Rating agencies have been battered by the media, but this is because people don’t understand how they work. Instead, they need to look more closely at the agencies’ rating methodology and why they are trusted by banks, brokers and investment houses

It concerns me that the media continues to perpetuate market myths and misconceptions about ratings of financial strength and the rating agencies. Perhaps it is time to redress the balance.

One myth touched upon in September’s Global Reinsurance is the lack of oversight of such ratings and the companies that provide them. However, rating agencies have been regulated for many years by the US Securities and Exchange Commission, and this is being taken to a higher level by the tightening of that oversight and the introduction of regulation from the EU.

In addition, many major banks and investment houses, as well as the larger insurance brokers, employ analysts who use ratings day in, day out. In doing so, they provide their own check on the output of the agencies: if they didn’t think the agencies were doing something right, they wouldn’t use them.

While rating agencies have come in for heavy criticism from various quarters, I would suggest that much of this talk is from people who don’t understand what a rating is. Indeed, in a way, a key role of ratings is taking the blame when things go wrong.

Starting with the rating of banks, there are two important factors for people to bear in mind.

The first is that a bank’s rating starts with a standalone credit profile, indicating the core strength of the bank. To this are added implicit ratings for systemic support; this is the belief that banks are essential to the economic system and that there will come a time when a government has to step in to prop up a bank. This is what we saw with so many institutions in the UK.

The second factor is that agencies talk about the transition profile of ratings, and make every effort to highlight the likelihood of rapid transition of a rating for institutions where this might be an issue. The transition profile – the speed with which the rating may change in a stressed scenario – has been gentle for most major motor manufacturers but steeper for companies in confidence-sensitive industries. This is something those of us from the reinsurance industry understand well.

Some critics complain about the accuracy of ratings, but there is an equal number who grumble about their stability.

As medium-term indicators, they are aiming for stability. My favourite analogy for describing how ratings tackle this issue is Hooke’s Law, which describes the elasticity of a substance.

Put simply, if you extend a spring it will return to its original shape, except when extended beyond a point of no return. Similarly, ratings need to be stable over the medium term, but rated entities change their financial characteristics by the day, or even the minute, so they are always “pulling on the spring” – challenging the elasticity of the rating.

Rating agencies start by looking at the underlying fundamentals of an entity, such as its financial position and the nature of its market, and then consider how the entity will cope with normal market stresses and whether the rating will survive such a situation.

Sometimes the markets move dramatically away from the fundamentals, for instance when there are big changes to share prices, credit default swap spreads or other indicators, which appear to suggest that an entity needs a different rating.

This has happened dramatically in some cases when, before a collapse, the short-term indicators suggest a rating much higher than the agencies’ current ratings. In other cases, the ratings remain high even though the markets have lost confidence and the spreads suggest much lower ratings.

The key is that in most instances the short-term indicators return to the fundamentals, but sometimes they stay out for so long that they shift the fundamentals: this is the point at which the spring of Hooke’s Law cannot return to its original shape and the rating needs to change.

It is all an issue of timing, and determining the point when a downgrade is necessary is taken seriously by the agencies.

Ultimately, understanding the approach that rating agencies take starts with the appreciation that these organisations are trying to predict the future. This can only involve degrees of probability – not the absolute certainty that some market-watchers seem to think they’re indicating.

Finally, it is easy – and important – to view the performance of the rating agencies by looking at their default and impairment statistics. These show that there is a high correlation between ratings and default, and that this applies even through the recent financial crisis. GR

Peter Hughes is founder of ratings consultancy Litmus Analysis