On the issue of higher limit loss development being less than lower limit loss development, this is a real possibility for WC. A priori, I would not assume that the higher limit loss development (i.e., $50,000 cap) would produce higher ldfs than the lower limit data (i.e., $25,000 cap). The reason is large losses.
WC (like most insurance lines) has a large number of small claims that represent a minority of the total dollars paid. WC, unlike other lines, has a statutorily determined benefit, whereby the most serious claims can actually be estimated earlier than the other claims, and more accurately than the other claims. For example, if the WC claim is for a fatal case with a 40 year old widow left behind, the claim will be valued using an annuity value, and this fact will be known immediately. But it the claim was for a back injury from lifting something, it may take months, or years, to know how long someone will be disabled and what the medical costs will be. Therefore with these two claims, the largest claim will see the least amount of percentage development, and any such development will take a lifetime to be recognized, while the "small" claim has much more potential to develop.
One thing that may help is if the impact of WC life table (a.k.a. life pension, a.k.a. fatal or permanent total) claims could be removed from the table. A fatal claim can't really get any worse (other than the dependent's living longer than the life tables expect). Also, permanent totals can get worse on the medical side, but this usually occurs on a gradual, over-time basis. Therefore these large claims can be thought of a ballast or dead weight in a loss development triangle.
One other item to mention. The large loss discussion mentioned above does not care about the volume of the data. It is possible, even under the largest data sets, for higher limit losses to have less percentage development than lower limit losses. I don't know if this will always happen, but I wouldn't rule it out.
P.S. Please let me know if I need to include more info on this email for it to be added to the discussion thread.
Visit the CAS Web Site at http://www.casact.org
To subscribe or unsubscribe from CASNET:
Send an e-mail to firstname.lastname@example.org
Type in the body join casnet to subscribe
or leave casnet to unsubscribe.