Do Credit Rating Agencies Detect Accounting Frauds?

Accounting fraud imposes severe costs on firms and their stakeholders. Firms at which fraud occurs often have inefficient resource allocation and face higher cost of capital and regulatory penalties. While shareholders suffer the brunt of these damages, frauds can also affect debtholders when firms miss contractual payments or declare bankruptcy. Credit rating agencies traditionally act as gatekeepers for debtholders, but their role and responsibility in detecting accounting fraud remain an open question.

On the one hand, credit rating agencies are in a unique position to detect accounting fraud. First, they have access to non-public information. While the SEC’s Regulation Fair Disclosure prevents managers from selectively disclosing information to other information intermediaries such as financial analysts, credit rating agencies are specifically exempt from this restriction. Second, credit rating agencies state clearly in their rating framework that a firm’s financial transparency affects their rating decisions. Last, credit rating agencies have reputational incentives to ensure that their ratings are accurate and thus should scrutinize firm financial reports.

To gain greater insight into the role of credit rating agencies in corporate governance, we interviewed the head of one of the top three rating agencies in the Asia-Pacific region, who confirmed that rating agencies have better information access compared with other information intermediaries and that they have incentives to incorporate fraud information in their ratings. Concerning information access, rating agencies’ privileged access to private information, as well as the use of non-disclosure agreements, allows them to directly observe proprietary information not available to other information intermediaries. For example, while equity analysts who follow issuers generate their own estimates for earnings forecasting models, issuers usually provide rating analysts with three to five years of earnings and cash flow forecasts. Moreover, rating agencies conduct site visits during the initial rating and again during each annual review. During these visits, rating analysts tour the firm’s operations and often have face-to-face talks with top executives, including CEOs and CFOs. Such private meetings allow rating analysts to observe executives’ vocal cues and facial expressions, all of which can help rating analysts better gauge executives’ management philosophy, personality, and integrity.

However, both investors and regulators have criticized credit rating agencies for failing to sufficiently assess issuers’ financial reporting quality, especially in the wake of the high-profile fraud cases of the early 2000s. The more recent financial crisis has cast further doubt on whether credit rating agencies have the expertise to analyze complex transactions and products and the incentive to expose frauds. Indeed, credit rating agencies specifically stated that they “heavily relied on the quality, completeness and veracity of information” disclosed in firms’ published financial statements, and that they do not focus on “searching for and exposing frauds.” Finally, even if they detect accounting fraud, they may withhold a negative rating action in order to stabilize ratings for contracting purposes or regulatory concerns.

To conduct our analysis, we use securities class-action lawsuits from Stanford’s Securities Class Action Clearinghouse database where the defendant firms have received credit ratings from Standard & Poor’s. We restrict our sample to 259 lawsuits brought under Section 10(b) of the 1934 Securities Exchange Act (manipulative and deceptive devices) to focus on cases most likely related to accounting frauds.

To answer whether credit rating agencies detect accounting frauds, we examine their rating actions against firms where fraud has occurred and before the fraud has been publicly revealed. Our findings show that these firms experience a greater number of negative rating actions during the four quarters prior to the public fraud revelation, including lower ratings, more rating downgrades, and more negative credit watch additions. This result holds regardless of whether we compare fraud firms with non-fraud firms that have similar economic fundamentals and stock performance, or with themselves but during non-fraud periods. In particular, we find that 38 percent of firms committing accounting fraud are downgraded and 30 percent of firms committing accounting fraud are put on negative credit watch during the two years before fraud revelation. On average, they experience a cumulative downgrade of a 0.63 notch during the four quarters before the fraud is publicly revealed, which equals 68 percent of the average downgrade magnitude during the quarter in which the fraud becomes public (a 0.92 notch). Our findings also show that such negative rating actions are not limited to fraud firms that are in financial distress, but also those that appear financially healthy. This suggests that credit rating agencies respond not merely to deteriorating performance of fraud firms, but also to accounting fraud per se.

In addition, our results show that credit rating agencies take more timely actions against more severe frauds, when other market participants such as short sellers also target fraud firms, and when frauds involve misstatements in accounts routinely scrutinized by rating agencies such as long-term tangible assets, intangible assets, and liabilities. This is consistent with credit rating agencies leveraging their expertise in analyzing long-term default risk and being more capable of detecting frauds involving accounting items of greater relevance to credit risk.

Last but not least, credit rating agencies’ warnings about fraud firms have real consequences. These actions not only provide relevant information to the equity markets and inform investors, but also contribute to the public identification of frauds and accelerate fraud discovery. Specifically, rating downgrades of one notch against fraud firms shorten fraud duration by 1.1 quarters, or 13 percent of the average fraud duration. Interestingly, this effect does not diminish when we control for short sellers, suggesting that short sellers and rating agencies play independent roles in fraud revelation, likely due to their different expertise.

Overall, we conclude that credit rating agencies possess private information about accounting fraud prior to its public revelation and that they incorporate this information into negative ratings actions, accelerating fraud discovery. This highlights the essential role they play in the capital markets.

This post comes to us from professors Allen Huang at Hong Kong University of Science and Technology, Pepa Kraft at HEC Paris, and Shiheng Wang at Hong Kong University of Science and Technology. It is based on their recent paper, “Credit Rating Agencies and Accounting Fraud Detection,” available here.  

1 Comment

Comments are closed.