Financial constraints and mergers and acquisitions : do financial constraints affect M&A significantly before and

(1)

Financial Constraints and Mergers

and Acquisitions

Do financial constraints affect M&A significantly before and

during the financial crisis?

L.F. de Jong - 6066887

February, 2014

University of Amsterdam

BSc in Business Economics, field Finance & Organization Supervisor: Dr. J.E. Ligterink

(2)

2

Abstract

This empirical study investigates whether financial constraints are of great influence on mergers and acquisitions. This study is based on mergers and acquisitions of NYSE, NASDAQ or AMEX listed companies within two periods: the first period from September 2003 until December 2006 and the second (financial crisis) period from September 2008 until December 2011. The data are from US companies only and from both periods the 30 largest M&A are taken into consideration. Characteristics such as the attitude of the takeover, the payment method and whether there are financing constraints are supplemented in the data. Financial constraints are determined by comparing -and combining- different methods in previous literature. The most important findings of this study confirm that the cumulative abnormal returns for target companies are affected significantly when a takeover is

announced. The announcement effects are 11.88% on average in the pre-crisis period and 17.79% in the crisis period for target firms. The coefficient of financial constraints is positive and the variable is found of significant value. Financially constrained firms thus have positive and substantial effect on the dependent variable. In other words, the cumulative abnormal returns are higher for financially constrained targets than for target firms that are not considered financially constrained in the announcement period. Besides this, there is strong evidence that the crisis period and payment method are of significant influence on the value of target firms in case of M&A.

(3)

3

1. Introduction

1.1 Background

M&A activity increased enormously during the past decades. It is assumed in many studies that M&A’s create value, especially for the targeted firms. The main motives to undertake M&A’s are to create synergies and solve agency problems and conflicts (Grinblatt and Titman, 1998). Agency problems and conflicts arise due to conflicts about the payout of free cash flow between managers and shareholders. Both parties are self-interested and there are conflicts between them over the choice of the best corporate strategy. Takeovers can reduce the agency costs and therefore it is considered as an important cause of M&A activity (Jensen, 1988). Besides this, both Jensen (1988) and Blasko et al. (2000) state that factors such as deregulation, taxes and increasing globalization of U.S. markets are forces that drive takeovers. However, M&A can also destroy value and often this is the consequence of the agency problem. Mergers can be pursued by managers because of motivations other than the ones in the interest of the shareholders (Blasko et al., 2000). Empirical evidence show that the

average M&A transaction creates value around the announcement date for the target

company’s shareholders, create negative synergies for shareholders of the acquiring company, and break-even, or earn a small positive abnormal return when evaluating the combined company after the merger (Blasko et al., 2000). Martynova and Renneboog (2006) have drawn the same conclusion from their study about M&A’s in Europe. From this you may wonder why companies continuously undertake M&A.

Gorton et al. (2009) have an explanation for this. In their eat-or-be-eaten theory they describe that firms not necessarily undertake M&A to create value, but to prevent being acquired by another firm. This can result in a takeover wave. The latest wave counts from 2003 to 2011 which includes the financial crisis. The period before the financial crisis (from 2004) was a booming period and since 2007 a bust period began. In the fall of 2008 the financial crisis became a worldwide fact. Since the crisis arrived, it has been the subject of numerous studies and books and is considered by many economists as the worst financial crisis since the Great Depression of the 1930s. Large financial institutions collapsed, the housing market suffered and consumer wealth declined. It isn’t mind-blowing to state that M&A activities could have suffered from the financial crisis. Many firms are closer to bankruptcy in the financial crisis and external financing gets more expensive. Financially constrained companies don’t have the financial possibilities that unconstrained firms do have. Already a lot of research has been done in the field of M&A but not so much relating it to the current financial crisis and the problem of financial constraints. I wanted to make my research a subject of the current environment and involve the financial crisis in it. This study analyzes M&A activity and value that is affected by financial constraints. Khatami (2011) did a research in financial constraints and M&A as well. As funding limitations lead to investments in the most profitable opportunities, bidding constrained companies acquisitions are more profitable. However the focus in this study is on target firms. The expectation is that acquiring

financially constrained firms are more profitable as well. The motivation behind this is that financially constrained target firms don’t have all the funding available to invest in profitable projects. They have unexploited opportunities and therefore more value can be generated.

(6)

6 Constrained target firms should be more valuable and according to Khatami (2011), acquiring constrained target firms offer higher premiums than for unconstrained companies.

In this study, two periods are compared to each other, to see if there has been a shift due to financial constraints in M&A. The first period is a non-crisis period from 2003 – 2006 and the second period contains the financial crisis, from 2008 – 2011. This challenged me to research the following question:

Do financial constraints affect M&A significantly before and during the financial crisis?

Apart from the main research question, stated above, there are some relevant sub questions to be answered. To form a fundamental base, with respect to answering the main question, the following questions need to be studied as well:

1. How to determine financial constraints?

2. Are there other important factors that have a substantial effect on M&A in crisis period?

It is interesting however that the relationship between financial constraints of companies and various factors of M&A’s have not been extensively explored in previous studies, purely for target firms. This study provides another insight in the field of M&A, as it only focusses on

target firms in takeovers. Previous literature mainly discusses bidding firms, but here the

focus is on target firms and relating their financial constraints to cumulative abnormal returns during the announcement period. Specifically, this study analyzes whether financial

constraints are of significant influence on target firms in case of an M&A. Only non-financial, large U.S. firms are taken into consideration in this paper whereas in previous studies often financial companies and/or smaller firms are discussed. Some of the important features, such as deal attitude, payment method and firm size, that are taken into account in other studies are included in this study as well.

The content of this paper is as follows. The second section – after this introduction section – contains a literature review in which the most influential literature on mergers and

acquisitions (gains), financial constraints, financial crises and the linkage among these topics are addressed. The literature review acts as a basis, on which the hypotheses are formulated. The hypotheses and the used model are described in the third section. After this, the outline of the retrieved data and the methodology of the research used is discussed in the fourth section. The different variables of the regression model are explained as well. In section five the empirical results are provided and will be analyzed and discussed. Besides this, the answer to the research question is formulated. Finally the conclusion addresses the most important findings of this research and some remarks will be made for future research about this topic.

(7)

7

2. Literature Review

This literature review will outline research that in my opinion is most important to the subject I analyze. It will provide a linkage between mergers and acquisitions and financial

constraints, due to the financial crisis of the past decade. The most important features of M&A are discussed. Then financial constraints are outlined, in which different methods of measurement will be addressed. Next financial crises are defined and the main focus is on the latest (current) financial crisis. This theoretical framework should be a solid basis for

understanding the relevance of the research question. 2.1 Mergers and Acquisitions

Mergers and acquisitions consist of two primary parties; the acquiring firm and the target firm. Besides these two parties there are multiple stakeholders such as banks, advisors, suppliers, customers, employees and so on. An acquisition is a purchase of the bidding firm to the target firm and the target firm legally ends to exist. A merger happens when two firms form a new company and continue on being that one company. In that case new stock is issued. In the literature there can be found a large amount of empirical evidence on M&A. One example of such an empirical paper is the work of Martynova and Renneboog (2006). Their work is of very high relevance to this research, as they describe very important features of mergers and acquisitions over the last decade, in Europe. Empirical evidence of event studies show that the average M&A transaction creates value for the target companies around M&A announcements (Khatami, 2011). However, M&A transactions destroy value for the acquiring company, and break-even when evaluating the combined company after the merger (Blasko et al., 2000). In the figure below it shows that in the two periods of interest for this paper, the opposite happens for the value of the M&A deals in the U.S. In the first period from 2003-2006, both the deal-value and the number of M&A rise substantially as for the second period from 2008-2011 the deal-value and the number of M&A deals collapse and begin to rise slightly again. This could be due to the fact that the financial crisis had its arrival in the second period. However, the number of M&A deals still was of a substantial amount in 2011 in the U.S.: over 10,000 mergers, as can be drawn from the figure on the next page.

2.1.1 Merger Waves

The 1990s produced the greatest wave of mergers in the U.S. history according to Gorton, Kahl and Rosen (2009). Between the period of 1995-2000, U.S. M&A raised every year and hit records. However, in 2002 M&A had dropped to nearly 500 billion U.S. dollars and after that it increased again until 2008 arrived (see Figure 1 on the next page). As the importance and growth of mergers and acquisitions increased over the years, more research was done. Gorton et al. (2009) describes two reasons that can give rise to merger waves. The first consists of an eat-or-be-eaten theory, in which they explain that defensive takeovers can create a chain of M&A transactions. The defensive acquisition of one firm makes other target firms more vulnerable and therefore the target firm will defensively acquire other target firms. The other explanation of the origin of a merger wave has a ‘positioning’ explanation. To increase firm size one company acquirers another and the anticipation of potential M&A

(8)

8 creates a race between firms, which causes the ‘wave’. Over the history of M&A activity, six large takeover waves were defined (Martynova and Renneboog, 2006). The first wave is of the 1900s, the second of the 1920s, the third of the 1960s, the fourth of the 1980s, the fifth of the 1990s and the last one took place from 2003-2007. Only this last wave was characterized by something else: the financial crisis. Due to this crisis the sixth wave was divided in a boom and a bust period; the two periods taken into consideration in this paper. There is a trend to discover relating the waves to economical seasons. The first wave arose after the Depression of the 19th century and the second wave took place in economic prosperity until the Great Depression. The third and the fourth wave were in a period of economic growth and the fifth wave as well, after the economic recession of 1990-1991. The last wave, from 2003-2007 again was in a period of economic prosperity until the financial crisis arrived. From this it can be stated that merger waves do not move together with periods of economic downturn. Considering that in periods of economic downturn more companies face financial constraints, merger waves have no clear relation with financial constraints.

Figure 1 – Number of U.S. M&A deals and their value [Source: www.imaa-institute.org] 2.1.2 Motives for M&A

As can be read in the previous part, Gorton et al. (2009) conclude two reasons for merger waves, which are motives to undertake M&A. The defensive and positioning techniques are used to increase firm size and to reduce the risk from being acquired by another firm. Managers in this sense carry out acquisitions to secure their own position and private benefits. Roll (1986) investigates in his article why firms undertake M&A activities when there is no – or little – value added after the transaction. Irrational behavior is an explanation for this. He describes how managers want to perform well and be a part of a takeover in their career and therefore act in their own benefit. Grinblatt and Titman (1998) identified six other sources of gains from mergers, which also can be seen as motives:

(9)

9  Operating synergies: when two firms combine their resources, they can reduce their

costs and increase their economies of scale.

 Tax motivations: two firms combined can create opportunities to reduce tax liabilities.  Mispricing motivations: this is the case when there is asymmetric information. For

example the bidding firm believes that the target firm is undervalued because of (wrong) information signals.

 Market-power: by taking over another company, one can gain monopoly power.  Disciplinary takeovers: when there are inefficient managers in the target company,

more efficient managers can be placed after the takeover to increase firm value and make better use of the firm’s resources.

 Earnings-diversification motivations: bankruptcy can be prevented by this activity, as diversifying the firm’s activities can generate more cash flows, while maintaining the same level of total risk.

2.1.3 Measurements of Profitability for M&A’s

Bruner (2002) sums up four approaches to measure M&A profitability. I will discuss these methods briefly in order to create some background information on the proxies for M&A profitability:

 Event studies are about examining the abnormal returns around an announcement date of a takeover. Abnormal returns are calculated by detracting a benchmark return (here called the normal return (NR)) from the return (R). Often a large market index is used as a benchmark, such as NYSE, NASDAQ or S&P500 (in the U.S.). Another common measure is the CAR: Cumulative Abnormal Returns, which determines the

profitability of shareholders based on abnormal returns. It is the summation of AR. R is measured during the event window, including the announcement day and the NR is measured in the estimation period before the announcement day (De Jong, 2007).

ARi,t = Ri,t – NRi,t

This approach has dominated since the 1970s.

 Accounting studies examine the reported financial results of an acquiring firm before and after the takeover to see how their financial performances have changed. Key indicators are net income, ROE (return on equity) or ROA (return on assets), EPS (earnings per share) and leverage and liquidity of the firm.

 Surveys of executives are seen as very simple and usable instruments. Managers are asked if they think value is created, on a standardized survey. By aggregating the results it can be generalized from the sample.

 Clinical studies are studies of very intensive research. Only one or a couple of small samples are taken into account and are extensively studied. This in-depth research approach is very time consuming but gives good insights about one transaction. In this study I chose to perform an event study, as this is the most common way to examine the returns from a drawn sample in case of M&A’s. Empirical papers from Khatami (2011) and Martynova and Renneboog (2006) are based on event studies as well. The method of an event study is described more briefly in the Methodology section.

(10)

10 2.2 Financial Constraints

After defining and discussing M&A’s in the previous section, financial constraints are briefly described next. The most important questions are what the definition of financial constraints is and how to proxy a financially constrained firm.

2.2.1 Definition

Capital market imperfections can generate financing frictions, which in turn can increase the cost of external financing (Alshwer et al., 2011). These constraints can have substantial effects on, for example, a major investment decision or capital structure choices. A firm is considered financially constrained if it is not able to invest at its optimal level (Khatami, 2011). This could be due to capital market frictions or inability to raise capital to invest in growth opportunities. Alshwer, Sibilkov and Zaiats (2011) define financial constraints as frictions generated by capital market imperfections that can increase the cost of external financing, which could lead to underinvestment by some firms. Firms that are more likely to be financially constrained are smaller firms, private firms, less profitable firms, firms that are less likely to pay dividends and firms with slightly lower growth prospects according to Campello, Graham and Harvey (2009). It seems logical that firms that are smaller, have less capital available to invest in opportunities for them to grow.

2.2.2 Proxies for Financial Constraints

In previous literature researchers are often in need of a measure to determine financial

constraints. As can be imagined, many methods can be found and the most important ones are discussed here in order to create some knowledge about measurement mechanisms for

financial constraintness. The different measures can take two different values: continuous or discrete. The described approaches below can be considered as measures with continuous values. This means it is possible to categorize the firms based on the median value of the measure into one of three or four tertiles. For example, firms with payout ratios in the bottom tertile are defined as financially constrained.

 Payout Ratio

The payout ratio is defined as the sum of dividends plus stock repurchases divided by operating income. The companies can be ranked based on the payout ratio and assigned in the bottom three tertiles of the annual cash payout ratio distribution of the sample firms to the financially constrained group (Alshwer et al, 2011). The observations with a positive payout and zero or negative cash flow earn the highest payout ratio scores. Fazzari et al. (1988) use this ratio in their paper. They argue that firms that are financially constrained have lower payout ratios. This is the consequence of the preference of constrained firms to pay lower dividends and retain the ability to finance their growth options. Following this, one can say that constrained firms prefer to hold more cash on average.

 Kaplan and Zingales Measure (KZ Measure)

Kaplan and Zingales ranked firms as constrained based on information from the financial statements. This measure was hardly applicable in large samples.

(11)

11 However, years later the measure was adopted by many other researchers and inserted variables from Compustat and formulated the ‘new’ KZ index as follows (Khatami, 2011):

KZ = (-1.001909*CashFlow) + (0.2826389*Tobin’sQ) + (3.139193*Lev) – (39.3678*Div) – (1.314759*CashHolding)

 Firm Size

Firm size is measured by taking the natural log of book value of total assets. It is assumed that larger firms are less financially constrained. Dummy variables can be created to categorize different groups based on financial constraints. Firms above the sample median are labeled financially unconstrained and firms below the sample median are labeled as financially constrained (Khatami, 2011).

 Whited and Wu Measure (WW Measure)

Whited and Wu created a method based on structural method of investment. Here, Compustat variables are used as well. The WW measure is a linear combination of six factors: cash flow, a dividend payer dummy, leverage, firm size, industry sales growth and firm sales growth (Hadlock & Pierce, 2009).

WW = (-0.091*CashFlow) – (0.062*DivPos) + (0.021*LTLev) – (0.044*LNTA) + (0.102*ISG) – 0.035*SG)

 Hadlock and Pierce Measure (HP Measure)

Since Hadlock and Pierce found that all financial constraints measures had it

shortcomings, they exploited the Cash Flow Sensitivity of Cash approach of Almeida et al. and the Kaplan and Zingales Measure (Hadlock & Pierce, 2009). Besides this, their measure is mainly based on firm size and firm age (SA). They use qualitative information and statements of executives to categorize firms in the constrained or unconstrained group. Consequently they examined the various characteristics of the firms between the two groups to discover a relation. Their index is as follows:

HP = (-0.737*Size) + (0.043*Size²) – (0.040*Age)

 Leverage ratios

It is stated in multiple researches that leverage is an indicator of the financial condition of the firm. High usage of debt refers to constraints in financing projects. Two ratios of leverage are the debt-to-assets ratio and the debt-to-equity ratio.

The above stated measurement methods for constraints are only a handful of all the methods available. Each method relies on certain empirical evidence and theoretical assumptions that may be valid for one sample, but not for the other. In Appendix I an overview can be found from different measures used for financial constraints and developed over the years from different researchers (see Table 1). For each measure there is a statement to make. Taken all the above measures into consideration to be part of my research, I have decided use the payout ratio and the leverage ratio (debt). Because of the strong evidence on holding cash (in multiple researches) and leverage levels in constrained firms it seems to be the most

trustworthy methods of financial constraints. This is discussed more thoroughly in the Methodology.

(12)

12

2.2.3 Previous Literature on Financial Constraints

Financially constrained firms ‘save’ more money in case of financial adversity. This is confirmed by Almeida, Campello and Weisbach (2004). In their paper they find that financially constrained firms hold on average more cash than unconstrained firms. From Fazzari et al. (1988) it follows that financially constrained firms have significantly lower payout ratios than unconstrained firms. Hadlock and Pierce found in their last research (2009) serious doubt on the validity of the Kaplan and Zingales Measure. Besides this they found that the Size measure and Age measure are particularly useful measurement methods of financial constraints. These two variables appeared to be closely related to financial

constraints in their research. In addition to this, prior research has also found that size and age of a firm are strong predictors of constraints in multiple settings. As the WW index

confirmed, smaller firms are more likely to be constrained and alternatively, size does matter. Khatami (2011) states that financially constrained bidders are more profitable as they are limited in their funding and therefore invest in the most profitable opportunities. Besides this, acquiring companies that are financially constrained would lead to more profits as it is likely that more resources are unexploited. Because of this, more value can be created and Khatami found that bidders offer higher premiums for acquiring financially constrained companies than for unconstrained companies. He also states that the most profitable acquisitions are those where the bidder firm, as well as the target firm are both financially constrained.

2.3 Financial Crises

It is important for this research to explain what a crisis exactly is, when there can be spoken of a crisis and the determination of the ‘beginning’ of the crisis. This subchapter is important because of different definitions of crises in previous literature. Some background information about this topic is of interest to link financial constraints with this. First of all, multiple definitions are given and subsequently the history of the past crises is given. After that there is paid some extra attention to the last (or current) financial crisis, as that crisis is of

importance to this paper.

2.3.1 Definitions

Reinhart and Rogoff (2010) define different sorts of crises. First of all an inflation crisis is described as a year where inflation exceeds over twenty percent. Besides this they define a currency crash as another type of crisis. A currency crash is an annual depreciation of the currency greater than or equal to fifteen percent per year. To date to beginning of a currency crash, exchange rate depreciation is used as an indicator. However it is often seen that a combination of the inflation crisis and a currency crash happen at the same time and thus move together.

A debt crisis is considered the third type of crisis. These crises are seen as rather dramatic periods, as external debt is subject to immediate default on payment of obligations. This could be due to foreign legal jurisdiction for example. The time the crisis can be assigned starting point is the date of the default.

The fourth type of crisis Reinhart and Rogoff describe, is a banking crisis. They mark a banking crisis by two types of events. The first is a run on banks which leads to a closure, a

(13)

13 merger or a takeover by the government. Panic arises among the public and many people withdrawal their accounts. Changes in deposits could be used to date such a crisis. The second event is when there is no run, however the bank is still in need of closure, merging or taken over by the government for financial support. This often leads to more similar cases in the same period. This could be due to a collapse in real estate prices or increased

bankruptcies in the non-financial sector. Changes in the amount of bankruptcies or substantial increase of nonperforming loans can be used as an indicator to date this type of crisis.

2.3.2 History

Besides the fact that M&A come and go in waves, financial crises also can be categorized in waves. Kindleberger and Aliber (2011) describe in their book four different crisis-waves. The first wave started in the early 1980s when developing countries, including Mexico, Brazil and Argentina, defaulted on U.S. dollar denominated loans. The second wave found its beginning in the early 1990s where an implosion of real estate and stock bubbles led to a massive failure of banks in Japan. Consequently Finland, Norway and Sweden had to suffer from this

collapse as well. The third wave began mid-1997 and is now known as the Asian Financial Crisis. Thailand was kept in an asset price bubble and the currency value decreased

enormously. Also Malaysia and Indonesia got a taste of this implosion and not much later South Korea and Russia among some other countries had to coop with this recession as well. The fourth and last wave started in 2007 where prices of real estate in the U.S. began to decline. Besides the U.S., this was also the case for the U.K., Spain and Iceland. Following that bond prices began to fall hard in Spain, Greece and Portugal.

Studying the history of financial crises, Kindleberger and Aliber (2011) found a trend. Prior to every crisis there is a so-called credit bubble. Rapid growths in loans as borrowers want more lending to finance real estate purchases. However, if prices in real estate fall and stock prices decline, also currencies depreciate. Consequently, borrowers don’t have enough cash to pay outstanding loans and therefore take new loans. Due to their indebtedness an

implosion occurs. Not only the borrowers loose, but also the lenders loose.

In literature there is not really an agreement on what the biggest financial crisis is so far. Many state the Great Depression after the Wall Street Crash in 1929 as the biggest crisis in history. Others dispute whether the Great Recession of 2008 dominates the Great Depression. Besides these two large crises, some other big crises are note-worthy as well in this section: the Oil Crisis in 1973, the Russian Crisis in 1998, the recession of Japan from 1990-2000 and the Asian Financial Crisis in 1997.

There is also a ‘model’ that is applicable for every crisis (Kindleberger et al., 2011). Prior to each crisis in history the same factors arose, like changes in supply of credit. This

pro-cyclical phenomenon increased when there was a booming period and decreased in case of an economic downturn. In a booming period, optimistic investors are more eager to borrow and financial institutions are more willing to lend and pay less attention to risk. When a downturn arrives the investors become less optimistic and more cautious. Presumably lenders become more cautious as well as loan losses increase and their capital declines. This sketch is applicable for most crises of the past era and also for the financial crisis of 2008.

(14)

14

2.3.3 Last Financial Crisis

As the first decade of the 21st century progressed, M&A activity was growing to massive proportions. However, during 2007 there were different factors that could have functioned as signals of the crisis. For example lending slowed down and as sources of financing dried up, M&A activity began to slow as well (Grave, Vardiabasis & Yavas, 2012). This could be seen as a pre-crisis period. In the book The Financial Crisis Inquiry Report (2011) is described thoroughly why the ‘actual’ beginning of the crisis was in the fall of 2008. Some state that 2007 was the year the last crisis found its origin. Ivashina and Scharfstein (2010) found in their research that in the beginning of 2007 new lending (of substantial proportions) was provided to companies and public to finance real investment and for restructuring purposes. In the figure below one can find the amount of lending in the U.S. in 2007 and 2008. During 2007 there arose a credit boom, followed by a meltdown of mortgages and other securitized products. This set the seed for some panic already. However in the fall of 2008, the crisis fully hit the economic market, resulting in global economic instability and a crashed credit market. Saqib et al. (2013) state that the 2008 financial crisis represents probably the most dramatic event that the financial system underwent since the Great Depression. To be more precise, the crisis reached enormous proportions in September 2008 with the failure of the U.S. banks Lehman Brothers and Washington Mutual and the collapse of large insurance companies such as American International Group. After this, panic arose by a lack of transparency of the balance sheets of major financial institutions. A run arose by short-term bank creditors creating a difficult position for the banks. Lending was suddenly cut. In the fourth quarter of 2008 new loans to large borrowers fell by 47% relative to the third quarter (Ivashina & Scharfstein, 2010). Following this, credit markets seized up and trading came to a halt. The stock market plummeted and the economy plunged into a deep recession. According to Reinhart and Rogoff (2010), this financial crisis can be categorized into the ‘banking crises’. To this day, the economic environment is still suffering from the financial crisis.

(15)

15

3. Hypotheses and Model

A literature review and a theoretical framework that examines mergers and acquisitions and financial constraints are used to formulate the hypotheses for this paper. Khatami (2011) documents in his paper significantly higher announcement abnormal returns to the financially constrained bidders compared to the unconstrained ones. This is in line with the limitation of investment choices of the constrained firms to the most profitable ones, as mentioned before. However, investigation on financial constraints in target firms only is hard to find. It seems less interesting to researchers. The expectation on the announcement returns of the target companies is positive as well as Khatami (2011) stated that returns are highest in case when bidding and targeted firms are financially constrained.

Furthermore the expectation is that there are more financial constraints in periods of financial crisis and that this has a relation with the cumulative abnormal returns of M&A.

3.1 Hypotheses

For all the hypotheses the tests are used only in relation to the target firms, as that is the focus in this paper. The first hypothesis is created to test whether M&A have a positive influence on the event window Cumulative Abnormal Returns. I expect the CAR to be positive during the announcement, as the target firms can be overestimated on the future efficiency of the combined value. This is in line with previous empirical papers (Martynova et. al, 2006 and Blasko et. al, 2000). The announcement period, or the event window, is stated to be 7 days prior to the announcement date and one week after it.

H1: M&A cause positive CAR for the target firm in the announcement period

The second hypothesis is stated to examine when a firm is financially constrained, this affects the CAR in the announcement period. In the paper of Khatami (2011) it is found that

constrained bidder firms realize similar abnormal returns around the announcement as unconstrained firms. That would mean that H0 in this case would not be rejected. However, I expect that financially constrained firms have less money available and thus are more critical towards their investment (M&A). Because of this well-thought takeover this can result in a higher CAR for financially constrained firms.

H2: Financial constraints have a significant influence on CAR in the announcement period Quotation: H0: α1(t=0) = 0 versus H1: α1(t=0) ≠ 0

The third hypothesis is created to whether financially constrained firms generate higher CARs in the crisis period than in the non-crisis period. This is based on the theory of Khatami (2011). Firms that are financially constrained have unexploited opportunities and are more cautious towards their investment choice. The expectation is that in periods of crisis firms in general are more careful when considering an investment such as an M&A. Because of this

(16)

16 cautious consideration and investing the best opportunity possible, more value can be

generated. This is formulated as follows:

H3: Financially constrained target firms generate higher CARs in the crisis period than in the

non-crisis period

3.2 Theoretical Framework

In order to make a clear distinction between the variables in this thesis, I created a simple theoretical framework. The financial constraints variable is the independent variable in the regression and is the point of focus. Together with the control variables it is tested what the effect is on M&A profitability, which is measured with the cumulative abnormal returns.

(17)

17

4. Methodology

In this paper an event-study methodology is employed, using the market model to measure the stock price effects associated with the announcements of acquisitions. First of all there is a brief description of an event study. Next (the retrieving of) the data for both M&A and financing constraints are explained. After this the variables of the regression model and the regression model itself are described in detail.

4.1 Event Study

De Jong (2007) states that an event study tests the significance of abnormal stock returns around event dates. An event study is chosen here, to test the hypotheses of the last section. It examines the abnormal returns to shareholders around the announcement date of the M&A deals. An event study is conducted based on three steps (De Jong, 2007):

 Identify the event of interest and in particular the timing of the event;  Specify a benchmark model for normal stock return behavior;

 Calculate and analyze abnormal returns around the event date.

The format of the event and estimation window that is used in this research looks as follows:

Source: De Jong, 2007.

4.1.1 Identify the Event

This first step is to define the data when the news of the event (the takeover) becomes public. In this paper, this is the case when M&A is announced according to Thomson One database. The event date is defined as t = 0. However, not only the announcement date itself, but also the period around this date is of importance. News about M&A can spread quickly and is sometimes anticipated on, therefore the Event Window [t1 , t2] is created.

4.1.2 Specify Normal Returns

For the calculation of the normal return, first there must be an Estimation Window defined, denoted as [T1 , T2]. This Estimation Window is the estimation period before the Event

Window. The stock return during the Estimation Window is considered to be the normal stock return. Martynova and Renneboog (2006) used an estimation period of 240 days, as is the case in this paper. The Estimation Window can be denoted as [-300 , -60], which means that the estimation period starts from 300 days until 60 days prior to the event date. The notation for the Event Window is as follows: [-7 , 7] – meaning one week prior to the announcement date and one week after it.

(18)

18 There is a substantial amount of time needed between the two periods, to make sure that the event has little influence as possible on the estimation period. The normal return of stock NRi

for each targeted company is calculated using the following formula:

This formula defines the normal return as the average return over the Estimation Window, where i is the stock index and T = (T2 – T1 + 1) which stands for the number of days in the estimation period.

Then the daily stock return is needed, so the normal stock return can be calculated. For the daily return, the following formula is used:

where Ri,t is the stock return for day t and stock i, and Pi,t is the stock price for day t and

stock i, both for target companies.

4.1.3 Calculate and Analyze Abnormal Returns

The abnormal return, as stated earlier in this paper, is the difference between the actual return and the normal return. In formula: The ARs are calculated as the actual returns minus the returns for the value-weighted index of NYSE, AMEX and/or NASDAQ stocks:

where ARi,t is the abnormal return of stock i on day t and Ri,t and NRi,t are defined in the

previous subparagraph.

Since there is dealt with a time interval here, the abnormal returns are aggregated into the cumulative abnormal returns:

Eventus is used to create an event study and to retrieve the returns and the returns of the market model. Using the returns the CAR of the estimation window is calculated in Eventus; a list of every firm’s CAR can be found in Appendix II.

(19)

19 4.2 Data and Sample

4.2.1 Data – M&A

The two samples consists of U.S. public firms that announced and completed an acquisition between the period of 1 September 2003 – 31 December 2006 and the crisis period of 1 September 2008 – 31 December 2011. The acquisitions were selected from the Mergers and Acquisition Database of Thomson One. The deals were selected based on the following criteria:

- The acquirer and target are listed on NYSE, NASDAQ or AMEX; - Financial and accounting data available for the target company; - Both firms in the deal are independent companies;

- Transaction value of the deal lies above $500 million dollar; - Bidder owns 100% of the shares after acquisition;

- No financial and/or regulated firms are allowed in the sample.

The required financial data and share prices are identified from WRDS. Methods of payments are retrieved from Thomson One: only deals that are fully paid in cash, fully paid with shares or a mix of both are taken into consideration. Besides this, also the industry in which the firms operate, and whether the takeover deal was friendly or hostile are retrieved from Thomson One. The thirty largest M&A deals (based on deal value) are considered from each period. In Appendix I one can find the two lists of the takeover deals from each period.

4.2.2 Data – Financial Constraints

In order to define financial constraints for the target companies the following formulas are used:

where the total dividends are paid at the end of the fiscal year, one year prior to the announcement date of the takeover transaction. The EBIT is taken at the end of the fiscal year, one year prior to the announcement date as well.

The debt-to-assets ratio is calculated using the following formula:

where total debt consists of short-term and long-term debt, for the fiscal year one year prior to the transaction. Book values of debt are used because market values are harder to obtain. Total assets from the end of the fiscal year, one year prior to the announcement date of the transaction are used.

For every firm in both periods the dividend payout ratio and the debt-to-assets ratio are calculated. Firms above the total mean of the 60 companies of the debt-to-assets ratio are considered as the more financially constrained ones. Firms under the total mean of the 60 companies of the dividend payout ratio are also considered as more financially constrained,

(20)

20 because firms paying less dividend to its shareholders are considered to have less internal funding available.

Following this, a list is created (see Table 5 & 6 in Appendix I), where firms that are both above the median of the debt-to-assets ratio and under the mean of the dividend payout ratio are listed as financially constrained. The data that are used for these calculations are retrieved via Thomson One and WRDS and are dated one year prior to the announcement date.

Because of this, target firms are indicated as financially constrained before M&A was under discussion.

4.3 Variables

4.3.1 Dependent Variable

The measurement CAR will be used to measure the profitability of M&A. CAR stands for cumulative abnormal returns. The CAR determines the profitability for shareholders derived from the event study. Abnormal returns are based on stocks and therefore it gives a direct measure for value creation.

The used date ranges in this even study are the following:

- The estimation window that consists out of 300 to 60 days prior to the announcement day;

- One week prior to the announcement day; - The announcement day (t=0);

- One week after the announcement day; - The total event window.

(see the format on page 17).

4.3.2 Independent Variable

The key explanatory variable here is ‘financially constrained’ (FC) for targeted companies. This variable is created as a dummy and the description of the measurement is stated in section 4.2.2. It is coded 1 if the firm is considered financially constrained and 0 otherwise. As M&A are mostly large investments, the financial crisis should be of great impact as the expectation is that less capital is available for most companies. The FC variable is therefore of major importance. When dealing with financial constraints, a firm should be less likely to invest a great amount of money in a project, or it should be more cautious towards a great investment. The expectation expands to when a firm acquires a financially constrained firm, it will be more profitable. This would be the case due to unused resources or possibilities that are unexploited in the target firm. Besides this it seems likely to be highly correlated with the dummy variable Crisis. An overview of the financially constrained firms can be found in Appendix I.

(21)

21

4.3.3 Control Variables

In this paper there are a couple of independent variables that are used as control variables. X2 = Cross-variable of financial constraints and crisis period;

X3 = Cash-only; the deal is paid with cash-only, no stocks or mixture in the payment. It is a

dummy variable, taking a value when payment method is in cash; X4 = Firm size; the natural logarithm of firm size, based on total assets;

X5 = Crisis; a dummy variable taking a value in period of crisis (2008-2011);

X6 = Cross-variable of firm size and crisis period;

X7 = Attitude; taking a value when the deal attitude is hostile and friendly otherwise;

X8 = Industry; taking a value when acquiring and target firm are from the same industry;

X9 = Cross-variable of financial constraints and cash-payment.

4.4 Regression Analysis

The method used for this study is regression analysis. This is a method, which is based on obtaining data and used to formulate the relationship between the dependent variable and independent variables. In this paper, the dependent variable is the cumulative abnormal returns (CAR), and the independent variables and control variables are listed above. The CAR is calculated using Stata with the retrieved data from Thomson One and Eventus. For every firm the cumulative abnormal returns are calculated, in the event period.

The regression model used is formulated as follows:

CARi,t = α + α₁FC + α₂(FC · Crisis) + α₃Cash+ α₄LSizei + α₅Crisis + α_{6 (}LSizei · Crisis) + α7 Att + α8 Ind + α9 (FC · Cash) + ԑi

To start the regression analysis using Stata, the input data file need to be prepared. This Excel file contains the data of CAR, the Financial Constraint dummy, the Crisis dummy, the

Payment method dummy, Assets to determine size, the Attitude dummy and the Industry

dummy. The Excel sheet is converted to the format that Stata recognizes. Next the regression can be executed.

To test the significance on behalf of the variables towards the CAR the test is used. The t-test statistical formula is:

(22)

22

5. Interpretation of Results

5.1 Announcement Period Returns

The announcement period returns are measured from one week prior to the announcement day until one week post the acquisition. The cumulative abnormal returns are calculated for each company around the announcement date. Two lists are created which separates the firms in two periods (crisis versus non-crisis). Table 3 in Appendix II shows these lists with the

CARs in percentages. These numbers are retrieved from Eventus via WRDS database. The

mean of the cumulative abnormal returns is higher in the crisis period, namely 17.79%. In the non-crisis period the mean CAR is 11.88%. This means that M&A’s have a substantial

positive influence on the announcement day. For the financially constrained target firms (separated from the non-financially constrained firms) the mean CAR for both periods is calculated as well. The CARs are 25.12 % and 18.80% respectively for the non-crisis and crisis period. This can be found in Table 3 of Appendix II. The third hypothesis in this study stated that financially constrained firms generate higher CARs in the crisis period. According to the calculations this is not the case. Therefore this hypothesis is rejected. There can be concluded that the CARs are higher for financially constrained target firms on average.

5.2 Summary Statistics and Regression Analysis

In this study only target firms with a deal value of higher than $500 million are taken into consideration. However, the financial constraints variable is of high significant value as was retrieved in Stata with the data output of Thomson and Eventus. The t-statistic and p-value are 2.09 and 0.042 respectively which means that financial constraints have a great influence on the value of target firms in the event window. This can be seen in Table 4 in Appendix II. The correlation between the FC variable and the Crisis variable is small, but positive (see all the correlations in Table 5 in Appendix II). This means financial constraints move together with the crisis. Though it was not as high as was expected. Due to the crisis more firms have to deal with financial troubles as their profits decrease due to less consumer spending. Furthermore the Crisis variable has a sufficient positive correlation with the Industry

variable, namely 0.1400. In a rational way this could be explained that in crisis periods firms take less risk and like more safe options. To take over a firm that is active in the same

industry as the buyer, this means less risk than if it were in another industry. Besides the deal attitude seems to have a relation with the crisis as well. None of the firms in the sample of this study in the crisis period were hostile takeovers, whereas in the non-crisis period there were. This can also be explained by less risk taking, more safe options to invest in when acquiring a firm. A friendlier takeover seems logical in this scenario. This is confirmed by the negative correlation between the Crisis variable and the Attitude variable, which is -0.1857. In Stata every variable has been estimated twice, once separately and once together with all the other variables. Many of the p-values are greater than 0.05, which indicates that the coefficients are not significantly different from zero. This means that the variables would not relate to the valuation change of the company. However, the t-tests give a whole other indication; namely that the variables are significant. Still, the results are not as expected and

(23)

23 are considered not significant due to the large p-values. This means that there cannot be concluded that the crisis-period, the attitude towards the deal and the industry have an influence on the cumulative abnormal returns.

As can be seen in the Table 5 (Appendix II), we see that FC, Size and the Cash have the strongest correlations with the cumulative abnormal returns. These correlations are all positive, meaning that as the value of one variable goes up, the value of the other variable tends to go up as well. Knowing that these variables are strongly associated with CAR, the prediction is that they would be statistically significant predictor variables in the regression model. When the payment to finance the takeover exists fully of cash, the cumulative abnormal returns are higher (the positive correlation). This is in line with the theory of Alshwer et. al (2011) in which they explain that more cash holding in a firm points strongly towards a financially constrained company. Following this, financially constrained

companies in case of M&A earn higher CARs as the investment in a financially constrained target firm can create possibilities as some resources are still unexploited.

Now, recall the main research question of this study:

Do financial constraints affect M&A significantly before and during the financial crisis?

Taken all previous interpretations together, it can be stated that financial constraints do affect the value of the target firms when there can be spoken of an M&A. There is evidence that there are more firms financially constrained in the crisis period and that there is a small, but positive correlation between the financial constraints and the crisis. From this it can be concluded that financial constraints are of a (slightly) bigger influence on M&A value in periods of crisis.

5.3 Hypotheses Evaluation

The first hypothesis stated whether M&A cause positive CARs for target firms in the

announcement period. The findings of the event study found significant positive cumulative abnormal returns for both the crisis and the non-crisis period during the announcement period. This is in line with the findings of Khamami (2011), Martynova and Renneboog (2006) and Blasko et al. (2000).

The second hypothesis was stated to test whether financial constraints have a significant influence on CAR during the event window. This was calculated with the regression in Stata. As stated before, from the output regression it can be concluded that financial constraints have a highly significant influence on the value of the target company during the

announcement period. In other words, the null hypothesis must be rejected as there is enough statistical evidence that financial constraints have influence on the stock prices of target firms in case of an M&A.

(24)

24 The third hypothesis was formulated to test whether financially constrained target firms generate higher CARs in the crisis period than in the non-crisis period. In the event study retrieved from Eventus, the mean CAR for the non-crisis period during the announcement period was higher than the mean CAR for the crisis period (25.12% and 18.80% respectively – Table 3 in Appendix II). This was not anticipated on, as the expectation was the opposite. In this study a statistical significance of 5% is used and a p-value of 5% as well. In many empirical papers this is the most common benchmark to use. If the p-value falls below 5%, this means that the hypothesis must be rejected and that there is statistical significance. The opposite applies for t-statistic, which has to be higher than 5% to be significant – in case of a positive hypothesis.

(25)

25

6. Conclusions and Remarks

As shown in previous studies, M&A transactions create positive value for target companies (Martynova & Renneboog, 2006). This study found the same positive relation during the announcement period. Positive abnormal returns show trust by its shareholders. If

shareholders are positive about future earnings, their demand for a firm’s shares should also increase. This could lead to an increase in abnormal returns.

The key explanatory variable ‘financially constrained’ is of great influence with respect to the

CARs of the target companies. It has a significant positive relation with the cumulative

abnormal returns during the announcement period. Thus firms with financial constraints earn higher abnormal returns on the announcement day. A rational interpretation could be that when target firms have to coop with financial constraints, it means not all chances can be exploited. After announcing the takeover it means that those chances can be exploited and more resources are made available. Combined with positive attitude towards an M&A this results in higher demand for a firm’s stock and results in higher returns. This is all in line with firms that have higher leverage (more financially constrained) earn higher abnormal returns.

The most important factors that influence the CAR in the announcement period individually are the size of the firm (based on assets), and the payment method. The other variables are not significant when tested individually with respect to the CAR, as their p-values are higher than 0.05. Among the most important findings ought to be the correlation between the crisis and financial constraints. In the output it is found that these two variables move together in the same direction. This is confirmed by the fact that there are (slightly) more financially constrained companies in the crisis-period than in the non-crisis period. Besides this, the cumulative abnormal returns are higher in the period of the financial crisis. This could be due to well-thought investments and more careful actions taken in periods of economic downturn. There can be concluded that in period of crisis, the cumulative abnormal returns during the announcement period of M&A is higher and it seems that financial constraints play an important role in this.

For future research in this field it would be interesting to compare more companies in crisis periods and non-crisis periods. Besides this to make the investigation more reliable, other crisis-periods (than the 2008 financial crisis) should be taken into consideration to compare to other non-crisis periods. Since this sample includes only large, non-financial firms, there might be concluded there is not enough data to offset results of large firms against small firms. The same problem yields for the financial constraints, as large firms are less likely to be financially constrained than smaller firms. The sample should be drawn not based on deal value, but perhaps on other variables that can be of interest. Taken this into account, future research should constitute a more diverse sample that could lead to more significant results.

(26)

26

Reference List

Almeida, H., Campello, M., Weisbach, M.S. (2004). The Cash Flow Sensitivity of Cash. The Journal of Finance, Vol. LIX, No. 4.

Alshwer, A.A., Sibilkov, V., Zaiats, N. (2011). Financial Constraints and the Method of

Payment in Mergers and Acquisitions. JEL Classification: G31, G34. SSRNWorking Paper Series.

Blasko, M., Netter, J.M., and Sinkey, J.F. (2000). Value creation and challenges of an

international transaction: The DaimlerChrysler merger. International Review of

Financial Analysis, Vol. 9, pp. 77-102.

Bruner, R.F. (2002). Does M&A Pay? A Survey of Evidence for the Decision-Maker. The Journal of Applied Finance, Vol. 12, pp. 48-68.

Campello, M., Graham, J., and Harvey, C.R. (2009). The Real Effects of Financial

Constraints: Evidence from a Financial Crisis. Working Paper No. 15552, NBER

Working Paper Series.

Fazzari, S.M., Hubbard, R.G., Petersen, B.C., et al. (1988). Financing Constraints and

Corporate Investment. Brookings Papers on Economic Activity, Vol. 1988, No. 1, pp.

141-206.

Financial Crisis Inquiriy Commision (book), (2011). The Financial Crisis Inquiry Report:

Final Report of the National Commission on the causes of the Financial and Economic Crisis in the United States. Publisher: Washington, DC: Financial Crisis

Inquiry Commission.

Gorton, G., Kahl M., and Rosen R. (2009). Eat or be eaten: A theory of mergers and merger

waves. Working Paper, The Rodney L. White Center for Financial Research, Wharton

School.

Grave, K., Vardiabasis, D., and Yavas, B. (2012). The Global Financial Crisis and M&A. Journal of Business and Management, Vol. 7, No. 11.

Grinblatt, M., and Titman, S. (1998). Financial Market and Corporate Strategy. Boston:

Irwin/McGraw Hill.

Hadlock, C.J., and Pierce, J.R. (2009). New Evidence on Measuring Financial Constraints:

Moving Beyond the KZ Index. Review of Financial Studies, Forthcoming. JEL

Classification: G31, G32, D92.

Ivashina, V. and Scharfstein, D. (2010). Bank Lending during the Financial Crisis of 2008. Journal of Financial Economics, Vol. 97, Issue 3, pp 319-338.

(27)

27 Jensen, M.C. (1988). Takeovers: Their Causes and Consequences. Journal of Economic Perspectives, American Economic Association, Vol. 2, No. 1, pp 21-48.

Jong, de, F. (2007) Event Studies Methodology. Tilburg Univeristy. Retrieved from blackboard University of Amsterdam.

Khatami, H. (2011). Returns from M&As and Acquisition Premiums: The Case of Financial

Constraints. University of Manchester, Manchester Business School.

Kindleberger, C.P., and Aliber, R.Z. (2011). Manias, Panics and Crashes: A History of

Financial Crises. Publisher: Palgrave Macmillan, sixth edition.

Martynova, M., and Renneboog, L. (2006). Mergers and Acquisitions in Europe. ECGI – Finance Working Paper No. 114/2006; CentER Discussion Paper Series No. 2006-06.

Reinhart, C.M., and Rogoff, K.S. (2010). From Financial Crash to Debt Crisis. Working Paper No. 15795, NBER Working Paper Series.

Roll, R. (1986). The Hubris Hypothesis Of Corporate Takeovers. Journal of Business

59(2): 197-216.

Saqib A., Bodt, de E., and Lobez, F. (2013). Bank M&A Activities and Financial Crisis

Vulnerability. IMF Working Paper No. 07/215; SKEMA ECCCS, Univ. Lille Nord

de France.

Walsch, J.P. (1988). Top Management Turnover Following Mergers and Acquisitions. Strategic Management Journal, Vol. 9, pp. 173-183

(28)

28

Appendix I

Table 1

–

Proxies for Financing Constraints

Author(s) Year Financing constraints variables

Fazzari, Hubbard & Petersen 1988 Payout ratio

Devereux & Schiantarelli 1990 Firm age / Size / Sector Hoshi, Kashyap & Scharfstein 1991 Group membership

Whited 1992 Bond rating / Debt to assets ratio / Interest coverage ratio Gilchrist & Himmelberg 1995 Commercial paper rating / Payout / Size / Bond rating Chirinko & Schaller 1995 Firm age / Concentration of ownership / Group membership Kaplan & Zingales 1997 Cash / Unused lines of credit / Leverage figures

Kadapakkam, Kumar & Riddick 1998 Firm size

Erickson & Whited 2000 Firm size / Bond rating Lamont, Polk & Saá-Requejo 2001 KZ Index

Almeida, Campello & Weisbach 2004 Payout ratio / Firm size / Bond rating / Comm. paper rating / KZ index Greenaway, Guariglia & Kneller 2005 Liquidity / Credit ratios

Whited 2006 Payout ratio / Size / Group membership

Cleary 2006 Payout ratio / Size

Acharya, Almeida & Campello 2007 Payout ratio / Asset size / Bond rating / Comm. paper rating Denis & Sibilkov 2010 Payout ratio / Firm size / Bond rating / Comm. paper rating

Used sources: Hadlock and Pierce (2009), Khatami (2011), Thomson One database

Table 2 – Variable Definition

Variable Definition

CARi,t Cumulative abnormal return for company i on time t

FC Dummy variable taking a value if target firm is financially constrained

Cash Dummy variable taking a value if the payment of the M&A is settled in cash only * LSizei Natural logarithm of target total assets

Crisis Dummy variable taking a value if M&A was effective between Sept 1, 2008 – Dec 31, 2011 if not, then the M&A was effective between Sept1, 2003 – Dec 31, 2006

Attitude Dummy variable taking a value if the M&A was a hostile takeover, otherwise it is friendly * Industry Dummy variable taking a value if target and bidding firm were active in the same industry * FC · Crisis Cross variable of dummy Financial Constraints times dummy Crisis

LSizei · Crisis Cross variable of dummy target Size times dummy Crisis

Cash · FC Cross variable of dummy Cash times dummy Financial Constraints

(29)

29 Table 3 – List 30 M&A deals Sept 1, 2003 – Dec 31, 2006 (U.S. companies)

Acquiror Name Target Name Date Announced Date Effective Value M&A ($mil)

AT&T Inc BellSouth Corp 03-05-2006 12-29-2006 72.671,00

Procter & Gamble Co Gillette Co 01-28-2005 10-01-2005 54.906,81 ConocoPhillips Co Burlington Resources Inc 12-12-2005 03-31-2006 35.395,01 Boston Scientific Corp Guidant Corp 12-05-2005 04-21-2006 27.861,29 Federated Department Stores May Department Stores Co 02-28-2005 08-30-2005 16.465,87 General Growth Properties Inc Rouse Co 08-20-2004 11-12-2004 12.588,64 KMart Holding Corp Sears Roebuck & Co 11-17-2004 03-24-2005 10.901,27 Cisco Systems Inc Scientific Atlanta Inc 11-18-2005 02-27-2006 6.865,72 Public Storage Inc Shurgard Storage Centers Inc 08-01-2005 08-22-2006 5.072,97 Simon Property Group Inc Chelsea Property Group Inc 06-21-2004 10-14-2004 4.861,07 Anadarko Petroleum Corp Western Gas Resources Inc 06-23-2006 08-23-2006 4.636,53

McClatchy Co Knight Ridder Inc 03-13-2006 06-27-2006 4.572,70

Kimco Realty Corp Pan Pacific Ret Ppty Inc 07-10-2006 10-31-2006 3.954,21 Occidental Petroleum Corp Vintage Petroleum Inc 10-13-2005 01-30-2006 3.566,79

Home Depot Inc Hughes Supply Inc 01-10-2006 03-31-2006 3.475,36

Allergan Inc Inamed Corp 11-14-2005 03-24-2006 3.412,27

Johnson Controls Inc York International Corp 08-24-2005 12-09-2005 3.287,98 Noble Energy Inc Patina Oil & Gas Corp 12-16-2004 05-16-2005 2.954,13

Whirlpool Corp Maytag Corp 07-17-2005 03-31-2006 2.653,71

Lyondell Chemical Co Millennium Chemicals Inc 03-29-2004 11-30-2004 2.546,45 DRS Technologies Inc Engineered Support Systems Inc 09-21-2005 01-31-2006 1.852,06

Valero LP Kaneb Services LLC 11-01-2004 07-01-2005 1.743,81

Crompton Corp Great Lakes Chemical Corp 03-09-2005 07-01-2005 1.549,79

Lee Enterprises Inc Pulitzer Inc 01-31-2005 06-03-2005 1.432,84

Yellow Roadway Corp USF Corp 02-27-2005 05-24-2005 1.268,27

Pennsylvania Re Invest Trust Crown American Realty Trust 05-14-2003 11-20-2003 1.214,74 Plains All American Pipeline Pacific Energy Partners LP 06-12-2006 11-15-2006 1.026,10 Camden Property Trust Summit Properties Inc 10-04-2004 02-28-2005 1.011,00

Yellow Corp Roadway Corp 07-08-2003 12-11-2003 965,55

Health Care REIT Inc Windrose Med Ppty Trust 09-13-2006 12-20-2006 806,86

(30)

30 Table 4 – List 30 M&A deals Sept 1, 2008 – Dec 31, 2011 (U.S. companies)

Acquiror Name Target Name Date Announced Date Effective Value M&A ($mil)

Pfizer Inc Wyeth 01-26-2009 10-15-2009 67.285,70

Exxon Mobil Corp XTO Energy Inc 12-14-2009 06-25-2010 40.298,14

Merck & Co Inc Schering-Plough Corp 03-09-2009 11-03-2009 38.615,31 CenturyLink Inc Qwest Commun Intl Inc 04-22-2010 04-01-2011 22.276,24 The Dow Chemical Co Rohm & Haas Co 07-10-2008 04-01-2009 15.513,13

CenturyTel Inc Embarq Corp 10-27-2008 07-01-2009 11.559,41

Schlumberger Ltd Smith International Inc 02-21-2010 08-27-2010 11.041,61 FirstEnergy Corp Allegheny Energy Inc 02-11-2010 02-25-2011 8.503,24

AMB Property Corp ProLogis 01-26-2011 06-03-2011 8.364,71

Ecolab Inc Nalco Holding Co 07-20-2011 12-01-2011 8.111,84

Caterpillar Inc Bucyrus International Inc 11-15-2010 07-08-2011 7.453,19 Ventas Inc Nationwide Health Properties 02-28-2011 07-01-2011 5.793,48

Danaher Corp Beckman Coulter Inc 02-07-2011 06-30-2011 5.782,22

PepsiCo Inc Pepsi Bottling Group Inc 04-20-2009 02-26-2010 5.421,63

Baker Hughes Inc BJ Services Co 08-31-2009 04-28-2010 5.240,49

AES Corp DPL Inc 04-20-2011 11-28-2011 4.708,63

The Stanley Works The Black & Decker Corp 11-02-2009 03-12-2010 3.469,75 Enterprise Products Partners TEPPCO Partners LP 04-29-2009 10-26-2009 3.283,25

Pulte Homes Inc Centex Corp 04-08-2009 08-18-2009 3.105,76

Holly Corp Frontier Oil Corp 02-22-2011 07-01-2011 2.853,86

AGL Resources Inc Nicor Inc 12-07-2010 12-09-2011 2.382,46

King Pharmaceuticals Inc Alpharma Inc 08-22-2008 12-30-2008 1.565,10 Windstream Corp Iowa Telecom Services Inc 11-24-2009 06-01-2010 1.153,60 Magellan Midstream Partners LP Magellan Midstream Hldg LP 03-03-2009 09-30-2009 1.147,91 Buckeye Partners LP Buckeye GP Holdings LP 06-11-2010 11-19-2010 1.139,01 National Oilwell Varco Inc Ameron International Corp 07-05-2011 10-05-2011 777,67

Kirby Corp K-Sea Transp Partners LP 03-13-2011 07-01-2011 573,75

Sprint Nextel Corp Virgin Mobile USA Inc 07-28-2009 11-24-2009 538,96 Vanguard Natural Resources LLC Encore Energy Partners LP 03-25-2011 12-01-2011 537,13 Raytheon Co Applied Signal Technology Inc 12-20-2010 01-28-2011 509,05

(31)

31

Appendix II – outputs

Table 1 – List Financially Constrained Target Firms 2003 – 2006

Note1 : Debt Ratio is highlighted when above the total mean (of the 60 companies) Note2 : Payout Ratio is highlighted when below the total mean (of the 60 companies)

Note2 : The dummy taking value of 1 means that the target firm is financially constrained, combining the two ratios

Target Name Target Debt Ratio Target Payout Ratio Dummy

BellSouth Corp 0,612 0,356794294 0

Gillette Co 0,770 0,304109589 0

Burlington Resources Inc 0,575 0,044783983 0

Guidant Corp 0,299 0,113260418 0

May Department Stores Co 0,642 0,266090298 0

Rouse Co 0,805 0,960537268 0

Sears Roebuck & Co 0,839 0,080605227 1

Scientific Atlanta Inc 0,194 0,021273104 0

Shurgard Storage Centers Inc 0,633 41,40681445 0

Chelsea Property Group Inc 0,650 1,051803951 0

Western Gas Resources Inc 0,629 0,058524675 1

Knight Ridder Inc 0,657 0,177050441 1

Pan Pacific Ret Ppty Inc 0,526 0,947453253 0

Vintage Petroleum Inc 0,708 0,061578058 1

Hughes Supply Inc 0,462 0,072811774 0

Inamed Corp 0,249 0,001340483 0

York International Corp 0,697 0,156695306 1

Patina Oil & Gas Corp 0,724 0,043964318 1

Maytag Corp 0,977 0,185666094 0

Millennium Chemicals Inc 0,809 0,376344086 0

Engineered Support Systems Inc 0,524 0,007894789 0

Kaneb Services LLC 0,676 0,698445468 0

Great Lakes Chemical Corp 0,544 0,304713805 0

Pulitzer Inc 0,367 0,228519673 0

USF Corp 0,510 0,106023517 0

Crown American Realty Trust 0,925 3,100634726 0

Pacific Energy Partners LP 0,514 0,999610895 0

Summit Properties Inc 0,556 2,408742859 0

Roadway Corp 0,717 0,064948575 1

Windrose Med Ppty Trust 0,615 2,246364883 0

MEAN Debt Ratio MEAN Payout Ratio

Financial constraints and mergers and acquisitions : do financial constraints affect M&A significantly before and