Update: An earlier version suggested that every accident year was inadequate. Reader GB points out that the article I cite almost certainly refers to the adequacy of the loss reserve on the balance sheet each year end. My misunderstanding also led me to wonder why the article dwelt on asbestos and environmental reserves, but with GB’s insight, the reasons for that become clearer now.
I’ve edited throughout to correct and clarify. Sorry.
Frustrated at my lack of posting lately, but work has taken one of those turns. I wanted to devote more time to this fine Contingencies article than I will be able to.
Every year the p/c industry estimates its loss reserves on the Annual Statement. And for every year since 1994, that estimate has proved too low, according to Milliman’s Susan Forray
. (She doesn’t have data before 1994.) This is surprising, since in recent years, calendar year development – what we usually read about in the trade press – has been favorable. But CY development is a mix of movements on all previous accident years. Favorable movements on recent accident years have more than offset unfavorable movements in older years.
The chart tells the story. It uses accident year data to re-estimate what the industry should have posted each year. (The color coding distinguishes the movement in the first, second and third development year, etc.) The years for which we have complete development (10 years, cause that’s what you can get from the annual statement) end with 2001.
Only subsequent years show favorable development overall, and, well, the jury is still out on those. Notice that 1995, 1996 and 1997 started out favorably, too, and those didn’t end so well.
This contradicts the conventional wisdom, which tells us that the industry overestimates some years and underestimates others. The CW is true for calendar years.
Forray points to two causes. One is medical inflation and its impact on workers comp. (<Wonky> I’d be curious if the runoff of WC tabular discounts affect this, but I don’t have time to check. </Wonky>) The other is development on asbestos and environmental reserves, which have been the bane of actuaries for more than two decades now.
Harkening back to the headline, though: Reserve estimates are the most public work that actuaries undertake. (Reserves are ultimately an executive management decision, but actuaries play a big role.) I think it’s interesting that actuaries – whom most execs will paint as seeing the glass half-empty – appear to have consistently underestimated losses.