Truck and Barter

Statisticians take note; The New York Times reports that the objective of government disaster relief is to minimize type I error (rejecting valid welfare recipients) at the expense of type II error (not rejecting invalid welfare recipients):

In the spring of 2002, as the weather warmed up for the first time since the 9/11 attack, federal and state officials announced a plan to reimburse New Yorkers who replaced air-conditioners damaged by dust and debris from the collapse of the World Trade Center...

It was an honor system of sorts, one that relied on the belief that people shaken by a national tragedy would not turn around and use it for personal gain - in this case, a free air-conditioner as warm weather approached.

But a federal audit released yesterday suggests that that is exactly what happened.

About 62 percent of the people who were reimbursed for air-quality products were not eligible, based on a sample of 4,435 applicants. Given that as many as 225,000 people asked for money back, the finding could mean that roughly 140,000 people used the program improperly...

Some people who applied for - and received - money were miles from ground zero. And a program that had been budgeted for $15 million ballooned to more than $45 million.

Of course we expect fraud, but fraud was permitted because officials feared denying anyone who could legitimately get an air conditioner welfare payment:

The audit also said there was no evidence that anyone who was eligible had been denied air-quality equipment.

Of course there would have been political hell to pay if a type I error was comitted, and somebody died because the government did not give them compensation for an air conditioner damaged on 9/11.