Fitch Ratings, by Solvency II Wire, under a CC-BY-2.0 licence

In a recently published article, we propose the introduction of a tailored liability regime to induce credit rating agencies (CRAs) to produce ratings as accurate as the available forecasting technology allows.

In a nutshell, credit rating agencies issue judgements on the quality of financial assets reducing asymmetric information on financial markets. In the recent years, their conduct has been closely scrutinised because many believe that their behaviour contributed to sparkling the global financial crisis. In this vein, scholars and policymakers noted that the CRAs business model is plagued by a fundamental conflict of interest, as they are paid by the same issuers that they have to rate (issuer-pays model).

In our article, we argue the problem is more complex. In a well-functioning market, ratings have a value only if investors consider them reliable. Therefore, if CRAs systematically misreported information in order to please their clients they would soon lose credibility and their ratings would become worthless, regardless of the paying scheme adopted. In addition to the conflict of interest, CRAs’ misbehaviour is determined by three concurring factors.

• First, high ratings produce relevant regulatory benefits and hence they have a value irrespective of how many people consider them reliable.
• Second, some investors are naïve and even ex-post they cannot discriminate between accurate and inaccurate ratings.
• Third, forecasting techniques are imperfect because of uncertainty.

Addressing the joint effect of the above market failures, especially when combined with the issuer-pays model, is very complex. On the one hand, regulators do not have enough information to specify how ratings should be calculated in order to correct these market failures. On the other hand, courts can hardly monitor CRAs’ conduct to determine whether it meets a standard of efficient care. Recent empirical evidence confirms that the quality of ratings has decreased after the introduction of more stringent regulations and fault liability in the United States.

Because regulation and negligence rules have proven ineffective, if not counterproductive, we propose a different approach. We argue that the incentives of CRAs could be improved introducing a tailored strict liability rule: CRAs should be liable to pay damages whenever a bond or a company they rate defaults. A rule of this kind poses three problems. In particular: (i) covering all the losses resulting from the default of any rated issuer would almost immediately bankrupt any CRA. (ii) CRAs cannot be expected to make absolutely accurate predictions.

In our article, we propose a solution for each of these problems. To address the first problem, we introduce a liability cap. More precisely, we propose to calculate the liability by multiplying the price paid by the issuer for the inverse of the highest probability of default associated with the cluster in which the issuer is included (higher rating implying lower probability of default, hence, given the issuer fee, higher lability if default occurs).

We show that under this liability rule a CRA that issues accurate ratings will make zero profits, which is the outcome in the presence of perfect competition. However, CRAs models cannot be assumed to be perfect because no forecasting method can be perfectly accurate. For this reason, we introduce a parameter, α, which is a number between 0 and 1 signifying the confidence of CRAs on their own forecasting methods. In our model, α must be disclosed along with the publishing of any rating. If the chosen α is equal to 1, the CRA will pay a sum equal to the liability cap and will avoid losses only if its predictions are perfectly accurate. For α=1/2 the CRA will pay half of the liability cap, and so on. At the other extreme, for α=0 the CRA will be immune to liability, but the market will know that the CRA does not trust its own judgement.

In other words, CRAs will be able to choose their own liability exposure, but in doing they it will also tell the market how trustworthy they believe their ratings are. The key characteristic of α is contractibility. Issuers will want α to be as high as possible, whereas CRAs will prefer lower values of α to minimise lability exposure. The opposing interest of CRAs and issuers will ensure that the both ratings and α will reflect all available information in a competitive market setting.

Finally, CRAs must be protected from systemic risk, which we define simply as correlated defaults. We solve the problem differently because corporate bonds and structured finance products have different exposure to systemic risk. For corporate bonds it is sufficient to set a time limit to liability exposure because corporate defaults are correlated only in the medium to long term. We argue that CRAs should be held liable only if the rated firm goes bankrupt three months (the typical “watchlist timing”) after receiving the rating.

Contrariwise, the defaults of structured finance products are strongly correlated also in the short term. For this reason, we develop a mechanism that ensures that CRAs are liable only if rating inaccuracy is confirmed by the law of large numbers. We propose that a public authority records the ratings issued by the CRAs, the fees they receive, and the frequency of defaults of each structured finance product. Within a given interval (e.g. 5 years) , the public authority will compare the predicted probability of defaults with the observed frequency of defaults. Only if this comparison reveals that the CRA underestimated the probability of default over a long time horizon (importantly including the period preceding the burst of a bubble), liability will be imposed on the CRA. This correction to our strict liability regime is effectively a countercyclical macro-prudential regulation.

♣♣♣

Notes:

• This article is based on the authors’ paper A Strict Liability Regime for Rating Agencies, in American Business Law Journal, Volume 52, Issue 4, Winter 2015 , Pages 673–720.
• The post gives the views of its authors, not the position of LSE Business Review or the London School of Economics.
• Before commenting, please read our Comment Policy

Alessio M. Pacces is Professor of Law and Finance at the Erasmus School of Law, Erasmus University Rotterdam. He is also a Research Associate at the European Corporate Governance Institute (ECGI) and the Director of the European Master in Law and Economics (EMLE). Alessio’s research focuses on the economic analysis of corporate law and financial regulation. On these topics, he published books, chapters and peer-reviewed articles. He has been invited to speak by policy institutions, such as among others the OECD and the European Commission. He has held visiting appointments at Berkeley and Columbia Law school, among others.

Alessandro Romano received a PhD in Law and Economics from Erasmus University. He also holds a Ba (University of Napoli Federico II), a MSc in Business and Administration and an LLM. He was a Visiting Researcher at U.C. Berkeley in 2012, at the Yale Law School in 2015 and at the University of Oxford in 2016. He has published on ISI Thomson journals in the areas of Law, Economics, Business, Planning and Development and Multidisciplinary and on American law reviews.

Print Friendly