A ratio explanation

This report is the first in a regular series of AM Best reports on technical aspects of the financial strength and performance of UK insurers.

Over the coming months various aspects of the balance sheets and income statements of UK insurers will be examined, with notes from AM Best's analysts providing a guide to what the numbers mean and how analysts use them.

We start with the combined ratio. As the universally used `shorthand' for underwriting performance, references to the combined ratio will appear in almost all analyses of insurers, with even the largest companies frequently expressing their own performance targets using the ratio.

This report looks at how different companies compare in practice, and, equally importantly, what to be aware when interpreting the ratio.

The numbers quoted here relate to the December 2000 year-end - 2001 numbers for most companies are still a few months away but this report will be updated once the data is available

What is the combined ratio?
The combined ratio is the classic measure of a non-life insurer's underwriting performance. It adds the loss ratio (paid and incurred claims/net earned premium) to the commission and expense ratio (net commissions paid + underwriting expenses/net written premium). Typically, while described as a `ratio', it is in fact expressed as percentage.

With the numerator of the ratio representing costs to the insurer and the denominator representing income, the lower the ratio the better. Roughly speaking, a ratio of less than one (below 100%) means an underwriting profit, more than one (above 100%) an underwriting loss.

In fact, the calculation is mathematically incorrect. The two ratios have different denominators (net earned premium and net written premium respectively) and so, shouldn't really be simply added together in this way. But, the point of the ratio is not mathematical purity, but to provide an easy shorthand term to capture the underwriting performance of the insurer.

The different components of the calculation reflect the main complexity of understanding a non-life insurer's underwriting performance. Namely, that premiums received, expenses incurred and claims paid during the accounting year do not tell you everything you need to know to decide whether the insurer's underwriting for that year is profitable.

The biggest uncertainty is claims. While claims actually paid, or at least notified, during the year should be a straightforward reporting exercise, this does not represent the total amount of claims that need to be booked. In any line of business there will be some `tail' (claims arriving after the end of the year on policies written during the year). For liability business, in particular, the `tail' can be many years and even decades.

This means the insurer needs to estimate future claims on the current year's policies and add the cost of these to its underwriting result.

Conversely, some of the claims paid will have been on policies written in previous years and for which reserves were set up in the past and charged to prior years' results. This needs to be credited back to the current year's result

So, the amount reported by the insurer for the current year's claims would be claims actually paid plus allocations to the loss reserve for future claims on this years business minus allocations from the loss reserves for claims from old years net of reinsurance. This represents net claims paid and incurred.

A further complexity is that the insurer will have learnt more through the year the likely development of its old year business. So, in addition to adding to reserves for future losses on the current year, it may also decide it needs to add more to reserves for old years.

It is the inherent uncertainty of this process of future reserving which, above all, makes an insurers' reported profits in any given year an opinion not a fact, something that is clearly highlighted by the entry for Independent Insurance in the year 2000 combined ratio rankings. In accounts issued just a couple of months before it ceased trading, the company was still reporting a very healthy underwriting result.

Premium income too has to be adjusted. For policies written on an annual basis on any day other than 1 January, some of the income received will not actually have been earned during the accounting year (that is, not until the policy expires). Consequently, `written' premium is adjusted by allocations to and from the `premium reserve'. Allocations to the reserve are for that part of business written in the accounting year, but not fully earned; allocations from the reserve are for that part of business written in previous years but earned in the current one.

In a growing company the amount being added to the reserve will usually be greater than the amount released and, hence, net earned premium will be smaller than net written. Only, however, if a company is growing, or shrinking very substantially, will the difference be very great since the amount set aside for reserves will be largely balanced by the amount released.

So, when calculating the `loss ratio' analysts look at the best estimate of losses on the premium income actually earned that is, paid and incurred claims to net earned premium.

However, when looking at the commission and expense ratio the analyst is interested in the operational efficiency of the insurer, not its loss experience. So, it is the amount of business actually put on the books that is relevant and, hence, the measure is net commissions and expenses to net written premium.

What drives the level of the combined ratio?
Clearly, insurers would rather make underwriting profits than losses. In practice, insurers that routinely report a combined ratio of below 100% are rare, especially outside of highly specialised lines of business.

So, why do insurers appear to find it so difficult to get their pricing right? There are essentially three reasons for this.

First, investment income. In addition to the opportunity of making money on underwriting, an insurer also holds the premium income, which it is free to invest. In times of high interest rates, and especially for long-tail lines, this can represent a very significant amount of revenue for each pound of premium. That, of course, makes getting hold of the business in the first place that much more attractive to competing insurers, thus driving down the price of insurance and increasing the probability of underwriting losses.

Second, fixed costs and market presence. Insurance pricing is famously cyclical. At times, especially after several poor years or a major catastrophe, premium rates can soar as insurers seek to recoup losses (a process often driven in practice by reinsurers driving up the prices they charge insurers). So, for an insurer to succeed long-term they will often feel a need to have sufficient infrastructure and market presence to attract significant business during the hard markets. But, inevitably, this serves only to make the soft markets longer and deeper since, faced with supporting their fixed cost base and a desire to maintain their market presence, insurers find themselves sucked into writing business during soft markets that they know they should not. Hence, the supply of insurance is greater than it should be given the price and prices, in turn, remain too low for too long.

Third, loss cost inflation. On top of the classical pricing problems of any cyclical industry, insurers face a problem that is pretty much unique. This is that the cost to them of the product they manufacture (insurance cover) can increase significantly years after they sold it. This, in the jargon, is `loss cost inflation'. Not surprisingly it is most common in intangible long-tail lines like liability, where the ultimate size of claims reflects opinions and judgment (for example, of courts) not some predictable and objective economic measure of value. The rapid growth in `no win, no fee' law firms in the UK, and increasing court awards are just such a trend. In addition, the frequency, or even existence, of potential sources of claims can change radically over time. No one in the 1960s thought asbestos was a serious problem for insurers. Today, AM Best estimates the total losses to US insurers alone will be $65 bn, much of this on business written before 1970.

How is the combined ratio used in rating an insurer's financial strength?
A financial strength rating is an opinion of the insurer's ability to meet its obligations to policyholders. That ultimately depends on an insurer's solvency and liquidity. These are aspects of the balance sheet not the income statement. But, a rating has to be a prospective opinion. A balance sheet is snapshot of the past. So, rating analysts seek to evaluate the quality of the insurers business as one guide to how strong the balance sheet will be in the future. In particular they are looking for signs that could indicate a future problem and, more generally, the extent to which an insurer will be likely to attract, retain and grow its capital in the future.

As highlighted above, there is a wide range of variables, which prevents any straight-line relationship being drawn from the company's reported underwriting performance to its financial strength.

That said, subject to a company reporting its results accurately, underwriting performance can be an important indicator of likely future strength.

Four factors though, above all, must be borne in mind when evaluating the quality of underwriting using the underwriting result: