There are few things that investors dislike more than surprises, especially bad ones. That is one reason why the stock market in general prefers life insurers to non-life, which have inherent volatility in earnings because of the nature of their business. It is not convinced by (or interested in) the proposition that, over the cycle, non-life can produce better returns than life, not least because experience indicates that this out-performance is seldom achieved in fact.
The half year results of CGNU at the beginning of August were therefore profoundly disappointing. The market registered its displeasure by knocking up to 10% off its share price during the day. The implications - for CGNU and for other listed insurers - go further than that.
The company reported interim profits of £800 million, down from £853 million in mid-1999, compared with analysts' forecasts typically of between £900 million and £1 billion. The difference was entirely due to two unexpected items. One was a £41 million charge for investment in e-commerce development, which at other times the market would no doubt have shrugged off, or even welcomed. (It would have reacted just as badly to a formal announcement that no investment in e-commerce was going to be made at all.)
Far more significant was £90 million in additional claims from winter storms Lothar and Martin. This obscured to the point of invisibility substantial good news, such as the fact that UK life sales were far more robust than those of CGNU's competitors, that UK personal lines had a combined ratio of under 100%, and some important new distribution agreements. By contrast, Royal & SunAlliance later produced operating results entirely in line with expectations, and rose about 5%. Few analysts thought the fact that it reported a loss for the half year of £50 million (after a profit of £209 million in mid-1999) was even worth mentioning.
The French storms came late in the year, the government stimulated additional claims, and builders profited, so reinsurance protection proved inadequate. Unfortunately, reminding the market at the outset that the new group is still sensitive to earnings from property/casualty business invited application in the long-term of the composite discount with which it traditionally burdened CGU.
CGNU was, of course, aware that this maiden set of results was important. It organised a special session for stock market analysts to explain how the new consolidated accounts would be structured. During the period before the announcement, it indicated to them that extravagant earnings forecasts were likely to be disappointed. But at no time did it indicate just how disappointing the results would be.
The company later explained how the increase in French storm notifications arose, but this did not feature in the announcement made to the Stock Exchange at 7 am. If it had, some of the headlong fall in the share price might possibly have been mitigated (but probably not). If some arcane Stock Exchange regulation forbids such explanation, it should be changed.
Some prior indication that the French storms claims situation was deteriorating would have made a difference. Escalation in the number of individual claims, claims creep, and late notifications on broked commercial business for which Group Victoire was not the lead did not, after all, happen overnight, and certainly not on the night before the announcement. Quite why the company did not choose to make a pre-emptive announcement is hard to fathom. Perhaps it had an entrenched fear of “profits warnings”?
Maybe the result would have been no different, although it is hard to believe that £90 million in extra non-life claims in France justifies a fall of £1.5 billion in the group's market capitalisation. But without doubt the fall would have been less sudden, and probably on announcement of the results the market would have focused on the good news instead of the bad.
Does it matter? It does, to at least two sets of people. CGU may have had a relatively small number of individual shareholders; CGNU has rather more - the former Norwich Union mutual members who have not yet sold their Norwich shares for cash. They will no doubt wish they had. And, although it matters, too, to the individuals whose savings are invested with institutional investors, it matters more immediately to the fund managers who are employed to invest their money, and are judged on short-term performance. Those most affected are those who have committed their clients' savings to their conviction that CGNU is a good business. It may well be a long time before they are so ready to back that conviction again - and that is why the stock market dislikes surprises.