We have consistently argued that the key to ratings reform lies in “the need to incentivize the rating agencies to be accurate, or disincentivize them from being inaccurate.”
John Patrick Hunt recently noted in reference to rating agency reform that “the SEC’s proposal calls for agencies to disclose how frequently credit ratings are reviewed, whether different models or criteria are used for reviews as opposed to initial ratings, whether changes to models and are applied retroactively to existing ratings… Although these disclosures seem unobjectionable, one might ask why the market hasn’t demanded them from the agencies already.”
Indeed, both the subsequent deal monitoring accuracy and the ratings transparency of how rating agencies monitor their deals, leave much to be desired. (For examples, click here.)
Today’s piece focuses on the dual (and sometimes opposing) interests rating agencies may face subsequent to the closing of the deal and the provision of an initial rating. We explain that the rating agencies do not only lack the incentive to be accurate when monitoring deals, but are in fact potentially incentivized to be inaccurate.
There are various explanations for this phenomenon:
(1) Transparency into monitoring operations and procedures creates an unwanted legal liability to the extent it is not followed;
(2) Monitoring provides a comparatively moderate financial incentive for rating agencies (as opposed to the highly lucrative, high-margin business of rating new deals) and they retain only the most limited liability for maintaining outstanding ratings that are inaccurate;
(3) The private sector does not force the ratings to be accurate – the intermediary broker-dealers are largely benefitted by incorrect ratings. The impressive Calomiris and Mason (2010) comment that “ratings reform of all types must make it profitable for rating agencies to issue high-quality, non-inflated ratings, notwithstanding the demand for low-quality or inflated ratings by institutional investors.”
An exploration of Calomiris and Mason (2010) provides additional foods for thought. They make two points on which I wish to expand:
Part 1:
Rating inflation relaxes prudential regulations on buy-side institutional investors in three ways: (1) Inflation allows banks and insurance companies to maintain lower required equity ratios against the purchased debts. (2) Inflation may fool unsophisticated clients of institutional investors by making it appear that their portfolios are earning higher than normal returns relative to risk. (3) Inflation increases flexibility in portfolio management by removing potential constraints that might restrict the purchase or force the sale of lower-rated debts.
Part 2:
The pursuit of private benefits by institutional investors weakens competition among corporate governance rating agencies and leads to entrenchment among established providers of low-quality ratings in at least three ways: (1) Avoiding legal liability for not having pursued an appropriate decision making process when selecting portfolio firms is one private benefit that can drive institutional investors’ demands for governance ratings. That consideration will favor entrenched rating firms with dominant market positions, since age and dominance may be valuable characteristics for institutional investors seeking to show due care in relying on experienced and widely used agencies. (2) Institutional investors that seek to avoid accountability to ultimate investors for poor performance of their investment choices will favor entrenched agencies for the same reason. (3) Potential private benefits that some institutional investors hope to gain at the expense of stockholders will tend to be larger when they form alliances with entrenched rating agencies, since those agencies have more leverage over the firms that they rate.
Several of these latter points can transcend the corporate governance rating agencies restriction and be befitting of the NRSROs, too.
But the realization that faulty ratings monitoring produces larger, more volatile ratings swings (usually downgrades) makes one ponder a different question, too. Does the knowledge of rating agencies’ poor reputation for monitoring deals leave the investor more vulnerable to a large rating shock? (Recall the vicious circle that is triggered by large rating downgrades.) And, if so, does this encourage investors to perform rating-agency-type due diligence, using rating agency models, to best anticipate the next major shock?
At this point, one might conjecture that the fact that the rating agencies sell data and analytics as a separate business to their ratings might discourage them from being transparent as to their methodologies – which increases the need to rely on their models – and enhances investor demand for purchasing their data and analytical products.
As such, John Patrick Hunt’s question may be answerable: the more accurate and transparent the ratings and methodologies, the easier it will be to externally effectively monitor the ratings, and the less interest there will be for the rating agencies’ other products. Compare this to the situation in which external risk managers, for example, might find it both burdensome and frustrating to continue adapting their models to ever-changing ratings methodologies and, finding their capacity to predict or estimate ratings changes jeopardized if they rely on their internal model, ultimately surrender to purchasing the monopoly product.
Moody’s Analytics group now accounts for approximately 30% of revenues generated by Moody’s Corporation (MCO). Irrespective of the level of customer satisfaction with their ratings, the following table – from their public filings – may provide you with an inkling as to the increased dependency on their alternative products, which include non-rating business such as risk management software, credit analysis tools, economic research and data services, training and other professional services.
Quarterly Revenue for Moody’s Corporation (in $mm)
| Three Months Ended March 31: | |||
| 2010 | 2009 | 2008 | |
| Moody’s Investors Service (MIS) | 335.5 | 270.2 | 298.2 |
| Moody’s Analytics (MA) | 141.1 | 138.7 | 132.5 |
| Subscriptions | 117 | 118.2 | |
| Software | 16.1 | 9.5 | |
| Consulting | 5.6 | 4.8 | |
| Research, Data and Analytics | 104.6 | 102 | |
| Risk Management Software | 33.3 | 32.1 | |
| Professional Services | 3.2 | 4.6 | |
| Total for Moody’s (= MA + MIS) | $476.6 | $408.9 | $430.7 |
I conclude with an excerpt from a piece by Cotter and Forstall (2010), which provides an extraordinary basis for requiring increased rating agency disclosure as to the services they provide. Their comparative research into the behavior of accounting firms is strongly suggestive of their conclusion that “people will act very differently if they are required to disclose their actions than if they are allowed to act in secret.”
Their analysis provides the following implication: “once the new disclosures were required, auditors received far less in fees described as ‘other,’ which could influence auditing services. Once they stopped receiving money that could influence their judgment, they might be deemed more objective and therefore more trustworthy in terms of their auditing services.” The outcome of having to disclose potential conflicts of interest – a heightened level of independence brought to the analysis, whether it is perceived or actual – helps restore investor confidence in the integrity of the ratings process, and ultimately helps unlock frozen markets.
June 10th, 2011 → 1:11 pm
[...] earlier piece “Barriers to Adequate Ratings Surveillance” explored the combative interests rating agencies face when ratings and monitoring deals. [...]
July 18th, 2011 → 1:57 pm
[...] for which they aren’t being held accountable, and for which the negative-feedback-looping reputation mechanism is rendered inapplicable (they’re essentially a monopoly). Do we expect them to say [...]
September 23rd, 2011 → 10:12 am
[...] they can be encouraged to compete on accuracy, the reputational model appropriately dismissed by JP Hunt comes back into [...]
January 26th, 2012 → 5:10 pm
[...] Last but not least, the complaint focuses on the effectiveness of ratings surveillance. This is a topic of importance to us, as we feel that proper surveillance, alone, may have substantially diminished the magnitude of the crisis. At the very least, certain securitizations that ultimately failed may not have been executed had underlying ratings been appropriately monitored, and several resecuritizations may have become impossible, limiting the the proliferation of so-called toxic assets. See for example: Barriers to Adequate Ratings Surveillance [...]