Debt reductions versus share
repurchases: differences in firm
characteristics and excess returns
M.T.A. van de Coterlet
July, 2009
UNIVERSITY OF GRONINGEN
Faculty of Economics & Business
MSc Finance
Supervisor:
Debt reductions versus share
repurchases: differences in firm
characteristics and excess returns
Abstract
This research compares debt reducing firms with share repurchasing firms in order to determine whether they differ on several firm specific characteristics and excess returns in the UK market. Firms conducting a debt reduction have relatively high debt levels, high interest expenses, non-investment credit ratings, less return on assets and less cash and equivalents compared to share repurchasing firms. Avoiding the debt overhang problem or financial distress plays an important role in the decision to perform a debt reduction. Debt reducing firms suffer large stock price declines one year before the reduction, performing in line with the market index after the debt reduction is conducted and receive positive excess returns one and two years after the reduction. Furthermore, the credit rating of debt reducing firms improves significantly in the following years. The results suggest that debt reducing firms are undervalued prior to the reduction and that a debt reduction is a value enhancing activity. Share repurchasing firms receive high excess returns after the repurchase. However, the excess returns in the years hereafter are much lower. There is evidence of undervaluation for share repurchasing firms. Firm size is positively related to the excess return following a debt reduction and there is some evidence that financial distress negatively influences the excess return following a debt reduction.
PREFACE
TABLE OF CONTENTS
PREFACE... 2
TABLE OF CONTENTS ... 3
2 THEORETICAL BACKGROUND ... 7
2.1 Incentives why firms perform debt reductions or share repurchases... 7
2.1.1 Capital structure adjustment ... 7
2.1.2 Financial distress... 8
2.1.3 Signaling ... 9
2.1.4 Other incentives ... 11
2.2 Types of debt reductions and share repurchases ... 13
2.3 Stock price reactions following debt reductions and share repurchases. 14 2.4 Comparing debt repurchases with share repurchases ... 15
3 HYPOTHESIS ... 17
3.1 Capital structure adjustment... 17
3.1.1 Debt to total assets ratio... 17
3.1.2 Interest charges ... 18
3.2 Financial distress ... 19
3.2.1 MORE Credit ratings ... 19
3.3 Signaling... 21
3.3.1 Return on assets ... 21
3.3.2 Relative amount of cash and equivalents... 22
3.3.3 Past stock price and future stock price performance ... 22
3.5 Control variables ... 24
3.5.1 Firm Size... 24
3.5.2 Relative amount of the repurchase... 24
4 DATA & METHODOLOGY ... 25
4.2 Methodology... 26
4.2.1 Testing the hypotheses ... 26
5 RESULTS ... 30
5.1 Descriptive statistics... 30
5.2 Logit analysis ... 33
5.3 Excess returns... 34
5.4 Credit Rating ... 36
5.5 Firm specific characteristics and excess returns ... 37
5.5.1 Excess return of debt reducing firms ... 37
5.5.2 Excess return of share repurchasing firms ... 39
5.5.3 Excess returns of debt reductions and share repurchases together ... 40
5.6 Conclusion... 41
6 CONCLUSION ... 43
6.1 Limitations ... 44
6.2 Recommendations for future research ... 45
1
INTRODUCTION
This paper investigates whether debt reducing firms differ from share repurchasing firms on several different financial characteristics in the UK market. Furthermore, annual excess returns and future excess returns following debt reductions and share repurchases are examined to determine whether those events are value enhancing activities and if they are driven by under- or overvaluation. Comparing debt reductions with share repurchases is interesting because both are ways to deal with excess cash and both are ways to adjust the capital structure of a firm. Performing a debt
reduction reduces the leverage ratio while a share repurchase increases the leverage ratio of a firm.
The existing literature has examined several incentives for performing share repurchases and debt reductions. Share repurchases are performed in order to adjust the capital structure of a firm (Dittmar, 2000; Mitchell and Dharmawan, 2007), transfer wealth from debt holder to equity holders (Maxwell and Stephens, 2003: Jun et al., 2009), signal positive information to investors (Vermaelen, 1984; Zhang, 2005; Jun et al., 2009) or substitute for dividends (Grullon and Michaely, 2002; Baker et al., 2003; Mitchell and Dharmawan, 2007; von Eije and Megginson, 2008). In contrast, the incentives for debt reductions are not well investigated yet. There are three possible incentives for firms to perform debt reductions. Debt reductions can be made in order to adjust the capital structure but also to avoid financial distress. Furthermore, if management believes that the share price is overvalued a debt reduction can be an attractive alternative for a share repurchase.
price reactions following debt reductions are not yet investigated. It is somewhat surprising that there is so little research about debt reductions since debt reductions are very common and the total amount involved exceeded 60 billion dollar for US firms in 2004 (Julio, 2007).
This research shows that debt reducing firms differ on several financial
characteristics compared to share repurchasing firms. Debt reducing firms have higher debt levels, higher interest expenses, lower credit ratings, have less cash & equivalents and are less profitable than share repurchasing firms. Debt reducing firms have suffer a large stock price decline one year prior to the reduction with an average excess return of -10.5% compared to the market index. After the debt reduction the excess return
improves and debt reducing firms are performing in line with the market index. One year after the debt reduction and two years hereafter the excess returns improves to 3.1% and 4.0% respectively. Furthermore, the credit rating of debt reducing firms significantly improves in the years after the reduction. The results suggest that debt reducing firms want to avoid the costs of the debt overhang problem and financial distress. After a debt reduction firms receive higher excess returns and their credit rating improves. Although share repurchasing firms receive a significant excess return of 13.2% after the repurchase is conducted, the excess return decreases to 3.6% one year after the repurchase and to 1.7% two years after the repurchase.
2
THEORETICAL BACKGROUND
In this section the theoretical background with respect to debt reductions and share repurchases is discussed. The first part describes the incentives firms have to perform debt reductions and share repurchases and why this should add value for the firm and its shareholders. The second part discusses different types of debt reductions and share repurchases. The third part reviews results with respect debt reductions and share repurchases. The final part will compare share repurchases with debt reductions and discuss important differences and similarities between them.
2.1
Incentives why firms perform debt reductions or share
repurchases
The finance literature discusses several incentives why firms perform debt reductions or share repurchases and why this should be value enhancing activities for both the firm and its shareholders. Three incentives are discussed in this paper; capital structure adjustment, financial distress and signaling. There are however also some other incentives which are also discussed but do not make part of this research.
2.1.1 Capital structure adjustment
An important incentive for firms to conduct a debt reduction or a share repurchase is to adjust their capital structure. Issuing debt can be attractive because of tax advantages since interest expenses are tax deductable and will therefore lead to lower taxes. However, if earnings are less than expected, a firm can have difficulties repaying its debt and
Kruse et al. (2005) argue that if a firm has drifted towards a capital structure with too much debt, a tender offer for debt will move the firm closer to its optimal capital structure. In this case, firms which are highly overlevered should receive positive market reactions. Moreover, according to Julio (2007) debt overhang increases the required rate of return to equity holders and makes it difficult for a firm to obtain external financing. A relatively high level of debt also increases the probability of bankruptcy when a firm is unable to meet its financial obligations, which is also harmful for bondholders. In this case, the option to repurchase debt is valuable for both shareholders and bondholders (Julio, 2007). Repurchasing debt with binding protective covenants may eliminate restrictions that prevent a firm from pursuing value-enhancing activities (Kruse et al., 2005). The net result may be an increase in the share price and is therefore valuable for the firm, its shareholders and bondholders.
A share repurchase can also be a consequence of a capital structure adjustment strategy. Mitchell and Dharmawan (2007) find that firms can have two incentives with respect to the capital structure to perform a share repurchase; excess debt capacity or a low debt to equity/assets ratio (relatively to the optimal capital structure). In addition, Dittmar (2000) reports that a firm is more likely to conduct a share repurchase when its current leverage ratio is below its target leverage ratio.
Moving towards a target debt ratio plays an important role if a firm can choose between a debt reduction and a share repurchase. This incentive is more important by reductions/repurchases than by debt and equity issues (Hovakimian et al., 2001). Therefore, bringing a firm closer to its optimal capital structure should increase shareholders value. If a firm wants to achieve a dramatic change in capital structure is mostly performs a tender offer while an open market repurchase does not appear to be motivated by a capital structure change (Baker et al., 2003).
2.1.2 Financial distress
obligations. However, according to Robichek and Myers (1966) the costs of financial distress are already incurred when a firm comes under threat of bankruptcy, even if bankruptcy is ultimately avoided.
Firms that are in financial distress may be able to repurchase their debt at a discount (Kruse et al., 2005). Bonds of financially distressed companies are highly illiquid and could therefore be mispriced. If firms are able to coerce bondholders into selling their debt at a discount to true value, then shareholders may gain. This can be seen as a wealth transfer from bondholders to shareholders. However, bondholders could also gain because they do not incur the costs that they would have in a costly bankruptcy process (Kruse et al., 2005).
There is however a problem implementing a debt reduction for a financial
distressed firm. According to Chatterjee et al. (1995) firms in financial distress try to do a workout instead of being involved in a costly bankruptcy procedure. These workouts are implemented by either a tender offer, in which debt is repurchased with cash, or by an exchange offer, in which old debt is exchanged for new securities including debt, equity or a combination of both. The problem is that bondholders have little incentive to participate in an offer which, if successful, will benefit the non-tendering bondholders and not the tendering bondholders (Chatterjee et al, 1995).
It can be stated that a debt reduction for a financial distressed firm should be an attractive option for both bondholders and stockholders. However, since non-tendering bondholders gain more than tendering bondholders such a workout is difficult to
implement. The bondholders might not gain from this operation or even loose wealth and are therefore not willing to comply in such a process.
2.1.3 Signaling
Firms can conduct share repurchases in order to signal positive news to investors. Several aspects can be related to the signaling effects such as future performance, excess
cash, prior price declines and undervaluation. Debt reductions can be performed because
Buying back shares can be interpreted that management believes the firm’s shares are undervalued and that management is positive about the future performance of the firm. If the firm finances the share repurchase with excess cash it signals that the
company will not spend its cash on value destroying investments. Furthermore, it signals managers are confident that future cash flows are strong enough to support future
investments and debt commitments (Koller et al., 2005). Stewart (1976) shows that share repurchasing firms have in many cases excess cash and lack of value enhancing
investment projects. In the first few years after the repurchase, share repurchasing firms slightly outperform non-repurchasing firms. Using excess cash to repurchase stocks can therefore be a value enhancing activity.
However, Jagannathan et al. (2000) and Guay and Hafford (2000) show that firms with permanent higher cash flows choice to increase dividends and firms with temporally higher cash flows chose to repurchase shares. Furthermore, Guay and Hafford (2000) show that a substantial dividend increase leads to a higher share price increase than a share repurchase because substantial dividend increases have a larger permanent
component with respect to higher profits. From this point of view it is doubtful whether a share repurchase is the most appropriate way of dealing with excess cash. However, it should however lead to a positive stock price reaction.
Firms who conduct a share repurchase have often suffered prior price declines. Jun et al. (2009) states that share repurchases are negatively related to prior price declines which makes the management of a firm believe that their stock is undervalued. Zhang (2005) also finds evidence for prior price declines of share repurchasing firms. Firms which conducted an open market repurchase have a significant drop in share price 20 days prior to the announcement. This suggests that firms tended to repurchase shares when their stocks relatively underperformed the market in the prior period.
There is also evidence of undervaluation of share repurchasing firms, D’Mello and Shroff (2000) show by using an earnings-based valuation model, that 74% of the firms that repurchase shares via a fixed-priced tender offer are undervalued.
significant net signaling benefits, partially because of preventing take-over bids and increasing the value of executive stock options. However, according to Jun et al. (2009) not all authors support this view, some find that firms earnings fall after a share
repurchase. An explanation for this is that managers are overconfident about the firm’s future prospects.
Excess cash can also be used to conduct a debt reduction. Myers and Majluf
(1984) suggest that managements have information which the market does not have. Management repurchase decisions are driven by whether the stock is perceived as over-or undervalued by the management. When a firm has lack of investment oppover-ortunities over-or the current share price is too high in the eyes of the management a debt reduction by, for example, buying back corporate bonds on the market could be an attractive option.
However, according to Koller et al. (2005) debt repayments with excess cash is not a wise thing to do for several reasons. First, it is hard to interpret this as an indication of undervaluation of bonds or debt. Relative to stocks, bonds are less likely to be
undervalued unless the company is in financial distress. Thus, buying back bonds is more likely to indicate to investors that management believes stocks are overvalued; otherwise, management should buy back shares. Second, it signals that future cash flows may not be sufficient to support current levels of debt and that management therefore needs to reduce the corporate debt burden now, while it has the cash to do so. Third, for all cash returns to investors, debt repayments could signal a lack of investment opportunities.
So performing a debt repurchase with excess cash can be a value enhancing activity for the firm if the shares are overvalued in the eyes of the management. The shareholders however will not gain from this operation and will probably face a decrease of the share price.
2.1.4 Other incentives
of their cash to shareholder through share repurchases (Grullon and Michaely, 2002). For the European Union, von Eije and Megginson (2008) find that over the period from 1989 to 2005 cash dividends have declined continuously with time while the probability of repurchasing shares has increased.
The dividend irrelevancy theory of Miller and Modigliani (1961) implies that given perfect and complete capital markets, dividends and share repurchases are perfect substitutes. In this line of reasoning, shareholders should prefer a share repurchases in stead of dividends when profits on shares have a lower tax then dividend gains. There is however much discussion whether the theory of Miller and Modigliani still holds
(Grullon and Michaely, 2002). Several papers confirm the view of Miller and Modigliani but others disagree. Investors still have a preference for cash dividends even if share repurchases or stock dividends will result in a higher net pay because of lower taxes (Shefrin & Statman, 1984). There are two reasons behind this, the ‘theory of self control’1and the ‘theory of choice under uncertainty’2.
There can be an underlying reason for performing a share repurchase as a substitute for dividends. According to Cornell (2005) firms do often perform a share repurchase in order to pay the exercised stock options from executives instead of being a substitute for dividends.
A share repurchase can also be made in order to transfer wealth from debt holders to equity holders. Repurchasing equity reduces a firm’s assets which produces a decrease in the value of claims that bondholders have and therefore transfers wealth from debt holders to equity holders (Jun et al., 2009). According to Maxwell and Stephens (2003) the losses bondholders incur because of the decrease in bond returns are offset by the gains to equity holders by positive stock price returns.
Firms also have an incentive to copy repurchase decisions made by competitors within their line of business. According to Cudd et al. (2006) repurchase events are significantly enhanced by the occurrence of recent similar choices made by the firm’s competitors. This is true for both debt and equity repurchases. The probability of a
1The theory of self control is introduced by Thaler and Shefrin (1981) and implies that investors are
willing to pay a premium for self-control reasons.
2
debt/equity repurchase is positively associated with the size of the firm conducting the largest percentage of debt/equity in the same industry in the prior year.
2.2
Types of debt reductions and share repurchases
There are three alternatives for a firm to decrease its current debt level and two alternatives to repurchase shares. A firm can decrease its debt by doing an exchange offer, a tender offer or an open market repurchase. A share repurchase can be done by either a tender offer or an open market repurchase. All of these repurchases are discussed below.
The first alternative is an exchange offer. When a company does not have the cash to repay some of its debt it can do a debt reducing exchange offer, which means that a company will exchange some of its debt with shares. An exchange offer is mostly undertaken by the management of a firm in an attempt to avoid bankruptcy, and thereby preserve value for shareholders (Lie et. al., 2001). A successfully completed exchange offer substantially reduces the likelihood that a firm will enter into bankruptcy. However, the average stock price reaction is negative, apparently because the announcement also conveys information that the firm’s financial situation is worse than publicly information indicates (Lie et al., 2001).
The second and third alternative to perform a debt reduction is by doing a tender offer or an open market repurchase. Those two methods can also be used to perform a share repurchase. With a tender offer a firm offers cash in order to pay off some of it debt or to repurchase some of its shares. The firm which wants to tender specifies the amount of debt/shares it is offering to repurchase, the offer price and the period during which the offer is in effect. Open market repurchases occur less often than tender offers. When a firm performs an open market repurchase is repurchases some of its debt or shares on the market against current market prices.
repurchases that firms with higher cash flows, poorer investment opportunities, larger insider shareholdings, higher dividend yields and greater volatility of returns are more likely to conduct a tender offer. In periods of market turbulence or weak business conditions, firms prefer the open market share repurchase.
2.3
Stock price reactions following debt reductions and share
repurchases
Simply stated a debt reduction is a leverage-decreasing event and therefore the market’s reaction to debt offers should be negative (Kruse et al., 2005). However, given the incentives above, several circumstances can make a debt repurchase a value adding event for shareholders. Today only announcement returns with respect to debt reductions are investigated, no research is done on the long term stock price effects of a debt repurchase.
Two recent studies find that a debt repurchase announcement is usually followed by a positive market reaction. Julio (2006) find evidence that debt repurchases are
motivated by target adjustment motivations. The probability that a firm will repurchase some of its debt increases when the firm deviates more from its target leverage. This also influences the amount of the repurchase. He also finds significant abnormal returns around debt repurchase announcements. The abnormal returns are larger for open market repurchases than for tender offers. Open market repurchases have an average cumulative abnormal return (CAR) of 2.49% and tender offers a CAR of 1.06% (measured in a 4-day announcement period (-1,0,1,2)).
In contrast to debt reductions, many academic literature focuses on share repurchases. An open market share repurchase typically leads to a positive abnormal return of the share price (Dann, 1981) and is seen as a positive signal by investors (Vermaelen, 1981). Also tender offers are followed by positive market reactions
(Vermaelen, 1984) and are significantly positively related to subsequent earnings forecast errors and negatively related to subsequent changes in equity market risk (Dann et al., 1991).
However, according to McNally and Smith (2007) earning long-run abnormal returns when buying shares following repurchase announcement does not generate robust abnormal returns. They conclude that there are no strategies for buying shares after repurchase program announcements that would yield abnormal returns for investors. This is confirmed by Zhang (2005) and Lee et al. (2005). They both find that there is no abnormal return in the long run following share repurchases in Japan and Korea.
2.4
Comparing debt repurchases with share repurchases
According to the existing literature, announcements of both debt reductions and share repurchases lead to positive abnormal returns. With respect to share repurchases, the abnormal returns are limited to the event window, no evidence is found that share repurchases create long run positive abnormal returns. With respect to debt repurchases only announcement returns are investigated, today there is no research which has also taken long term returns into account.
When looking at the incentives for doing a debt reduction or share repurchase some similarities and differences are found. First, both debt reductions and equity repurchases are alternatives to deal with excess cash. Doing a debt reduction with excess cash will probably be followed by a negative stock price reactions while performing a share repurchase will lead to positive stock price reactions. Both are the results of
signaling effects, a debt reduction with excess cash gives a bad signal to investors while a share repurchase will give a positive signal to investors.
returns when a firm is away from its optimal capital structure. According to Hovakimian et al. (2001) adjustments to the capital structure play an important role in determining whether to do a debt reduction or a share repurchase.
This research is the first who investigates differences in firm specific characteristics between share repurchasing firms and debt reducing firms. There is however some literature which investigates differences between share repurchasing firms and non-repurchasing firms. Guffey & Schneider (2004) find several differences in financial characteristics; firms who perform a share repurchase have significantly larger total assets, higher net sales, more shares outstanding, higher market values, lower debt to equity ratios, higher return on assets, higher income and higher growth than
3
HYPOTHESIS
As stated in the theoretical background section, this research is one of the first which compares debt reductions with share repurchases. It deals with two main issues. First, it investigates whether debt reducing firms differ from share repurchasing firms on several firm specific characteristics and annual excess returns. This explains differences between debt reducing and share repurchases and determines whether debt reductions and share repurchases are value enhancing activities. Second, it investigates which firm specific characteristics influence the excess returns for both debt reductions and share repurchases.
Three incentives for firms to perform a debt reduction or a share repurchase are investigated. Those are capital structure adjustment, financial distress and signaling. To measure each of those incentives several variables are taken into account. The incentives are discussed below as well as the variables to measure them. There are also some control variables taken into account which are also discussed.
3.1
Capital structure adjustment
As stated in the theoretical background section, firms do often perform a debt reduction or a share repurchase in order to adjust their capital structure. Debt reductions are made to decrease the leverage ratio while share repurchases can be made in order to increase the leverage ratio. The variables for measuring capital structure adjustment are debt to total assets ratio and interest charges. Both variables are discussed below.
3.1.1 Debt to total assets ratio
to the debt overhang problem or even financial distress. Furthermore, firms with higher leverage ratios may have their cash flows discounted at higher rates, suggesting debt reductions would be preferred over share repurchases (Fama and French, 1992). According to Julio (2006) a debt reduction for highly levered firms can bring the firm more close to its optimal capital structure. The above arguments suggest that firms with high leverage ratios have an incentive to reduce some of their debt.
For share repurchasing firms it is the other way around. Repurchasing shares can also be done in order to bring a firm closer to its optimal capital structure because the firm reduces its equity (Baker et al., 2003) while its debt remains the same. Therefore, firms with low leverage ratios have an incentive to repurchase some of their shares. Given the arguments above I expect that firms who conduct a debt reduction will have relatively high debt to asset ratios and firms who conduct a share repurchase will have relatively low debt to asset ratios.
Within this line of reasoning, the debt to assets ratio could also have a positive effect on the excess return after a debt reduction and a negative effect after a share
repurchase. Other literature shows that firms who are in financial distress or suffer from a debt overhang problem receive a positive market reaction after a debt reduction (Kruse et al., 2005; Julio, 2006). Furthermore, firms who have a relative high debt level and
therefore reducing their debt will move towards their optimal capital structure and should receive a positive market reaction (Kruse et al., 2005). The opposite is true for share repurchasing firms, firms with a relatively low debt level will move towards their optimal capital structure when repurchasing equity, the share price reaction should therefore be positive (Baker et al., 2003).
3.1.2 Interest charges
obligated to reject value enhancing investment opportunities (Julio, 2007) and have a higher chance of financial distress. Firms with relatively high interest expenses therefore have an incentive to reduce their debt. Therefore I expect that debt reduction firms will have higher relative interest charges than share repurchasing firms.
Because high interest expenses make a firm more risky, a debt reduction should be a value enhancing activity and this should lead to an increase in the share price. In contrast, firms who have low interest expenses or even net interest income could move to a more suitable capital structure when repurchasing shares and should receive a positive stock price reaction. Therefore I expect interest charges to have a positive effect on the excess return following a debt reduction and a negative influence on the excess returns following a share repurchase.
3.2
Financial distress
If a firm is in financial distress it has a strong incentive to reduce its current debt level. The indicators for financial distress are actually the same as the indicators for capital structure adjustment; however it is difficult to determine whether a firm is in financial distress when only looking at the debt to total assets ratio or the interest charges. In order to determine financial distress the MORE (Multi Objective Rating evaluation) credit rating is taken.
3.2.1 MORE Credit ratings
A firm’s credit rating is an indicator for the firm’s creditability and its access to the debt market. A (very) high or low rating can also be seen as an indication that a firm is away from its optimal capital structure. If a firm has got a low credit rating it can have difficulties obtaining new funds or can only issue debt at high interest rates. Firms with low ratings therefore may have to reject value enhancing projects because they are unable to raise funds for those projects. A firm with a low credit rating therefore has an incentive to reduce its debt to acquire a better rating.
its optimal capital structure (Mitchell and Dharmawan, 2007). Therefore I expect that debt reducing firms will have relatively low credit ratings and that share repurchasing firms will have relatively high credit ratings. MORE credit ratings are obtained from the AMADEUS database.
The AMADEUS database makes a clear distinction between different ratings. Firms with ratings of BBB and higher have an ‘investment grade rating’, firms with a rating between CCC and BB have a ‘non-investment grade rating’ and firms with a rating of CC or lower are said to be ‘distressed’. In first instance a linear relationship between different ratings could be a way of implementing those ratings. For example; give a C rated company a value of 1, a CC rated company a value of 2, a CCC rated company a value 3 etc. The problem with this method is that this implies that the difference between a BB rated firm and a BBB rated firm is equal to the difference between an AA rated firm and a AAA rated firm which is of course not true.
In order to solve this problem two dummy variables are included to measure credit ratings. The first dummy is given a dummy variable of 0 when a firm has a so called investment grade (BBB rating or higher) and a value of 1 when a firm has a non-investment grade credit rating or a distressed rating (BB rating or lower), this dummy is stated as the credit rating dummy and is an indication that the firm is away from its optimal capital structure.
The other dummy variable measures financial distress. AMADEUS explains that a company with a CC rating shows signals of high vulnerability, a C rated company shows considerable pathological situations (the company's capacity to meet its financial commitment is very low) and a D rated company has no longer the capacity to meet its financial commitments. In order to determine financial distress firms with a distressed credit rating (rating of CC or lower) are given a dummy variable of 1 and non-distressed firms are given a dummy variable of 0 (rating of CCC or higher), this dummy is stated as the distressed credit rating dummy.
reduction should resolve this. The opposite is true for a share repurchase, a (high) investment grade credit rating can indicate that increasing its leverage ratio by
repurchasing shares will move the firm closer to its optimal capital structure and should therefore results in a positive market reaction.
I expect that debt reducing companies more often have a ‘distressed’ credit rating than share repurchasing firms. Furthermore, I expect that financial distressed firms will receive a positive market reaction following a debt reduction because the incentive to adjust their capital structure is even larger than for firms who only suffer from a debt overhang problem. The distressed credit rating dummy should therefore be positively related to the excess return after a debt reduction.
3.3
Signaling
Signaling can be an important incentive to perform a share repurchase. The management of a firm can signal several things with a share repurchase. This can be undervaluation of the share price, high expected future profits or that excess cash will not be spend on value destroying activities and future cash flows are high enough to support current debt levels. A debt reduction however can sometimes be interpreted as a bad signal if it is performed with excess cash. Investors could see this as a signal that the shares are overvalued (otherwise management would have conducted a share repurchase), future expected cash flows are too low to support current debt levels or the firm has lack of valuable investment opportunities (Koller et al., 2005). To measure signaling effects three variables are used: return on assets, cash & equivalents and past and future stock price performance which are all discussed below.
3.3.1 Return on assets
In contrast, if a highly profitable firm reduces its debt it is actually signaling that its current share price is overvalued otherwise the firm would have bought back shares. This should lead to a negative market reaction (Koller et al., 2005). Highly profitable firms therefore have an incentive to repurchase shares in stead of buying back debt and therefore I expect that share repurchasing firms and debt reducing firms will differ on this aspect.
I also expect that return on assets are negatively related to the excess return following a debt reduction and positively related to the excess return following a share repurchase. The reason behind this is also the signaling hypothesis: if a firm with high profits reduces its debt (repurchases shares) it signals that the management believes that the current share price is overvalued (undervalued) (Koller et al., 2005). Moreover, Guay and Harford (2000) find that profitable firms who perform a share repurchase are
rewarded with positive excess returns.
3.3.2 Relative amount of cash and equivalents
Performing a share repurchase or buying back debt are both ways to deal with excess cash. Investors can interpret a debt reduction with excess cash as a signal that the current share price is overvalued and a share repurchase with excess cash as a signal that the current share price is undervalued. Therefore, Koller et al. (2005) suggests that using excess cash to perform a debt reduction will lead to negative market reactions while using excess cash to repurchase shares will lead to positive market reactions.
In line with this, the relative amount of cash and equivalents should be positively related to the excess return after a share repurchase and negatively related to the excess return following a debt reduction. I expect that share repurchasing firms will have a relatively higher amount of cash and equivalents than debt reducing firms. The relative amount of cash is measured as total cash & equivalent divided by total assets.
3.3.3 Past stock price and future stock price performance
that the share price is undervalued and will therefore perform a share repurchase.
According to Zhang (2005) firms who have recently seen their stock price decreased have an incentive to conduct a share repurchases. The opposite should be true for debt
reducing firms; firms who have received an increase in the stock price have an incentive to buy back debt because the stock price could be overvalued (Myers and Majluf, 1984). Because a share repurchase is unattractive in this case, a firm can decide to reduce debt in stead of buying back shares.
To determine undervaluation the excess return after the debt reduction/share repurchase should be higher than the return one year prior to the repurchase. For
overvaluation it is the other way around, the excess return after the reduction/repurchase should be lower than the return one year prior to the reduction/repurchase. I expect that debt reducing firms are on average overvalued and that share repurchasing firms are on average undervalued.
Four measures are taken into account; past stock price performance, stock price performance after the reduction/repurchase, the stock price performance one year hereafter and the stock price performance two years hereafter. For debt reducing firms I expect to find overvaluation, returns should be smaller after the debt reduction (compared to one year before the debt reduction) or even become negative but should be back to normal in the future years. For share repurchasing firms I expect to find undervaluation. I expect that the excess return after the repurchase is higher than the return prior to the repurchase. In the following years I expect that the excess return will be back to normal because other authors (McNally and Smith, 2007; Zhang, 2005; Lee et al., 2005) did not find positive future excess returns following share repurchases.
3.4
Comparing capital structure adjustment incentives with the
signaling effects for debt reductions
hypothesis however suggests that debt reducing firms are firms who should have high share prices and good performance and that they conduct a debt reduction because a share repurchase is unattractive given the high share price. The capital structure adjustment arguments are therefore somewhat contradictory to the signaling hypothesis. The results should indicate which of those two incentives is important for the decision to reduce debt or repurchase equity.
3.5
Control variables
Besides the variables mentioned above, two control variables are also taken into account. Those are firm size and relative amount of the repurchase.
3.5.1 Firm Size
According to Kruse et al. (2005) larger firms have lower positive excess returns following debt reduction announcements than smaller firms. The same is true for share repurchases, according to Zhang (2005) smaller firms acquire a higher excess return following a share repurchases than larger firms. Therefore I expect to find no difference between share repurchasing firms and debt reducing firms with respect to size and that firm size is negatively related to the excess return of both debt reductions and share repurchases. Total sales are taken as a proxy for size, simply because larger firms have higher sales than smaller ones.
3.5.2 Relative amount of the repurchase
4
DATA & METHODOLOGY
4.1
Data
This research investigates annual debt reductions and share repurchases made by UK listed firms. A dataset is gathered from DATASTREAM which includes the increase in long term debt (Long Term Borrowings), the reduction in long term debt (Reduction In LT Debt), net proceeds from issuing equity (Net Proceeds From Sale/Issue of Com & Pfd) and the amount of equity repurchased (Com/Pfd Purchased, Retired, Converted, Redeemed) for all UK listed firms between 2000 and 2006. In total 6861 observations are found that have made at least one of the above reductions/issuances in the given time period.
To determine whether a firm has made a debt reduction the reduction in total long term debt is subtracted from increase in total long term debt. If this difference is negative the firm has made a net debt reduction. There is a large difference between the amounts of the debt reductions. In order to include only debt reductions that have real influence on the capital structure of the firm the debt reductions should be at least 1% of total assets in order to be included in the final dataset. Furthermore, In order to make the dataset as homogenous as possible only debt reducing firms who have not issued or repurchased equity in the same year as the debt reduction are included. In this case the excess return and the decision to the reduce debt are not influenced by equity repurchases or issues. This leads to a total of 755 net debt reductions made by UK firms between 2000 and 2006.
To determine a net share repurchase the amount of equity repurchased is
reducing/equity repurchasing firm is taken at the end of the year prior to the debt reduction/equity repurchase because the decision to reduce debt or repurchase shares is made in compliance with this data. Financial institutions are excluded because of their different balance sheets. For 97 debt reducing observations it is not possible to obtain all data from DATASTREAM, 68 observations where not listed (anymore) one year before the debt reduction and/or one year after the debt reduction, 155 did not have a MORE credit rating and 38 where financial institutions. For the equity repurchases these numbers are 6, 2, 23 and 2 respectively. The final dataset includes 397 debt reduction observations and 67 share repurchase observations. For the variable firm’s size the natural logarithm of the total sales is taken. All other variables are winsorized at the 1% and 99% percentile.
Because debt reducing firms may not have repurchased or issued equity and share repurchasing firms may not have reduced or issued debt this dataset contains no exchange offers. This research will therefore only look at tender offers and open market
repurchases but will not make a distinction between those two. To calculate the excess returns the FTSE All shares index is taken as the market index. This index is preferred above the FTSE 100 because the firms in the sample are very different in size and market capitalization and only a small part of those firms are listed on the FTSE 100.
4.2
Methodology
4.2.1 Testing the hypotheses
First, the different firm specific characteristics and excess returns discussed in the hypothesis section are tested with a Wilcoxon Rank test and a T-test to determine
whether debt reducing firms differ on those aspects compared to share repurchasing firms. The Wilcoxon Rank test compares the medians of the variables while the T-test compares the means.
characteristics are, as stated in the data section, taken at the end of the year prior to the debt reduction/share repurchase decision. For the variable size the natural logarithm of the total sales is taken. All other variables are winsorized at the 1% and 99% percentile. The logit analysis is as follows:
[DEDUMMY]it = +it [DTOA]1 i(t1)+ [IC]2 i(t1)+ [CRD]3 i(t1) 4
[DCRD]i(t1)+ [ROA]5 i(t1)+ [CASH]6 i(t1)+ [ER-1]7 i(t1)+ (1) 8
[SIZE]i(t1)+it
The debt/equity dummy (DEDUMMY) has a value of 1 for a debt reduction and a value of 0 for a share repurchase, debt to total assets is stated as DTOA, the relative interest charges are stated as IC, the MORE credit rating dummy between investment grade and non-investment grade firms as CRD, the credit rating dummy between distressed and non-distressed firms as DCRD, return on assets as ROA, the relative amount of cash & equivalents as CASH, the past stock price performance as ER-1 and firm size as SIZE. All variables are described in detail in Appendix 1.
Third, excess returns of both debt reductions and share repurchases are examined. The returns of the stocks and the market index are calculated by using continuously compounded returns: 1 0 , t t t i P P Ln R (2) 1 , 0 , , t m t m t m P P Ln R (3)
Where Ri,t is the return of the share, Rm,t is the return of the market index, Pt0 is
the price of the share at the end of this year, Pt1 is the price of the share at the end of
the previous year, Pm,t0 is the value of the market index at the end of this year and Pm,t1
is the value of the market index at the end of the previous year. The excess returns are calculated by correcting for the changes in the market index
Where ERi,t is the excess return.
To test for undervaluation and overvaluation the excess return after the debt reduction/share repurchase (ER0) is compared to the excess return one year prior to the reduction/repurchase (ER-1), one year hereafter (ER1) and two years hereafter (ER2). A T-test is used to compare the means and a Wilcoxon Rank test is taken to compare the medians.
The last part of this research will look at which firm characteristics are
influencing the excess returns after a debt reduction and a share repurchases. This is done by a multivariate cross sectional ordinary least squares (OLS) regression.
The regression analysis is as follows (for both debt reductions and share repurchases):
[ER0]it = +it [DTOA]1 i(t1)+ [IC]2 i(t1)+ [CRD]3 i(t1)+ [DCRD]4 i(t1)+
5
[ROA]i(t1)+ [CASH]6 i(t1)+ [ER-1]7 i(t1)+ [SIZE]8 i(t1)+ [RAR]9 i+ (5)it
All variables are described in detail in Appendix 1.
4.2.2 Multicolinearity
When performing a logit analysis or a regression analysis it is important to look at multicolinearity between the different independent variables. A correlation matrix for the independent variables is given in Table 1.
Table 1: Correlation matrix for the independent variables.
DTOA IC CRD DCRD ROA CASH ER-1 SIZE RAR
DTOA 1 0.558 0.390 0.120 -0.057 -0.395 -0.041 0.218 -0.169 IC 1 0.233 0.095 -0.101 -0.455 0.041 0.040 -0.051 CRD 1 0.149 -0.472 -0.149 -0.216 0.007 -0.070 DCRD 1 -0.148 0.019 -0.081 -0.096 -0.120 ROA 1 -0.038 0.269 0.246 0.057 CASH 1 0.000 -0.151 -0.068 ER-1 1 0.062 0.050 SIZE 1 -0.032 RAR 1
5
RESULTS
This chapter discusses the results derived from the dataset. First the descriptive statistics are presented. Second, the logit analysis which compares the debt reductions with the equity repurchases is discussed. Third, the excess returns are examined to determine whether they significantly differ from zero and hereafter the factors
influencing those excess returns are presented and discussed. Finally, a conclusion based on the results is given.
5.1
Descriptive statistics
The descriptive statistics are in table 2. The table also presents whether debt reductions and share repurchases differ on their financial characteristics by performing a T-test for the mean and a Mann-Whitney U-test for the median.
Table 2: Descriptive statistics of the debt reductions and share repurchases. In order to test for differences among debt reductions and share repurchases a T-test and a Mann-Whitney U-test are performed. The values given by the T-test and the Mann-Whitney U-tests are the T-statistic and the Z-value respectively for the two tests.
Average Median T-test Mann-Whitney U-test
Debt Share Debt Share
DTOA 0.2345 0.0502 0.2210 0.0013 9.108*** -10.355*** IC 0.0167 -0.0114 0.0108 -0.0050 7.750*** -9.889*** CRD 0.6272 0.1940 1 0 6.936*** -6.608*** DCRD 0.0327 0 0 0 1.503 -1.501 ROA 0.0019 0.1307 0.0402 0.1327 -5.891*** -7.140*** CASH 0.0925 0.2859 0.0549 0.2601 -11.083*** -8.336*** SIZE 4.7133 4.7894 4.6664 4.9316 -0.722 -0.859 ER-1 -0.1052 0.0448 -0.0754 0.0822 -1.974** -2.535** ER0 0.0004 0.1317 0.0372 0.1808 -1.872* -1.957** ER1 0.0310 0.0359 0.0520 0.1028 -0.079 -0.275 ER2 0.0407 0.0174 0.0519 0.1104 0.321 -0.062
Total debt reductions 397
Total share repurchases 67
***,**,* statistically significant at the 1%, 5% and 10% level respectively. DTOA is debt to total assets, IC are the interest charges, CRD is the credit rating dummy, DCRD is the distressed credit rating dummy, ROA is the return on assets, CASH is cash & equivalents, SIZE is firm’s size, ER-1 is the stock price performance one year prior to the reduction/repurchase, ER0 is stock price performance after the
Table 2 shows some differences between the average and median values of firm specific characteristics of debt reducing and share repurchasing firms. Another interesting difference is that the dataset contains 397 debt reductions and only 67 share repurchases. Debt repurchases seems to be more common than share repurchases in the UK. Debt reducing and share repurchasing firms differ significantly for both tests on their debt to assets ratio, interest charges, credit ratings, return on assets, relative amount of cash & equivalents, past stock price performance and current years stock price performance.
Firms which conduct a debt reduction have on average a debt to assets ratio of 0.24 while share repurchasing firms have an average debt to assets ratio of 0.05. Debt reducing firms have to use on average 1.67% of their sales to pay their interest
commitments while share repurchasing firms receive on average 1.14% of their sales as interest income. Because of the relatively low debt levels and high cash & equivalents levels, share repurchasing firms even have net interest income on average in stead of interest expenses.
For the Credit rating, 62.7% of the debt reducing firm have a non-investment grade credit rating (credit rating below BBB) while only 19.4% of the share repurchasing firms has a non-investment grade credit rating. There are only 12 firms that have a distressed credit rating (rating of CC or lower) which are all debt reducing firms. Debt reducing firms have on average a lower credit rating than share repurchasing firms but are, in most cases, not in financial distress according to their credit rating.
Debt reducing firms have on average a slightly positive return on assets of only 0.2% while share repurchasing firms have a much higher return on assets of 13.1%. Debt reducing firms also have significantly less cash & equivalents than share repurchasing firms. The differences between debt to assets, interest charges, the credit rating dummy, return on assets and the relative amount of cash & equivalents are all as predicted. As stated in the hypothesis section no difference is expected for firm size because other literature state that size is negatively related to the excess returns following debt
reductions and share repurchases. Table 2 shows there is no significant difference in size between debt reducing and share repurchasing firms.
reductions could have received positive share price reactions one year before the reduction and why share repurchasing firms could have suffered share price declines before the repurchase. Share repurchases could be made because of undervaluation and debt reductions could be made because of overvaluation. However, the opposite is true, debt reducing firms have suffered a decline in share price one year prior to the
reduction/repurchase of -10.5% while share repurchases firms have received an increase of 4.5%. This difference is significant at the 5% level for both tests. Given this result it can be stated that debt reductions are performed by ‘bad’ performing firms while share repurchases are performed by ‘good’ performing firms (with respect to the share prices). The results show that a share price increase in the past is not an important incentive to conduct a debt reduction. This is further discussed in the excess returns section.
After the debt reductions or the share repurchase are performed both test show a significant difference in excess returns. Share repurchasing firms perform significantly better after the reduction/repurchase for both tests. Debt reducing firms only have a very small excess return of 0.04% while share repurchasing firms have an excess return of 13.2%. In the hypothesis section was expected that both should receive positive excess returns because other literature indicates this. Debt reducing firms do not receive a positive market reaction which is in contrast to Kruse et al. (2005) and Julio (2006) who both find positive announcement returns. It is possible that debt reducing firms already have incurred a decline in share price just before the reduction announcement and this offsets the positive announcement returns. Another possibility is that the excess announcement returns are lost in the period after the event window of Julio (2006) and Kruse et al. (2005).
5.2
Logit analysis
The second test performed is a logit analysis which tests whether the different characteristics influence the decision to do a debt reduction or a share repurchase. Because debt to assets, interest charges and the credit rating dummy are all highly
correlated with each other and they are all measures for capital structure adjustment, three separated logit analyses are performed which each include only one of those variables. Because the distressed credit rating dummy only contains 12 distressed observations (out a total of 464) it was not possible to include this variable in the logit analysis. The results are in table 3.
Table 3: Logit analysis testing whether the different firm specific characteristics influencing the decision to do a debt reduction or a share repurchase (DEDUMMY). Because of high correlation between debt to assets, interest charges and the credit rating dummy three regressions are done each with one of those variables included and excluding the other two.
DTOA IC CRD
Coefficient Z-statistic Coefficient Z-statistic Coefficient Z-statistic
DTOA 10.258 4.907*** IC 45.211 3.454*** CRD 1.136 2.688*** CASH -5.924 -4.739*** -5.450 -3.964*** -8.363 -7.263*** ROA -6.002 -4.101*** -6.628 -4.745*** -4.801 -3.243*** SIZE -0.535 -2.152** -0.421 -1.845* -0.359 -1.636 ER-1 -0.059 -0.153 -0.509 -1.377 -0.183 -0.528 C 4.451 3.560*** 4.955 4.241*** 4.746 4.165*** McFadden R² 0.441 0.375 0.352 LR statistic 168.893 143.512 134.975 Prob(LR stat.) 0.000*** 0.000*** 0.000*** Share rep. = 0 67 Debt red. = 1 397 Total obs. 464
***,**,* statistically significant at the 1%, 5% and 10% level respectively. DTOA is the debt to total asset ratio, IC are the interest charges, CRD is the credit rating dummy, CASH is total cash & equivalents, ROA is return on assets, SIZE is the size of the firm, ER-1 is the stock price performance one year prior to the reduction/repurchase. All variables are explained in detail in Appendix 1.
variables are statistically significant at the 1% level, it can be concluded that capital structure adjustment incentives play an important role in the decision to reduce debt or to repurchase shares.
There is also evidence that signaling plays an important role in the decision to reduce debt or repurchase shares. Return on assets and cash & equivalents are both significant at the 1% level. Debt reducing firms have lower return on assets and less cash & equivalents than share repurchasing firms. High return on assets or excess cash is therefore an incentive to perform a share repurchase instead of a debt reduction. The past stock price performance however has no significant influence in the decision to reduce debt or repurchase shares. Size is significant for two of the three tests, there is some evidence that larger firms have an incentive for performing a share repurchase instead of a debt reduction.
5.3
Excess returns
This section will investigate whether the stock price reactions of both debt reductions and share repurchases are subsequent to overvaluation or undervaluation. A T-test (for the mean) and a Wilcoxon Rank test (for the median) are performed to determine whether the excess returns after the debt reduction/share repurchase is
conducted (ER0) differs from the excess return one year prior to the reduction/repurchase (ER-1), one year hereafter (ER1) and two years hereafter (ER2). Table 4 present the results.
The excess return one year prior to the debt reduction differs statistically significant at the 1% level for both test compared to the excess return after the debt reduction is conducted. In this light, debt reductions are not the consequence of a relatively high share price. In fact, if the prior tables are also taken into account, debt reducing firms have, compared to share repurchasing firms, significantly higher levels of debt, higher interest expenses, less profits (measured as return on assets) and in 62.7% of the cases a non-investment credit rating. Debt reductions are not made because
makes the firm less risky. The excess return one year after the reduction and two year after the reduction do not significantly differ compared to excess return after the reduction.
Table 4: T-test (for the mean) and a Wilcoxon Rank test (for the median) to determine whether the excess returns after the debt reduction/share repurchase is conducted (ER0) differs from the excess return one year prior to the reduction/repurchase (ER-1), one year hereafter (ER1) and two years hereafter (ER2). All excess returns are corrected for the market index. The results given for the tests are the T-values (T-test) and Z-values (Wilcoxon Rank test).
Debt reducing firms Share repurchasing firms
ER0 & ER-1 ER0 & ER1 ER0 & ER2 ER0 & ER-1 ER0 & ER1 ER0 & ER2 Difference in mean 0.1056 -0.0306 -0.0403 0.0870 0.0959 0.1144 T-test 2.698*** -0.894 -0.975 1.36 1.653 1.769* Negative ranks 183 216 202 27 30 23 Positive ranks 214 181 195 40 37 44 Wilcoxon -2.570*** -0.150 -0.493 -0.974 -1.718* -1.655* Total obs 397 397 397 67 67 67
***,**,* statistically significant at the 1%, 5% and 10% level respectively. ER-1 is the stock price performance one year prior to the reduction/repurchase, ER0 is stock price performance after the
reduction/repurchase, ER1 is the stock price performance one year after the reduction/repurchase and ER2 two years after the reduction/repurchase.
This results show that there is no evidence of overvaluation for debt reducing firms prior to the reduction decision. In fact, there is evidence of undervaluation because the excess returns improve right after the debt reduction and the share price even shows an increase in the following years. If table 2 is also taken into account there is some evidence that a debt reduction is a value enhancing activity, debt reducing firms receive a stock price decline of 10.5% one year prior to the reduction, are performing in line with the market index after the reduction and even receive excess returns of 3.1% and 4.0% one and two year after the reduction.
the repurchase there is only little evidence of undervaluation. If table 2 is also taken into account there is evidence that a share repurchase is a value enhancing activity, share repurchasing firms show excess return of 13.2% compared to the market index after the share repurchase is conducted.
5.4
Credit Rating
The results above show that the excess returns of debt reducing firms improves after the debt reduction is conducted and in the next two years compared to one year before the reduction. Furthermore, the results show that debt reducing activities are performed in order to avoid the debt overhang problem or financial distress. It is
therefore interesting to investigate if the credit rating of debt reducing firms improves as well in the next years. To test if the credit rating improves after the debt reduction is conducted, one year hereafter and two years hereafter a Mann-Whitney test (for the median) and a T-test (for the mean) are performed. Once again a firm with an investment grade credit rating (rating of BBB or above) receives a dummy value of 0 and a non-investment credit rating firm (rating of BB or below) receives a dummy value of 1. If the credit rating of a firm is not available for a particular year this firm is excluded from the dataset (especially data over the credit ratings over 2008 are often not yet available on AMADEUS at this time3). The results are in table 5.
Table 5: T-test and a Mann-Whitney U-test to determine whether the Credit rating dummy (CRD) of debt reducing firms improve after the debt reduction is conducted (t=0), one year after the debt reduction (t=1) and two years after the debt reduction (t=2) compared to the base year (t=-1). The value given for each test is the Z-value. A non-investment credit rating (rating of BB or lower) has a value of 1 and a investment credit rating (rating of BBB or higher) has a value of 0.
Total observations Average T-test Mann-Whitney U-Test
CRD t=-1 397 0.627
CRD t=0 395 0.597 -0.802 0.677
CRD t=1 395 0.559 -1.884* 1.601
CRD t=2 351 0.523 -2.610*** 2.221**
***,**,* statistically significant at the 1%, 5% and 10% level respectively.
Table 7 shows that the credit rating of the debt reducing firms improves every year. One year before the debt reduction 62.7% of the firms has a non-investment grade
credit rating. After the debt reduction is conducted this number decreases to 59.7%, one year after the debt reduction to 55.9% and two years after the debt reduction to 52.3%. Only the Credit rating dummy from two years after the debt reduction differs statistically significant from the base year for both tests. Given the results above, it can be concluded that conducting a debt reduction has a positive influence on the credit rating of a firm.
With respect to distressed credit ratings no interesting differences were found. One year before the debt reduction 12 of the 397 observations have a distressed credit rating. In the year de debt reduction is conducted 13 firms have distressed credit rating, one year after the debt reduction this number becomes 10 and two years after the debt reduction 11 firms have a distressed credit rating.
5.5
Firm specific characteristics and excess returns
The next test performed is a multivariate cross sectional OLS regression analysis which tests the influence of several firm specific variables on the excess returns for both debt reductions and share repurchases after the debt reduction/share repurchase is performed. Because some of the expected signs for the independent variables differ between debt reductions and share repurchases two separate regressions are done; one for the excess return of debt reducing firms (table 6) and one for the excess return of share repurchasing firms (table 7), table 8 shows the results of regression analysis which takes debt reductions and share repurchases together. Because of the high correlation of debt to assets, interest charges and the credit rating dummy three regression analyses are
performed which all have only one of those variables included. Because no share repurchasing firm has got a non-investment credit rating this variable is excluded from the share repurchase regression analysis.
5.5.1 Excess return of debt reducing firms
cross terms) is conducted (see Appendix 2). The White’s test shows evidence of
heteroscedatisity in the sample. Therefore, a regression analysis with heteroscedasticity-robust standard errors is performed and stated in table 6.
Table 6: Multivariate OLS Regression analysis with heteroscedasticity-robust standard errors testing the influence of several firm specific variables on the excess return at t=0 of firms who have conduct a debt reduction. All independent variables are measured at the end of the year before the debt reduction is conducted (t=-1). DTOA, IC and CRD are taken separately in three regressions because of the high correlation between those variables.
DTOA IC CRD
Coefficient T-statistic Coefficient T-statistic Coefficient T-statistic
DTOA -0.172 -0.831 IC 1.365 1.097 CRD -0.017 -0.266 DCRD -0.357 -1.516 -0.392 -1.681* -0.371 -1.595 CASH -0.374 -0.988 -0.220 -0.648 -0.313 -0.859 ROA -0.228 -1.106 -0.205 -0.971 -0.243 -1.076 SIZE 0.092 2.794*** 0.082 2.592*** 0.084 2.594*** RAR -0.596 -1.726* -0.476 -1.337 -0.531 -1.548 ER-1 0.084 1.527 0.078 1.429 0.081 1.454 C -0.373 -2.368** -0.395 -2.483** -0.368 -2.330** R² 0.049 0.051 0.047 Adjusted R² 0.032 0.034 0.030 F-statistic 2.856 3.005 2.758 Prob(F-stat) 0.006*** 0.004*** 0.008***
***,**,* statistically significant at the 1%, 5% and 10% level respectively. DTOA is the debt to total asset ratio, IC are the interest charges, CRD is the credit rating dummy, DCRD is the distressed credit rating dummy, CASH is total cash & equivalents, ROA is return on assets, SIZE is the firm’s size, RAR is the relative amount of the reduction, ER-1 is the past year stock price return. All variables are explained in detail in Appendix 1.
All three measures for capital structure adjustment (debt to assets, interest charges and the credit rating dummy) do not have a significant relation with the excess return after a debt reduction (ER0). The distressed credit rating dummy is significantly
For the signaling hypothesis, the excess return one year prior to the debt reduction, return on assets and cash & equivalents do not have a significant influence on excess return so there is no evidence that signaling influences the excess returns after the debt reduction is conducted. The relative amount of the repurchase has got a significant negative influence at the 10% level for one of the three tests so there is some evidence that the amount of the repurchase has a positive influence on the excess return.
Firm size is positively related to the excess return at the 1% level for all three tests. A larger firm receives a higher positive excess return than a smaller firm. This is
contradictory to the research of Kruse et al. (2005) who find that larger firms receive a lower share price increase than smaller firms. An explanation for this finding could be that larger firms are better monitored by investors; a debt reduction performed by a larger firm can receive more attention from the media and investors. If investors see a debt reduction as positive information an increase in the share price is a logical consequence.
5.5.2 Excess return of share repurchasing firms
Table 7 presents the OLS regression results for the share repurchasing firms.
Table 7: Multivariate OLS Regression analysis testing the influence of several firm specific variables on the excess return at t=0 of firms who have conduct a share repurchase. All variables are measured at the end of the year before the share repurchase is conducted (t=-1). DTOA, IC and CRD are taken separately in three regressions because of the high correlation between those variables.
DTOA IC CRD
Coefficient T-statistic Coefficient T-statistic Coefficient T-statistic
DTOA -0.118 -0.327 IC -0.042 -0.013 CRD -0.008 -0.059 CASH -0.357 -1.424 -0.343 -0.840 -0.342 -1.360 ROA -0.171 -0.503 -0.166 -0.486 -0.179 -0.439 SIZE -0.029 -0.438 -0.025 -0.377 -0.026 -0.379 RAR 0.064 0.149 0.071 0.165 0.067 0.156 ER-1 0.083 0.832 0.090 0.833 0.089 0.896 C 0.403 1.124 0.372 1.030 0.381 0.987 R² 0.066 0.064 0.064 Adjusted R² -0.028 -0.030 -0.030 F-statistic 0.702 0.683 0.684 Prob(F-stat) 0.649 0.664 0.663
