financing. Perhaps it is possible to find more observations with the use of other databases than Thomson One and CRSP. This will solve the problem of the diluted observations and results in findings that are more in line with the related literature. Another direction for future research could be with the use of a different dependent variable instead of the takeover premium. This variable should also reflect the return for the target’s shareholders like the takeover premium. The target’s -1/1 BHAR (the buy and hold abnormal return for the period of 1 day prior to the announcement date till 1 day after the announcement date) is a possibility. Next to that, finding good instruments for the methods of payments and methods of financing to use in an IV-regression is a good topic for future research. This could solve the problem with common factors between the methods of payment and reduce the endogeneity problem.

The last suggestion for future research is related to the control variables. In both this paper as in other literature it is clear that the toehold has an effect on the takeover price. It might be useful to investigate this effect more, perhaps with instead of a toehold dummy the percentage of toehold. In other words what is the effect on the takeover price when the acquirer has 1% additional toehold in the target company.

28

References

Berk, J. B., DeMarzo, P. M., Harford, J. V., Sarkar, S., Bhanot, K., & Stuart, D. (2009). Fundamentals of corporate finance. Upper Saddle River, NJ: Pearson Prentice Hall.

Betton, S. E. Eckbo, and K. Thorburn, 2008. Corporate takeovers. In: Handbook of Corporate Finance: Empirical Corporate Finance, Amsterdam: Elsevier/ North-Holland.

Boateng, A., & Bi, X. (2013). Acquirer Characteristics and Method of Payment: Evidence from Chinese Mergers and Acquisitions. Managerial and Decision Economics.

de La Bruslerie, H. (2013). Crossing takeover premiums and mix of payment: An empirical test of contractual setting in M&A transactions. Journal of Banking & Finance, 37(6), 2106-2123.

Faccio, M., & Masulis, R. W. (2005). The choice of payment method 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. Journal of Finance 44, 41-57.

Frank, M. Z., & Goyal, V. K. (2003). Testing the pecking order theory of capital structure. Journal of financial economics, 67(2), 217-248.

Hansen, R. G. (1987). A theory for the choice of exchange medium in mergers and acquisitions. Journal of Business 60, 75-95.

Hennessy, C. A. (2004). Tobin's Q, debt overhang, and investment. The Journal of Finance, 59(4), 1717-1742.

Jensen, M. C. (1986). Agency costs of free cash flow, corporate finance and takeover, American Economic Review 76, 323-329.

Martin, K. J. (1996). The method of payment in corporate acquisitions, investment opportunities, and management ownership. The Journal of finance, 51(4), 1227-1246.

Moeller, S. B., Schlingemann, F. P., & Stulz, R. M. (2005). Wealth destruction on a massive scale? A study of acquiring‐firm returns in the recent merger wave. The Journal of Finance, 60(2), 757-782.

Myers, S. C. (1984). The capital structure puzzle. The journal of finance, 39(3), 574-592.

Myers, S. C., & Majluf, N. S. (1984). Corporate financing and investment decisions when firms have information that investors do not have. Journal of financial economics, 13(2), 187-221.

29 Pinkowitz, L., Sturgess, J., & Williamson, R. (2013). Do cash stockpiles fuel cash acquisitions?. Journal of Corporate Finance, 23, 128-149.

Schlingemann, F. P. (2004). Financing decisions and bidder gains. Journal of Corporate Finance, 10(5), 683-701.

Shleifer, A., & Vishny, R. W. (2003). Stock market driven acquisitions. Journal of financial Economics, 70(3), 295-311.

Stulz, R.M., 1990. Managerial discretion and optimal financing policies. Journal of Financial Economics 26, 3– 27.

Vladimirov, V. N. (2014). Financing bidders in takeover contests. Available at SSRN 2023576.