Published on Nov 30, 2023
The link between credit risk and return patterns on equity markets has increasingly become an area of interest. In this thesis we investigate the existence of a systematic relationship between credit ratings, as indicators of credit risks, and abnormal equity returns. In particular, we investigate the announcement effect on equity returns associated with credit rating changes.
Furthermore, we contribute to the understanding of the observed announcement effects by relating them to various components of the rating process. We base our study on a sample of credit rating changes from March 1990 to February 2006 by Moody’s and Standard and Poor’s for companies listed in the Nordic countries. We find that downgrade announcements on average are associated with negative abnormal share price reactions, whereas no systematic reaction is associated with upgrades. Through sub sample and cross-sectional analyses we gain a deeper understanding of the driving forces behind the characteristics of the observed announcement effects.
In general, we argue that variations in announcement effects are driven by various event and issuer specific characteristics and that these can be related to the relevance and implication of the information as well as the degree of market anticipation. Specifically, rating updates driven by changes in profitability and market position are more pricing relevant than those motivated by changes in capital structure. Also, rating events preceded by official opinions of the likely direction of the rating update have less pricing impact. Based on these two dimensions we identify several additional aspects of the credit rating process with implications for the impact on equity returns. These explanatory factors provide the foundation for a comprehensive analysis of the asymmetric reactions between upgrades and downgrades as well as for the cross-sectional variations for both rating events.
The aim of this thesis is to provide further insights to the link between credit risks and the corresponding impact on equity returns. In particular, our aim is to study the impact of credit rating changes on abnormal equity returns around the time of the announcement. For the purpose of identifying the factors of relevance for potential links between indicators of credit risks and return patterns on equity markets, various aspects of the credit rating process are analysed and interpreted with focus on the impact on equity investors.
Based on this approach, the purpose of the thesis isi) to investigate whether there is a systematic and robust link between indicators of credit risk and return patterns on equity markets as well as ii) to explain the dynamics of a potential relationship based on observable characteristics.
Hence, the purpose can be summarized by two research questions:
Is there a systematic link between credit ratings, as indicators of credit risk, and the return patterns on equity markets?
How can variations in issuer and event specific characteristics explain potential announcement effects on equity returns associated with credit rating changes?
The link between credit risk and bond returns is very intuitive, whereas the effect on equity returns is less clear. Several related studies have studied the existence of announcement effects on equity returns associated with credit rating transitions. The results, which are primarily based on US data, are rather inconclusive. Nevertheless, a majority of previous studies have concluded that the announcement reaction associated with downgrades is considerably larger than that found for upgrades. Appendix B provides a more comprehensive overview of related previous research.
Our main approach to test the abnormal share price reaction associated with credit rating changes is to perform an event study over relevant event windows and test for significance using regular parametric tests. For the purpose of estimating the parameters for the calculation of normal returns we have selected an estimation window involving 180 trading days from t = +60 to t = +240 , where the day of the rating announcement, the event day, is denoted t = 0 . The choice of a post-event estimation period is based on the general finding that the period prior to the announcement generally is associated with a downward (upward) share price drift for downgrades (upgrades).
To study the timing of potential share price reactions we have defined several event windows. The event window used to study the announcement effect is defined from t = 0 to t = +1. The following trading day is included since the announcement of an updated credit rating may not affect the closing price until the following day if the trading on the event day had stopped at the time of the announcement. In order to test whether the choice of event window affects the results we also used an alternative definition of the event window including the announcement only (t = 0).
The pre-event window covers the period from t = -10 to t = -1. By studying the abnormal returns during this period we are able to draw conclusions about the development of the company’s abnormal share price prior to the announcement. Finally, we define the post event window as the period between the trading days t = +2 and t = +10. The choice of post-event period is consistent with several previous studies. Also, studying a relatively short post-event window increases the ability of the test to isolate the effect associated with the announcement of the rating update.
Credit ratings are predictions of potential credit losses due to failures of making payments, delay in payments or partial payments. A credit loss is defined as the difference between what the issuer has promised to pay and what is actually received. Both Moody’s and Standard and Poor’s credit ratings measure total credit loss, including both the probability that an issuer will default as well as the expected severity of the loss if default occurs. The rating assignments for Moody’s and Standard and Poor’s are shown in Table A.1 in Appendix A.
For Standard and Poor’s the highest credit rating is AAA and the equivalent rating for Moody’s is Aaa. Bonds with ratings lower than BBB- or Baa3 are said to have non-investment grade status, implying exposure to speculative elements and substantial credit risks. The CRAs emphasize that credit ratings are fundamentally different from buy, hold or sell recommendations issued by equity and fixed income analysts. Rather than an investment advice, a credit rating should merely be viewed as the CRA’s opinion about the creditworthiness of a particular issuer.
Also, the aim of a credit rating is to measure the long term default risk rather than short term fluctuations, which are primarily driven by cyclical developments in the economy. The ultimate goal is to provide a through-the-cycle rating. Hence, credit ratings are not expected to react instantaneously to changes in default risk, but rather to exhibit a large degree of rigidness.
A common feature of publicly traded bond issues is the importance of limited information asymmetry between the issuer and the investors. In this respect the CRAs may play a pivotal role for the existence of public debt markets. By means of specialization in information gathering and access to non-public information CRAs facilitate the access of borrowers to debt markets. This function is generally referred to as signaling, which involves interpretation and provision of new information. One could argue that without this function markets for a number of debt securities would fail since it would not be efficient for individual investors to invest the required amount in reducing informational asymmetries. Rather than absorbing the costs of communicating directly to the market, potential issuers would instead find it more profitable to finance themselves with ordinary bank debt.
In addition to the signaling function, CRAs are generally assumed to have a certification role, which involves the formalization of a professional credit risk opinion. Many investors and regulators require credit rating coverage in order for issuers to achieve their confidence or approval. It is likely that participants on equity markets also are concerned with the information associated with credit ratings. The functioning of equity markets is perhaps not as closely linked to the existence of credit ratings. Nevertheless, the potential incremental information provided by the CRAs through their signaling function may also contribute to reduce informational asymmetries on equity markets.
For the rating mechanism to work it is crucial that CRAs maintain their reputation as reliable and objective sources of information. Increasingly, this has become an issue of debate. For instance, the collapse of the American utilities giant Enron was followed by an especially turbulent period for the CRAs. As late as one month before the collapse both Moody’s and Standard & Poor’s assigned the company solid investment grade ratings. A credit analyst at the time of the collapse said “Investors have been burned by rating moves so many times I'm not sure why anyone still pays attention
The intensified criticism aimed towards CRAs may reflect the increased dependence on credit ratings due to the process of disintermediation on financial markets. Also, as the market for credit ratings has expanded, larger competition between the leading CRAs may lead to temptations of exploiting the business’ inherent conflicts of interests. Nevertheless, the main impact on the announcement effect on equity returns should primarily relate to the amount of new information of relevance for shareholders that is released through the credit rating event.
The rating process starts when the CRA obtains a mandate from the issuer. A rating team is thereafter formed with a lead analyst responsible for managing the process. The team meets with the issuer’s management to review key parameters. Following the fundamental analysis, a rating committee discusses the rating recommendation by the lead analyst. The rating committee is composed by 5 to 10 credit analysts who decide with a majority vote whether to support the proposed rating opinion.
The issuer can appeal the assigned rating and must in such cases also provide support for that. The committee decides with a majority vote whether to sustain or reject the appeal. In most markets outside the US the issuer has the choice of whether to publish the rating. If the issuer decides to do so, a press release with the rationale for the rating is sent to the media. Ratings of public debt issues are monitored for a minimum of one year and thereafter the issuer can request additional surveillance. A rating change is a consequence of the surveillance process and the rating decision follows the same procedure as for the initial rating.
Thesis By : Thomas Bergha & Olof Lennströmb, Stockholm School of Economics
https://www.moodys.com
https://www.factiva.com
https://ratingsdirect.com
