But how fungible are the categories applied to a given cost source? Suppose a company finds a way to target customers in a way they predict is likely to outperform the actuarial models using “expensive ad tech”. They can shave off some of the claim cost, but avoid underwriting profitably by inflating the cost of customer acquisition (including big bonuses for their “industry leading” customer acquisition performance ...)
To me some variation on this kind of scheme seems likely to be common - am I wrong headed to think these kinds of tricks occur often?
No I don’t think that wouldn’t work long term at a public insurer. Expense measures are fairly well understood by analysts, investors and regulators. If expected claims got out of wack with pricing it would become a problem for an insurer.
Many insurers have near fixed acquisition costs due to distribution channel realities. Agents take a commission, Google and Ad networks take a referral fee, etc. and those with large direct to consumer portfolios may experience challenges in underwriting profitability as is. Insurers will certainly try to manage those acquisition expenses efficiently and to optimize underwriting performance. Executive comp would be anchored on those things rather than allowed to underperform and compensated for.
If you call beating actuarial models a trick, then yes that certainly can occur but that’s kind of the name of the game. You’d rather companies figure this out because models become more efficient, pricing becomes more segmented and risks are priced accordingly.
To me some variation on this kind of scheme seems likely to be common - am I wrong headed to think these kinds of tricks occur often?