Actuarial Outpost
 
Go Back   Actuarial Outpost > Actuarial Discussion Forum > Health - Disability
FlashChat Actuarial Discussion Preliminary Exams CAS/SOA Exams Cyberchat Around the World Suggestions



Reply
 
Thread Tools Search this Thread Display Modes
  #1  
Old 08-16-2018, 02:41 AM
acemanhattan acemanhattan is offline
Member
SOA
 
Join Date: Sep 2014
Posts: 379
Blog Entries: 1
Default Need To Justify Company's Provision for Adverse Deviation

Hi all.

Every 3 years our company reviews the PAD we add to our IBNR estimates, and this year it falls on me to write the memo.

Historically we've added a PAD that, in our opinion, ensures reserves are sufficient 90% of the time. Previous memos state that this is our method, but give no justification other than to say that "it's reasonable by industry standards."

This time around I'm being asked to support that claim and I'm looking for help finding something like any of (1) an industry survey of PAD showing that aiming for 90% is reasonable, (2) some sort of text that says aiming for 90% is reasonable, (3) some sort of GAAP rule that would justify our method, or (4) anything else you can think of to help me make the case.

Thanks in advance.
Reply With Quote
  #2  
Old 08-16-2018, 07:46 AM
Gabriel's Avatar
Gabriel Gabriel is offline
Member
CAS AAA
 
Join Date: Mar 2007
Location: New York, NY
College: New York University
Posts: 219
Default

What sort of variability (CV) underlying your reserves does your PAD = 90 percentile imply? Is this CV reasonable?
Reply With Quote
  #3  
Old 08-16-2018, 01:24 PM
Dr T Non-Fan Dr T Non-Fan is offline
Member
SOA AAA
 
Join Date: Sep 2001
Location: Just outside of Nowhere
Posts: 93,112
Default

Any idea how wrong (and in which direction) your un-PADded (ha!) IBNR Liabilities are, on a monthly basis?

Perhaps you're always over-estimating. Then your added provision might be too high for some tax collection agency.
Perhaps you're always underestimating, because some stock analyst complains about low profits and downgrades the stock as a result and the company needs to assuage him.

Has there been an actual need for PAD in the past? Some giant (and this is relative to the size of IBNR) claim that no one estimating IBNR Liabilities at the time could have foreseen? If it never happens, you probably are estimating for a "black swan" (dark grey, at best) event, which sucks to defend.

DISCLAIMER: I'm no expert. Just watched other experts.
__________________
"Facebook is a toilet." -- LWTwJO
Reply With Quote
  #4  
Old 08-16-2018, 09:24 PM
cincinnatikid cincinnatikid is offline
Member
SOA
 
Join Date: Jul 2005
Posts: 2,183
Default

Are you being asked to justify whether the "90% of the time" is reasonable, or whether the current PAD factor effectively achieves that?

The former is a bit more theoretical, you might justify the assumption with impacts on rates or surplus if your reserves are off in a negative direction by more than whatever the CTE of the 90th percentile is. I think a sound logical approach would be as well received as an industry standard, which may or may not exist, to justify this.

The latter is a bit more technical, justified with a demonstration that restated reserves are below original estimates + PAD approximately 90% of the time.
Reply With Quote
  #5  
Old 08-22-2018, 03:03 PM
youngorchard youngorchard is offline
Member
CAS SOA
 
Join Date: Nov 2008
Studying for CAS
Posts: 264
Default

Not sure which country you are working out of, but from the top of my head, I know that Australia, Singapore, and Indonesia all use 75% confidence level for setting PAD/Risk Margin purposes. This information may serve as an objective benchmark to use.

Simple google search will yield numerous results on the topic related to methods/confidence levels of estimating PAD.
Reply With Quote
  #6  
Old 08-22-2018, 03:13 PM
mathmajor's Avatar
mathmajor mathmajor is online now
Member
SOA AAA
 
Join Date: Dec 2010
Location: Nowhere in particular
Studying for Japanese
College: B.S. Applied Math
Favorite beer: La Croix Grapefruit
Posts: 9,608
Default

You could study the variability of individual lags over time. Obviously not lag 0. That will be more subjective.
__________________
FSA
Opinions are provided for entertainment purposes only and are no substitute for professional guidance.
Reply With Quote
  #7  
Old 08-22-2018, 04:57 PM
exponentialpi's Avatar
exponentialpi exponentialpi is offline
Member
SOA AAA
 
Join Date: Nov 2008
Location: Po
Studying for ACA
College: Graduate of hard knocks
Favorite beer: Cold Smoke
Posts: 1,455
Default

I will say you are in good company. I have a similar request on the docket.
Reply With Quote
  #8  
Old 08-22-2018, 08:57 PM
T-roy's Avatar
T-roy T-roy is online now
Member
SOA
 
Join Date: May 2002
College: UW Whitewater
Favorite beer: FFF GBH/NG Moon Man
Posts: 8,763
Blog Entries: 1
Default

This sounds pretty easy to me. Look back at the last 50 "monthly" reserve estimates and their ultimates (say with an additional 12 months runout) and if 45 of them are AOK...ok, you do the math. Anything else you can PM me and we can discuss a consulting fee.

Bottom line, you are now the boss/leader of this project. Are you an ASA? FSA? What did you learn if so? You da pro! Go!
__________________
Quote:
Originally Posted by ORLYLOL View Post
Beer is ****ing amazing
Reply With Quote
  #9  
Old 08-23-2018, 10:54 AM
echo echo is offline
Member
 
Join Date: Nov 2002
Posts: 875
Default

I would think you need to justify it based on your own experience if it is credible. As others have said, looking back at how your best estimate IBNR compared to what was actually needed would seem to be your best bet. For short tail claims you probably would have a lot of data. For long tail claims you may have limited data and might consider a model of some kind to justify the PAD.
Reply With Quote
  #10  
Old 08-23-2018, 03:52 PM
exponentialpi's Avatar
exponentialpi exponentialpi is offline
Member
SOA AAA
 
Join Date: Nov 2008
Location: Po
Studying for ACA
College: Graduate of hard knocks
Favorite beer: Cold Smoke
Posts: 1,455
Default

Quote:
Originally Posted by T-roy View Post
This sounds pretty easy to me. Look back at the last 50 "monthly" reserve estimates and their ultimates (say with an additional 12 months runout) and if 45 of them are AOK...ok, you do the math. Anything else you can PM me and we can discuss a consulting fee.

Bottom line, you are now the boss/leader of this project. Are you an ASA? FSA? What did you learn if so? You da pro! Go!
That's the easy part. Justifying that 45 is the right number to use as the benchmark and not 40 or 49 for the auditors is the trickier part.
Reply With Quote
Reply

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off


All times are GMT -4. The time now is 05:45 PM.


Powered by vBulletin®
Copyright ©2000 - 2018, Jelsoft Enterprises Ltd.
*PLEASE NOTE: Posts are not checked for accuracy, and do not
represent the views of the Actuarial Outpost or its sponsors.
Page generated in 0.47666 seconds with 11 queries