• No results found

Share repurchase announcements in the United States and the free cash-flow hypothesis

N/A
N/A
Protected

Academic year: 2021

Share "Share repurchase announcements in the United States and the free cash-flow hypothesis"

Copied!
44
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

SHARE REPURCHASE

ANNOUNCEMENTS IN THE

UNITED STATES AND THE

FREE CASH-FLOW

HYPOTHESIS

Abstract

In this thesis I will investigate the market reaction surrounding share repurchase announcements in the U.S. between 2006 and 2016 using the free cash-flow hypothesis. I find a significant cumulative abnormal return of 1.4452% in the three-day period surrounding share repurchase announcements. Also, the market reaction is explained by several free cash-flow hypothesis variables. However, during the financial crisis there was no evidence supporting the free cash-flow hypothesis.

Berend Leupen (10504478)

Economie en Bedrijfskunde

Financiering en Organisatie

Supervised by J. E. Ligterink

31-01-2017

(2)

Statement of Originality

This document is written by Berend Leupen who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.

The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

(3)

Contents

1. Introduction ... 3

2. Literature review ... 5

2.1. Tradeoff in payout policy ... 7

2.1.1. Flexibility hypothesis ... 8

2.1.2. Substitution hypothesis ... 8

2.2. Methods of share repurchases ... 9

2.3. Motives for share repurchases ... 10

2.3.1. Signaling hypothesis ... 11

2.3.2. Optimal leverage ratio hypothesis ... 11

2.3.3. Management incentive hypothesis ... 11

2.3.4. Takeover deterrence hypothesis ... 11

2.3.5. Free cash-flow hypothesis ... 12

2.4. Empirical results ... 13

2.5. Contribution to existing literature ... 16

2.6. Hypotheses ... 16 3. Data description ... 16 4. Methodology ... 17 4.1. Event studies ... 18 4.2. Regression analysis ... 19 5. Results ... 24 5.1. Event Studies ... 24

5.1.1. Short term results ... 24

5.1.2. The 81 day period result ... 25

5.1.3. Conclusion ... 26

5.2. Regression results ... 27

5.2.1. Multiple regression analysis... 27

5.2.2. Conclusion ... 28

6. Conclusion and discussion ... 29

6.1. Conclusion ... 29

6.2. Discussion ... 31

7. Reference list ... 32

(4)

1. Introduction

Payout policy is divided into two categories. The classical dividend payout, which has been the primary kind since it was invented approximately 150 years ago. Not much of its characteristics have changed in time. Dividend tends to be sticky and rigid (Lintner, 1956). Then came the extraordinary share repurchase or stock buyback. Mensa et al. (2014) find that in 1997 the number of firms repurchasing shares surpassed the number of dividend payers for the first time, and show that share repurchases have been the most popular way of payout since. Jagannathan et al. (2000) view that the flexibility inherent in share repurchase programs is one reason why they are sometimes used instead of dividends. A second reason is the ability to take advantage of share undervaluation by repurchasing shares.

There are a few practical reasons to repurchase shares. Firstly, a firm might want to improve its earnings per share by repurchasing shares, even though this is not necessarily value increasing. The number of shares outstanding will be reduced, dividing earnings over a smaller amount of shares. Secondly, share repurchases can be used to enhance the liquidity of the shares and stabilize its shares. Third, by keeping shares in treasury, dilution can be limited when employees realize their share options (Dittmar et al., 2003).

However, apart from the more obvious (and sometimes invalid) reasons, firms have more complex motives to repurchase shares. The literature puts forward many theories trying to explain a positive abnormal return at the announcement of share repurchases. The primary theories are, the signaling hypothesis, optimal leverage ratio hypothesis, takeover deterrence hypothesis, management incentive hypothesis and the free flow hypothesis. In this thesis, the emphasis is on the free cash-flow hypothesis. The free cash-cash-flow hypothesis by Jensen (1986) is based on agency theory. Agency theory explains how in the relation between agent and principal, unaligned goals or different levels of risk aversion can lead to agency cost. The free cash-flow hypothesis states that managers in control of high levels of free cash-flow are more likely to invest in bad investment opportunities, with motives like empire building for example. Also, if free cash-flow is kept within the firm, decreasing marginal utility of the investments available will cause the share’s return to deteriorate. However, when managers distribute free cash-flow to shareholders

(5)

instead, the market will react positively and share price will go up, and this can cause abnormal returns (Grullon and Michaely, 2004).

Recent developments have changed the economic environment. The financial crisis increased corporate governance (Bainbridge, 2012), and installment of new banking requirements like Basel III reduced the availability of cash. Uncertainty has grown, availability of cash is limited, and profits decreased. These changes have implications for the existing theories. Primary factors on which the free cash-flow hypothesis is based have changed a lot as well in recent years. Increased corporate governance could control over-investment by managers better, reducing the need of distributing excess cash. Also, the reduction in profitable investment opportunities may have increased free cash-flow, resulting in more risk of over-investment. Lastly, the decrease in the availability of cash may have increased the importance of retaining cash, which in turn discourages the distribution of cash. Since the impact of the different changes is unknown, I think it is of value to investigate the announcement effect of share repurchases, and how it relates to the free cash-flow hypothesis. The results can be compared to previous research, to see what changed, and what caused these changes.

The research question of this paper is: how does the announcement of share repurchases affect cumulative abnormal returns of shares in the United States? This study will analyze the share repurchase announcement of 1090 U.S. firms between 2006 and 2016. The choice for this period is because it is most recent and therefore can have useful implications for business policy and additional research. My research will be further divided into two parts. First, event studies are conducted to see how the share repurchase announcements influence the abnormal returns. Second, a multiple regression analysis is performed to investigate if the free cash-flow hypothesis has any explanatory power on the market reaction following share repurchase announcements. By doing so, this paper adds to the existing literature in several ways. First, I use a sample which is very recent, enabling the possibility to make a comparison over time. Second, I conduct event studies using the developed Carhart four-factor model (Carhart, 1997), to get more accurate abnormal returns (Table 4). And third, I provide additional variables that significantly impact abnormal returns in the regression analysis.

(6)

repurchases in the U.S. The found cumulative abnormal return is lower in my research compared to the more earlier studies. However, the cumulative abnormal return that I find is well in line with the more recent studies. In the multiple regression, return on assets and the interaction of high cash and market-to-book ratio are significant (Table 6). With two of the explanatory variables being significant, there is evidence for the free cash-flow hypothesis. However, during the financial crisis, none of the explanatory variables are significant, supporting the prediction that the effect of the free cash cash-flow hypothesis at the announcement of share repurchases is marginalized in this period.

The remainder of this thesis is organized as follows. First, I will review the existing literature, looking at the trade-off between dividend and share repurchases, methods and motives of share repurchases, the existing empirical results and the hypotheses. Second, I explain the way in which I collected my data in a data description. Third, in my methodology, the conducted event studies and regression are described. Fourth, the results of the event studies and regression are listed and analyzed. And lastly, my conclusions regarding this research are drawn.

2. Literature review

Share repurchases have increased in popularity over the past decades. Grullon and Michaely (2002) state that expenditures on share repurchase programs (relative to total earnings) rose from 4.8% in 1980 to 41.8% in 2000. Also, annual expenditures on share repurchases grew at an average annual rate of 26.1% per year over the period 1980 to 2000. Dividends only grew at an average annual rate of 6.8%. Mensa et al. (2014) find that in 1997 the number of firms repurchasing shares surpassed the number of dividend payers for the first time, and show that share repurchases have been the most popular way of payout since (Graph 1).

(7)

Graph 1

This figure shows the relation between the number of publicly listed payout-paying firms and the payout method from 1971 through 2012. Reprinted from Payout Policy (p. 81), by J. Farr-Mensa, R. Michaely, & M. Schmalz, 2014

The fraction of public firms engaging in share repurchases was highest in 2012 at more than 45%, compared to less than 20% in 1980 (Graph 2).

Graph 2

This figure shows what share of total U.S. public firms paid out a dividend, repurchased shares, and the amount of firms with positive total payout from 1972 through 2012. Reprinted from Payout Policy (p. 82), by J. Farr-Mensa, R. Michaely, & M. Schmalz, 2014

(8)

Also, since 1997, the total amount of cash spent on share buybacks exceeded that of dividend payout every year, except 2009. The sum of share buybacks was highest in 2007, total spending on share buybacks reached 560 billion of real 2012 dollars that year, forming over 73% of cash payout by publicly listed companies in the U.S. At the time of the Great Recession, share buybacks decreased strongly but regained its strength afterward. In 2012 the total amount of cash spent on buying back shares was 364 billion dollars, covering 58% of cash payout by public firms in the U.S. that year (Graph 3).

Graph 3

This figure shows the total volume of cash spend by public U.S. companies on dividend payment, repurchasing shares, and the combination of the two from 1972 through 2012. Reprinted from Payout Policy (p. 83), by J. Farr-Mensa, R. Michaely, & M. Schmalz, 2014

The remainder of this section is constructed in the following way. First, the choice between share repurchases and dividend payout is discussed. Second, the primary methods of share repurchases are explained. Third, motives for share repurchases are explained. Fourth, previous empirical results are listed. Fifth, my contribution to the existing literature is explained. Sixth, I state the predictions of my hypotheses.

2.1. Tradeoff in payout policy

The number of share repurchasing firms have grown rapidly since 1980, surpassing dividend-paying firms in 1997 for the first time (Mensa et al., 2014). They show that

(9)

share repurchases have been the most popular way of payout since. Within the literature, I have found two insights that try to explain the change in payout policy, the flexibility hypothesis and the substitution hypothesis.

2.1.1. Flexibility hypothesis

The flexibility hypothesis explains how a combination of share repurchases and dividend payouts can provide a more flexible way of increasing and decreasing payout, compared with solely dividend payout.

Lintner (1956) explains that dividends tend to be sticky using his observations of dividend policy. His first observation is that firms generally set long-term dividends-to-earnings ratio objectives in relation to the positive net present value projects available to them. Moreover, firms only change dividend policy when they are sure that the increases in earnings are sustainable.

Jagannathan et al. (2000) find that share buybacks and dividend payouts are used at different times, and by various sorts of companies. Share buybacks are strongly pro-cyclical, dividend payout, on the other hand, rises more smoothly over time. Dividend payout is used by businesses with higher permanent operating cash flow, while firms use buybacks with short-term cash holdings, not needed for operations. Also, evidence in their study shows that, in general, companies buying back shares have more volatile free cash-flow. Lastly, companies buy back shares when their shares are underperforming, and expand dividend payout when shares perform well. These findings are in line with the assumption that share buybacks are more flexible than dividend payment, and could explain the preference over dividends in some circumstances.

2.1.2. Substitution hypothesis

Fundament of the substitution hypothesis is built on the dividend irrelevance theory by Miller and Modigliani (1961). Their theory implies that, given perfect market conditions, dividends and share repurchases are perfect substitutes. Meaning that cash can be distributed by dividend payment or share repurchase without a difference. The way in which cash is distributed also does not matter for the agency theory of Easterbrook (1984) and Jensen (1986), which explains how over-investment by managers can be controlled by limiting excess cash. Also, the signaling models by Bhattacharya (1979) and Miller and Rock (1985) imply that

(10)

dividends and share repurchases are interchangeable. However, the model by John and Williams (1985) takes the taxes on dividend into account, suggesting that share repurchases and dividends are not perfect substitutes. Also, in the model by Allen et al. (2002) dividend payouts and share repurchases are not substitutes because the former attracts institutions. According to them, institutional investors have a better information-gathering ability, and therefore are more capable of discovering if a firm is undervalued or not. So when firms want to signal that they are undervalued, the firms will increase dividend payout. This signaling equilibrium is not reached with share repurchases and, therefore, they are not interchangeable.

Skinner (2008) finds that firms progressively use share buybacks rather than dividend payments, even companies that go on paying a dividend. Even though other components determine the timing of share buybacks, the total amount of share buybacks is decided based on the level of earnings. His empirical findings support the fact that share buybacks are the prevalent payout method at the moment.

Grullon and Michaely (2002) investigate if dividend payouts are substituted by share repurchases. They research if the increases in dollar volume of share repurchases have been used as a substitute for dividends. Consistent with their substitution hypothesis, Grullon and Michaely (2002) find empirical evidence that the increase in share repurchase activity in the U.S. has been financed with potential increases in dividends. A survey by Brav et al. (2005) finds that firms continue to pay dividends mostly because of history. These companies have paid dividends for many years and are obligated to continue paying a dividend. Also, they find that firms would rather solely repurchase shares as a payout form, if free from the tradition to pay dividends.

2.2. Methods of share repurchases

In the literature, I find four primary methods a firm may use to repurchase its equity. Explanation of the various share repurchase methods provides additional information so that different concepts addressed in this thesis will be better understandable. Firstly there are tender-offer repurchases. A fixed-price offer specifies a single purchase price, a number of shares sought, and an expiration date in advance. If the offer is oversubscribed, the firm may buy a number of shares bounded by the stated number sought (Comment and Jarrell, 1991). Secondly, Dutch auction repurchases.

(11)

The Dutch-auction offer also specifies the number of shares sought. Instead of a single offer price, however, the Dutch auction specifies a range of prices within which each tendering shareholder chooses his or her minimum acceptable selling price. The highest bids adding up to the number of shares on offer will be considered as winning bids. The price of the shares is set to the lowest winning bid, and even the shareholders who bid at a price higher than the lowest winning bid gets to pocket the shares at the lowest winning bid price (Comment and Jarrell, 1991). Thirdly, open-market repurchases. Where shares are bought on the open open-market, just like an individual investor would, at the market price (Comment and Jarrell, 1991). And lastly, private repurchases. Here, the firm negotiates with the shareholder directly to buy a large amount of shares (Vermaelen, 1981).

Of the possible buyback methods, open market share buybacks are the dominant one. One example is that Ikenberry et al. (1995) show a 90% share of open-market buybacks relative to the total amount of buybacks from 1985 to 1992. Dann (1981), Vermaelen (1981), Asquith and Mullins (1986), and Comment and Jarrell (1991) have identified significant, positive abnormal returns associated with the announcement of each of the four types of buyback arrangements.

2.3. Motives for share repurchases

Several theories try to explain share repurchases, which can lead to different hypotheses regarding the abnormal return. Five of the most important motives I find for share repurchases in the literature are listed. First, the signaling hypothesis (Vermaelen, 1981). Second, the optimal leverage ratio hypothesis (Bagwell and Shoven, 1988). Third, the management incentive hypothesis (Dunsby, 1994, Jolls, 1996, and Fenn and Liang, 1997). Fourth, the takeover deterrence hypothesis (Bagwell, 1991). Finally, the free cash-flow hypothesis (Jensen, 1986). These theories can explain share repurchases independently, but a combination of the theories is possible as well. A firm can, for example, try to optimize its leverage ratio and signal information with a share repurchase announcement at the same time. Also, apart from share repurchases, special dividends can reach the same goal in the signaling- and free cash-flow hypothesis (Miller and Rock, 1985, and Grullon and Michaely, 2004, respectively).

(12)

2.3.1. Signaling hypothesis

The signaling hypothesis analyzes the share repurchase decision in a world with asymmetric information. The signaling hypothesis suggests that management gives an information signal when it repurchases its shares. The direction of the signal is unknown. It might be that due to a lack of profitable investment opportunities the firm decides to distribute cash. Conversely, the firm might perceive that it is undervalued. Especially when the firm offers to repurchase its shares at a premium, the company might perceive being undervalued. The tender offer is then meant to transfer the value of this inside information to the shareholders (Vermaelen, 1981).

2.3.2. Optimal leverage ratio hypothesis

Share repurchases can be financed with debt. The interest paid on debt is tax deductible, creating a tax shield (Vermaelen, 1981). Bagwell and Hoven (1988) claim that there is an optimal leverage ratio to minimize the cost of capital. Consequently, a firm can repurchase shares when its leverage ratio is different from the optimal leverage ratio.

2.3.3. Management incentive hypothesis

When management holds share options, it is in their interest to use share repurchases to distribute cash. The reason is that share repurchases do not dilute the per-share value of the shares. When management repurchases shares, the price per share will increase, which will also increase the value of the share options that management holds. Also, shares provided to managers when they exercise options are often treasury shares. Thus, it might be beneficial for a firm to repurchase shares when management holds a large amount of share options (Dunsby, 1994, Jolls, 1996, and Fenn and Liang, 1997).

2.3.4. Takeover deterrence hypothesis

Firms can become the potential target of a takeover. When the firm has an upward-sloping supply curve for shares, it can decide to buy back its share to increase the cost of an acquisition by a potential buyer. Shareholders that are willing to tender their shares in the buyback are systematically those with the lowest valuations, leaving a pool of shareholders which values their shares more. So by increasing the

(13)

cost of acquisition through share buybacks, the company can defend against a takeover (Bagwell, 1991).

2.3.5. Free cash-flow hypothesis

Free cash-flow hypothesis rests on agency theory. One implication of the agency theory suggests that management of a firm with free cash-flow in excess over its investment opportunities (opportunities with positive net present values), will lead to over-investment. This behavior occurs because of misalignment of interest between managers (the agents) and shareholders (the principals). Managers can for example focus on empire building, instead of creating shareholder value. As a consequence, Jensen (1986) argues that if the excess cash that management controls can be cut by share repurchases, over-investment is reduced.

Jensen (1986) and Lang et al. (1991) use the free cash-flow hypothesis to predict that shareholders will choose to limit managers’ access to free cash-flow to mitigate agency conflicts over its deployment. What is key in these papers is the tradeoff between providing sufficient internal capital for managers to efficiently fund all good projects, while not providing excess internal capital that allows managers to spend on negative net present value projects.

Dittmar et al. (2003), Kalcheva and Lins (2004), and Pinkowitz et al. (2004) investigate the relation between cash holdings and shareholder influence as suggested by Jensen (1986) and Stulz (1990) in a cross-country model. Harford et al. (2008) do the same for the U.S. Cross-country evidence shows that greater shareholder influence is associated with lower cash holdings. This suggests that shareholders want managers to pay out cash and that they force them to do so when they are so empowered (Dittmar et al., 2003, Kalcheva and Lins 2004, and Pinkowitz et al., 2004). For the U.S., direct evidence of the fact that more shareholder influence will lead to higher cash payout is provided by Harford et al. (2008).

Décamps et al. (2011) investigate the relation between agency problems and cash payout as well, using the asymmetric volatility phenomenon. The asymmetric volatility refers to the observed negative correlation between share market movements and the volatility of share returns. In the model by Décamps et al. (2011) based on asymmetric volatility, the increasing level of cash in companies leads to a deteriorating marginal value of cash. In their numerical investigation, more severe

(14)

agency costs lead to less volatile share returns. Less volatile share returns means that the marginal value of cash increases, because of a decrease in its cash holdings. This effect reflects the fact that firms with higher agency costs tend to accumulate less cash before distributing to shareholders.

Gao et al. (2013) examine the drivers for cash policies for private firms to help understand public firms’ cash policies. They show that private companies hold, on average, about half as much cash as public firms do. And, can attribute the difference to the much higher agency costs in public companies. This is indicated by their empirical results that show public firms use excess cash in shortsighted ways and in ways that strongly reduce operating performance compared to the way private firms use cash.

2.4. Empirical results

Much prior research has already been done about the effects of share repurchase announcements on the share price. Almost every study finds a positive cumulative abnormal return surrounding a share repurchase announcement for the event window [-1;1]. Table 1 presents an overview of the results found in papers about the U.S. short-term cumulative abnormal returns. Consequently, I will expect a positive cumulative abnormal return in my study as well.

Next, I will discuss the empirical evidence found in the previous studies (Table 1) in more detail. Vermaelen (1981) finds that the signaling hypothesis explains the abnormal returns after a tender offer best. Firms offer their shares at a premium when they want to signal positive information about future earnings. Comment and Jarrell (1991) find that share repurchases by fixed-price self-tender offers, Dutch auctions, and open market repurchase programs are associated with significant, positive abnormal returns at the announcement. Of the three methods, fixed-price self-tender offers are associated with the highest abnormal returns. Overall their study provides the most support for the hypothesis that share repurchases increase share price because they can credibly provide a signal that the shares of the firms are undervalued. Also, they find evidence that the takeover deterrence implications of share repurchases contribute to the positive returns on the announcement. The study by Ikenberry et al. (1995) investigates the underreaction to open market share buybacks. In their study, they buy shares and hold them for four years. As a result,

(15)

their abnormal return after the share buyback announcement for this period is over 12%. On the other hand, the immediate market reaction after a firm announces to buy back shares was 3.5% averagely. Stephens and Weisbach (1998) study the actual share repurchases in open market buyback programs. They estimate that firms in their sample repurchase 74% to 82% of their announced target buyback level. Also, they find that actual share repurchases are negatively related to a firm’s prior quarter share return. This is consistent with the signaling hypothesis. The study by McNally (1999) proves the existence of a signaling equilibrium for open market repurchases. Also, support for the optimal leverage ratio hypothesis is found. Grullon and Michaely (2004) provide two main empirical findings. First, they do not find evidence that firms buying back shares experience a growth in profitability, and that firms buying back shares decrease their investments over time. This is in contrast with the expectations of the signaling hypothesis. Second, they find that the positive market response after announcing share buybacks are caused by the cutback in free cash-flow and undiversifiable risk. And, the market reacts more strongly to firms announcing share buybacks when the probability of these firms to overinvest is higher. This evidence provides support for the free cash-flow hypothesis. I will use this evidence to investigate the free cash-flow hypothesis in my study. Peyer and Vermaelen (2005) find that repurchases at a premium can be modeled as signals, while other repurchases are just wealth transfers between the company and the selling shareholders. Lie (2005) finds evidence that the market reacts positively to earnings announcements after a share repurchase program announcements. Also, both the operating performance improvement and the positive earnings announcement return only apply to firms that actually repurchase shares during the same fiscal quarter. Chan et al. (2007) find that there is no evidence for pseudo-market timing as an explanation for the positive long-term abnormal share returns for firms buying back shares. Instead, they find evidence that managers possess market timing abilities when announcing and executing buyback decisions. Billet and Xue (2007) find new evidence for the takeover deterrent hypothesis. They have empirical findings that firms’ repurchase activity increases when they face a probability of being taken over. Wang et al. (2009) empirically find that the average market response to actual share repurchase announcements is positive and significant for firms with over-investment problems, indicated with relatively low Tobin’s Qs. This result supports the free cash-flow hypothesis. In line with this study, I will use Tobin’s

(16)

Q to investigate the free cash-flow hypothesis in my research. Chang et al. (2010) find that upon the announcement of share repurchases, the market responds more positively to those announcements by firms that previously have made a better record on actual repurchases. This result suggests that experience of prior share repurchases plays a major role in the value assessment of subsequent information signals through share repurchase announcements. In line with signaling hypothesis, the study by Babenko et al. (2012) find evidence that the market reacts more positive to share repurchase announcements by firms when insiders recently purchased shares.

Table 1

Previous empirical results of studies investigating cumulative abnormal returns surrounding the announcement of share buyback programs. From left to right are listed; reference to the study, country investigated, period investigated, the number of days in the event window with the announcement at time 0, and cumulative abnormal return. All results are significantly different from zero.

Study Country Period Event

window

CAR

U.S. short-term abnormal returns

Vermaelen (1981) U.S. 1970 - 1978 -1;0 3.6%

Comment and Jarrell (1991) U.S. 1985 - 1988 -1;1 2.3% Ikenberry et al. (1995) U.S. 1980 - 1990 -2;2 3.5% Stephens and Weisbach (1998) U.S. 1981 - 1990 -1;1 2.7%

McNally (1999) U.S. 1984 - 1988 -1;1 1.1%

Kahle (2002) U.S. 1993 - 1996 -1;1 1.6%

Grullon and Michaely (2004) U.S. 1980 - 1997 -1;1 2.7% Peyer and Vermaelen (2005) U.S. 1984 - 2001 -1;1 2.4%

Lie (2005) U.S. 1981 - 2001 -1;1 3.0%

Chan et al. (2007) U.S. 1980 - 1996 0;11 4.9%

Wang et al. (2009) U.S. 1997 0;1 1.3%

Chang et al. (2010) U.S. 1986 - 2005 0;2 1.9%

(17)

2.5. Contribution to existing literature

This paper will analyze the share repurchase announcement of 1090 U.S. firms. The research will be further divided into two parts. First, event studies are conducted to see how the share repurchase announcements influence the abnormal return. Second, a regression analysis is performed to investigate the relation between the free cash-flow hypothesis and cumulative abnormal returns. By doing so, this paper adds to existing literature in several ways. First, I use a sample which is very recent, this enables the possibilities to make a comparison over time. Second, I conduct event studies using the Carhart four-factor model (Carhart, 1997), to get more accurate abnormal returns (Table 4). And third, I supply additional variables that significantly impact abnormal returns in the regression analysis.

2.6. Hypotheses

In accordance with the related literature, my expectation is that share repurchase announcements will be followed by positive cumulative abnormal returns explained by the free cash-flow hypothesis, except during the financial crisis. My hypotheses therefore are:

H1: Share repurchase announcements in the United States between 2006 and 2016 are followed by positive cumulative abnormal returns.

H2: Share repurchase announcements in the United States between 2006 and 2016 are followed by positive cumulative abnormal returns explained by the free cash-flow hypothesis.

H3: Support for the free cash-flow hypothesis has decreased during the financial crisis of 2008 and 2009.

3. Data description

This thesis investigates the share price reaction after a share repurchase announcement. Also, it investigates if the market reacts as predicted by the free cash-flow hypothesis. This dataset will consist of share repurchase announcements made by firms in the U.S. in the period January 1, 2006, to December 31, 2015.

The share repurchase announcements are collected using the ZEPHYR database. ZEPHYR contains information on Mergers & Acquisitions, IPO’s, private

(18)

equity and venture capital deals, and rumors. A limitation of this research is that the ZEPHYR database is not entirely complete, but I do not complement by other databases. The following search criteria are used: share repurchases of firms in the U.S., in the period January 1, 2006, to December 31, 2015. This resulted in an initial sample of 5361 share repurchase announcements of firms in the U.S. This data is cleaned of announcements with an unknown value sought, an invalid ticker symbol or ISIN number, and the wrong type of announcement. Additionally, if a firm has multiple share repurchase announcements, only the oldest one is kept to prevent share repurchase announcement effects from being mismatched. A sample of 1572 share repurchase announcements is kept. And finally, the program Eventus collects the information needed to calculate the abnormal returns using the market model and the Carhart four-factor model. Keeping the firms announcing to repurchase shares for which this information was available, reduced the sample to 1356.

For every firm in the sample, the historical financial data necessary for this research is found using the static request option of the Datastream tab in Excel. The historical financial data needed for the regression analysis is collected for all firms, at the date of the share repurchase announcement. The following data was obtained from Datastream: market-to-book ratio, market value, total assets, cash and short-term investments, and net income before preferred dividend (shown in Appendix A). Share repurchase announcements for which all necessary variables were available were kept, leaving a sample of 1145. A limitation of my study is that this sample is relatively small if the extent of the chosen period is taken into account. The complete list of ticker symbols of firms used for this research can be found in Appendix B.

4. Methodology

This research consists of two parts. The first part is event studies, in which the abnormal returns of shares surrounding share repurchase announcements are investigated. The second part is a multiple regression used to examine the free cash-flow hypothesis.

(19)

4.1. Event studies

Event studies are conducted to research the market reaction following a share repurchase announcement. My event studies are computed using the program Eventus on Wharton Research Data Services.

The first step is specifying the length of the estimation period in trading days, then the event window, and the gap between the end of the estimation period and the beginning of the event window. The estimation period is the period of which the parameters in the Carhart four-factor model are estimated. The event window is the period in which the market reaction is investigated. The gap is meant to avoid overlap of the expected- and normal share returns of the firms announcing to buy back shares. Following the procedure of conducting an event study as defined by Mackinlay (1997), the event window used is three days [-1;1], the gap between the end of the estimation period and the beginning of the event window is 15 days, and the estimation window is 110 days [-116;-16]. Eventus now creates the trading calendar, which accounts for holidays, weekends and other non-trading days. This ensures the specified length for each firm-event window.

In step two, Eventus first extracts the CRSP daily returns between the beginning of the estimation period and the end of the event window according to the trading calendar. Then, it adds the following risk factors. The monthly return of the CRSP value-weighted index less the risk-free rate return (market), the monthly premium of the book-to-market factor (HML), the monthly premium of the size factor (SMB), and the monthly premium on winners minus losers (momentum). And lastly, it calculates the returns adjusted for delisting.

In step four, Eventus first runs the Carhart four-factor model. Then, the program calculates the (cumulative) abnormal returns for the Carhart four-factor model analogously.

In step five, the t-statistic of the cumulative abnormal return is calculated. However, sometimes multiple companies are announcing to repurchase shares at the same time, which causes an event clustering problem. Therefore, the “standardized cross-sectional test” of Boehmer et al. (1991) takes into account information from both the estimation- and the event windows and allows for event-induced variance shifts. The test applies the cross-sectional technique used for the ordinary cross-sectional test, but uses standardized abnormal returns (SARs) rather than abnormal return (ARs) as in the ordinary cross-sectional test:

(20)

The three-day event study is used to provide an answer to the first hypothesis, and for further analysis. For illustration purposes, an additional event study is conducted with the same time window as Vermaelen (1981). Namely, from 40 days prior to the announcement, to 40 days after the announcement [-40;40]. And lastly, to compare the contribution of using the Carhart four-factor model for this research, instead of the market model, the three-day event window study is also conducted using the market model.

4.2. Regression analysis

The second and third hypotheses are based on the free cash-flow hypothesis. Next, I will describe the different variables used to analyze the market response after firms announce to buy back shares, in accordance with the free cash-flow hypothesis. The variables are defined in Table 2. This will be followed by the formulation of the multiple regression. And lastly, the eight assumptions of a multiple regression and evidence of fulfillment of the assumptions are shown.

Cash: The free cash-flow hypothesis predicts that firms with high levels of cash and short-term investments, will limit their amount of cash and short-term investments available to their managers to reduce over-investment. One way to reduce available cash and short-term investments is by repurchasing shares. This is why repurchasing firms are expected to have high levels of cash (Grullon and Michaely, 2004). So when firms with high levels of cash and short-term investments repurchase shares, the risk of over-investment is reduced, which is expected to be followed by a positive abnormal return. The relation between repurchasing firms with high levels of cash and short-term investments and cumulative abnormal returns is, therefore, expected to be positive. Additionally, an interaction variable is created that

(21)

is equal to the market-to-book ratio if the value of cash is among the highest 25% values, and 0 otherwise. In line with Lie (2000), the coefficient of this interaction variable is expected to be negative.

Market-to-book ratio: Following Lie (2000), I include in my regressions the market-to-book ratio as an approach to Tobin’s Q. A low Tobin’s Q is associated with relatively high free cash-flow and low growth opportunities. Accordingly, the free cash-flow hypothesis expects a negative relation between the market-to-book ratio and cumulative abnormal returns. Also, an interaction variable is created that is equal to the value of cash and short-term investments if the market-to-book ratio is among the lowest 25% values, and 0 otherwise. In accordance with the free cash-flow hypothesis, I expect a positive relation between the cumulative abnormal returns and this interaction variable.

Return on assets: Firms that repurchase their shares should experience a decline in profitability according to the free cash-flow hypothesis (Grullon and Michaely, 2004). Consequently, the return on assets is expected to be negatively related to cumulative abnormal returns.

Number of shares sought: This variable is used as a control variable. The size of the share repurchases announced is expected to be positively related to the cumulative abnormal returns.

Financial crisis dummy: This dummy variable, used as part of interaction variables, states which share repurchase announcements are within the financial crisis period. All interaction variables of crisis and an explanatory variable are expected to have an insignificant impact on the cumulative abnormal return. The fall of Lehman Brothers on September 13, 2008, is seen as the start of the crisis (Fernando et al., 2012). Even though it was not the far beginning of the financial crisis, it is the initial real shock to the financial market. The end of the Great Recession came in June 2009 according to the U.S. National Bureau of Economic Research (2012), the financial crisis ended at about the same time.

(22)

Table 2

Definition of the variables used in the regression

Variable Stata abbreviation Definition

Dependent variable Cumulative Abnormal Return

CAR3 Calculated as the three-day event window cumulative abnormal return

Independent variable

M/B MB The ratio between market value and book

value of the firm

Cash CASH The ratio between cash and short-term investments, and total assets

Return on assets

ROA The ratio between net income before preferred dividend and total assets Shares

sought

SOUGHT The ratio of shares announced to be bought back and the number of shares outstanding Dummy

variable Financial crisis dummy

CRISIS It takes on the value of 1 for variables within the financial crisis period, and the value of 0 if it falls outside of this period

Low M/B dummy

LowMB It takes on the value of 1 for the lowest 25% M/B values, and the value of 0 for the highest 75% M/B values

High cash dummy

HighCASH It takes on the value of 1 for the highest 25% cash values, and the value of 0 for the lowest 75% cash values

Interaction variable

(23)

low M/B of cash and low M/B High cash

and M/B

HighCASH_MB Interaction variable that multiplies the values of high cash and M/B

Cash, crisis and low M/B

CASH_LowMB_CRI SIS

Interaction variable that multiplies the values of cash, crisis, and low M/B

High cash, crisis and M/B

HighCASH_MB_CRI SIS

Interaction variable that multiplies the values of high cash, crisis and M/B

Return on assets and crisis

ROA_CRISIS Interaction variable that multiplies the values of return on assets and crisis

Shares sought and crisis

SOUGHT_CRISIS Interaction variable that multiplies the values of shares sought and crisis

M/B and crisis

MB_CRISIS Interaction variable that multiplies the values of M/B and crisis

Cash and crisis

CASH_CRISIS Interaction variable that multiplies the values of cash and crisis

The conducted ordinary least squares (OLS) regression is a multiple regression. The regression is constructed to represent the free cash-flow hypothesis, and is shown below: 𝐶𝐴𝑅3 = 𝛼 + 𝛽1∗ 𝐶𝐴𝑆𝐻 + 𝛽2∗𝑀 𝐵 + 𝛽3∗ 𝐶𝐴𝑆𝐻 ∗ 𝐿𝑜𝑤 𝑀 𝐵 + 𝛽4∗ 𝐻𝑖𝑔ℎ 𝐶𝐴𝑆𝐻 ∗ 𝑀 𝐵+ 𝛽5 ∗ 𝑆𝑂𝑈𝐺𝐻𝑇 + 𝛽6∗ 𝑅𝑂𝐴 + 𝛽7∗ 𝐶𝐴𝑆𝐻 ∗ 𝐿𝑜𝑤𝑀 𝐵 ∗ 𝐶𝑅𝐼𝑆𝐼𝑆 + 𝛽8∗ 𝐻𝑖𝑔ℎ 𝐶𝐴𝑆𝐻 ∗𝑀 𝐵 ∗ 𝐶𝑅𝐼𝑆𝐼𝑆 + 𝛽9∗ 𝑅𝑂𝐴 ∗ 𝐶𝑅𝐼𝑆𝐼𝑆 + 𝛽10∗ 𝑆𝑂𝑈𝐺𝐻𝑇 ∗ 𝐶𝑅𝐼𝑆𝐼𝑆 + 𝛽11∗ 𝑀 𝐵 ∗ 𝐶𝑅𝐼𝑆𝐼𝑆 + 𝛽12∗ 𝐶𝐴𝑆𝐻 ∗ 𝐶𝑅𝐼𝑆𝐼𝑆 + 𝜀

To be allowed to perform a multiple linear regression, eight assumptions have to be fulfilled. In Table 3 I will go over the assumptions, providing evidence that they are satisfied.

(24)

Table 3

Multiple linear regression assumptions, and evidence of fulfillment of the assumptions

Assumption Evidence

The dependent variable should be measured at the continuous level

My dependent variable is a continuous variable

There are two or more independent variables, which should be measured at the continuous or categorical level

My independent variables are independent, and continuous or categorical

There should be independence of observations (i.e., independence of residuals)

Durban-Watson d-statistic(8, 1145) = 2 At k=8 and n=200

Lower Cutoff is 1.592 Upper cutoff is 1.757

Since 2>1.757, so no indication of serial correlation

Also, I use robust standard errors, which corrects for serial correlation

There needs to be a linear relationship between (a) the

dependent variable and each of the independent variables, and (b) the dependent variable and the

independent variables collectively.

An augmented component-plus-residual plot was graphed for every explanatory variable

All seemed extremely close to linear, except the variable SOUGHT

However, SOUGHT did not look nonlinear enough to become worrisome

Data needs to show homoscedasticity

I use robust standard errors to correct for heteroskedasticity

Data must not show multicollinearity Variance inflation factor (VIF) command used Worrisome if VIF value>10

Highest VIF value found 2.05 So no indication of multicollinearity There should be no significant

outliers, high leverage points or highly influential points

Cook’s D and DFITS measures are performed Cook’s D cut-off point is d>4/n

Which makes d>4/1145

(25)

looking at them more closely I decided to remove them

For DFITS the cut-off point is 2*sqrt(k/n) Which makes d>2*sqrt(7/1145)

Four more observations above cutoff point and after looking at them more closely I decided to remove them

The residuals (errors) should be approximately normally distributed

Analyzed using the kernel density estimate, a standardized normal probability plot, and a plot of the variables quantiles against the quantiles of a normal distribution

The distribution looked normal

5. Results

5.1. Event Studies

The event studies conducted show the (cumulative) abnormal share returns surrounding the announcement of a share repurchase program. Using the Carhart four-factor model (Carhart, 1997), the expected share returns are estimated in case the event would not have happened. By subtracting the actual share returns from the expected share returns, the abnormal returns are calculated. In the first paragraph the short term results, in the second paragraph the 81-day period result, and in the third paragraph, the conclusion will be discussed.

5.1.1. Short term results

Table 4 shows the results of the short-term event studies. As seen in the table the three-day event window cumulative abnormal return is positive and highly significant. Also, abnormal returns surrounding share repurchase announcements can be compared to previous studies. When my event study is compared with the more earlier studies by Stephens and Weisbach (1998), Grullon and Michaely (2002), Kahle (2002), Peyer and Vermaelen (2005), Lie (2005), and Chan et al. (2007), the

(26)

1.4452% cumulative abnormal return is substantially lower. However, my event study is well in line when compared to the more recent studies by Wang et al. (2009), Chang et al. (2010), and Babenko et al. (2012). And lastly, Table 4 shows that the Carhart four-factor model calculates the cumulative abnormal return more accurately, compared to the market model.

Table 4

The table below shows from left to right, the risk model, the number of share repurchase announcements, the abnormal returns for the three days in the event window individually, the three-day cumulative abnormal return, and its t-statistic under the two-sided zero hypothesis µ = 0. The ** and * indicate a t-statistic significantly different from zero at the 1% and 5% level, respectively.

Risk-model obs AR [-1] AR [0] AR [1] CAR [-1;1] t-statistic

Carhart four-factor model

1355 -0.20% 0.67% 1.45% 1.4452% 8.76513**

Market model 1355 -0.17% 0.69% 1.5% 1.5013% 8.61135**

5.1.2. The 81-day period result

Table 5 and Graph 4 show the 81-day period cumulative abnormal return, as proposed by Vermaelen (1981). This graph is intended to show if a delayed announcement effect occurs, not visible in the short term window. As shown in the graph, the abnormal return decreases before the announcement, and at the moment of the announcement, the abnormal return jumps. Subsequently, the abnormal return increases slightly more in the day 3 to 40 period after the announcement. The movement is about the same as observed by Vermaelen (1981), except that the observed jump in abnormal return is not as big. The decreasingly negative abnormal return, previous to the share repurchase announcement, suggests that this is not a good control period (Vermaelen, 1981). As a consequence, the period should not be included in the event window.

(27)

Table 5

The table below shows from left to right, the risk model, the number of observations, the 81-day cumulative abnormal return, and its t-statistic under the two-sided zero hypothesis µ = 0. The cumulative abnormal return is tested for significant difference from zero at 1% and 5% level and is marked with ** and *, respectively.

Risk-model obs CAR [-40;40] t-statistic

Carhart four-factor model 1355 -1,4652% -1.38664

Market model 1355 -1.9617% -1.67463

Graph 4

Cumulative abnormal return using the Carhart four-factor model for the 81-day event window around the share repurchase announcement at time zero.

5.1.3. Conclusion

Based on the results above can be concluded that if the sample is selected similar to most of the other papers on the matter, the three-day cumulative abnormal return is positive and highly significant. Remarkable is that the 1.4452% cumulative abnormal return is substantially lower compared to the more earlier studies by Stephens and Weisbach (1998), Grullon and Michaely (2002), Kahle (2002), Peyer and Vermaelen (2005), Lie (2005), and Chan et al. (2007). However, in the same range when compared to the more recent studies by Wang et al. (2009), Chang et al. (2010), and

(28)

Babenko et al. (2012). The 81-day cumulative abnormal return is much lower, but not a proper measure of cumulative abnormal return in this study. Since the pre-announcement period is not a good control period, it should not be included in the event window (Vermaelen, 1981). As a consequence, my first hypothesis, stating that share repurchase announcements in the U.S. between 2006 and 2016 are followed by positive cumulative abnormal returns, is not rejected.

5.2. Regression results

In this part, the free cash-flow hypothesis, which is supposed to explain the cumulative abnormal returns following a share repurchase announcement, will be tested. Next, the multiple regression analysis is discussed, after which a conclusion is drawn.

5.2.1. Multiple regression analysis

In Table 6 the results of the multiple regression, conducted to test the free cash-flow hypothesis, are listed. As shown in the table, the explanatory variable return on assets is significant (p-value of 0.014), and the explanatory interaction variable of high cash and market-to-book ratio is significant (p-value of 0.047). Also, the control variable number of shares sought is significantly different from zero (p-value of 0.001). And lastly, none of the interaction variables of crisis and an explanatory variable is significantly different from zero at the 10% level.

Table 6

Multiple regression of different explanatory variables on the cumulative abnormal return following a share repurchase announcement and its t-statistic between parentheses under the two-sided zero hypothesis µ = 0. The ** and * indicate a t-statistic significantly different from zero at the 1% and 5% level, respectively.

Additionally, the number of observations, the R-squared, and the F-value are stated.

Variable Coefficient

Constant .0081293

(2.76)** Independent variable

(29)

(-1.51) Cash .0116554 (1.27) Return on assets -.0327814 (-2.47)** Shares sought .0757682 (3.41)** Interaction variable

Cash and low M/B .0003384 (0.02) High cash and M/B -.0002571

(1.83)* Cash, crisis and low M/B .0453639

(0.91) High cash, crisis and M/B -.0022108

(-1.24) Return on assets and crisis -.0573607

(-1.10) Shares sought and crisis .2971075

(3.22)**

M/B and crisis .0023384

(1.44) Cash and crisis -.0010538

(-0.05)

Observations 1090

𝑅2 0.0434

F 14.94

5.2.2. Conclusion

The multiple regression is conducted to test the free cash-flow hypothesis on the cumulative abnormal returns following share repurchase announcements. And, to test if the impact of the free cash-flow hypothesis marginalized during the financial

(30)

crisis. In the multiple regression, the explanatory variable return on assets and the explanatory interaction variable of high cash and market-to-book ratio is significant. The negative coefficient of return on assets indicates that firms with declining profitability are rewarded by a positive cumulative abnormal return when a share repurchase is announced (Grullon and Michaely, 2002). The negative coefficient of the interaction of high cash and market-to-book ratio suggests that the market reacts positively when firms with high levels of cash and low investment opportunities (proxied by Tobin’s Q) announce to repurchase shares (Lie, 2005). With two significant explanatory variables concluded can be, that share repurchase announcements in the U.S. between 2006 and 2016 are followed by positive cumulative abnormal returns explained by the free cash-flow hypothesis. Consequently, the second hypothesis will not be rejected. Also, none of the interaction variables of crisis and an explanatory variable is significant. This evidence supports the decreased effect of the free cash-flow hypothesis during the financial crisis. As a consequence, the third hypothesis will not be rejected either.

6. Conclusion and discussion

6.1. Conclusion

Existing literature has thoroughly investigated share repurchases. And, especially in the U.S., studies have found significant positive cumulative abnormal returns. However, since share repurchases themselves do not cause abnormal returns, theories have been constructed trying to find the cause of the abnormal returns. I am most interested in the free cash-flow hypothesis. The reason for this is that the literature confirming the hypothesis is not up-to-date. In recent years an integrated web of changes has impacted the economic environment and with that the results of previous research. The financial crisis increased corporate governance (Bainbridge, 2012), and installment of new banking requirements like Basel III, reduced the availability of cash. Primary factors on which the free cash-flow hypothesis is based also changed a lot in recent years. Increased corporate governance could control over-investment by managers better, reducing the need of distributing excess cash. The reduction in profitable investment opportunities may have increased free

(31)

cash-cash may have increased the importance of retaining cash-cash, which in turn discourages the distribution of cash. This thesis can, therefore, contribute to the existing literature on the free cash-flow hypothesis by Jensen (1986), Grullon and Michaely (2002), and Harford et al. (2008), by investigating the impact of the different changes on the free cash-flow hypothesis using new research tools and a new model.

Using a sample of 1090 different U.S. firms, share repurchase announcement effects are examined. Firstly, a significantly positive three-day cumulative abnormal return is found of 1.4452%. Cumulative abnormal returns surrounding share repurchase announcements can be compared to previous studies. When my event study is compared with the more earlier studies by Stephens and Weisbach (1998), Grullon and Michaely (2002), Kahle (2002), Peyer and Vermaelen (2005), Lie (2005), and Chan et al. (2007), the 1.4452% cumulative abnormal return is substantially lower. However, my event study shows similar results when compared to the more recent studies by Wang et al. (2009), Chang et al. (2010), and Babenko et al. (2012). Also, the 81-day period cumulative abnormal return should not be used to test the first hypothesis. Since the pre-announcement period is not a good control period, it should not be included in the event window (Vermaelen, 1981). Its only use is observing if a delayed reaction to the share repurchase announcement takes place. As a result, my first hypothesis, stating that share repurchase announcements in the U.S. between 2006 and 2016 are followed by positive cumulative abnormal returns, is not rejected.

In the multiple regression, the explanatory variable return on assets and the explanatory interaction variable of high cash and market-to-book ratio is significant. The negative coefficient of return on assets indicates that firms with declining profitability are rewarded by a positive cumulative abnormal return when a share repurchase is announced (Grullon and Michaely, 2002). The negative coefficient of the interaction of high cash and market-to-book ratio suggests that the market reacts positively when firms with high levels of cash and low investment opportunities (proxied by Tobin’s Q) announce to repurchase shares (Lie, 2005). As shown, two of the explanatory variables significantly predicted the cumulative abnormal return, in line with the free cash-flow hypothesis. So concluded can be, that the second hypothesis, stating that share repurchase announcements in the U.S. between 2006

(32)

cash-flow hypothesis, is not rejected. Also, none of the interaction variables of crisis and an explanatory variable are significant in the multiple regression. This evidence supports that the effect of the free cash-flow hypothesis during the financial crisis was marginalized. As a consequence, the third hypothesis will not be rejected either.

Evidence for the free cash-flow hypothesis implicates that agency costs still have a noticeable impact on a firm. Meaning, that corporate governance can still use improvement, to prevent managers from making bad net present value decisions. Moreover, share repurchase programs can be used to limit managers access to free cash-flow, resulting in a positive market reaction to the announcement.

6.2. Discussion

A limitation of the study is that the database used to collect share repurchase announcements, ZEPHYR, is not complete. When compared to other information databases, some announcements are missing. And secondly, some coefficients in my study are not significant because the sample size is too small. So for further research, the use of multiple sources of data and an increase in the sample size can give a more accurate, complete result.

(33)

7. Reference list

Allen, F., & Michaely, R. (2003). Payout policy. Handbook of the Economics of Finance, 1, 337-429.

Asquith, P., & Mullins Jr, D. W. (1986). Signalling with dividends, stock repurchases, and equity issues. Financial Management, 27-44.

Babenko, I., Tserlukevich, Y., & Vedrashko, A. (2012). The credibility of open market share repurchase signaling. Journal of Financial and Quantitative Analysis, 47(05), 1059-1088.

Bagwell, L. S., & Shoven, J. B. (1988). Share repurchases and acquisitions: An analysis of which firms participate. In Corporate takeovers: Causes and

consequences (pp. 191-220). University of Chicago Press.

Bagwell, L. S. (1991). Share repurchase and takeover deterrence. The Rand journal of economics, 72-88.

Bainbridge, S. M. (2012). Corporate governance after the financial crisis. Oxford University Press.

Bhattacharya, S. (1979). Imperfect information, dividend policy, and “the bird in the hand” fallacy. Bell journal of economics, 10(1), 259-270.

Billett, M. T., & Xue, H. (2007). The takeover deterrent effect of open market share repurchases. The Journal of Finance, 62(4), 1827-1850.

Boehmer, E., Masumeci, J., & Poulsen, A. B. (1991). Event-study methodology under conditions of event-induced variance. Journal of financial economics, 30(2), 253-272.

(34)

Bonaimé, A. A., Hankins, K. W., & Harford, J. (2013). Financial flexibility, risk management, and payout choice. Review of Financial Studies, hht045.

Bozanic, Z. (2010). Managerial motivation and timing of open market share repurchases. Review of Quantitative Finance and Accounting, 34(4), 517-531.

Brav, A., Graham, J. R., Harvey, C. R., & Michaely, R. (2005). Payout policy in the 21st century. Journal of financial economics, 77(3), 483-527.

Carhart, M. M. (1997). On persistence in mutual fund performance. The Journal of finance, 52(1), 57-82.

Chan, K., Ikenberry, D. L., & Lee, I. (2007). Do managers time the market? Evidence from open-market share repurchases. Journal of Banking & Finance, 31(9), 2673-2694.

Chang, S. C., Chen, S. S., & Chen, L. Y. (2010). Does prior record matter in the wealth effect of open-market share repurchase announcements?. International Review of Economics & Finance, 19(3), 427-435.

Comment, R., & Jarrell, G. A. (1991). The relative signalling power of Dutch‐auction and fixed‐price self‐tender offers and open‐market share repurchases. The Journal of Finance, 46(4), 1243-1271.

Dann, L. Y. (1981). Common stock repurchases: An analysis of returns to bondholders and stockholders. Journal of financial Economics, 9(2), 113-138.

DeAngelo, H., DeAngelo, L., & Skinner, D. J. (2008). Corporate payout policy. Foundations and trends in finance, 3(2-3).

Décamps, J. P., Mariotti, T., Rochet, J. C., & Villeneuve, S. (2011). Free cash flow, issuance costs, and stock prices. The Journal of Finance, 66(5), 1501-1544.

(35)

Dittmar, A., Mahrt-Smith, J., & Servaes, H. (2003). International corporate governance and corporate cash holdings. Journal of Financial and Quantitative analysis, 38(01), 111-133.

Dunsby, A. (1995). Share repurchases, dividends, and corporate distribution policy.

Easterbrook, F. H. (1984). Two agency-cost explanations of dividends. The American Economic Review, 74(4), 650-659.

Farre-Mensa, J., Michaely, R., & Schmalz, M. (2014). Payout Policy. Annual Review of Financial Economics, 6(1), 75-134.

Fenn, G. W., Liang, N., & Prowse, S. (1997). The private equity market: An overveiw. Financial Markets, Institutions & Instruments, 6(4), 1-106.

Fernando, C. S., May, A. D., & Megginson, W. L. (2012). The value of investment banking relationships: evidence from the collapse of Lehman Brothers. The Journal of Finance, 67(1), 235-270.

Gao, H., Harford, J., & Li, K. (2013). Determinants of corporate cash policy: Insights from private firms. Journal of Financial Economics, 109(3), 623-639.

Grullon, G., & Michaely, R. (2002). Dividends, share repurchases, and the substitution hypothesis. the Journal of Finance, 57(4), 1649-1684.

Grullon, G., & Michaely, R. (2004). The information content of share repurchase programs. The Journal of Finance, 59(2), 651-680.

Harford, J., Li, K., & Zhao, X. (2008). Corporate boards and the leverage and debt maturity choices. International Journal of Corporate Governance, 1(1), 3-27.

Ikenberry, D., Lakonishok, J., & Vermaelen, T. (1995). Market underreaction to open market share repurchases. Journal of financial economics, 39(2), 181-208.

(36)

Iyer, S. R., & Rao, R. P. SHARE REPURCHASES AND THE FLEXIBILITY HYPOTHESIS.

Jagannathan, M., Stephens, C. P., & Weisbach, M. S. (2000). Financial flexibility and the choice between dividends and stock repurchases. Journal of financial

Economics, 57(3), 355-384.

Jensen, M. C. (1986). Agency cost of free cash flow, corporate finance, and

takeovers. Corporate Finance, and Takeovers. American Economic Review, 76(2).

John, K., & Williams, J. (1985). Dividends, dilution, and taxes: A signalling equilibrium. the Journal of Finance, 40(4), 1053-1070.

Jolls, C. (1996). The role of incentive compensation in explaining the stock-repurchase puzzle.

Kahle, K. M. (2002). When a buyback isn’t a buyback: Open market repurchases and employee options. Journal of Financial Economics, 63(2), 235-261.

Kalcheva, I., & Lins, K. V. (2004). International Evidence on Cash Holdings and Expected Managerial Agency Problems. ECGI. Finance Working Paper, 42.

Lang, L. H., Stulz, R., & Walkling, R. A. (1991). A test of the free cash flow

hypothesis: The case of bidder returns. Journal of Financial Economics, 29(2), 315-335.

Lie, E. (2000). Excess funds and agency problems: an empirical study of incremental cash disbursements. Review of Financial Studies, 13(1), 219-248.

Lie, E. (2005). Operating performance following open market share repurchase announcements. Journal of Accounting and Economics, 39(3), 411-436.

(37)

MacKinlay, A. C. (1997). Event studies in economics and finance. Journal of economic literature, 35(1), 13-39.

McNally, W. J. (1999). Open market stock repurchase signaling. Financial management, 55-67.

Miller, M. H., & Modigliani, F. (1961). Dividend policy, growth, and the valuation of shares. the Journal of Business, 34(4), 411-433.

Miller, M. H., & Rock, K. (1985). Dividend policy under asymmetric information. The Journal of finance, 40(4), 1031-1051.

NBER. (2012). US Business Cycle Expansions and Contractions. Retrieved from http://www.nber.org/cycles.html#announcements

Peyer, U. C., & Vermaelen, T. (2005). The many facets of privately negotiated stock repurchases. Journal of Financial Economics, 75(2), 361-395.

Pinkowitz, L., & Williamson, R. (2004). What is a dollar worth. The market value of cash.

Stephens, C. P., & Weisbach, M. S. (1998). Actual share reacquisitions in open‐ market repurchase programs. The Journal of finance, 53(1), 313-333.

Stulz, R. (1990). Managerial discretion and optimal financing policies. Journal of financial Economics, 26(1), 3-27.

Vermaelen, T. (1981). Common stock repurchases and market signaling: An empirical study. Journal of financial economics, 9(2), 139-183.

Wang, C. S., Strong, N., Tung, S., & Lin, S. (2009). Share repurchases, the

(38)

8. Appendix

A. Used Datastream variables in this study with name and description

Datastream Code Name Description

PTBV Price-to-Book Value This is the share price divided by the book value per share

MV Market Value The share price multiplied

by the number of ordinary shares in issue. The amount in issue is updated

whenever new tranches of stock are issued or after a capital change

WC02999 Total Assets The sum of cash,

marketable securities,

accounts receivable, prepaid expenses, inventory, fixed assets, intangible assets, goodwill, and other assets

WC02001 Cash and Short-term

Investments

The sum of cash and highly liquid investments that are readily convertible to known amounts of cash

WC01651 Net Income Before

Preferred Dividends

Represents income after all operating and non-operating income and expense,

reserves, income taxes, minority interest and extraordinary items

ISIN International Securities

Identification Number

Referenties

GERELATEERDE DOCUMENTEN

Het is evenwel ook geschikt als introductie voor de beginnende aios radiologie, maar zou in mijn optiek Felson niet kunnen vervangen, omdat Felson toch nog iets nadrukkelijker in

Additionally, announcing firms on Hong Kong market experience a period of share price underperformance before the initial announcements, while China mainland markets are observed

Therefore in situations of high uncertainty where information asymmetries are increased, as measured by higher cash flow volatility or higher R&D expenses, Continental

The abnormal returns to bidder firms acquiring private targets are reported to be positive and significant for both cash as well as equity deals, where the latter have

As argued earlier, institutions are taxed less on their dividends in the US and since these institutions invest predominantly in large firms, the effect of a share

Credit rating announcements published before the financial crisis show large negative cumulative average abnormal returns for downgrades in all event windows except for

Here UPR represents the variable unexpected change in the risk premium, UTS the variable unexpected change in the term structure, UI the variable unanticipated change in rate

Mean annual return on the FTSE All Share Index by quartile of the prior six months share of equity issues in total equity and bond issues.. Figure 3 shows the mean returns