Defending the indefensible

Written by Chris Huhne MP and published in Financial News on Fri 14th Sep 2007

One of my few claims to fame is that I was one of - perhaps even the first - City analyst ever to confess publicly to error. I was running the sovereign risk side of what is now Fitch ratings during the Asian crisis of 1997, and it seemed sensible at the time. For a start, it was such a contrast with the pretended omniscience of most analysts that it was bound to get some media attention (and it did). And I reckoned we might get some credit for an honest post mortem that would help rebuild confidence in our work. We highlighted the key role of short term debt and illiquidity.

The best thing the rating agencies could now do is publish honest post mortems of their work on structured investment vehicles and the sub-prime mortgage market, as Moody's grudgingly did on Wednesday. Honest error is not reprehensible, particularly as a major contributor to the crisis has been market panic and illiquidity. If someone has behaved badly, they should walk the plank. Perhaps Kathleen Corbet of Standard & Poor's falls into that category. Certainly, the press release announcing her departure was so dead pan as to titillate every conspiracy theorist in London and New York.

In general, though, much of the criticism levelled at the rating agencies is just silly and misinformed. I have no interest in defending rating agencies, but it seems to me that most media, political and even academic commentary clearly does not understand the agencies or the capital markets.

The most common charge is that the ratings agency clearly got it wrong because structures based on sub-prime mortgages are now seriously impaired. But ratings are about defining the chance of default. So the agencies are not necessarily wrong if a AAA borrower or issue defaults, but only become wrong once the number of those defaults clearly exceeds the past probabilities set out in ratings default histories.

The second charge is similar to the first: the ratings agencies were too slow on the uptake and should have warned investors that the issues were going to default. But investors and issuers are happy to say precisely the opposite when rating agencies downgrade ratings quickly: I remember being accused of helping to foment the Asian crisis by cutting ratings on a number of countries without enough consideration.

Which do the markets want? Speed or careful consideration, due deliberation and lots of opportunity for the issuer to comment? You certainly cannot have both. In line with the role of defining default probabilities, the agencies generally have substantial internal checks and balances to stop them jumping to conclusions. But those checks and balances also mean that ratings do not bounce around as much as investment bank analysts' opinions.

The rating agencies are also said to be riddled with conflicts of interest. In an ideal world, their work would be paid for solely by investors, but anyone who has tried to get investors to fork out much for research knows that it is easier to sell sand to the Saudis. Since the actual rating is out there on the wires immediately, the agencies can only extract money from investors for the research itself. In reality, getting issuers to pay for ratings is essential to fund what is pretty much a public good.

That does set up a conflict of interest, but it is substantially less acute than many others in the markets. Unlike investment banks, after all, the agencies do not underwrite deals or put up their own capital. Moreover, any given issuer fee is insignificant in the total turnover of any of the agencies. They would be stark, staring, raving mad to compromise their reputation - and their undoubted power to move the market - for a measly rating fee or three.

Of course, individual analysts and groups may have bonus pressures. But the agencies in my time were on the case there too: no rating was issued without a committee decision in which the analysts who visited the corporate or the country were in a minority. It was an old practice, but it increased the cost of bribery to prohibitive levels.

Then there is the charge that, since the regulators increasingly use the agencies to assess risk, they should be more regulated themselves. Watch what you wish for. The trend toward using ratings as a risk benchmark is convenient for central banks and others, but it sets up potential public pressures that are dangerous. Regulation would slow ratings down even more. And it would bureaucratise a process in a way that would ultimately undermine the credibility of ratings.

That is exactly what has happened in Japan. Japanese ratings were widely used as a regulatory requirement, which in turn seemed to set up incentives for the Japanese agencies to provide higher ratings rather than credible ratings. The result is that Japanese rating agencies' judgements are informed more by the milk of human kindness than the mean hearts of Moody's, Standard & Poor's and Fitch. And no-one in the markets takes Japanese ratings seriously.

So if you want to destroy ratings agencies as a useful market tool - but probably provide even higher returns to their shareholders - make it a policy requirement that everyone uses ratings and that the ratings agencies are closely regulated.

The more sophisticated charge is that the agencies are oligopolistic. Well, yes. Having worked for the smallest - and we thought we were trying hardest - there is no doubt that a market dominated by three players has too few competitors. But at least it is better than the market dominated by two players, which was the case ten years ago. And it is not easy building up the credibility to break into the market: you need a niche as IBCA had with non-US banks or as A.M.Best has with insurance.

Still, why spoil the fun? Ratings agencies, like financial regulators, are there to be kicked and take the flak when things go wrong. Politicians blame regulators. Bankers blame ratings agencies. It is a tough job taking the rap for the ups and downs of capitalist asset markets. But someone has to do it. Meanwhile, somewhere out there in la-la-land is the finance minister of an economy where risk has gone, debtors always pay you back, rewards are plentiful and shares only ever rise. Dream on.

Chris Huhne is a Liberal Democrat MP and former head of sovereign ratings at Fitch.