New York – It was 1991 when I got the offer to join S&P Ratings Group. S&P’s ratings business was still relatively small, around $350 million in revenues, a fraction of the $2 billion business it would become. Moody’s was a subsidiary of Dun & Bradstreet. Fitch had just been bought by an ex-S&P employee and was shaking up the industry, raiding S&P analysts, expanding offerings, promoting themselves to investors. Things were changing fast for the ratings agencies, and I had an opportunity to part of it.
I was working directly for Leo O’Neill, the head of S&P Ratings. My job had two parts: help launch new ratings products, and set up a calling program on institutional fixed income investors. The calling program was a direct response to Fitch, which was promising to make its analysts more accessible than those of S&P and Moody’s. S&P’s response was to get analysts out to meet with investors.
The new product component focused on helping the various ratings groups get funding for new ratings initiatives. Leo’s intent was to try to reinvest some of the profits that were being hoovered by the parent, McGraw-Hill. There were many promising new ratings opportunities-bank loans, private placements, mutual funds, among others-but no mechanism to get the necessary resources to pursue them.
Not that the ratings business wasn’t growing. My predecessor had left NY to open the London office, pursuing one of the biggest opportunities for the ratings business–internationalization. This began a race among ratings agencies to affiliate with local ratings agencies, often with investments or outright acquisitions. This was a successful strategy–nearly 50% of revenues are now from outside the U.S.
The largest future driver of growth–rating securitized transactions–was important but not the behemoth it would become. When I joined S&P, structured finance was still part of the financial institutions group, but that would change quickly. The guy heading structured had just been poached by Fitch (he would eventually head up Fitch after making Fitch a contender in rating securitized transactions.)
Residential mortgage transactions had been around since the early 80’s and credit card receivables were a few years old when I joined. Still ahead were a plethora of different asset-backed securities, commercial mortgage transactions, student loans, CDO’s and, of course, sub-prime.
The challenge with all new ratings opportunities was establishing an effective methodology. This was especially sensitive because of the issuer-paid business model. The more liberal the ratings approach, the more ratings you can perform. But this is counterbalanced by the reputational risk of putting your ratings franchise at risk. If a new ratings approach undermined the credibility of the base business, it wasn’t worth it, no matter how big the opportunity might appear. These days, in the wake of the sub-prime meltdown, the assumption is that greed got the better of the ratings agencies, contributing to the failure of the ratings. When I joined S&P there was a very tangible and wide-spread concern for protecting the credibility of the ratings, which powered the company’s checks & balances on new ratings activities.
One of the most important reputational controls was a formalized process for reviewing and approving new ratings ‘criteria’, or methodologies. Each group within ratings had their own criteria committee which reviewed all major ratings actions such as significant upgrades or downgrades, as well as any changes to ratings policies. Then there was the Ratings Policy Board (RPB), comprised of some of most senior analysts from each ratings group. Members had unforgiving dispositions and long memories. Rank and file analysts dreaded going to the RPB, which could be a withering experience.
The system worked pretty well. New ratings initiatives had to make business sense, and they also had to be reviewed analytically. The analytic reviews were rigorous, and there were plenty of new initiatives that died in the RPB.
Nevertheless, there were failures. The most notable during my time was Orange County which declared bankruptcy in 1994. But for the most part, the default track record for ratings was very good. Enron and Parmalat occurred after I had left the ratings group in 1999, and I left S&P altogether in 2005, missing the collapse of sub-prime.
When I think back on my experience at S&P Ratings, it strikes me how proactive management was in mitigating reputational risk. Leo O’Neill chaired the Ratings Policy Board, and he instilled a culture which required scrutiny of all major ratings activities, including new initiatives. His leadership was reinforced by a generation of managers which had joined S&P when the ratings group was still a small and fragile business, and understood the importance of intangibles like reputation and trust. Most of that generation has retired. It was easy for their successors to take the ratings franchise for granted. Spectacular success makes it hard to envision failure, casting reputational concerns as somewhat old-fashioned and cassandra-like.
There were doubtless other issues, such as the changing dynamics of the duopoly which began with Fitch’s aggressive and successful bid to become the third agency, intensifying competitive pressures on the ingrained checks and balances. These pressures escalated when Moody’s was spun out as a separate entity in 2000, making it easier for S&P’s parent to squeeze the ratings business using Moody’s comparables as leverage.
I also wonder if the scale of the business had an impact. The structured finance group had a lot of hubris when I was around, and I can only imagine how much autonomy they received as their size and profitability grew. I suspect it became a business within a business, isolating itself from the rest of the ratings group.
Nevertheless, if I had to point to one area, it is management. It is inconceivable to me that decisions that S&P apparently made on its sub-prime criteria could have been made during the Leo O’Neill era. Credibility is an intangible, and the requisites for maintaining it are similarly intangible. It is these intangibles of leadership and culture that apparently were lost, and it is not clear that they can be regained.