The real costs and benefits of credit ratings : the case of mergers and acquisitions

(1)

The Real Costs and Benefits of Credit Ratings:

the Case of Mergers and Acquisitions

University of Amsterdam

Master Thesis

June 2018

Student name:

Anastasia Kuzmina

Student number: 11713429

Email:

ang.kuzmina@gmail.com

Faculty:

Economics and Business

Program:

MSc in Finance

(2)

2

Statement of Originality:

This document is written by Student Anastasia Kuzmina 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)

3

Abstract

Currently, there are a lot of articles that explain the role of credit ratings, as well as describe the positive benefits for the firm. Meanwhile, there is a gap in the literature – as both benefits and real costs of credit ratings in case of M&A were barely researched. This paper takes an endeavor to establish a check of a likelihood of being an acquirer in case of being close to the credit threshold, which can involve the possibility of the downgrade of the credit rating. For the acquirer, a possibility to perform an acquisition in such case can lead to negatively associated costs of both surpassing the investment opportunity and the increase in the cost of debt. I show that the acquirers indeed do fewer acquisitions in a year which reflect the real costs of credit ratings. Furthermore, the method of payment and CAR effects were tested for acquirers in relation to the credit rating itself. As a result, the credit ratings influence the decision about the method of payment, while the announcement returns are not affected by the credit rating.

(4)

4

1. Introduction

Credit rating agencies (CRAs) play a vital role in assessing the creditworthiness of a firm, security or bond. CRAs disclose information to the public and reduces the complexity of assessing the creditworthiness of a firm into a grade ranging from AAA to D. In fact, credit ratings are found to be an important indicator for financial markets and the most important CRAs, S&P, Moody’s and Fitch, provide companies that issue a bond with a credit rating. In the literature credit ratings have multiple functions: mainly (1) reducing asymmetric information between the firm and the outside investors, (2) give access to a larger pool of capital, (3) acting as an instrument for public regulations and (4) allows to measure the creditworthiness of firms. Boot et al. (2006) argue that credit ratings can act as “focal points”, meaning that it positions a monitoring role through credit rating methodology and guide investors. Boot et al. (2006) confirm that credit ratings have a real impact, as most of the investors indeed treat the announced ratings as valuable and reliable, as a result, they take their investment decisions based on it. In addition, Dallas (1997) concludes that credit ratings allow corporations to get a public access to wider range of financing opportunities within the debt markets. However, the credit ratings also have some negative effects such as: (1) which may take a form of agency problems (Hermalin & Weisbach, 2012), (2) as well as surpassing of investment opportunities (Almeida, Cunha, & Ferreira, 2017), (Aktas, Karampatsas, Petmezas, & Servaes, 2018) to get a higher credit rating. (3) It can tricker an increase in coupon rates (Bhanot & Mello, 2006). (4) It can decrease the provided trade credit (Klapper, Laeven, & Rajan, 2011). While the literature on the benefits of credit ratings is extensive, the costs of credit ratings is not. In relation to merger and acquisitions, the literature of credit ratings is not extensive, in particular, only a few papers describe the costs of credit ratings, such as Hermalin & Weisbach (2012), Aktas, Karampatsas, Petmezas, & Servaes, (2018) and Almeida, Cunha, & Ferreira (2017). The benefits of credit ratings have been more extensively studied but the research has still some limitations.

In particular, prior literature finds that credit ratings provides a monitoring role (Boot, Milbourn, & Schmeits, 2006). Furthermore, credit ratings give access to a larger pool of capital which increases the easiness of using cash in acquisitions. In addition, prior M&A literature provides evidence that the majority of acquisitions is funded by cash and therefore debt (Faccio M. & Masulis, 2005; Harford, Klasa, & Walcott, 2009). Furthermore, some studies also find that credit ratings have an impact on the quality of the acquisitions and therefore act as a disciplinary measure (Aktas, Karampatsas, Petmezas, & Servaes, 2018). In contrast, prior literature finds that there are also costs to credit ratings. Begley (2015) finds that if a firm is issuing a bond and it is

(6)

6

close to the Net debt / EBITDA thresholds, it is more likely to reduce R&D investments and SG&A costs. In addition, he finds that the gain in the short-run has a negative impact in the long-run due to reduction in patent citations and economic value. These findings implicate that firms are willing to give up investments that can cause economic value in the long run to gain in the short-run. This leads me to believe that there might be real costs to credit ratings. Acquisitions are overall value-enhancing and are large investments (Aktas, Karampatsas, Petmezas, & Servaes, 2018). An acquisition can strongly affect a firms credit rating and therefore it might be a kind of investment that a firm wants to do less off when it is close to a downgrade. Moreover, since credit ratings also have a monitoring role, they can influence the acquisition behavior of a firm. These results lead me to the following research question:

Do credit ratings influence the acquisition behavior of a firm?

As a result, the paper shows that the likelihood of doing an acquisition is significantly negatively influenced by the relative proximity of the firm to a credit rating threshold in case of the bond issuance in the same year. It means that if a firm is issuing a bond in a year it is less likely to do an acquisition to avoid a downgrade of his credit rating. My findings indicate that firms are willing to give up potentially valuable M&A investments to sustain their credit rating. This finding indicates that credit ratings have real costs. Furthermore, the paper shows that the relation between the method of payment and credit rating is curvilinear. This indicates that the highest rated acquirers are more afraid of a downgrade of their credit rating and therefore use more stock in the acquisition. While the medium graded firms are using more cash since they are less afraid of a downgrade and the downgrade will lead to relatively less damage for the manager's and firm’s reputation. The lowest rated firms are using more stock because the credit rating reflects the financial constraints of the firm. These results indicate that credit ratings have indeed a monitoring role and are able to identify the creditworthiness of a firm. Regarding the short-term performance of the acquisition, no significant results were found. The status of the credit rating signal, as it was described previously by Kerwer (2005), provides investors with creditworthiness, meaning that the potential acquirer may face bad reputation costs and complications to access the capital market. The findings, however, do not suggest that credit ratings have a disciplinary effect on firms in regards to the quality of the acquisitions.

This study contributes to the credit ratings and M&A literature. First, I identify the real costs of credit ratings by testing if a firm, that is in the proximity of a credit rating threshold and is issuing a bond in the same year, on the likelihood of doing an acquisition. This methodology is used

(7)

7

because when firms are close to the threshold of the credit rating, they are more incentivized to change their financial decisions to sustain or improve their credit rating. Furthermore, a credit rating is particularly important if a firm issues a bond in a year since an (up)downgrade will lead to a (lower) higher cost of debt. It is expected that firms that are in proximity to the threshold and issue a bond are doing fewer acquisitions. Furthermore, the methodology used leads to more robust results since using credit ratings itself does not account for variation within a credit rating. Second, I identify the benefits of credit ratings. In particular, the monitoring role and assessment of the creditworthiness of credit ratings. I test the credit rating on the method of payment and the short-term performance of acquirers. I expect to find a positive relationship between the credit rating and the percentage of stock. Furthermore, I expect to find a negative relationship between the credit rating and the short-term performance of acquirers, mainly due to the increased use of stock and the less reputational damage a credit rating will have on the firm and manager. The obtained findings have some direct implications for academics and practitioners. In particular, the findings implicate that managers focus too much on the gains in the short-term at the costs of long-run returns by giving up valuable acquisitions. Furthermore, it shows that using the credit rating as a linear function might lead to biased results.

This study builds upon the methodology of Begley (2015), who investigates the costs of credit ratings in regards to R&D and SG&A costs, and Aktas et al. (2018), which research the effect of credit ratings on acquisition likelihood. In particular, I use the same methodology as Begley (2015) in regards to the use of credit rating thresholds, based on the Net debt / EBITDA ratio, and the issuance of a bond in a difference-in-difference framework. Although Begley (2015) uses all the credit rating thresholds, my study only uses the speculative graded thresholds because I believe that investment graded firms are less likely to surpass an acquisition which is usually a one time oppertunity. This study implements his methodology for acquisitions. The methodology of Aktas et al. (2018) is used to test the effect of credit ratings on the method of payment and short-term performance. In particular, I use the 𝐶𝑅2 measure to adjust for a non-linear effect. To sum up, this study contributes to the credit rating literature by studying both the costs as well as benefits of credit ratings.

The article is organized as follows: Section 2 I describe the recent literature about the costs and benefits of credit ratings, Section 3 I provide insight into methodology and data collection, Section 4 I describe the obtained results, Section 5 is devoted to the discussion of results, Section 6 provides the final overview of the study and suggestions for the future research.

(8)

8

2. Literature review

2.1 The history of credit ratings

John Moody was concerned about the lack of information for investors and was the first who published the bond rating report for railroad bonds in 1909. His success was followed by Poor’s Publishing Company in 1916, which provided potential investors and market participants with an independent opinion about credit risks. The Standard Statistics Company published their credit rating in 1922 and Fitch Publishing Company in 1924 (White, 2010). Prior to the existence of credit ratings, the banks performed a role of credit rating agency, as they provided potential investors with information about corporate borrowers. Obtaining inside information and lending money to corporate borrowers, banking institutions were able to match financial abilities confirming the deal with their reputation. The reputation of the bank allowed to retain and attract clients, but they may cause reputational challenges in case of providing a bad borrower with a loan. The bank should make the borrower follow obligations and certify the bonds to maintain trustworthy reputation. The Glass-Steagall act of 1933 required to disclose the information and increased the need for more transparency of information, as a result, led to the creation of the Securities and Exchange Commission (SEC). This regulation led banks to restriction of investing only in safe securities and obliged them to follow the credit rating agencies. Considered regulations, as well as globalization, increased the importance of credit ratings.

Based on the information of BIS (2000) (Bank for International Settlements) there are 130 to 150 credit rating agencies across the word, which mostly focused on local markets. All in all, there are three the most prominent credit rating agencies in the world: Standard & Poor’s, Fitch’ and Moody’s Ratings. Each credit rating agency has its own methodology based on which the rating is developed to establish the credit risk. In order to provide companies with an independent opinion about its creditworthiness, S&P uses both quantitative and qualitative approaches. They analyze ten categories and rate them from 1 to 6, where 1 means the highest score. Moody’s methodology is similar to S&P’s, as both qualitative and quantitative approaches are used. The procedure is the following: the recent trends are analyzed, the weights to variables are distributed based on the following indicators: institutional stability, the level of development and income rate. Moody`s evaluate four sovereign categories: economic structure and performance, fiscal indicators, external payments and debt, monetary and liquidity factors. The third pronounced credit rating agency is Fitch Ratings, which covers debt securities offered by financial institutions across 75 countries. To collect information for the rating, Fitch Ratings sends an informational request corrected specifically to the geographic area to sovereign officials regarding the debt level

(9)

9

and repayment ability. Fitch Ratings uses non-public information to assess financial distress, as a result, it incorporate the information in risk models and examines results for sovereign borrowers.

As was considered, credit rating agencies examine various factors to be able to issue the rating for sovereign borrowers. Cantor and Packer (1996) systemize the most essential components for credit rating issuance on the country level: stage of economic development, per capita income, GDP growth, default history and debt restraint. This research was complemented by Alfonso (2003), that state that GDP per capita is the most crucial indicator for credit rating formation. Credit ratings are a forward-looking instrument in a way that it allows both investor and market participants to foresee potential events. In the same time, not all events could be considered before, it leads to the conclusion that even the high quality of credit rating cannot guarantee the outcome of potential events.

Credit ratings are instruments used to assess the creditworthiness of a firm and are used to make investment decisions Klapper et al. (2011). Based on the S&P credit rating, the AA rank means the higher credit quality or a lower probability of default in comparison to BBB bond. The investment grade AAA in S&P rating means the highest credit rating and strongest ability of the considered company to meet financial commitments. The grades from AAA to BBB- are called investment, which related to issuers with prudent creditworthiness, while grades from BB+ to D are called speculative and referred to agents which face significant obstacles in debt repayment. To construct the credit rating of the particular issuer, the CRA should assess its ability to repay obligations and obtain both financial and business information to study the ability to conduct repayments. In addition, credit rating agencies examine trends of the industries and anticipates possible economic fluctuations.

2.2 The benefits of credit ratings

As previously was considered, there are costs associated with credit ratings for firms, but at the same time, investors and firms perceive credit ratings as a useful instrument for taking decisions. Several of the most crucial benefits of credit ratings will be discussed furthermore.

First, nowadays investors face a broad range of investment opportunities and information plays the most crucial role in the selection process. In this regards, credit ratings allow to overcome the asymmetric information problem and allow investors to assess the risk of the companies in their portfolio. At the same time, credit ratings are essential instruments for firms to obtain financing, which allow achieving corporations more attractiveness and obtain allocation of resources from

(10)

10

the investor. Graham et al. (2005) discover the effect of risk raking avoidance by investors and surpass of economic value to obtain smooth earnings. Credit ratings confirmed to be used to perform capital structure decisions. In this regards to promote the company and risk reduction, managers start to disclose information voluntarily. Furthermore, Kaminsky et al. (2001) also confirm that investors due to the lack of available information are using credit ratings for taking investment decisions. The downgrade of credit rating hurts the business from gaining, as portfolio managers in this situation are restricted in investment activities due to the legal requirements. On the opposite, the upgrade of credit rating allows to speed up the capital and possibly lead to over-lending.

In addition, credit ratings allow investors to measure the creditworthiness of firms. Credit ratings have an essential role in understanding the information about the financial health of the companies, what guides financial decisions. For example, Surendranath et al. (2016) state that firms benefit from being rated by having the additional checks and balances of the credit rating agencies. Recent research has addressed the difference between firms that have different access to capital markets and study their investment behavior in this regard. Kerwer (2005) explores the functions of the credit rating and tests them using a qualitative approach, the main contribution is that companies with higher credit ratings capture lower costs of debt and obtain a higher debt capacity. In addition, the credit ratings allow to explore creditworthiness, provide investors with reliable information based on the high quality of rating, that guarantees unproblematic access to liquid capital markets.

In this regards, firms are intended to take manipulative actions associated with credit ratings to enact attractiveness and obtain funds from investors. Alissa et al. (2013) discover the actions of firms that are above and below the expected credit rating. The study shows that in order to obtain a higher credit rating and manage income-increasing earnings, the firms tend to change their accounting decisions and push the expected credit rating up. Besides, Roychowdhury (2006) found out that managers adjust operational activities within their company to avoid losses. As activities of the firm affect cash flows, managers are taking initiative to adjust real actions to meet earnings targets by decreasing R&D expenses. Further, Begley (2015) justify that the firms are also changing their R&D and SG&A expenses in case of proximity to the credit threshold to be able to sustain the current credit status to reduce costs of debt capital.

Second, credit ratings reduce the asymmetry of information between the investors and portfolio managers, which allows the rated firm to gain access to a larger pool of financing possibilities. In addition, credit ratings allow to limit the risk of the investment portfolio and overcome the problem

(11)

11

of interests between managers that are investing in higher risk securities and investors. In the literature, the importance of credit ratings was recently discussed in several research papers. Boot, Milbourn and Schmeits (2006) argue that credit ratings improve the information base about firms for investors and coordinate beliefs regarding the state of the firms. The article shows that credit ratings have a real monitoring effect and investors are making investment decisions based on them. Czarnitzki and Kraft (2007) conducted a study to test whether the ratings improve the predictive power concerning default risks. To test whether the credit ratings contribute to the unique information on the creditworthiness of the firms, the authors assess the probability of default risk and tests it on the threshold criteria. Besides, Whited (1992) and Almeida et al. (2004) explore the role of credit rating and claim the associates improvement the asymmetry of information and decrease of borrowing constraints. As credit ratings allow investors to choose portfolio companies based on their “investment grade”, they also help to maintain risks. Moreover, Rauf & Sufi (2010) explore the influence of the debt structure of firms with low credit quality, the results explain that firms with low credit rating experience more asymmetric information and issue different types of debt, both secured and subordinated. As a result of considered studies, the higher credit rating allows firms to overcome the problem of information asymmetry, and leads to a lower cost of debt as well as the increase of borrowing capacity.

Third, credit ratings, being a credible instrument, provide companies with a guidance regarding potential financial decisions. Tang (2009) explore the credit rating influence the information asymmetry and validates the significant influence on a firm’s investment decisions. Due to the revealed information about credit quality, it leads to lower costs of borrowing and increase of debt issuance in case of an upgrade of credit rating, the opposite happens in case of a downgrade. In case of a credit rating upgrade the firms have an opportunity to do more investments in comparison to downgraded firms. The credit rating upgrade allows to reduce information asymmetry and postpone investment decisions.

Fourth, high credit ratings allow to ease access to capital markets and helps to perform investment decisions concerning targets. Judge & Korzhenitskaya (2012), as well as Kisgen (2009), continue the topic of credit ratings in relation to the capital structure choice. Jughe and Korzhenitskaya (2012) find that credit ratings have a significant influence on the level of leverage for firms. In addition, the higher the rated bidder – the easier the access to debt markets and more deals are sponsored by cash in comparison to lower rated bidders. The research of Hovakimian, Kayhan, and Titman (2009) establishes that if the credit rating is lower than target the firms are considering the repurchasing decision and decrease their leverage. They are also more likely to reduce their

(12)

12

dividend payouts when they have ratings below their target. In addition, firms participate in more acquisitions when they are above the rating of the target. This paper finds that firms base their investment decisions on their credit rating. Harford and Uysal (2014) contribute to this study by studying the influence of the credit rating on acquisition activity, the study shows that the credit rating influences the ability of the firm to take place in acquisitions and in addition improves its quality. The significant contribution of the study is that firms are more likely to do an acquisition when they have a higher credit rating. Moreover, the takeover premiums are lower when a firm is rated compared to non-rated firms (Surendranath, Thanh, & Wang, 2016). Harford and Uysal (2014) indicate that credit ratings allow to decrease financial constraints and improve the likelihood of performing acquisitions.

Fifth, credit ratings signal the financial constraints of the firm to the market and in addition, they act as a disciplinary measure. Aktas et al. (2018) examine the role of the credit rating of the firm in the acquisition process and influence the stock returns, in addition, the following impact of the acquisition on rating changes is tested. The paper explores the factors that have an influence on mergers and acquisitions, in addition, it focuses on the likelihood of a credit rating change for both high and low rated firms. The article claims that an increase in a firm’s credit rating positively influence the probability of performing an acquisition, except for the highest rated firms, and lowers the announcement returns, which is consistent with financial constraints hypothesis. In addition, the acquisitions lead to a downgrade of the highly-rated firms, the opposite effect is true for firms with a low rating. The article found that firms that are close to the credit rating threshold behave in a similar manner. The main contribution of the article is that the firms with high credit ratings avoid acquisitions to escape a downgrade of their credit rating whereas the lower credit rated firms more actively participate in acquisitions as deals do not pressure their ratings. The lowest rated firms perform fewer amount of acquisitions because they are financially constraint.

2.3 The costs of credit ratings

Credit rating agencies provide potential investors with credible information and support corporations to raise money in capital markets by borrowing investments through the issuance of bonds or notes. Furthermore, due to credit rating instruments, corporations are able to obtain recognized credit status, which is used by an investor as an initial screening procedure to perform a decision and manage the portfolio to match credit risk or issue debt for a target based on their own risk preferences.

Firstly, the high credit rating allows to signal the market about the creditworthiness position in the market and possess wider access to commercial opportunities, so the downgrade of credit rating

(13)

13

decreases trade contracts with suppliers. Klapper et al. (2011) explores the credit terms and concludes that the largest and most creditworthiness buyers tend to obtain the contracts with longer maturities. This study shows that a potential downgrade of the credit rating may jeopardize contracts between buyers and sellers.

Secondly, a downgrade of the credit rating can potentially lead to an increase in the coupon rate. Bhanot and Mello (2006) state that the downgrade of a credit rating can lead to an early prepayment or an increase in coupon rate if the bond has a “rating trigger clause”. Kraft (2015) shows that credit ratings provide adjustments to rating-based contracts to issuers and allow firms to get wider access to liquid markets in case of obtaining an investment grade.

Thirdly, the credit rating downgrade can give provide the market with a signal about declined managerial quality to the market, more broadly the reputation of manager can be worsened by credit rating downgrade. Based on the research of Boot et al. (2006), credit rating agencies are acting as monitors, such CRAs allow investors receive information about the quality of the management team, it means that the reputation of manager can worsen and lead to costly signaling to the market. Credit ratings lead to a better ability of management control, but Hermalin and Weisbach (2012) conclude that it increases the cost of compensation and leads to agency problems. The article demonstrate that costs may outweigh benefits, which leads to a reduction of firm value. Hermalin and Weisbach (2012) argue that managers are incentivized to manipulate disclosure of financial information and prefer less disclosure ex-post but always disclose in case of voluntary actions. The same result was found by Graham et al. (2005): managers prefer to disclose information to meet short-run earnings targets and reduce the information asymmetry. They state that there are three main reasons for disclosure, which is to (i) promote transparency in reporting, (ii) reduce information risk and (iii) to face the lacks of mandatory reporting. In the same time, as such disclosure of policy has its own benefits, firms also face complications to maintain the same policy in the future, that may influence the reputation, as well as costly to share information with competitions. As a result, the article shows that CFOs are willing to surpass positive NPV projects in order to meet short-term financial goals which lead to an upgrade of credit rating. This tends me to believe that firms are willing to give up valuable M&A investments to sustain a higher credit rating.

In addition, the adverse effect of information disclosure was discussed by Kliger and Sarig (2000), they show the negative effects of credit rating reports, stating that paying for credit ratings the firms may not disclose inside information into the public. Private information may harm the welfare of firms and make competitors benefit, the article sheds a light on the credit rating private

(14)

14

information, the unique contribution in developed methodology, exploring the change of pricing reaction based on the credit rating changes. The result of the study shows that the update of credit ratings does not affect the value of the firm, but at the same time is valuable. To conclude, the article states that credit ratings allow investors to be updated about the state of companies. Verrecchia (1983) also explores the costs of disclosure of information from the perspective of managers, stating that the degree of disclosure has costs. He states that disclosure is costly due to three main reasons: (i) communication of information can be costly, (ii), there are costs associated with producing information, (iii), disclosure can have side effects, as competitors receive valuable information. This is also related to credit ratings as credit ratings are costly for the firm to get and may influence the reputation of the firm and manager. Edmans et al. (2013) continue the research of Verrecchia (1983) regarding the cost of disclosure of financial results, showing that the process of providing a credit rating is costless, but this policy leads to costs due to influence on real investment. Increase in growth opportunities leads to decrease of disclosure, as investments are more important for firm rather than disclosure benefits

Additionally, as the previous research shows that decisions of firms are highly correlated with credit rating, the investment behavior can also be influenced by the current credit rating. Aktas et al. (2018) found a connection between the credit rating level and acquisitions. The higher credit rating leads to decrease of the likelihood of acquisitions and increase returns, but in case of a medium credit rating lead to more acquisitions. In addition, the article states that acquisitions are more likely to decrease the current credit rating for highly-rated firms and increase the rating for low rating firms. The main contribution is, that firms with high credit rating prefer to surpass investment opportunities due to the unwillingness to being downgraded. Campello et al. (2010) state that firms with low credit ratings face financial constraints and are more likely to surpass investment opportunities. Whited (1992) shows that the financial position of the firm influences its investment decision. He studies the ability of financially constrained firms to make investments, and focuses on the information asymmetry in debt markets. The study confirms the previous research results that information affects corporate investment decisions. Firms which are dependent on a credit rating decline investments due to associated cost of credit rating downgrades. Almeida et al. (2017) explore the costs of credit ratings, showing that firms decline investments and dependence on credit ratings, which happens in case of a downgrade of sovereign rating, leading to an increase in the cost of debt. Jorion et al. (2005) confirms that credit ratings could be the real cost of debt for firms and shows that both downgrade and upgrade of the credit rating have a significant influence on stock prices.

(15)

15

Lastly, the paper by Begley (2015) explores in detail the real costs of credit ratings and focuses his study on firms in proximity to the Net Debt/EBITDA thresholds to understand whether firms are indeed changing their investment behavior based on the current credit rating to meet short-run goals. In case the firm is close to the threshold of Net Debt/EBITDA ratio for credit ratings, the firm is more likely to decline the R&D level and SG&A before the bond issuance. In addition, it is shown that considered firms are also facing a decline in profitability, innovation output, and Tobin’s Q. The main goal of the study is to examine how do firms change their investment behavior in the short-run based on the credit rating criteria at the cost of economic value in the long run. As a result, the paper finds that firms that are just below the Net Debt/EBITDA threshold reduce their R&D investment by 4.7% and SGA costs by 2.3%. Firms that are near a threshold of Net Debt/EBITDA are 18 percentage points more likely to cut R&D investment and 6 percentage points more likely to reduce SG&A. In addition, the article finds that firms around the credit rating threshold experience a sharp decline in patents and paper citations. The reduced cost in the short run come at the cost of innovation, profitability and economic value in the long run.

2.4. Hypothesis

The capital structure has a great influence on corporate financial decisions during the M&A process, the acquirer considers either stock or cash to sponsor the M&A investment when in some cases uses a mixture of cash and stock means of payment. Fazzari et al. (1988) show that financial constraints affect investment decisions of firms, as they are more limited in the credit markets. Martin (1996), Faccio and Masulis (2005) found that acquiring firms tend to less pay in cash because of information asymmetry, Alschwer (2011) explored that acquiring firms which face financial constraints prefer to use more cash in their payments as the equity method leads to higher costs. Furthermore, Whited (1992) explores the information asymmetry in debt markets and posit that financial position of the firm affects their financial decision. Almeida et al. (2004) show the importance of credit rating for reduction of the asymmetric information for the value of the firm. This literature helps to understand that firms with a stable financial position are allowed to issue required funds in short convenience to be able to sponsor investment needs. Billett et al. (2011) showed that firms with low credit ratings face higher costs of debt. Frank and Goyal (2009) argue that the information asymmetry is reduced with higher credit rating. In addition, Rauf and Sufi (2010) stud the connection between the debt structure and credit quality, they mention that firms with low credit rating experience higher asymmetry of information and therefore rely on more costly forms of payment, secured bank-debt and subordinated public debt. Thus, these findings indicate that credit ratings help to expand the range of options to find M&A opportunities.

(16)

16

The form of payment was discussed in the previous literature mostly for financially constrained firms, but in fact, the connection between the credit rating and the method of payment was not broadly studied in the literature. Aktas et al. (2018) shed a light on the method of payment in M&A process based on the credit rating, they show that high credit ratings have low influence on the stock-financed acquisition. As Aktas et. al (2018) focus on the consideration of the method of payment, the methodology involves some endogeneity issues, so my analyses use different models to predict effects without any related biases. In this regard, I wonder to know how the credit ranking influences the method of sponsoring of M&A opportunity. I have an expectation that by obtaining the high credit rating the firm allows to obtain cheaper debt, it means that the increase of credit rating helps to reduce the cost of debt. On the opposite, the downgrade of credit rating leads to increase of costs of obtaining debt, what means that higher status reduced associated costs. As the credit rating is getting lower, it means that the firm does not provide a credible signal to the market and it is much more expensive to use cash in M&A, making the stock payments less expensive. As a result, the following hypothesis was constructed:

H1: Firms that have a higher credit rating are more likely to use cash than lower credit rated firms.

The CAR effect was discussed for acquirers by Scanlol et al. (1989), which show that acquisitions lead to a significant positive influence of shareholder’s wealth of acquiring companies. Walker (2000) established a positive significant connection between announcement returns for acquirers and the method of payment. Moreover, Faccio et al. (2006) show the negative abnormal returns for acquirers when the target is listed and significant positive return when the target is unlisted. The credit rating has an influence on the abnormal returns was and not broadly discussed in the previous literature, but it is important to look over this relation. The more positive returns should be obtained by firms with higher credit rating because such firms have access to the best investment opportunities and participate in a broader range of acquisitions. In addition, as the firm with high credit rating obtains fewer costs it would be eager to maintain the same credit rating. In addition, as the firm is highly rated it is expected to use more cash in case of M&A and therefore debt. An acquisition affects the credit rating of the highly rated firm more than low rated because a downgrade causes higher side effects such as reputational damage of the firm and manager. As a results, the following hypothesis was developed:

H2: The credit rating has a positive effect on the short-term performance of acquirers

As investors face a large amount of investment opportunities credible information is essential in the decision-making process. Myers (1984), Whited (1992) and Almeida et al. (2004) show that

(17)

17

credit ratings allow to reduce the asymmetry information and indeed improve investment decisions. Moreover, Boot et al. (2006) show that credit ratings are indeed perceived by investors as a credible signal in assesing of potential investment opportunities. Similar result was obtained by Graham et al. (2005), which show that investors indeed use credit ratings as the base for investment decisions. Similarly, Roychowdhury (2006) found out that managers are adjusting their activities in order to avoid associated losses. Alissa et al. (2013) explored that firms are very sensitive to a position above or below the expected rating level and tend to change accounting decisions to avoid credit rating downgrade. It leads to the conclusion, that credit ratings are important for both groups – the investors use credit ratings as a reliable instrument for the decision-making process, firms – take actions to be able to stay attractive for investors and have access to capital markets. Furthermore, Kerwer (2005) showed that firms which obtain higher credit ratings do have lower costs of debt, this result was expanded by Rauf and Sufi (2010), they show that firms with low credit ranking face difficulties in access to capital markets and high costs of debt. In addition, Begley (2015) state that firms that are close to the credit threshold (Debt/EBITDA) reduce their investments in R&D and SG&A prior to bond issuance. As the firm has an incentive to avoid a credit rating downgrade to keep costs lower and possess a broader range of investment opportunities. In order to be able to keep the credit ranking on the existing level, the firm is less likely to restrict itself from participation in an M&A opportunity. As discussed, to be able to sustain the current credit rating, the firm should keep costs and expenses down, it is especially crucial when the firm is issuing the bond to perform M&A since the credit rating will determine the yield of the bond. To summarize the previous literature, the following hypothesis was constructed:

H3: Firms that are close to the threshold of their credit rating are less likely to do an acquisition.

3. Methodology

3.1 The benefits of credit rating

The benefits of credit ratings will be tested on the method of payment and short-term performance of the firm. If firms indeed are changing their investment behavior based on the credit rating, it is most likely that they will change their method of payment and short-term performance as well. If the acquisition is very valuable, the firms do not want to pass it, they might complete the deal with more equity to sustain the higher credit ratings if their credit rating is high or to not become even lower rated if the credit rating is already very low. The advantage of using a payment in equity is that it does not affect the Net Debt but does affect the EBITDA since the EBITDA of the target will

(18)

18

be added to its own EBITDA. Therefore firms can sustain their current credit rating or even improve it by using equity in paying for the acquisition. Furthermore, if indeed firms do better acquisitions when they have a very high or low credit rating, it should be reflected in the announcement returns of the firm. The method of payment will be tested using a Probit, Tobit and OLS regressions with yearly fixed effects. The short-term performance will be tested based on OLS with and without yearly fixed effects. Firms that have the highest credit ratings are afraid to lose it and because of this reason they will pay less in cash and do better deals, while firms that are medium rated will pay the most in cash since the costs of a downgrade are relatively less. The methodology of my study regarding the method of payment and short-term performance is based on the paper of Aktas et al. (2018). His paper uses credit ratings to determine the effect of the acquisition likelihood on the method of payment.

The variable of interest is the credit rating variable. The credit ratings are converted to a numerical scale as is a common practice in the literature. The credit ratings are from 1, for the AAA rated firms, to 19, for the CCC- rated firms, which is the same method Aktas et al. (2018) use. The variable 𝐶𝑅 indicates the numerical scale for the credit ratings in the regressions. In addition, it is believed based on the study of Aktas et al. (2018) there might be a curvilinear, U shape, effect of credit rating the variable 𝐶𝑅2_{is added.}

The difference-in-difference methodology used in section 3.2 is not used to test the method of payment and short-term performance, because the requirement for a difference-in-difference model is that the treated and not treated group have to be balanced before the event occurs. As the treated and untreated group in the sample are not similar, because only a subsample of credit rating thresholds is used, it is unlikely that they will pay the same amount of cash and stock before the event. To be able to test the effect credit ratings have on the method of payment and short-term performance, the credit rating variable itself is added.

Dependent variable

The method of payment is defined as the percentage of stock used in an acquisition. The percentage of stock is derived from Thomson One and is constructed by using the formula 1 − % 𝐶𝑎𝑠ℎ if the percentage of stock is missing. The method of payment is regressed using a Probit model, Tobit model and normal OLS with yearly fixed effects. The control variables are based on the article by Faccio & Masulis (2005).

To measure the short-term M&A performance an event study will be used. The performance will be measured by the cumulative abnormal returns over two event windows. The event windows

(19)

19

will be three days and eleven days surrounding the announcement date. The cumulative abnormal return is constructed as follows

𝐶𝐴𝑅(𝑡1, 𝑡2) = ∑ 𝐴𝑅𝑖𝑡 𝑡₂

𝑡1

𝐶𝐴𝑅𝑖𝑡 is equal to the cumulative abnormal return over the period 𝑡1, 𝑡2. 𝑡 = 0 is the announcement date of the acquisition. 𝐴𝑅𝑖𝑡 is the abnormal stock return of a company on certain date. The market return model is used to construct the abnormal stock return.

𝐴𝑅𝑖𝑡 = 𝑅𝑖𝑡− 𝑅𝑚

The market return (𝑅𝑚) is the value-weighted return as defined in CRSP. 𝑅𝑖𝑡 is the return of company i at time t. The event window is equal to one trading day before the announcement of the acquisition to one trading day after the announcement of the acquisition for the three-day event window. For the eleven-day event window variable, the variable is equal to five trading days before the acquisition till five trading days after the acquisition.

Control variables method of payment

To control for the omitted variable bias, additional control variables are added. Faccio and Masulis (2005), using a sample of 3,667 firms, find that the variables 𝑀𝑎𝑟𝑘𝑒𝑡 𝑡𝑜 𝐵𝑜𝑜𝑘 𝑟𝑎𝑡𝑖𝑜, 𝐶𝑜𝑙𝑙𝑎𝑡𝑒𝑟𝑎𝑙, 𝐴𝑐𝑞𝑢𝑖𝑠𝑖𝑡𝑖𝑜𝑛 𝑖𝑛𝑡𝑒𝑛𝑠𝑖𝑡𝑦, 𝑆𝑎𝑚𝑒 𝑖𝑛𝑑𝑢𝑠𝑡𝑟𝑦 and 𝐶𝑟𝑜𝑠𝑠 𝑏𝑜𝑟𝑑𝑒𝑟 significantly influences the method of payment. They state that firms that are overvalued are more likely to pay with stock since it is cheaper for the existing shareholders to do so. Hansen (1987) also supports the argument that are firms that are overvalued are more likely to pay with stock. Furthermore, he finds that the probability of the transaction financed by stock increases with the debt level of the acquirer and decreases with the debt of the target. In addition, he states that future growth opportunities increase the likelihood of a stock payment. The 𝑀𝑎𝑟𝑘𝑒𝑡 𝑡𝑜 𝐵𝑜𝑜𝑘 𝑟𝑎𝑡𝑖𝑜 captures the effect of the future growth opportunities since it is forward-looking as well as the overvaluation of the firm. Martin (1996) contributes to the topic, exploring the effects of uncertainty of the value of the transaction and the method of payment, he is exploring such determinants as size and investment opportunities. The study informs that investment opportunities of acquirer, not size, are important for the method of payment. The results are consistent with the findings of Hansen (1987) which show that there is a negative connection between the stock payment method and firm size, because of the need to attract debt. For this reason, the control variable 𝐿𝑛(𝑎𝑠𝑠𝑒𝑡𝑠) is added. This

(20)

20

on the other hand also relates to the credit rating since the attractiveness of buying bonds from firms, among others, is determined by the credit rating of the firm. Collateral tends to make bonds more attractive since the costs of bankruptcy will be less. Indeed, Faccio and Masulis (2005) find that collateral influences the method of payment and acquirers are paying more with cash if they have more collateral. To control for the effect of collateral the variable 𝑃𝑃&𝐸

𝐴𝑠𝑠𝑒𝑡𝑠 is added.

Since there are limits to the amount of debt that can be borrowed, which may determine the method of payment significantly, the variable 𝑇𝑟𝑎𝑛𝑠𝑎𝑐𝑡𝑖𝑜𝑛 𝑣𝑎𝑟𝑖𝑎𝑏𝑙𝑒 is added. Faccio and Masulis (2005) and Schwieringa and Schauten (2007) find that the relative size of the target decreases the use of cash in an acquisition.

Faccio and Masulis (2005) find that a cross-border deal influences the method of payment due to the volatility of the exchange rates and the home bias of firms. Furthermore, Faccio and Masulis (2005) and Fishman (1989) argue that if firms encounter multiple bidders or a hostile bid, they get paid more in cash. The reason is that a stock bid has more risks and in case of a hostile bid, the acquirer wants to maximize the possibility of succeeding the takeover. Firms in the same industry are more likely to pay with stock because there is less asymmetric information and therefore targets are more likely to accept stocks.

Based on the literature review, the following regression is designed to test the first hypothesis that credit ratings firms that have a higher credit rating are more likely to use cash than lower credit rated firms: % 𝑆𝑡𝑜𝑐𝑘 = 𝛽1𝐶𝑅 + 𝛽2𝐶𝑅2 + 𝛽3𝐴𝑐𝑞𝑢𝑖𝑠𝑖𝑡𝑖𝑜𝑛 𝑖𝑛𝑡𝑒𝑛𝑠𝑖𝑡𝑦 + 𝛽4𝑀𝑎𝑟𝑘𝑒𝑡 𝑡𝑜 𝑏𝑜𝑜𝑘 𝑟𝑎𝑡𝑖𝑜 + 𝛽5 𝑃𝑃&𝐸 𝐴𝑠𝑠𝑒𝑡𝑠 + 𝛽6𝐶𝑜𝑚𝑝𝑒𝑡𝑖𝑛𝑔 𝑏𝑖𝑑𝑑𝑒𝑟 + 𝛽7𝐶𝑟𝑜𝑠𝑠 𝑏𝑜𝑟𝑑𝑒𝑟 + 𝛽8 𝑆𝑎𝑚𝑒 𝑖𝑛𝑑𝑢𝑠𝑡𝑟𝑦 + 𝛽9 𝐻𝑜𝑠𝑡𝑖𝑙𝑒 + 𝑦𝑒𝑎𝑟 𝑓𝑖𝑥𝑒𝑑 𝑒𝑓𝑓𝑒𝑐𝑡𝑠 + 𝜀𝑖𝑡

Control variables for short-term performance

As with the method of payment, the short-term performance may also be influenced by other variables than the credit rating. Travlos (1987) finds that acquirers that pay in equity experience strongly negative announcement returns, this can be influenced by the overvaluation of the stock or that the deal has risks. While cash bidders experience a positive announcement returns. This could be driven by solid growth opportunities and less risk which do not have to be shared. To control for this, the percentage of stock is added.

The size of the deal naturally also has an effect on the short-term performance, an equally bad deal that is 1% of the equity that will destroy 50% of value will experience a 0.5% decline in value

(21)

21

of the acquirer. A deal that is 50% of the acquirers' equity will, in that case, destroy 25% of the value of the acquirer. These arguments are confirmed by many pieces of research such as Scanlon, Trifts, & Pettway, (1989); Jarrell and Poulson, (1989) and Asquith, Brunner, and Mullins, (1983). To control for this effect the variable 𝐴𝑐𝑞𝑢𝑖𝑠𝑖𝑡𝑖𝑜𝑛 𝑖𝑛𝑡𝑒𝑛𝑠𝑖𝑡𝑦 is added.

Georgen and Renneboog (2004) find that acquirers experience a negative abnormal return of 2.5% from a hostile offer in contrast to the positive abnormal return of 2.5% in a friendly offer. To control for this the variable 𝐻𝑜𝑠𝑡𝑖𝑙𝑒 is added. In case of multiple acquirers, Georgen and Renneboog (2004) state that target has a higher bargaining power and therefore get paid more premium. In theory, this should lead to a lower announcement return of the bidder. To control for this effect the variable 𝐶𝑜𝑚𝑝𝑒𝑡𝑖𝑛𝑔 is added.

Denis et al. (2002) explore the effects on international operations and valuations, and using U.S. financial information for their research and they state that firms that have active international operations have a significantly greater return from international operations, what tells about the global diversification effect. The firms that tried to operate globally obtain a downward in their excess value, but firms that end up being international obtain an increase in excess value. This indicates that firms in the same industry might experience higher announcement returns. Therefore, the variable 𝑆𝑎𝑚𝑒 𝐼𝑛𝑑𝑢𝑠𝑡𝑟𝑦 is added. To control for the merger waves where overenthusiasm might influence the announcement returns yearly fixed effects are added. To sum up, in order to test whether the credit ratings have a positive effect on the announcement returns of the acquirer, the following regression is developed:

𝐶𝐴𝑅 = 𝛽1𝐶𝑅 + 𝛽2𝐶𝑅2 + 𝛽3𝐴𝑐𝑞𝑢𝑖𝑠𝑖𝑡𝑖𝑜𝑛 𝑖𝑛𝑡𝑒𝑛𝑠𝑖𝑡𝑦 + 𝛽4𝑀𝑎𝑟𝑘𝑒𝑡 𝑡𝑜 𝑏𝑜𝑜𝑘 𝑟𝑎𝑡𝑖𝑜 + 𝛽5

𝑃𝑃&𝐸 𝐴𝑠𝑠𝑒𝑡𝑠 + 𝛽6𝐶𝑜𝑚𝑝𝑒𝑡𝑖𝑛𝑔 𝑏𝑖𝑑𝑑𝑒𝑟 + 𝛽7𝐶𝑟𝑜𝑠𝑠 𝑏𝑜𝑟𝑑𝑒𝑟 + 𝛽8 𝑆𝑎𝑚𝑒 𝑖𝑛𝑑𝑢𝑠𝑡𝑟𝑦 + 𝛽9 𝐻𝑜𝑠𝑡𝑖𝑙𝑒 + 𝛽10𝐿𝑛(𝑎𝑠𝑠𝑒𝑡𝑠) + 𝛽11% 𝑆𝑡𝑜𝑐𝑘 + 𝑦𝑒𝑎𝑟 𝑓𝑖𝑥𝑒𝑑 𝑒𝑓𝑓𝑒𝑐𝑡𝑠 + 𝜀𝑖𝑡

3.2 The real costs of credit ratings

As the study focuses the costs and benefits of credit ratings, I will be focusing on the incentive of an upcoming issuance of debt and the corresponding incentives from the credit rating and the probability of becoming an acquirer follows the methodology of Begley (2015). The methodology is used to determine if companies are changing their behavior when they are close to the threshold and issuing a bond in a year. The issuance of a bond is an incentive for a firm to pay attention to its credit rating. This is because, among others, the cost of debt is determined based on the credit rating. Furthermore, since the issuance of a bond is, in general, a major event for a firm, it is most

(22)

22

likely planned in advance. The firm, therefore, has an incentive to take actions in advance to meet a certain credit rating criteria. The Net Debt/EBITDA ratio thresholds per rating are particularly important and the study will focus on the incentives around this threshold as stated by Begley (2015). The Net Debt / EBITDA is a major criterion used as a methodology to determine the credit rating and is therefore used in this study as an instrument for treated and untreated groups. My paper uses the difference-in-difference framework for the firms that are in proximity to the credit rating threshold as the treated group and for the firms that are not in proximity to the threshold as a in the control group. The treatment is tested when the firm is issuing a bond in a year compared to not issuing one. This methodology is used because it tests if the firm behaves differently when they have a high incentive to adjust their earnings to sustain and/or improve a certain credit rating. An issuance of a bond is a moment where the credit rating is the most important. Furthermore, firms that are in proximity to the threshold can more easily change their Net debt and/or EBITDA due to the ability to increase its credit rating or sustain it. This methodology tackles the variation within a certain credit rating. Therefore the methodology that I use can test the costs of credit ratings, by testing the surpassing of valuable M&A opportunities to sustain a certain credit rating.

The paper will use a difference-in-difference design to test the likelihood of being an acquirer or a target if the firm is in proximity to the credit rating threshold. For the difference-in-difference design, I will use a Net Debt/EBITDA with 10% lower bound and 40% upper bound of the threshold as the treated group and outside that bound as the untreated. This lower and upper bound region will be called from now on, rating-incentive zone (R-I Zone/HI Index). The firms in this zone are more incentivized to manage their Net Debt and EBITDA to sustain their current credit rating than in comparison to firms not in this zone. This can be explained as the firms as they are closer to the threshold and they therefore can obtain easier a higher credit rating. I will use multiple thresholds of the HI Index to test my research question. This design follows the paper of Begley (2015). The treated group in my study is the firm that is in the 𝐻𝐼 𝑖𝑛𝑑𝑒𝑥 as described above. The rest of the firms will be the control group. The thresholds are based on Standard & Poor’s and Moody’s Net Debt/EBITDA ratios and corresponding credit ratings, following the paper of Begley (2015). Begley (2015) focuses, like me, his research on decreasing costs and investments, I believe that high rated credit ratings will have less incentive to change their acquisition behavior to maintain a higher credit rating. Because decreasing costs have a lower impact on a company than passing an acquisition. I believe that firms in the higher credit ratings are less likely to give up valuable M&A investment opportunities than non-investment grades mainly because higher

(23)

23

rated firms are able to finance the acquisition with cheaper debt and are less financially constraint. Therefore, the thresholds used are 2.5, 3.0, 4.0 and 5.0, which correspond to non-investment grades. As stated in Table 1 in the Appendix, these thresholds correspond to the credit ratings of mostly non-investment graded credit ratings. The 𝐻𝐼 𝑖𝑛𝑑𝑒𝑥 variable is as one if the Net debt/EBITDA falls between the following lower bounds. The lower bound is the threshold minus 10% of the gap difference between the current threshold and the next threshold. With a threshold of 2.5, the lower bound is 2.45 (the threshold before is 2.0). The upper bound is the threshold plus 40% difference to the next threshold. For a threshold of 2.5, the upper bound is 2.70. The range of 𝐻𝐼 𝑖𝑛𝑑𝑒𝑥 is [2.45-2.70] in this case. The 𝐻𝐼 𝑖𝑛𝑑𝑒𝑥 is zero if the firm not in this range. In table 1*, the upper and lower bound of the 𝐻𝐼 𝑖𝑛𝑑𝑒𝑥 can be found.

Table 1*: The Net debt/EBITDA ratio incentive zone.

Net Debt/EBITDA 2.5 3.0 4.0 5.0

Next threshold 3.0 4.0 5.0 6.0

Difference thresholds 0.5 1.0 1.0 1.0

Lower bound (10%) 2.45 2.9 3.9 4.9

Upper bound (40%) 2.7 3.4 4.4 5.4

For the robustness check, the following thresholds will be used are 1.25, 1.50, 2.0, 2.5, 3.0, 4.0 and 5.0 to test where the likelihood is driven by non-investment graded bonds, these thresholds are used. The methodology for these thresholds is the same as the non-investment grade thresholds in the table above.

The 𝑇𝑖𝑚𝑒 variable in my study follows the article of Begley (2015) and is equal to one if the company is issuing a bond in a year. The variable 𝑇𝑖𝑚𝑒 is equal to zero if the company is not issuing a bond in a year. It is used because when a company issues a bond in a year the credit rating is more important than if it is not using a bond. The yield of the bond will directly be influenced by the credit rating and therefore the year of the issuance creates a period where a downgrade has more consequences. Furthermore, it is more likely that a bond is issued when a large amount of debt is due, in these circumstances a firms is more likely to plan the issuance of the bond in advance and therefore have more incentive and time to improve their Net Debt / EBITDA ratio. The difference-in-difference measure developed as 𝐷𝑖𝐷 = 𝐻𝐼 𝑖𝑛𝑑𝑒𝑥 ∗ 𝑇𝑖𝑚𝑒. The difference between the treatment group and the control group will be taken to get the average

(24)

24

treatment effect on the treated. This effect is captured by the 𝐻𝐼 𝐼𝑛𝑑𝑒𝑥 ∗ 𝑇𝑖𝑚𝑒 variable. The Table 2* below summarizes the methodology of the acquisition policy of a firm.

Table 2*: The methodology of the acquisition policy

𝑡 ≠ 0 𝑡 = 0 𝐷𝑖𝑓𝑓𝑒𝑟𝑒𝑛𝑐𝑒

𝑁𝑒𝑎𝑟 𝑡ℎ𝑟𝑒𝑠ℎ𝑜𝑙𝑑 (𝐻𝐼 𝑖𝑛𝑑𝑒𝑥 = 1) 𝑌𝑡≠0𝑡𝑟𝑒𝑎𝑡𝑚𝑒𝑛𝑡 𝑌𝑡=0𝑡𝑟𝑒𝑎𝑡𝑚𝑒𝑛𝑡 ∆ 𝑡𝑟𝑒𝑎𝑡𝑚𝑒𝑛𝑡 𝑁𝑜𝑡 𝑛𝑒𝑎𝑟 𝑡ℎ𝑟𝑒𝑠ℎ𝑜𝑙𝑑 (𝐻𝐼 𝑖𝑛𝑑𝑒𝑥 = 0 ) 𝑌𝑡≠0𝑐𝑜𝑛𝑡𝑟𝑜𝑙 𝑌𝑡=0𝑐𝑜𝑛𝑡𝑟𝑜𝑙 ∆ 𝑐𝑜𝑛𝑡𝑟𝑜𝑙

𝐴𝑣𝑒𝑟𝑎𝑔𝑒 𝑡𝑟𝑒𝑎𝑡𝑚𝑒𝑛𝑡 𝑒𝑓𝑓𝑒𝑐𝑡 ∆ 𝑡𝑟𝑒𝑎𝑡𝑚𝑒𝑛𝑡 − ∆ 𝑐𝑜𝑛𝑡𝑟𝑜𝑙

For analysis, the variables 𝐴𝑐𝑞𝑢𝑖𝑠𝑖𝑡𝑖𝑜𝑛 𝑙𝑖𝑘𝑒𝑙𝑖ℎ𝑜𝑜𝑑 and 𝐴𝑐𝑞𝑢𝑖𝑠𝑖𝑡𝑖𝑜𝑛 𝑖𝑛𝑡𝑒𝑛𝑠𝑖𝑡𝑦 were generated and tested. The 𝐴𝑐𝑞𝑢𝑖𝑠𝑖𝑡𝑖𝑜𝑛 𝑙𝑖𝑘𝑒𝑙𝑖ℎ𝑜𝑜𝑑 is constructed as a dummy variable equal to one if the firm is conducting an acquisition in a year and zero if not. The 𝐴𝑐𝑞𝑢𝑖𝑠𝑖𝑡𝑖𝑜𝑛 𝑖𝑛𝑡𝑒𝑛𝑠𝑖𝑡𝑦 is a variable of the sum of transaction values normalized by the market value of equity of the acquirer at the end of the previous year. The fixed effects are used to control for merger waves (Harford J. , What drives merger waves, 2005).

Based on the previously considered factors, the following regression will be used to test whether the firms that are close to the threshold of their credit rating are less likely to do an acquisition:

𝑃𝑟𝑜𝑏𝑎𝑏𝑖𝑙𝑖𝑡𝑦 𝑜𝑓 𝑏𝑒𝑖𝑛𝑔 𝑎𝑛 𝑎𝑐𝑞𝑢𝑖𝑟𝑒𝑟

= 𝛽1 𝐻𝐼 𝐼𝑛𝑑𝑒𝑥 + 𝛽2𝑇𝑖𝑚𝑒 + 𝛽3(𝐻𝐼 𝐼𝑛𝑑𝑒𝑥 ∗ 𝑇𝑖𝑚𝑒) + 𝛽4𝐷𝑖𝑠𝑡𝑎𝑛𝑐𝑒 𝑡𝑜 𝐷𝑒𝑓𝑎𝑢𝑙𝑡 + 𝛽5𝑀𝑎𝑟𝑘𝑒𝑡 𝑡𝑜 𝑏𝑜𝑜𝑘 + 𝛽6𝐿𝑛(𝑓𝑖𝑟𝑚 𝑎𝑔𝑒) + 𝜀𝑖𝑡

The 𝐷𝑖𝑠𝑡𝑎𝑛𝑐𝑒 𝑡𝑜 𝐷𝑒𝑓𝑎𝑢𝑙𝑡 is added to control for the credit rating and risk to a default of the acquirer. Because adding the credit rating could lead to a bad control bias, because it should be highly correlated with the 𝐻𝐼 𝐼𝑛𝑑𝑒𝑥, I assume that firms that are closer to default are less likely to do an acquisition. Aktas et al. (2018) find that the credit rating influences the likelihood of doing an acquisition. The variable firm age is based on the article of Harford and Uysal (2014). They use size, but I find that size is highly correlated with 𝑇𝑖𝑚𝑒, while firm age is less. The variable 𝑀𝑎𝑟𝑘𝑒𝑡 𝑡𝑜 𝐵𝑜𝑜𝑘 𝑟𝑎𝑡𝑖𝑜 is added to control for investment opportunities of the firm (Harford & Uysal, 2014). While Aktas et al. (2018) use more control variables, most of them seem to be highly correlated with the bond variable and therefore are not included in the regression. The market to book ratio is defined as the market value of the firm plus book value of assets minus the book value of equity divided by the book value of the assets. The 𝑙𝑛 (𝑓𝑖𝑟𝑚 𝑎𝑔𝑒) variable is the age of the firms from 1979 in the dataset. 𝐷𝑖𝑠𝑡𝑎𝑛𝑐𝑒 𝑡𝑜 𝐷𝑒𝑓𝑎𝑢𝑙𝑡 is based on the article of Merton (1974)

(25)

25

and later the article of Bharath and Shumway (2008). For simplicity I will use the 𝐷𝑖𝑠𝑡𝑎𝑛𝑐𝑒 𝑡𝑜 𝐷𝑒𝑓𝑎𝑢𝑙𝑡 by Bharath and Shumway (2008), which is the following:

𝑁𝑎𝑖𝑣𝑒 𝐷 = 𝐹 𝑛𝑎𝑖𝑣𝑒 𝜎𝐷= 0.05 + 0.25 ∗ 𝜎𝐸 𝑛𝑎𝑖𝑣𝑒 𝜎𝑉= 𝐸 𝐸 + 𝐹𝜎𝐸+ 𝐹 𝐸 + 𝐹(0.05 + 0.25 ∗ 𝜎𝐸) 𝑛𝑎𝑖𝑣𝑒 𝐷𝐷 =ln [ 𝐸 + 𝐹 𝐹 ] +(𝑟𝑖𝑡−1− 0.5 𝑛𝑎𝑖𝑣𝑒 𝜎𝑉2)𝑇 𝑛𝑎𝑖𝑣𝑒 𝜎𝑣√𝑇

Where 𝐷 is the market value of debt, 𝐹 the face value of debt, 𝜎𝐷 is the standard deviation of the return on debt of the last year, 𝜎𝐸 is the standard deviation of the stock return in the last year, 𝐸 is the market value of equity, 𝑟𝑖𝑡 is the return of the stock of the last year.

3.3 Regressions

The regressions which are used in this paper are the Probit, Tobit and OLS regression. The Probit and Tobit regressions are used for the 𝐴𝑐𝑞𝑢𝑖𝑠𝑖𝑡𝑖𝑜𝑛 𝑙𝑖𝑘𝑒𝑙𝑖ℎ𝑜𝑜𝑑 and the method of payment. They are used because they test the effect in fixed boundaries. As the firm can do an acquisition or not, this means that the 𝐴𝑐𝑞𝑢𝑖𝑠𝑖𝑡𝑖𝑜𝑛 𝑙𝑖𝑘𝑒𝑙𝑖ℎ𝑜𝑜𝑑 has fixed boundaries and therefore can be tested by the Probit and Tobit regressions. The method of payment also has fixed boundaries namely, pay 100% with cash or 100% with stock or in between. The OLS regression is used to test the regression with fixed effects since that is not possible with a Probit and Tobit regression.

The Probit & Tobit regressions are computed using the maximum likelihood estimation, which means that the regression is fitted so that it is the closest to the real data. The basic model looks like

𝑦𝑖 = 𝛼 + 𝛽𝑖𝑋𝑖+ 𝜇𝑖 𝑢𝑖 ~𝑁(0, 𝜎2)

Where 𝛽𝑖 is a vector of all independent variables used in the regression. It shows a dependent variable, 𝑦𝑖 with boundaries of 0 to 1 for the acquisition likelihood and 0 to 100 for the percentage of stock. The Tobit regression has the same form but does not use a maximum likelihood estimator. The Tobit regression is designed to find estimations for a linear relationship for dependent variables where only a certain range is known. While constrained observations are clustered near the constraint and therefore can lead to biased results in normal OLS regressions. For this reason, the Tobit regression is chosen.

(26)

26

4. Data

4.1 Data collection

Previously considered methodology will be applied on a set of data, obtained from ThomsonOne dataset for M&A events and bond data, in addition, the information about the returns of stocks value-weighted market returns are obtained from CRSP and fundamentals and credit ratings from Compustat. The research is focused on the available information on U.S. listed acquirers available from 1979 to 2017. Targets can be from any country. In addition, financial firms (SIC codes 6000-6999) and regulated firms (SIC codes 4900-4999), in addition, a firm with a low rating (D (default), SD (selective default), DD) were excluded from the sample, following the article of Aktas et al. (2018). All the variables are winsorized at the 1% level to eliminate the influence of outliers. Credit ratings are only used for S&P because COMPUSTAT does not provide credit ratings from Fitch and Moody’s.

4.2 Sample selection

For each of the firms, the financial information (Method of payment, Bond data and Acquisitions) was collected from Thomson Financial SDC Mergers and Acquisitions Database. The information on acquirers is based on public firms, but targets can be from private, public or subsidiaries. The percentage of acquired shares is higher than 50%, the deal data consists of tender offers, acquisitions of remaining interest, exchange offers, self-tenders and disclosed value M&A. For the current study, the S&P rating was obtained from Compustat. Firms with rating with the rating of D (default), SD (selective defaults) and DD are excluded from the consideration following the study of the Alissa et al. (2013). Based on the article of Aktas et al. (2018) I drop the acquisitions where the sum of transaction value is divided by the acquirer’s market value is smaller than 1% because the acquisition is to be assumed not meaningful.

4.3 Descriptive statistics

Table 2 represents the descriptive statistics of the publicly listed U.S. firms over the period 1979 to 2017 for annual data. Table 1A presents the summary statistics just below a credit rating threshold and just above, it contains the number of observations, mean and standard deviation. The results show that there is an almost the same amount of observations for all variables, meaning the consistency of the dataset.

In Table 1A in the Appendix, the summary statistics are presented for the 𝑇𝑖𝑚𝑒 and 𝐻𝐼 𝐼𝑛𝑑𝑒𝑥 variable, the dependent variables and the control variables for the acquisition likelihood dataset.

(27)

27

The dataset contains 10.88% firms in the treated group, where 𝐻𝐼 𝐼𝑛𝑑𝑒𝑥 is equal to one, and 89.12% of firms in the control group. Only 4.24% of the firms in the dataset issue a bond in a year. Furthermore, 7.55% of the firms are doing an acquisition in a year according to the dataset. When 1.41% of the firms are a target in a year. On average firms do for 35.58% of their market value deals. The average Net Debt / EBITDA ratio is 1.226. This can explain the low number of firms that are in near the threshold since on average firms are not even close to the first threshold of 2.5. Furthermore, the average market to book value is 2.0001, indicating that firms on average have a higher market value than their value of assets and are creating higher value from their assets. On average a firm’s balance sheet consists of 17.33% of cash. Firms have an average market value of equity of USD 3,840 million. The average natural logarithm of firm age is 2.4689. In Table 1B presents the summary statistics for the dataset of the method of payment, premium, and short-term M&A performance. The sample consists of 1,408 observations with only the premium deviating from the sample with 691 observations. The average credit rating in the dataset is equal to 10.326 indicating an average credit rating of BBB-. This is slightly lower than the average credit rating found by Aktas et al. (2018). Furthermore, firms pay on average 36.07 % with stock. 44.25% of the acquisitions are a full cash acquisition and 18.75 % are full stock acquisitions. Furthermore, firms spend 31.45% of their market value of equity on an acquisition. An acquirer has on average a 12.1% higher CAR for the three-day window and 13.82% for the eleven-day window. This is not comparable with other studies. Furthermore, 1.42% of the deals are hostile and 3.2% of the deals have a competitive bidder. This is comparable to the summary statistics of Malmendie et al. (2016).

Table 1C shows the correlation matrix. I do not find that the control variables used in my study are strongly correlated with other independent variables. The variable Ln(firm age) replaces the variable Ln(firm size) since it is less than 0,2 correlated with the 𝑇𝑖𝑚𝑒 variable. The rest of the variables are not correlated more than 0.2.

Based on Table 2, it is seen that the data is equally distributed from the point of acquisition likelihood, it means that in a sample there are firms that are participating equally in M&A activity except for the B- and Below B- credit rating. The results lead to a conclusion that the acquisition intensity for AA+ is not high and it is seen as an outlier due to the small number of observations. In addition, the credit grades could be explained by the firm size. The 𝐴𝑐𝑞𝑢𝑖𝑠𝑖𝑡𝑖𝑜𝑛 𝑖𝑛𝑡𝑒𝑛𝑠𝑖𝑡𝑦 variable indicates a curvilinear relation which can be caused by a lower market value of firms which lower credit ratings or firms do relatively larger deals.

The real costs and benefits of credit ratings : the case of mergers and acquisitions