Introduction

To improve credit ratings quality and protect investors, the Credit Rating Agency Reform Act of 2006 was enacted in the United States (US) (US Securities and Exchange Commission [SEC] 2009). In response to the requirements of this act, some new rules relating to the oversight of credit rating agencies registered as Nationally Recognized Statistical Rating Organizations (NRSROs) were recently adopted by the SEC (SEC 2007). One of the proposed rules attempts to prohibit the practice of issuing some unsolicited ratings (SEC 2007). Specifically, SEC states that “it would be unfair, coercive, or abusive to issue an unsolicited credit rating and communicate with the issuer or obligor to induce or attempt to induce them to pay for the credit rating or another product or service of the NRSRO or its affiliates” (SEC 2007: 178). The practice of issuing unsolicited ratings by some NRSROs might have raised sufficiently serious public concerns (from issuers and/or investors) to draw the attention of major regulators like the SEC.

These growing concerns and controversies might have been triggered by the observation that unsolicited credit ratings, on average, tend to be lower than the solicited ratings (see, for example, Poon 2003; Gan 2004; Shimoda and Kawai 2007; Fairchild, Flaherty, and Shin 2009; and Poon, Lee, and Gup 2009). Some researchers asked whether unsolicited ratings were lower than they deserved. Some issuers with unsolicited ratings questioned whether they had been fairly treated by the rating agencies and whether the creditworthiness of these ratings was lower than those of the solicited ratings (Behr and Güttler 2008). Shimoda and Kawai (2007) found that the differences between solicited and unsolicited ratings were smaller than before, but they believed that strong and deep-rooted concerns about the reliability of unsolicited ratings still remained among issuers.

Amid increasing debate on the issue of unsolicited ratings, the SEC has recently made an attempt to seek more information from NRSROs on how they define “unsolicited credit ratings” and their practices, procedures, and methodologies (if any) with respect to these ratings (SEC 2007). The SEC defines an unsolicited rating as “one that is determined without the consent and/or payment of the obligor being rated or issuer, underwriter, or arranger of the securities being rated” (SEC 2008).

Probably as a consequence of the Credit Rating Agency Reform Act and the related subsequent rules, the three major global NRSROs—Standard & Poor’s Ratings Services (S&P), Moody's Investors Service (hereafter Moody's), and Fitch Ratings, Ltd. (hereafter Fitch)—have issued updated policy statements with respect to unsolicited ratings. Specifically, S&P declares that “this rating(s) was initiated by S&P. It may be based solely on publicly available information and may or may not involve the participation of the issuer's management” (S&P 2007: 1). Moody's states that “this (unsolicited) rating was initiated by Moody's and was not requested by the issuer” (Moody's 2006a: 1, 2006b: 1). Under Fitch's policy, Fitch asserts that “the decision to issue these (unsolicited) ratings must meet the same standards for information and analysis as the decision to issue solicited credit ratings. Since 2001, Fitch has publicly disclosed such ratings as having been initiated by Fitch” (Fitch 2007: 1–2)¹. In sum, credit ratings that are initiated and paid by issuers are called “solicited ratings,” and credit ratings that are not paid for by the issuing firm are called “unsolicited ratings.” We use these definitions throughout the paper.

In addition, the dispute over unsolicited ratings might become a discussion topic for bank regulators and supervisors worldwide in relation to the calculation of the minimum capital requirements of Basel II (Fitch 2006; Behr and Güttler 2008). Basel II (Basel Committee on Banking Supervision 2006) provides a framework for the international convergence of capital measurement and capital standards. Specifically, the First Pillar of Basel II suggests that the regulatory capital of banks should be based on risk-based assets and that banks can use eligible external credit assessment institutions, such as some global credit rating agencies, to determine their credit risk. However, Basel II allows each country's national bank supervisory authority to decide whether its country's banks can use unsolicited ratings in addition to solicited ratings in determining their credit risk in relation to regulatory capital (Basel Committee on Banking Supervision 2006). This indirectly draws distinctions between the solicited and unsolicited ratings of eligible external credit assessment institutions and may put pressure on national bank regulators to make a decision on whether their banks can use unsolicited ratings in the same way as solicited ratings for capital adequacy purposes according to Basel II.

This paper examines whether unsolicited credit ratings are lower than solicited ratings using a global sample of nonfinancial firms. Our study is related to Poon (2003), Poon and Firth (2005), and other literature, but is distinct in two aspects. First, we use a global sample of nonfinancial firms. Given the regulatory structure of financial firms, the financial characteristics that affect financial firms' decisions to seek credit ratings and credit rating levels are very different from those affecting nonfinancial firms. A cross-country comprehensive study of nonfinancial firms can clarify the issue of solicited versus unsolicited credit ratings. Our study also encompasses a longer sample period than Poon (2003). Second, given that the decision to seek credit ratings and credit rating determination are endogenous in the rating process, the literature (e.g., Poon and Firth 2005) primarily uses Heckman's two-step method to account for the sample-selection bias. Heckman's procedures rely on a well-specified rating decision (i.e., the decision to seek credit ratings) equation in the first step. The credit rating literature, however, does not have a theoretical foundation in choosing specific financial characteristics to explain a firm's decision to seek credit ratings. The decision on whether to include a particular financial characteristic seems to depend on some conceptual arguments as well as on data availability. That is, specification errors may exist in Heckman's procedures. To resolve this concern, we used Wooldridge's (2002) instrumental-variable approach in our analysis. Wooldridge (2002) shows that his instrumental-variable approach does not require a perfectly specified rating decision equation. Hence, our research method is an improvement over similar studies in the literature.

We offer several interesting findings. First, the firms with unsolicited ratings, on average, are financially weaker than firms with solicited ratings—in some respects. Second, there is indeed a sample-selection bias in credit ratings. That is, the decision to seek a credit rating is not independent of the determinants of credit rating levels. Third, after controlling for some key financial ratios and sample-selection bias, unsolicited ratings seem to be lower. Our results are consistent with the literature and they are robust to the full sample as well as to the Japan subsample.

Section 2 presents the recent literature on unsolicited ratings. The research method used is described in Section 3 and the results are discussed in Section 4. Section 5 includes our conclusions.

Download this Paper [ PDF 293.6KB| 26 pages ].

Comment(s)

There are [0] comment(s) for this entry. Post a comment.