FDA economic model used in Regulatory Impact Analysis underestimates benefits by ignoring short term effects of smoking

I just submitted this public comment to the docket on the FDA deeming rule.  The analysis they did this time makes all the same mistakes they did in 2010 when they analyzed the proposed warning label rule.

RE: Economic model used in Regulatory Impact Analysis underestimates benefits by ignoring short term effects of stopping and starting tobacco use

The Regulatory Impact Analysis for the proposed rule bases its assessment of benefits on the old 2004 book, The Price of Smoking, by Sloan, et al (Ref 68).  The analysis in this book, while reasonable at the time it was published, is badly out of date because it does not account for the rapid changes in risk of several diseases (most notably heart and lung diseases and well as complications of pregnancy) that happen when people stop or start smoking.  Even cancer risks begin to fall much more quickly than reflected in Sloan, et al.  Because of the (appropriate) use of time discounting, the FDA’s failure to account for these short-term effects leads the RIA to substantially underestimate benefits and so substantially overestimate the break-even point in terms of years of life saved.

Because we provided a detailed critique of using this book as the key reference in our public comment submitted to FDA in response to its Notice of Proposed Rulemaking on Cigarette Warning Labels (Docket No. FDA–2010–N–0568, 75 Fed. Reg. 69524 et seq., November 12, 2010) on January 10, 2011, I am resubmitting that comment to the current docket.

There is also a discussion of the FDA’s overestimate of the costs of complying with warning label requirements which are relevant to the current rule (although the details are different).

This comment also outlines reasons that it is inappropriate to include a consumer surplus discount in estimating net benefits of the proposed deeming rule.  We will also be submitting an additional more detailed critique of the use of consumer surplus before the comment period closes.