Dear friend,

It’s amazing how uniform the pricing is amongst reinsurers right now. I keep hearing that in today’s new disciplined reinsurance market, that if you go out with a programme that is a little light on pricing, you struggle to get it placed.

But as soon as you nudge the pricing up into market range, everyone jumps all over it and you’re signing everyone down like there’s no tomorrow.

This is worrying — after Katrina the old orthodoxies were thrown away and we returned to a proper underwriters’ market where pricing was subjective and therefore varied wildly from shop to shop. It was a bit like Mad Max!

Now we’re finding that because all the capital models and ERM lines of reporting have been re-tweaked everyone suddenly agrees with everyone else and there’s only a tiny margin between what one player says over another. More General Hospital than Mad Max.

This must be frustrating for brokers — there’s nothing worse that a whole bunch of underwriters ganging up on you and agreeing on everything!

But the one thing I do know is that when everyone agrees on something it is usually wrong!

None of these new capital models or information-gathering systems has been tested in anger yet, so we still don’t know if they work any better than they did in 2005.

Run back through old copies of the magazine in early 2005 and you see what with hindsight was an extraordinarily complacent time. The market was saying “of course, we’re so much more sophisticated these days, the data quality is just chalk and cheese compared to what we used to get in the 1990s… blah, blah, blah.”

(Or should I say baa, baa, baa)

Did the world stop turning after 2005? No, of course not.

What we have now, at absolute best, is modelling that takes the world as it was in 2005 into account.

You can’t beat the market, but if you follow it blindly, you usually regret it. In its most extreme manifestation this reliance on models could metamorphose into a systemic and corrosive form of mass blindness, or even hysteria.

One of the reasons why those enormous computer-traded ‘Quant’ funds all started losing high percentages of their funds under management earlier in the summer was that they all used the same methodology. One day they all said “sell” simultaneously – and low and behold, there was no-one to sell to!

Prices plunged for no good reason and human traders waded in and took the computers to the cleaners.

The fun thing is that since demand is plummeting, capacity is abundant and the Quant jocks at the hedge funds are now offering proper insurance-style coverage with most of the nasty basis risk taken out, competition is intense and prices are going to fall below the point at which the computer says “no” rather emphatically.

I don’t believe that this cheery consensus can last — there is always someone who cracks and breaks ranks.

Back in the 1990s I remember an underwriter who really wanted to write me a line on a programme. He got out his curve and started crunching the numbers – but the pesky little curve just wasn’t co-operating. None of the layers seemed to do the business for him.

Eventually he seemed to give up, and wrote a decent line on the (cheap) top layer on which we had been struggling.

“How come you like the top layer so much?” I asked (I should really have shut up and ran away before he changed his mind).

“To be honest, I don’t really like it. But it’s the only place where I can bend my curve enough to be able to write you a line!”

I suppose curves are bent before you start, so what’s a little tickle here or there?

But it kind of puts ERM into a little perspective.

I sincerely hope we are all right this time, but I seriously doubt it.