July 17, 2003

Friends and Colleagues:

I recently sent out for comment a preliminary draft of a paper, “Deferred Taxes in Public Finance and Macroeconomics”. Much of what is in that paper - the asset of about $3 trillion of taxes on future withdrawals from retirement accounts that the federal government has accrued; the fact that accrual accounting for deferred taxes would offset roughly half the increase in the national debt from 1981 to 1992 and a multiple thereafter; the addition of a third side to the future political economy of deferred taxes with pressure to lower taxes coming from the larger number of voters withdrawing funds; the understatement of historical estimates of real returns to stock and bonds as current BLS methods would substantially lower historical CPI inflation used to deflate nominal values, to name a few - remains valid. However, the part of the paper projecting the future contained a programming error that in turn resulted from a word-processing error which I should have but did not catch. The term for the share of after-tax withdrawals consumed was inadvertently dropped from the investment equation. While the qualitative discussion remains valid, this did lead to a considerable overstatement of future deferred taxes. I am in the process of correcting this problem, but as the preliminary draft, circulated for comment, has gotten into the public domain (given the Internet age, people have been citing and writing about it before I have had a chance to correct it), I especially want to alert you well before the revised draft is ready for Working Paper status. Please do not cite or quote anything from the original paper without explicit permission. I apologize for any inconvenience this may have caused you. The fault is entirely my own, not that of anyone who may have quoted the preliminary draft.

Best,

Michael Boskin