The idea behind ratings is a simple one: take a complex entity, undertake an established and consistent approach to understanding the credit quality of the company and then publish a very simple indication of its relative financial strength. But the very simplicity of that indicator can lead to misunderstanding and misinterpretation – it’s all too easy for ratings users to make assumptions about what ratings mean which can lead to problems down the line.
The aim of this paper is to explore the suggestion that rating agencies are always too late (or too early), and to help insurance practitioners understand how important it is to read the words published by the agencies rather than just looking at the headline rating. I’m going to do this by using my favourite analogy – Hooke's Law.

Many will remember their Physics teacher talking about the concept of elasticity of a substance, and demonstrating this with a spring. If you extend a spring it will, in most circumstances, spring back to its original shape, but all springs have a point of no return – the point at which the spring stays in its stretched shape rather than springing back. This is an explanation of Hooke’s law.

This concept works really well when thinking about ratings. Firstly it’s important to remember that ratings are described by the agencies as being medium term indicators – terms such as “rating through a normal market cycle” are often used, and the concept of “ratings stability” is very important to them, and to the investors who use ratings.

If ratings are medium term indicators, then they need to display stability. But the rated entities themselves 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. 

The rated insurer’s financial characteristics change all the time, but does that mean that the fundamentals of the organisation have changed?

Can we be sure that this graph is showing a long term trend, or will the insurer ‘bounce back’ before the end of the year?  When is the right time to downgrade?

So how do the agencies create that stability? Simply they look at the underlying fundamentals, the very beating heart, if you like, of the rated entity – its financial position, of course, but also the nature of the market it operates in, its position in that market, the management and sources of capital amongst other qualitative issues. And they ‘stress’ these factors – in other words they consider the impact of normal market stresses with a view to understanding how the entity would cope with these stresses and whether the rating would survive such a situation.

There can be situations where the markets move dramatically away from the fundamentals – when Credit Default Swap ('CDS') spreads (prices), the share price and other short term indicators reflect a rating which is largely different from the level of the actual rating. This has happened very dramatically in some cases, when, before a collapse, the short term indicators suggested a rating level which was much higher than the ratings that the agencies were giving – i.e. the ratings were never very high and the markets thought they should be higher. And it has happened the other way in other cases, where the ratings have remained high although the markets have lost confidence and the spreads reflected much lower ratings.

The key here from an analytical perspective is to understand that in the majority of instances the short-term indicators return to the fundamentals (i.e. the prices begin to reflect the actual rating level), but that sometimes they stay out for so long that they actually shift the fundamentals – the point at which the ‘spring’ cannot return to its original shape and the rating needs to change.

Is there ever a right time to downgrade?
The right time to downgrade is the point at which the spring isn’t going to return to its original shape – the point at which the rated entity’s characteristics have fundamentally changed, when the ‘stresses’ which I mentioned earlier have gone past the level which the rating can withstand. That’s a tough call that the analysts spend a lot of time deliberating. And it’s also a call that, given the chance, is discussed in detail with the management of the organisation, and very clearly signalled by the agencies – if you know where to look…

The early warning signs
If the rating’s like a stressed spring, then how do you tell when it’s reaching breaking point? The agencies do their best to make this clear. Firstly, there’s the use of the “outlook” that attaches to the rating. The outlook refers to the likely direction of the rating – so a negative outlook tells that, all things remaining equal, if the fundamentals continue moving in the direction they are, then the rating’s likely to be downgraded.

But the clearest indicators are in the words that accompany the rating: the rationale.

Your expectation is my expectation…
I used the word expectation above, and it’s a very important word when it comes to understanding how stressed the ratings spring is. Imagine yourself in the shoes of the analyst – spending a lot of time with each of the management teams of a group of peers in the sector, hearing their views, plans, ideas, pressures, and, most importantly, expectations. It should become clear whether these are achievable, how they compare with those of the competitors, and what direction they are likely to take the company. 

If there's a strong enough argument to persuade the rating committee (ratings are determined by committees of analysts) that the expectations of the company are within the realms of achievability, then there’s every chance that the company’s expectations will become those that drive the rating.

So how does this feed into the rationale?

Don’t be irrational – read the rationale
Rating agencies aim to be as transparent as they can about the drivers of a rating. So they publish a rationale clarifying their thoughts, and in the rationale will usually be an identification of their expectations. It gets really interesting when there’s a negative outlook and some clear issues facing the company, as we have seen with a number of insurers in recent years.

The rationale should paint a very clear picture. The various plans, if public, may be listed in the rationale, with a message from the analysts giving their views on the execution risk of delivering what is proposed. And a clear bi-directional statement should be made, along the following lines:
“Should the company manage to execute these plans, then the rating will be affirmed, but with a negative outlook. Should the company fail to achieve any one of their proposals, then the rating will be downgraded by a notch to a rating of ‘A’ with a stable outlook”.

This statement should be explicit, and it should be very clear what will happen to the rating. So here we have a good idea of how close the spring is to breaking point, what needs to be done to stop it from breaking, and what the new spring will look like if the existing one does break.

About the authorPeter Hughes worked for 20 years in the London Reinsurance Market (largely for Sedgwick Group where he was Senior Partner in charge of Europe for Sedgwick Re) before joining S&P in 1998.  Peter held a senior role in the management of S&P.  

Peter established Litmus Analysis last year in order to help the insurance markets understand ratings and use them more effectively.  Litmus helps brokers and buyers in their market security work, monitoring the actions of the rating agencies and assessing exposures to unrated carriers; insurers manage their relationships with the rating agencies, bringing them through to the successful publication of a rating; and training to insurance personnel.