In a previous post, I discussed the highly inappropriate Pentagon response to Linda Bilmes' study of the potential financial costs stemming from servicemembers wounded or injured in Iraq and Afghanistan. In this post, I'm going to talk about a very serious flaw that I found while examining that report. I've kept the two posts separate because I don't want the inappropriate Pentagon response to cloud an objective look at the paper itself.
There are actually a number of places where I think her analysis does not adequately consider various factors, but one of the errors is so large that its effects will pretty clearly eclipse all the rest. This particular error is clear cut, and clearly resulted in estimated costs that were far higher than they should have been.
In both her LA Times op-ed and in the paper itself, Dr. Bilmes states that 44% of Persian Gulf War veterans filed disability claims. That figure struck me as being a bit high, so I thought is might be worth taking a closer look at that figure.
Dr. Bilmes provides the basis for the 44% calculation in footnote 26 (page 9) of her paper:
For Gulf War, the total claims filed to date are 271,192, of which 205,911 have been approved, 20,382 were denied and 34,899 are still pending (GWVIS, August 2006, p.7: Granted Service Connection +Denied Service Connection +Claims Pending)
I was only able to access a Google cached version of the August, 2006 GWVIS report, but I was able to download some of the earlier reports at another site. Looking at the table on the referenced page, I found a major error in her calculations.
The numbers of claims filed that she provided were correct. Unfortunately, in calculating the percentage of veterans filing claims, she appears to have divided by the wrong number. If I divide 271,192 by 613,526, I get 44.2% - just about what she got. Unfortunately, 613,526 is not the number of veterans of the conflict. It's the number of living veterans of the conflict. The number of claims, on the other hand, is not based on the number of currently living veterans.
From the glossary on page 17 of the same document:
The number of Claims Granted Service Connection includes veterans who have subsequently died. The total number of claims granted remains constant or increases each quarter.
Since the number of claims granted includes people who have since died, and since the number of claims granted + the number denied + the number pending do add up to the total number of claims filed, it seems clear that the percentage of veterans filing claims should not have been calculated based on the number of currently living veterans. Instead, it should have been calculated based on the total number of conflict veterans. That number (696,842) is found on the same table of the same document, one line above the number of living veterans. Dividing the number of claims by the total number of veterans, I get 38.9% - a figure that is substantially lower than the 44% used by Dr. Bilmes.
Why is that important? Well, in the paper, when calculating the financial costs of caring for Iraq and Afghanistan veterans, Dr. Bilmes used the 44% rate she calculated for the Gulf War as the basis for her assumptions about the likely rate of claims from the current conflicts. Her most optimistic scenario used 44%, and her two more pessimistic scenarios used 50%, on the assumption that the more intense current conflicts would result in a somewhat higher rate. If those figures are adjusted downward to account for the miscalculation of the Gulf War rate, her estimated financial costs will drop by quite a bit.
This might not seem like much to get worked up about. After all, the costs of caring for the new generation of veterans is still going to be higher than any of the estimates coming from the White House. It is clear that the VA does not have the resources needed to provide the influx of new veterans with all the benefits that they are entitled to. The situation is grim no matter how you look at it.
That's actually why I'm worked up about this. The White House and its (rapidly decreasing) pool of allies will fight against anyone who dares contradict their views, and they will fight with every weapon that they get. A flaw in a study is rapidly siezed upon as an excuse to completely ignore not just that study, but any other study suggesting anything similar. There's really no excuse for sloppy mistakes at the best of times, and under current circumstances they can be exceptionally harmful.
- Log in to post comments