The Informational Content of Credit Ratings in Brazil: An Event Study
| Autor | Flávia Cruz de Souza Murcia - Fernando Dal-Ri Murcia - José Alonso Borba |
| Cargo | Universidade Federal de Santa Catarina. Santa Catarina, SC, Brazil - Universidade de Sao Paulo, Sao Paulo, SP, Brazil - Universidade Federal de Santa Catarina, Santa Catarina, SC, Brazil |
Page 503
The Informational Content of Credit Ratings in Brazil: An Event Study
(Conteúdo Informacional de Ratings de Crédito no Brasil: Um Estudo de Evento)
Flávia Cruz de Souza Murcia*
Fernando Dal-Ri Murcia**
José Alonso Borba***
Abstract
This study analyzes the effect of credit rating announcements on stock returns in the Brazilian market during 1997-2011. We conducted an event study using a sample of 242 observations of listed companies, 179 from Standard and Poor’s and 63 from Moody’s, to analyze stock market reaction. Abnormal returns have been computed using the Market Model and CAPM for three windows: three days (-1, +1), 11 days (-5, +5) and 21 days (-10, +10). We find statistically significant abnormal returns in days -1 and 0 for all the three types of rating announcement tested: initial rating, downgrades and upgrades. For downgrades, consistently with prior studies, our results also showed negative abnormal returns for practically all windows tested. Overall, our findings evidence that rating announcements do have information content, as it impacts stock returns causing abnormal returns, especially when they bring ’bad news’ to the market.
Keywords: credit rating; Brazilian market; event study.
JEL codes: G32.
Resumo
Este estudo analisa o efeito de anúncios de rating de crédito nos retornos de açôes no mercado brasileiro no período de 1977 a 2011. Foi conduzido um estudo de evento utilizando uma amostra de 242 observaçôes de companhias abertas, sendo 179 da Standard and Poor’s e 63 da Moody’s, para analisar a reação do mercado
Submitted 27 June 2013. Reformulated 29 November 2013. Accepted 5 February 2014. Published on-line 17 March 2014. The article was double blind refereed and evaluated by the editor. Supervising editor: Ricardo P. C. Leal. The authors would like to thank CAPES, CNPq and IAAER/Deloitte for the financial support.
*Universidade Federal de Santa Catarina. Santa Catarina, SC, Brazil. E-mail: flavia_c_souza@hotmail.com
**Universidade de São Paulo, São Paulo, SP, Brazil. E-mail: murcia@usp.br ***Universidade Federal de Santa Catarina, Santa Catarina, SC, Brazil. E-mail: jalonso@cse.ufsc.br
Rev. Bras. Finanças (Online), Rio de Janeiro, Vol. 11, No. 4, December 2013, pp. 503–526 ISSN 1679-0731, ISSN online 1984-5146
c
[circlecopyrt]2013 Sociedade Brasileira de Finanças, under a Creative Commons Attribution 3.0 license -http://creativecommons.org/licenses/by/3.0
Page 504
Murcia, F., Murcia, F., Borba, J.
acionário. Os retornos anormais foram calculados utilizando o Market Model e o CAPM para três janelas de evento: três dias (-1, +1), 11 dias (-5, +5) e 21 dias (-10, +10). Foram encontrados retornos anormais estatisticamente significativos nos dias -1 e 0 e para todos os três tipos de anúncios testados: ratings iniciais, downgrades e upgrades. No caso dos downgrades, consistente com os estudos anteriores, os resultados também mostraram retornos anormais negativos para praticamente todas as janelas testadas. De modo geral, os resultados evidenciaram que os anúncios de rating possuem conteúdo informacional, uma vez que impactam nos retornos das açôes, especialmente quando eles trazem ’más notícias’ ao mercado.
Palavras-chave: rating de crédito; mercado brasileiro; estudo de evento.
1. Introduction
One strand of credit rating studies examine whether ratings announcements affect stock and bond prices of public companies (Poon & Chan, 2008). On an efficient market, rating changes will only have an effect if they contain new information. Using regression models and/or event studies, these studies examine the instantaneous effect of rating changes in the prices of securities.
According to Li et al. (2003), studies on the informational content of ratings seek to answer two main questions: Do rating announcements have impact on the stock market? And if so, how does the market reacts to different types of credit rating announcements? For Barron, Clare and Thomas (1997), the motivation of these studies has been to evaluate the relevance of credit ratings to capital market efficiency.
In this line of thinking, our study aims to analyze the impact of credit rating announcements in the stock prices of public companies in the Brazilian market. The motivation for conducting this study is based on inconclusive findings in the existing literature, due to mixed prior results, as well as the uniqueness presented by the Brazilian market, as most studies do not address emerging economies.
Prior literature evidences a lack of consensus regarding the information content of credit rating. For Li et al. (2003) the informational value of rating agencies is a controversial and inconclusive issue. Previous research on the informational content of announcements issued has shown different results (Pinches & Singleton, 1978, Jorion et al., 2005, Creighton et al., 2007).
According to Calderoni et al. (2009) some studies in the 80s and 90s conducted in the United States presented evidence that only downgrades
504 Rev. Bras. Finanças (Online), Rio de Janeiro, Vol. 11, No. 4, December 2013 [notdef]
Page 505
The Informational Content of Credit Ratings in Brazil: An Event Study
have reliable information content. In other words, there seems to be an asymmetric effect, where negative news (downgrades) affect stock prices, but good news (upgrades) don’t.
For Vassalou & Xing (2003), these results are considered confusing, because there would be no reason for stock returns to react to downgrades and upgrades asymmetrically. According to Jorion et al. (2005), it is unclear why only the negative information has value.
On the other hand, the empirical evidence on downgrade is also imprecise. Griffin & Sanvicente (1982), for example, found that poor stock performance after credit rating downgrades emerges constantly in the four weeks after the announcement. Holthausen & Leftwich (1986) also ob-served a decrease in stock price, but entirely focused on the announcement day and the day after the announcement.
For Creighton et al. (2007), despite the fact that studies reveal that security prices react to rating announcements, the magnitude of the response is usually very small, and the vast majority of adjustments in prices around announcements ratings occurs in the weeks or months prior to the announcement. According to Richards & Deddouche (2003), many studies have found statistically significant abnormal returns in the announcement periods, which are often small, especially in comparison to the abnormal returns for periods of pre-announcement. As stated, existing literature on the subject is quite controversial.
The justification for conducting our research also relies on the fact that the great majority of the previous studies have been developed in the U.S., UK and Australia; so the effect on other countries, especially in emerging markets, is still unclear (Li et al., 2003). According to Creighton et al. (2007) most previous studies have used data from the United States, where there is a more significant role for credit ratings.
Furthermore, Han et al. (2009) argue that credit ratings, particularly those issued by Standard & Poor’s and Moody’s, are critical to international investors in corporate debt from emerging markets because:
I. financial information in emerging markets are much less transparent than in developed markets;
II. there are no reliable financial organizations in emerging markets that can certify the eligibility of a debt to international investors;
III. many foreign institutional investors are not allowed to invest in speculative-grade bonds in emerging markets; and
[notdef] Rev. Bras. Finanças (Online), Rio de Janeiro, Vol. 11, No. 4, December 2013 505
Page 506
Murcia, F., Murcia, F., Borba, J.
IV. bank regulators use ratings for financial regulation and supervision as well as capital adequacy rules. We found a single study that used only Brazilian companies in the existing literature: Callado et al. (2008) analyzed the relationship between the stock returns of banks listed on the BM&FBovespa and the public announcement of the first risk assessment issued by credit rating agencies, i.e., the initial rating. They found no significant evidence supporting the hypothesis that the initial credit rating caused abnormal returns in the analyzed sample.
However, the study’s sample was composed of only seven companies. Also, authors only used descriptive statistics, like means of abnormal returns to analyze their results. In this sense, our paper expands Callado et al. (2008) study as we have used a larger sample, abnormal returns calculated by both Market Model and CAPM and more robust statistical methods as we conduct an event study and multivariate analysis. We have also tested all three types of rating announcement (initial rating, upgrades and downgrades) and three windows (-1, +1; -5, +5; -10, +10).
For the reasons outlined, we believe there is a gap in existing literature, especially regarding the Brazilian market, which has specific characteristics that can lead to different results than those founded in other markets. Note, for example, that Brazil adopts a Code Law legal system, mainly derived from Portugal. According to prior research regarding “Law and Finance” (La Porta et al., 1997, 1999, 2000, 2002), code law countries have: (i) less developed equity markets; (ii) firms with more concentrated control; (iii) lower number of publicly traded companies and smaller number of initial public offering each year; (iv) more companies penalized by investors in the valuation process; and (v) companies that pay less dividends.
For Lopes & Walker (2012), the Brazilian market is characterized by low enforcement, incentives for manipulation of financial statements due to tax influence, instable capital market and poor governance standards. In this sense, the role of credit rating is not quite clear.
Moreover, Brazil has...
Para continuar a ler
PEÇA SUA AVALIAÇÃOCOPYRIGHT GALE, Cengage Learning. All rights reserved.
