Essays on credit ratings

Abstract

This PhD thesis consists of three independent empirical research papers about credit ratings and the connection between credit ratings and the financial market. A credit rating agency (CRA) is a third-party financial institution providing assessments and recommendations for other market participants about the performances of firms, the default risk of financial instruments issued by those firms and the credit quality of sovereign countries who issue government debts. On account of the essentiality of information transparency in the modern financial system, the credit rating industry is gaining importance in terms of its role as an information-provider for market participants and of its function to reduce the information asymmetry in the financial market. Due to this significant importance, CRAs may impact the behaviors of market participants by offering signals about the firms or financial instruments and hence have a strong influence on the stability of the financial market. The aim of this PhD thesis is to discuss the relationship between credit ratings and the financial market from three perspectives: the link between sovereign ratings and firm’s performances in the context of the European debt crisis, the role of credit ratings in the pricing of asset-backed securities (ABS) before and after the global financial crisis, and the motivation of self-interests of the CRA industry. In the first empirical paper, I investigate the shock of sovereign downgrades and the association between them with the performances of listed banks which are registered in the downgraded countries. The previous literature shows the connection between sovereign ratings and bank performances and hypothesizes some potential channels (such as the government debt and government guarantee) by which sovereign ratings impact bank performances. Nonetheless, individual bank ratings are neglected in these analyses. I fill in this gap by studying the role of bank ratings in the transmission of effects of sovereign downgrades on bank performances. I find that sovereign rating downgrades followed by bank rating changes have a stronger impact on the bank stock returns and Z scores (a proxy of the bank insolvency risk) than those which are not followed by bank rating changes. To further tease out the independent effect of sovereign and bank rating downgrades, I take advantage of ‘sovereign-ceiling policy’ which creates the semi-passive bank rating downgrades when the sovereign rating ceiling has been downgraded. The empirical tests of the banks downgrades triggered by the sovereign ceiling policy help us to conclude that bank rating downgrades provide extra information to the market even if they occur no more than two days after the sovereign ratings are downgraded. In the second empirical paper, I focus on the ABS market and the role of CRAs in the determination of ABS prices. I investigate whether the reactions of ABS investors to credit ratings have significantly changed since the shock of the global financial crisis. To empirically test the ABS investors’ behavior, I run a series of regressions to study the correlation between ABS issuance spread and the issuance credit ratings in pre-and post-crisis periods. I find evidence of a weaker reaction of ABS investors to the ratings offered by credit rating agencies after the financial crisis. To supplement the static-regression analysis, I apply event-analysis methods to identify the ABS price reactions to the rating events in the two periods and identify weaker price reaction degrees after the crisis. The conclusion is that before the 2008 crisis, ABS investors’ decisions, reflected by both the issuance spread and transaction prices, were significantly associated with the ratings offered by CRAs while the post-crisis period has seen a weaker link between spread/prices and CRAs’ announcements, indicating a smaller influence of CRAs on the ABS market. The third empirical paper discusses CRA self-interests from the perspective of the association between rating solicitation status and the conservatism as well as the quality of rating services. It contributes the literature by applying the gap between unsolicited and solicited ratings as a measurement to investigate the motivation of CRAs. Based on a simplified theoretical model, I raise the hypothesis of self-selection to demonstrate the motivation of firms who do not solicit rating services and the rationality of rating agencies’ decision to offer ratings with more conservative levels for unsolicited rating recipients. Firms who realize that their future performances will deteriorate are less likely to solicit ratings from rating agencies. Rating agencies capture this signal and offer unsolicited ratings to those firms with a more conservative rating reflecting the self-selection effect. I empirically test this hypothesis using Moody’s unsolicited rating data and I obtain two findings. The first finding is that controlling for fundamental factors, Moody’s unsolicited ratings are lower than solicited ones. The second finding is that the rating qualities of both types of ratings are not significantly different from each other. My research contributes to both the academic literature and the practical field of the credit rating industry by providing a comprehensive discussion on the connection between a variety of credit rating services (sovereign ratings, firm ratings and bank ratings) and the market reactions (issuance and transaction prices, insolvency risk and default risk) in different cases (the structured finance products, the sovereign ceiling policy and the issuance of unsolicited ratings). The study also investigates the shock of two recent financial crises (global financial crisis and European debt crisis) on the credit rating industry, which provides suggestive findings for the regulators who are willing to take some actions to intervene in the CRA industry in order to enhance the stability of financial markets in the post-crisis era.