Interesting. I’ve done a lot of work where my customer wanted a credentialed actuary, but I’ve never done any work where it was a legal requirement. Of course, I can count on my hand the number of hours I’ve done work for a state. And hey, maybe the state did legally require me to be credentialed for that handful of hours, but since I was, it never came up.
There are handful of states that require a signed rate application or so for all filings where there is a rating component (whether rate filing or new product).
1 Like
I have always been working on admitted business. It would be nice not to have to work on filings as they can be draining. However, since my first position was in the NY DOI as a reserving actuarial assistant, I do get the DOI perspective.
I’ve mostly worked in reserving and large account pricing, with excursions into reinsurance and analytics. I’ve never actually been responsible for a rate filling.
I’d say lucky you but I’ve been traumatized by reserving before so I’d pick filings work any day over that.
I did reserving for four years. I would definitely take filings over it.
There are a lot of interesting problems to solve in reserving. “What does this mean?” I enjoy it.
Not sure what you are asking. I was talking about my personal experience. Allocating IBNR to the contract level at midnight when it’s due the next morning was quite unpleasant.
I wasn’t asking anything. I was describing the kind of puzzle that routinely shows up in reserving. “There’s a weird pattern over here. What does it mean? What can it tell us about what’s driving our costs, and what we will pay next year?”
Allocating ibnr to the contact level isn’t any fun. Luckily for me, that’s never been a significant fraction of my job.
Allocating IBNR to contracts is an interesting concept. When I was doing IBNR work it was calculated at the contract level. There was no “allocation” of a calculated aggregate number… you calculated it for all of the contracts and added up the results to get the aggregate.
I do recall the NY DOI getting salty when we failed to include sex and DOB (as explicitly required) on our contract-level IBNR reserves… on clearly identified group contracts.
1 Like
As far as calculating a dropout rate… I assume that for SOA they have all the data as we all have unique ID numbers that don’t change when our name / address / employer changes.
You certainly have to define what it means to drop out… maybe length of time since an attempted exam?
What about people who attain Associateship, try for Fellowship, quit Fellowship exams, and stay Associates? Are they dropouts? What if they never attempt any Fellowship exams / modules? Pay the fee for a module and never complete it?
Agree that it will be harder for CAS because people will start SOA exams intending for a CAS designation and never getting to a CAS exam.
That makes it muddy for the SOA side too, actually, for the successful CAS folks who quit taking SOA exams because they are on the casualty side.
You’d have to aggregate the SOA and CAS data. Do CAS students retain their SOA-issued student numbers for registering for CAS exams?
I don’t think so. I don’t think the SOA shares that, and i don’t think the CAS id numbers are large enough to include all the people who’ve ever signed up for SOA exams.
(The SOA used to share data from the then-joint exams, and stopped when they stole the exams.)
1 Like
It’s a headache. Calculating IBNR in aggregate is intuitive, once you start allocating, you reach nonsensical results because some IBNR could breach policy limits, and if you account for that, you need to figure out where to put the excess reserves. Then rinse repeat.
Ah. Mine is only used internally, and while it helps with monitoring the book to make a reasonable allocation, especially as to how much will emerge in the next year or two, it’s not a huge deal. Also, doing the allocation is literally less than half a day of my year. So it’s not what I think of when I think “what do I do for this paycheck?”
1 Like
So then it would be difficult to differentiate people who randomly happened to drop out at the end of joint exams vs those who went on to take CAS exams and achieve ACAS and/or FCAS.
Back in the era when 1-4 were joint and then 5+ were separate, I actually know a lot of folks who dropped out with 4 exams.
I guess they would mostly have at least one failed attempt at the SoAs version of 5 and/or 6 (often both since you had to alternate sittings). Probably not too many pass 1-4 and then make no further attempts at exams on either side of the aisle. If you assumed that 100% of the folks who passed 1-4 then attempted nothing further on the SOA side were sitting for CAS exams that wouldn’t be too far off.
I think it would be more complicated now. Especially during the era when the CAS was accepting either the SOA exam 3 OR their own exam 3.
Seems like a shitty idea, so actuaries should rise up and a put a stop to it.
If I recall we had 400k Excel rows with Sumifs and we had to make sure it mapped to contracts and you would get errors where IBNR landed to policies when there was no treaty year that year since the selections were done by accident year. I recall pressing F5 and waiting to 12:30 am before I could leave the office. Literally sitting around for 2.5 hours hoping that everything allocated with no errors. Brutal