The Problems with Estimates

The article by Sherry Glied, Dahlia Remler, and Joshua Graff Zivin is a useful summary and comparison of estimates of the impacts of proposals to subsidize the purchase of health insurance (private or public) as a way to reduce the number of uninsured persons. I will make one technical comment at the end of this discussion, but generally I agree with the article. What I want to discuss here is the contention (in Glied and her colleagues’ article and elsewhere) that these “cost” (really, spending) estimates either are or should be “central to the legislative prospects” of such proposals. I hasten to add that I am not arguing that the government can or should function without budgetary estimates. Of course, some numbers must be entered and then summed in order to provide estimates of total spending to be compared with estimates for total revenues. Instead, I will offer some reasons why, especially in the case of this policy intervention, such numbers need not and ought not to be central to the policy discussion.

Author(s): Mark V. Pauly

Read on Wiley Online Library

Read on JSTOR

Volume 81, Issue 1 (pages 155–159)
DOI: 10.1111/1468-0009.t01-1-00045
Published in 2003