The Micawber Principle
I used to be an Underwriter. Actually I still am – it’s a state of mind rather than a job you retire from.
An Underwriter, I used to explain, while a succession of bosses winced in the background, is "Someone who guesses right, most of the time." We design the rates and rules under which some people get insurance - at various rates – and some do not.
The Micawber Principle is our measure of success. Adjusted slightly for modern times (a certain Mr Charles Dickens originally wrote this nearly 200 years ago) it is: "Income £20, outlay £19.50 – result.. happiness. Income £20, outlay £20.50 – result.. misery."
See: what we charge for insurance depends on how likely someone is to make a claim, and how much we’re going to have to pay out if and when that claim happens.
Genteel traders
Insurance companies are not charitable institutions (surprise!) – they evolved from the gentlemanly tea and coffee shops frequented by shipowners anxiously waiting for the return of their vessels from some far-flung portion of the Empire, laden with exotic goods and spices.
Every so often a violent storm (or even more violent pirates) might delay or destroy a boat, which would sink both the goods and -often- the anxious shipowner. The hundreds of pounds (this was a while ago!) invested to own or charter the ship and fund the cargo were likely the entire fortune of a single family. Success meant fame and fortune; failure meant ruin.
So some bright spark came up with the idea - "Look lads" (he may have phrased it rather differently…) "There are hundreds of us setting out on these trading ventures, and mostly we’re successful. But every so often we lose someone because their boat doesn’t come back. I reckon it’s one out of every 100. There are a hundred of us here. (I like easy sums)
"So suppose we all chip in 1% of our next investment, and the one amongst us who loses out gets to receive the whole amount of the fund in compensation for their loss. If there’s no loss, the whole amount rolls over like a Lottery..." (Like I said he may have used different words…) in case in one year we have a really bad storm and no-one comes home!"
(Pause for polite applause and rattling coffee cups)
You get the point? A loss might happen, and if it does, the small contributions by the many will make up for the losses of the few. It’s very Spock. (Mr, not Dr.)
But this guy had it easy. Loss of a ship was a (relatively) unusual event. It could be costed pretty easily, and the pool of people paying into the fund all had expenses that could be verified. (Side note – that didn’t stop a number of enterprising people from inventing ingenious ways in which a ship could be sunk more than once, or maybe sunk with all hands – and then sold quietly afterwards…)
Balancing the Books
So welcome to the present day. There are (n) million cars on the road. We know mileages, accident rates by age of drivers, costs of repair and have a complete breakdown of the people we insure. And we have a lot of information about how much we’ve been paid, what claims have been reported, and how much the repairs to those vehicles actually cost at today’s prices.
How much do we charge this year?
It’s not a straight application of Micawber. We might insure 20,000 people – but they didn’t all pay their annual premium on the same day. (Yes I know there are monthly pay policies – do you want to explain this, or can I?)
Logically around 1650 new policies will join (or renew) each month over a year. So at the extremes, some people have just paid, but haven’t actually driven anywhere yet – they’re definitely a risk. Others just reached the end of their 12 months insurance and have no accidents – that we know about. Yet.
These are called incurred, but not reported – IBNR claims – and can happen up to 6 months -the usual policy condition- after this policyholder moved elsewhere. We still get to pay...
And we know how much it cost to replace a car bonnet a month ago (‘cause we paid for it) but how much will it cost in a year’s time when the guy, who just started out on his policy year, has his accident and -eventually- reports it in 14 months time…
So working out how much money we’ve actually been paid for a risk that we’ve actually "earned" (that money’s ours – no more unreported accidents or unexpected expenses… apart from the ones that we already know about) – can be quite difficult.
As will be estimating what total costs we’re going to have to pay for the
- accidents we do know about but haven’t inspected or repaired yet
- ones that have already happened but didn’t get reported
- ones that are going to happen to our happy customers going forward...
If the net result at any point is going to be ‘misery’ then look out for higher insurance premiums!
The cherry on the cake
And that (he said, after a dramatic pause,) is only the beginning – because some bright spark then thought: out of the aforementioned (n) million cars, a lot of them are being driven by ‘older’ drivers who are (allegedly) more responsible and (slightly) less boy-racer inclined, and tend not to drive around so much; so we could offer them a cheaper insurance!
Older drivers were definitely happy to pay less – unfortunately with fewer numbers paying into the now separate ‘accidents can happen’ pool which was also paying out for more frequent accidents, the cost per person went up.
Which is what happens when any cherry-picking insurance deal offers a discount. Someone, somewhere is paying for that….
Anyway. Underwriters. We have to untangle all that and decide "if this person has an insurance with us, will they make a claim? If so, how much is it going to cost?"
And then there’s marketing...
The happy chappies (and -esses) of the Marketing Department are often charged by the Powers That Be in the company to come up with some way to make the dull boring world of insurance incredibly enticing to Joe/ Josephine Public out there.
The frequent Marketing go-to is: "give ‘em a discount"..which means, unless we’re prepared for this to be a loss-leader (often indistinguishable from an unmitigated disaster) we need to get more money from somewhere else.
OK - who’s got the calculator? - Time to re-jig those careful calculations. Again.