days of medical claims payable

[Dr T Non-Fan]

From Oxford earnings release:Shares of Oxford fell more than 7 percent to $39.98.
Advest analyst Robert Mains said possible explanations for the drop could include the company's planned entry into the New Hampshire market or an increase in days of medical claims payable.
Mains said Oxford exited the New Hampshire market a few years ago because it was unprofitable, but now he said the current management team knows that market very well. Also, an entry into New Hampshire would not be large enough to have a significant impact on the results.
Also, Oxford's payables rose as a result of an increase in membership, Mains said.
(end)

I've tried to explain to people how irrelevant "days of medical claims payable" is. I've gotten nowhere.

I'll assume it's me, and request some help.

My issues:
1. Need a claims paid per day stat. This fluctuates a lot, so choosing the appropriate number is very important. Using a raw historical number might mislead.
2. Claims payable, AKA "IBNR Liability," is not directly related to claims paid. In fact, there's more of an inverse relationship.
3. This inverse relationship will cause the days of claims payable to vary a lot more than a Wall Street analyst (or investor) would like it to.

So, Mains says that investors saw days claims payable rise and dumped the stock? Sounds to me like an opportunity.

Any comments?

[josie]

I agree with you. Days claims payable is an irrelevant statistic in which these investor analysts, who we give all our money to, use to decide if they should invest in a company. It fluctuates much more than the investor analysts believe it should. Just the number of days used in the denominator (i.e, 90, 91, or 92) has an impact on the result beyond their acceptance.

That being said, since we have to explain and rationalize this irrational statistic, how do companies manage this. It seems that companies have detailed explanations regarding why the statistic changes. My specific questions are:

1) Before fianlizing the IBNR, do companies look at this statistic and possibly adjust the IBNR (within reason) so that the statistic is more acceptable?

2) What are your standard explanations that you use to help you explain the change in the statisitc?

3) Do you have any standard reports that you run that help with this explanation?

Thanks!

[Dr T Non-Fan]

I can't answer any of your questions. I don't think an actuary could explain it with a straight face.

I think the level of acceptable change is too strict.

Perhaps it needs a conjoined statistic: analyzing it in conjunction with unusual cash flow? Normally you'd expect an increase in positive cash flow with an increase in DOCP. But if there isn't, then maybe there's something fishy going on.
If this is the case, then disclosing it as an independent statistic is not proper.

Activities and situations that increase DOCP, off the top o' noggin:
1. Slowdown in paying claims.
2. Growth.
3. A few very large claims pending.
4. Systems conversion.
5. Three-day snowstorm.
6. 9/11.
7. Overstatement of IBNR Claim Liability.
8. Sudden large increase in average benefits covered.

[Gates Is Antichrist]

I don't believe in this case that IBNR is even of remote interest to an investor. They could have sage reasons to perk up for increase in DOCP (and possibly sell/drive down price):

1) When a DOCP increase actually reaches public airwaves, there's a chance of lapse meltdown - a "panic" (of some size) if you will. IOW a chain reaction of lapses may be set off under some circumstances, justifying vamoose from the company's stock :P

2) It's a small B- company and there's an immediate cash flow issue, or it's there may be an impending cash scrape (I'm ignoring Oxford's actual size & rating just to make a point). At least for small companies, this tactic is not necessarily fictional.

3) There are operational problems. (The conversion blew, and a whole line of business can't be processed.) (The new claims check system is garbage.) (etc.)

Irrelevant or irrevelant? Well, as stock pricing goes, it's not that a definitive cause/effect must be established; but when the effect part happens, there has to be suspicion raised as to the cause. And when DOCP goes UP, I can't come up with a good cause.

[Dr T Non-Fan]

So, when DOCP rises with cash flow, there shouldn't be an unexplainable issue. Claims weren't paid, and IBNR went up. IBNR up = increase in cash flow (from net income, that is).

Mostly, it sounds like people trying to pass themselves off as smarter than they really are. And, it sounds like an investment opportunity on the rebound. Wait a few days for the suckers to leave, buy and wait for the suckers to return.

[Toll Free]

I don't necessarily agree. Some IBNR fluctuation is normal, but dramatic increases in IBNR often signal big trouble - not the least of which because, in most standard IBNR calculations, a slowdown in claims payment increases completion factors, causing a "double-whammy" effect in the reserve, that in my experience has often been closer to a "triple-whammy" or worse.

Also, Oxford has already gone bankrupt once (I believe - at the least, they were in deep trouble), and I believe that understated IBNR (and contracting?) were the problems the first time. It's not suprising at all that investors are skittish.

I do use reserves/exposure and reserves/claims pmpm in setting reserves each month - more as a reasonability check, but for some coverages (drugs) it's really quite useful.

[alex]

Days of Claims Payable has been a hot topic among investors the past two days...

[Dr T Non-Fan]

Too bad these particular investors control a lot of the flow.

Perhaps a session at an SOA meeting could help actuaries explain the folly of "Days of Claims Payable" as a stand-alone statistic.

[josie]

I do think the IBNR is something that the investor analysts are worried about because it is such a large estimate. The IBNR has a huge impact on reported earnings and I can understand why investor analysts wouldn't like that. They are trying to get their hands around the IBNR with DOCP, but we all know that DOCP makes no sense. If we are not happy with DOCP, isn't it our responsibility as actuaries to come up with something that would help investors feel more comfortable about the IBNR. I agree that this topic should be addressed by the SOA because it is an extremely hot topic and I don't see it going away. A working session at the next meeting would be very helpful.

[Dr T Non-Fan]

I don't think it's our responsibility (sounds so important). We have the ability to come up with better analyzing tools, though.

Ideas:
1. Always restate past earnings as part of analysis, as opposed to the annual one-time changes. This might be a tax issue, though, if one rearranges earnings for timing purposes.

2. Instead of letting analysts believe that health insurance companies' earnings can be estimated with such precision (to the nearest penny per share), create a range of actual earnings for each restated period, with the near past having a wider range than the far past. And note that this range is acceptable for the industry. Smaller companies would have larger ranges (in terms of per-share earnings). Part of the risk of investing in a smaller company.