Accounting for the cost

Getting the accounting figures right in insurance is of vital importance, as ALBERT MUSHAI and HUGH-DAVID HUTCHESON explain

The September/October edition of SHEQ MANAGEMENT dealt with the California Springfellow case in which the Californian Supreme Court ruled that insurers were liable for policies issued decades ago. This article addresses the question of how insurers should account for the cost of these long-tail liabilities. To put it simply: where does the money used to settle claims on policies long since closed and charged to insurance years long since forgotten come from?

This is not an abstract question. A similar problem now arises in South Africa as a consequence of the Mankayi decision against AngloGold Ashanti (the judgment described as huge victory for former mineworkers with lung disease): the Constitutional Court ruled that s35 of the Compensation for Occupational Injuries and Diseases Act (Act 130 of 1993) does not preclude employees in the mining industry from suing employers. One can accept that employees will start suing former and current employees. In many instances, these liabilities are excluded from insurance policies, but in some cases these claims may be passed onto insurers.

No profits and no money sitting around           

This decision was unprecedented in South African legal history so it can be accepted that mining houses and insurers do not have money sitting in accounts waiting for the day that a court introduces new forms of liabilities. Contrary to popular belief, short-term insurers, worldwide, do not have much money. What insurers do is collect premiums for claims that arise during the year and pay claims from the collected premiums. Usually there is very little money, if any, left over at the end of the year. The accompanying graph indicates the total underwriting (the profits from insurance business) for the United States and South Africa for the period 1975-2005. Unfortunately we don’t have more recent figures for the US.

From the graph, it can be seen that the US market ran at a loss for the entire period with the exception of two years. The South African market fared a bit better, but if averaged out over the full period, it can be seen that the profit is virtually zero. The insurance markets survive because of investment profits. As a rule, the investment profits subsidise the underwriting profits.

It is correct that insurers hold a certain amount in reserve, but that is held to satisfy regulatory requirements – not future unprecedented claims. In South Africa, insurers would be unable to pay billions of rands and meet the regulatory requirements for the holding of reserves.

The real question is not where insurers get money from to pay these claims, but how an insurer is supposed to account for these unexpected claims. The answer to this question is interesting – insurers don’t know how to account for these claims and will no doubt incorrectly account for them with possibly disastrous consequences, as happened to Lloyd’s in the late 1980s continuing into the 1990s. Lloyd’s faced the greatest financial crisis in its history and was saved from liquidation by the concerted efforts of many parties. The cause of the crisis was in part attempting to fund asbestosis risks (long tail risk of the Springfellow nature) and was in the end averted by the establishment of a separate company, Equitas, which took over all Lloyd’s long-tail liabilities, enabling Lloyd’s to survive.

International accounting standards bodies are currently revising the accounting standards for insurance companies – the so-called IFRS 4 Part 2 project. The outcome is not expected for some years, but it is unlikely these standards will get it right because the draft issued for public comment covers both life and short-term companies, and these are two completely different operations.

What is the extent of the liabilities?

The first question that needs to be answered when an insurer faces a new liability that must be accounted for is this: what is the extent of the liability? Now, it should be very clear that the answer to that question cannot be determined. For example, when Lloyd’s became aware that the American courts were reinterpreting insurance contracts to hold insurers liable for unprecedented liabilities – what was the extent of that liability? Insurers work out their liabilities based on previous claims, and since there were no previous claims, the liability could not be determined. What happened is that actuaries were asked to provide the figure. This does not help because actuaries also need data, and there was no data. So all the actuaries could do was provide an estimate. Actuaries, like everyone else, are conservative, so they would provide a conservative estimate; so an accurate estimate could not be determined and a conservative figure was used. Taking the Mankayi decision into consideration, the question is how much money should be set aside for possible liabilities? Since no-one knows, one will have to pick a figure – say R50 billion. So that’s the first problem: we need to pick a figure, and no-one can say with certainty what that figure in fact is.

When will the money be required?

When will the money be required is a question no-one asks. Will the full amount be required in 2013 or over the next five or 10 or even 20 years? It is an important question. The answer makes a world of difference in practice, yet no-one answers this question.

How is it accounted for?

The most important question is how this amount is to be accounted for. Now most people, even accountants, believe that the most important item of expenditure of an insurance company is the amount spent on claims paid; that is, claims actually paid. This is incorrect, and has been since sometime in the 1300s when the accrual system of accounting was introduced. Expenditure is accounted for when notification of a claim is received, not when the claim is paid. The time of payment is largely irrelevant. So the most important item on the expenditure part of an insurance company’s financial statement is not the value of claims paid, but the estimate of claims reported. Once reported, the corresponding figure appears on the balance sheet as a liability. Now, the question becomes, should the
R50 billion be treated as a claim?

And what happens to the R50 billion figure? Those who don’t know better will treat it as “claims reported” and then include it in the income statement, resulting in the insurer showing massive losses as happened with Lloyd’s, then showing a corresponding massive increase in liabilities without sufficient assets to back these liabilities. This too is what happened at Lloyd’s. In short, the insurance company appears to be insolvent – which it may well be in years to come.

These massive new liabilities cannot be brought through the income statement without the insurer appearing to be insolvent. Unfortunately, modern accounting practices are heading in the direction of taking all movements on balance sheet items through the income statement. It should be clear that if this approach is used to account for long-term liabilities such as possible future claims, solvent insurers will be shown to have incurred losses they haven’t, and to be insolent when they are solvent. The
R50 billion should be seen as the amount needed to be held in reserve. Companies will not have that amount but can over years build up the reserve. It is for this reason that the question of timing is important. Getting the accounting entries right is of vital importance.

Legally Speaking is a regular column by Albert Mushai and Hugh-David Hutcheson, both in the school of Economics and Business Sciences, University of the Witwatersrand. Albert holds a master’s degree from the City University, London, and was the head of the insurance department at the National University of Science and Technology in Zimbabwe before joining Wits University as a lecturer in insurance. Hugh-David holds a PhD from Wits and is a lecturer in insurance.