To consider the impact that these cycles may have on the financial statements and solvency positions of insurers there has to be an understanding of the magnitude of any change in ultimate loss and the likely timing of the recognition of that change. The profit or loss in any financial year is a combination of the profit and loss from that accident year and also any recognized changes in the reserves from prior years.
A simplified example of a cycle shows how the emergence of information flows through into the financial results. In the figure below, if there is an assumption that financial results are about to be published in Year 10, which is indicated by the black line, what information exists about the likely ultimate position of prior year reserves at that time? This information is shown by the red bars. There are reserve releases from prior years 6 to 9 and recognized increases in reserves from Accident Years 1-5.
The balance sheet will reflect this level of anticipated liability by a recording of this as the best estimate of liabilities for all years. From an earnings perspective in Financial Year 10 it is only the change in prior years, over the course of the last year, that will have an impact. The change information that comes into Financial Year 10 is shown by the blue bars in the chart below.
In this simplified example of a perfectly symmetrical cycle and a constant volume of business, the reader would be forgiven for not being overly concerned about such a cycle scenario. The increases in the more distant back years have been mainly cancelled out by the releases in the more recent history. The real life situation, unfortunately, is rarely so simple. The choices made at each point in time in terms of business volumes and reinsurance buying strategy can be hugely influential on the outcome.
Following a period of sustained reserve releases, what would result if in Accident Years 0 to 5 volumes were dramatically increased across all lines? At Years 6 to 7, there is then some recognition of likely deterioration in Years 0 to 5, so the company contracts sharply. An assumption can be made that the increase in the reserves in Years 0 to 5 in this case is driven by large claims. At that time, however, it was thought that pricing in the excess of loss market for long-tail claims was too expensive. This leads to a decision to dramatically increase retentions. The result of these decisions would be a perfect storm. The company is heavily weighed down by reserve increases on years of expansion without having the full offsetting benefit of releases in more recent years. Reinsurance is set at a level that means it is not sufficient to dampen the large claim volatility.
Is this scenario far-fetched? It appears not. A cursory search reveals examples of companies who have fallen foul of misreading the cycle to some extent or another.