Aaltodoc publication archive (Aalto University institutional repository)
School of Business | Department of Finance | Finance | 2013
Thesis number: 13135
Do credit rating announcements have informational value?: European evidence
|Title:||Do credit rating announcements have informational value?: European evidence|
|Year:||2013 Language: eng|
|Department:||Department of Finance|
|Index terms:||rahoitus; financing; luotto; credit; arviointi; evaluation; osakemarkkinat; stock markets; informaatio; information; tieto; knowledge|
» hse_ethesis_13135.pdf size:799 KB (817633)
|Key terms:||credit rating; stock market response; rating announcement|
OBJECTIVES OF THE STUDY The main objectives of this thesis are to study how the stock prices of European companies that experience a rating change behave around the announcement day and to what extent the observed abnormal returns can be attributed to the rating announcement itself rather than contaminating information. I will also be studying the effect of sample-selection criteria on the results since the use of uncontaminated samples may lead to underestimating the market response. I will also present additional analysis about factors that affect the market response.
DATA AND METHODOLOGY The credit rating data consists of issuer-specific long-term ratings by S&P, Moody´s and Fitch between January 2000 and December 2011 for companies that belong to seven major European indices. Stock market reactions to rating changes are studied by calculating mean cumulative abnormal returns (MCARs) for different time periods inside [-25,+25] day window around the announcement day. I use three different samples, all built using different criteria, to examine the effect of sample-selection criteria on the results. The information content of a rating, controlling for other sources of information, is measured by calculating the difference between CARs of [-4,-2] and [-1,+1] windows.
RESULTS When studying the combined samples I find no market response during the announcement period [-1,+1] using the uncontaminated sample. When using the clustering sample there is a statistically significant market reaction to downgrades but not for upgrades. The results of the unconditional sample suggest that the market reacts to both announcements but the reaction to upgrades remains weakly significant. However, when examining the informational content of the announcement itself it seems that downgrades suffer more from contaminating news from other sources of information and the announcement itself explains smaller proportion of the market reaction than in the case of upgrades.
The results of my additional analysis suggest that multiple notch actions lead to stronger market reactions than single notch actions and that single notch actions are more heavily anticipated. Also the level of original rating seems to affect the market reaction as companies that are initially rated below investment grade experience stronger market reactions than investment grade companies. When studying each of the seven markets separately the results vary significantly. For example a reaction to downgrades but not to upgrades can be observed when using UK data and vice versa when using Nordic data.
Electronic publications are subject to copyright. The publications can be read freely and printed for personal use. Use for commercial purposes is forbidden.