The Harvard Project on U.S. Pharmaceutical Regulation, a.k.a.


Daniel Carpenter

Department of Government

Harvard University


Summary: A large-scale theoretical, historical and statistical analysis of pharmaceutical regulation in the United States as it is carried out by the U.S. Food and Drug Administration (FDA). This project has been funded by the National Science Foundation (SES-0076452, for "Formal and Empirical Analyses of Bureaucratic Delay: The Case of FDA Drug Review") (2000-2003), the Robert Wood Johnson Foundation Scholars in Health Policy Program (1998-2000), a Robert Wood Johnson Foundation Investigator Award in Health Policy Research (2004-2006), the Edmund J. Safra Center for Ethics at Harvard University, the National Institutes of Health through Harvard Catalyst (grant 1UL1RR025758-01), and the PriceWaterhouse Coopers Endowment for the Business of Government. None of these foundations or entities in any way endorses, or is otherwise responsible for, the contents of this Web site or Web page, or for any papers produced or claims found herein. Professor Carpenter neither seeks nor accepts research funding or any other form of compensation from the FDA, from private entities that may sponsor product applications to the FDA or that are otherwise regulated by the FDA, from patient advocacy groups or from consumer advocacy groups (e.g., Public Citizen). In addition, Professor Carpenter does not directly own stock or any other form of asset in any company regulated by the FDA. [A note about this policy.]



Historical and Theoretical Studies

Book: Reputation and Power: Organizational Image and Pharmaceutical Regulation at the FDA (Princeton: Princeton University Press, 2010). [More comprehensive page of materials on the Reputation and Power for this site.]

Awarded The 2011 Allan Sharlin Memorial Award of the Social Science History Association for Outstanding Book on Social Science History

Early Reviews and Reactions:

Tyler Cowen's treatment at Marginal Revolution

Steven Teles' review in the Washington Monthly

James Surowiecki's discussion in The New Yorker

Data and Documentation

NME Approval Time Data [.csv] -- This data set, which was used for the book and for a Fall 2010 article in the Journal of Law, Economics and Organization, represents what I believe to be the most comprehensive historical data set on FDA drug approval. At present it runs from 1950 to 2006, and we will be updating it to relfect more recent drug approvals. Susan Moffitt was my primary collaborator in setting up this data, and thanks to David Dranove and David Meltzer for sharing some of their data. Comments and corrections are most welcome.

Codebook [brief descriptions of variables]


Over the coming months I will create a new page where various archival documents of interest from the book are available (particularly where these are not protected by copyright, as in the case of federal archives).

Other Historical Studies

Bioequivalence: The Regulatory Career of a Medical Concept (with Dominique Tobbell), Bulletin of the History of Medicine 85 (1) (Spring 2011) 93-131. Presented at the AIHP Conference on Modern Medicines, Madison, Wisconsin, October 2008; and at "Unbranding Medicines: The Politics, Promise, and Challenge of Generic Drugs," Harvard University, December 2008.

Policy Tragedy and the Emergence of Economic Regulation: The Food, Drug and Cosmetic Act of 1938, Studies in American Political Development 21 (2) (Fall 2007) 149-180 (with Gisela Sin).

"Robust Action and the Strategic Use of Ambiguity in a Bureaucratic Cohort: FDA Scientists and the Investigational New Drug Regulations of 1963," (with Colin Moore). In Formative Acts, ed Stephen Skowronek and Matthew Glassman, University of Pennsylvania Press, 2007.

"The Ambiguous Emergence of an Organizational Identity: Pharmaceutical Regulation at the FDA, 1947-1961"; manuscript presented at the Boston University American Political History Seminar, Fall 2005; to be presented at Rutgers University Health Policy Seminar, January 2006; Yale University American Politics Workshop, January 2006; Stanford University Law School, March 2006.

"The Continuous Crystallization of an Organizational Reputation: The FDA in Public, Scientific and Congressional Audiences, 1962-1990." Presented at the CARR Conference on Risk and Regulation, Plenary Lecture, Centre for the Analysis of Risk and Regulation (CARR), London School of Economics, March 2007; presented at the Sociology Department Workshop, Northwestern University, April 2007.

Essays and Policy Relevant Studies

Reputation and Precedent in the Bevacizumab Decision (with Aaron Kesselheim and Steven Joffe), New England Journal of Medicine 365 e3 (July 14, 2011). [A July 2011 interview on the FDA's regulation of bevacizumab (Avastin) for HER-2 negative metastatic breast cancer appears at the Pharmalot site (interview in pdf).]

Reputation, Gatekeeping, and the Politics of Postmarket Drug Regulation, AMA Virtual Mentor (The American Medical Association), 8 (June 2006): 403-6.

The Political Logic of Regulatory Error (with Mike Ting), Nature Reviews – Drug Discovery, 4(10) (October 2005): 819-23. [PubMed link]

A Modest Proposal for Financing Drug Safety Regulation through User Fee Augmentation, Health Affairs (Web Exclusive) W5-469, October 18, 2005. For some relevant discussion, see Simon Frantz’s articles “How to Avoid Another Vioxx and Vioxx Fears Prompt Call for User Fee Evaluation” at Nature Online.

Gatekeeping and the FDA's Role in Human Subjects Protection, AMA Virtual Mentor (The American Medical Association), November 2004.

Accelerating Approval Times for New Drugs in the United States, The Regulatory Affairs Journal – Pharma, 15 (6) (June 2004): 411-417.

The Political Economy of FDA Drug Approval: Processing, Politics and Lessons for Policy, Health Affairs 23 (1) (January/February 2004): 52-63.

Staff Resources Speed FDA Drug Review: A Critical Analysis of the Returns to Resources in Approval Regulation, Journal of Health Politics, Policy and Law, 29 93) (June 2004), 431-442.

Approval Times For New Drugs: Does The Source Of Funding For FDA Staff Matter?, with Michael Chernew, A. Mark Fendrick, and Dean Smith. Health Affairs (Web Exclusive) December 17, 2003, W3-618-624.

Statistical Output for Selected Model Runs for this Piece

Models using Time-Varying Covariates: The CDER Staff Effect Persists even when Submission-Year Averaging is Relaxed

Amicus Curiae Brief for Wyeth v. Levine case (Supreme Court, 2008-2009 term)

Technical Papers


Regulatory Errors with Endogenous Agendas, American Journal of Political Science 51 (4) (October 2007) 835-853 (with Michael Ting). This paper also contains empirical tests of the model. [This is a pre-print version of the paper; previous version available as Regulatory Errors under Two-Sided Uncertainty, OR, The Political Economy of VIOXX, Working Paper #30 in the Robert Wood Johnson Foundation Scholars in Health Policy Program Working Paper Series] [Online Appendix with Proofs and Extreme-Value Regression Results.]

Protection without Capture: Product Approval by a Politically Responsive, Learning Regulator. American Political Science Review 98 (4) (November 2004), 613-631.

Robert Wood Johnson Foundation Scholars in Health Policy Working Paper #13.

Technical Notes for “Protection without Capture” [to be posted]

The Trouble with Deadlines (with Justin Grimmer). Working paper (2009), presented at Petrie-Flom Health Law and Policy Workshop, Harvard Law School, November 2006.

"Why Do Bureaucrats Delay? Lessons from a Stochastic Optimal Stopping Model, with Applications to FDA Drug Review," in Kenneth Meier and George Krause, Politics, Policy and Organizations: Scientific Approaches to Bureaucratic Politics (Ann Arbor: University of Michigan Press, 2003)

“A Theory of Approval Regulation,” with Mike Ting.


The Complications of Controlling Agency Time Discretion: FDA Review Deadlines and Postmarket Drug Safety, American Journal of Political Science 56 (1) (January 2012) 89-114 (with Jacqueline Chattopadhyay, Susan Moffitt, and Clayton Nall). [A replication archive with code and auxiliary materials can be found here.]

Early Entrant Protection in Approval Regulation: Theory and Evidence from FDA Drug Review, Journal of Law, Economics and Organization, 26 (2) (Fall 2010) [doi: 10.1093/jleo/ewp002] (with Susan Moffitt, Colin Moore, Ryan Rynbrandt, Ian Yohai, and Evan James Zucker). [This paper also contains a mathematical model of sequential regulation of a category of therapies.]

Drug-Review Deadlines and Safety Problems (with Evan James Zucker and Jerry Avorn), New England Journal of Medicine 358 (13) (March 27, 2008) 1354-61. [MEDLINE Link] [NEJM Exchange with FDA Officials, including Errata and Corrected Estimates] [Auxiliary Page for this Study, including Replication Datasets and Memorandum with Robustness and Sensitivity Analyses]

Deadline Effects in Regulatory Review: A Methodological and Empirical Analysis, available as Working Paper #35 in the Robert Wood Johnson Foundation Scholars in Health Policy Program Working Paper Series.

Groups, the Media, Agency Waiting Costs, and FDA Drug Approval, American Journal of Political Science, 46 (3) (July 2002) 490-505. [This link is to a stable URL at the J-STOR webpage; you may need J-STOR access to read this file.]

Winner, The 2001 Pi Sigma Alpha Award for Best Paper presented at the 2000 meetings of the Midwest Political Science Association [over 1,800 papers presented]

Nominated by the Midwest Public Administration Section for the Robert Durr Award (best methodological paper, 2000 convention).

Robert Wood Johnson Foundation Scholars in Health Policy Working Paper #21.

Replication results using shared (inverse Gaussian) frailties for primary indication, and fixed (coefficient vector) effects for sponsoring firm.

Replication data for Table 1, full model. [Note: these data are intended for read to S-Plus or to STATA (Version 7.0 or later). None of the drugs, nor firms, nor primary indications, is identified. All NME data collected from FDA sources and newspaper reports by Daniel Carpenter. To replicate Table 1, Model 1 of the AJPS article, download these data, save into STATA 7.0 or later, and then redefine the set variables (that is, rename _t, _t0, and _d as something else). Then stset the data (declare “acttime” to be the survival variable, and the variable previously named “_d” to be the failure or inverse-censoring variable. Then issue the following command:

streg hcomm hfloor scomm sfloor prespart demhsmaj demsnmaj orderent stafcder prevgenx lethal deathrt1 hosp01 hospdisc hhosleng acutediz orphdum mandiz01 femdiz01 peddiz01 natreg natregsq wpnoavg3 vandavg3 condavg3, dist(lognormal) frailty(invg)

Replication data for Tables 2 & 3, reduced models [Note: None of the drugs, nor firms, nor primary indications, is identified. All NME data collected from FDA sources and newspaper reports by Daniel Carpenter. Also, I just recently produced this reduced-model data and cannot yet retrieve the exact estimates reported in AJPS 2002, though the ones you get here are very close. It may be that I updated the data after the galleys went to AJPS, without saving the original data (with 1 or 2 flaws). If so, I should have documented this more carefully. I regret the error.]

Why Do Bigger Firms Receive Faster Drug Approvals? (with Colin Moore, Marc Turenne, Ian Yohai and Evan James Zucker). Prepared for presentation at the Harvard-MIT Workshop on Positive Political Economy, March 16, 2001. Currently under revision.