Really appreciate your response. You make some great points about there being so many possibilities for variation with GIWL mortality improvement. That’s a good idea about looking at Limited Underwriting tables from different eras and comparing.
One follow-up question regarding your statement here:
“Maybe you could reflect meaningful improvements in the (far) out years, but probably not so much in the early years.”
Are you referring to reflecting meaningful improvements over-and-above the improvement assumption we’d have for fully underwritten business, or no? I don’t think that’s what you’re saying, but just wanted to clarify. When you say “probably not so much in the early years,” it sounds like you’re saying you might expect little-to-no mortality improvement in early years for GIWL at all. Do you perhaps have a hunch or reason to believe the GI population isn’t subject to any of the mortality improvement experienced by the general population and fully underwritten population?
Speaking of which… a thought I had was: for lack of more information, I was trying to think of a way to use the mortality improvement table developed by the SOA based on U.S. population mortality :
https://www.soa.org/resources/experience-studies/2020/mortality-improvement-scale-mp-2020/
This SOA table shows lower mortality improvement for older ages in general than the table developed by my company (the one from my company being based on fully underwritten business). At the very least this does seem to prove that less underwriting means less mortality improvement, though I don’t know how linear that relationship is… but maybe there’s something I can do with these two tables.
The other thing that makes this tricky is mortality improvement tables (both from the SOA and the one developed by my company) are typically calendar-year tables, but of course anti-selection is a phenomenon related to policy-year.