I dunno. What’s the line of business? How much development do you see after 48 months? What’s the statute of limitations on the type of claim and how long did they usually take to be resolved after they are brought to the attention of the courts?
Fwiw, i usually use at least 10 years in my triangles. Sometimes 15. But if you are looking at auto physical damage 4 years is probably plenty. Jurisdiction matters, too.
I would say it’s not better to use a 48 month triangle. It’s better to use a 60 or 72 month triangle (plus any older AY’s you have available) and then use the data to convince yourself that a 1.0 factor is/isn’t appropriate at 60-ultimate.
I frequently assume that development after a certain point isn’t material and set factors to 1.0, but that’s after reviewing historic activity and looking at the reserving impact of that choice.
Are you running out of memory or something?
You still have to reserve for the all the claims that happen even if they are uncommon or rare.
No… just hypothetical question
Are you doing it by hand?
If not, make the triangle as big as you want it, then see how different your IBNR is. Then, show-and-tell time.
There might be losses beyond the agreed amount …
Do you mean covered losses in excess of the policy limit?
Sure. Allocated loss adjustment expense is a common reason, which are often in addition to the policy limit and are unlimited…
Perhaps there is a per person limit of liability, but no aggregate limit.
Perhaps there is a per coverage limit of liability, but no aggregate limit.
Perhaps there is some sort of punitive damage award that is above the policy limit and is deemed to be covered by the insurance.
There is often a good story behind such claims. For pricing, the question is whether you should throw these claims out because they are flukes that will never happen again, or are they symptomatic of something that can very well happen again.
i think executive who has no background in actuarial science might ask why the loss triangle has a long development period and ask whether if the loss triangle with less development triangle will produce a better result
That is what show-and-tell time is for.
You tell people you’ve done the comparison, and you show your work.
I don’t think there is one answer here. Depends on the line of business and the company.
FWIW, if you’re setting reserves, you don’t ignore a claim just because it’s inconvenient (for whatever reason).
Nor is using the assumption that "all claims are ‘fully developed’ at x months when you know you’ll have some claims that develop beyond that time frame going to be very useful no matter what your business philosophy is . . . at least, if you want to remain solvent long-term.
You don’t have to only look at an incurred loss triangle.
You can look at a claim count triangle (or claims outstanding triangle)
You could look at paid. You could look at a paid as a percent of incurred triangle. You could look at case reserves.
I have worked with a casualty triangle with open case reserves at 40 years. 40 freaking years. And that wasn’t A&T. With A&T, all bets are off. That’s much much worse.
If you have open case reserves after N months, you have very likely loss development after N months.
I just noticed that what had been the oldest open claim in my data finally closed – it was over 55 years old.
— Meanwhile at the claims adjuster desk —
Well guys, today’s my last day before retirement… oh shit lemme just close this rq
How does that even happen?
Work Comp is forever (until the injured employee dies). Just a guess.
Can’t speak to Maphisto’s situation, but Michigan has had unlimited Personal Injury Protection coverage.
So, if someone is involved in an auto accident in MI, their policy would generally be on the hook for the medical bills for its occupants. If one occupant is, say, 12 years old and sustains a serious injury that requires medical care for the rest of their life, that claim remains open until either they die or Jesus comes back and heals them completely.
We just stopped paying on our last WWII era claim a few years ago.
This tempts me to make one of those chad vs loser guy memes but it’s the supremacy of Property over other lines of business
Even some property claims can be surprisingly long lived. I know of some WTC and Katrina first party claims that stayed open a really long time.
I would add that those sorts of property claims are “long lived” primarily because the claimant has filed a law suit against the insurer about how the claim was settled. The impact there will be how ALAE will develop more than what the final indemnity payment to the policyholder will be.