This article was published in an Elsevier journal. The attached copyis furnished to the author for non-commercial research andeducation use, including for instruction at the author’s institution,sharing with colleagues and providing to institution administration.Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third partywebsites are prohibited.In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further informationregarding Elsevier’s archiving and manuscript policies areencouraged to visit:http://www.elsevier.com/copyrightAuthor's personal copyJournal of Health Economics 26 (2007) 865–876Altered states: The impact of immediate craving onthe valuation of current and future opioids夽Gary J. Badgera, Warren K. Bickelb, Louis A. Giordanoc,Eric A. Jacobsd, George Loewensteine,∗, Lisa Marschf,gaBiometry Facility, University of Vermont, Untied StatesbDepartment of Psychiatry, University of Arkansas, Untied StatescDepartment of Psychiatry, Duke University, Untied StatesdDepartment of Psychology, Southern Illinois University, Untied StateseDepartment of Social & Decision Sciences, Carnegie Mellon University, Pittsburgh, PA 15213, Untied StatesfNational Development and Research Institutes, Untied StatesgDepartment of Psychiatry, St. Luke’s-Roosevelt Hospital Center, New York, NY, Untied StatesReceived 9 June 2006; received in revised form 4 January 2007; accepted 4 January 2007Available online 16 January 2007AbstractBased on prior research showing that people underestimate the influence of motivational states they are notcurrently experiencing, we predicted and found that heroin addicts would value an extra dose of the heroinsubstitute Buprenorphine more highly when they were currently craving (right before receiving BUP) thanwhen they were currently satiated (right after receiving BUP)—even when the extra BUP was to be received5 days later. If addicts cannot appreciate the intensity of craving when they are not currently experiencingit, as these results suggest, it seems unlikely that those who have never experienced craving could predictits motivational force. Under-appreciation of craving by non-addicts may contribute to initial decisions toexperiment with drugs.© 2007 Elsevier B.V. All rights reserved.JEL classification: Addiction; Utility measurementKeywords: Addiction; Craving; Behavioral economics; Hot-cold empathy gaps夽The National Institutes of Health supported this research through grants 1R01DA11692, 1R01DA122997,R37DA06526 and T32DA07242 from the National Institute on Drug Abuse.∗Corresponding author. Tel.: +1 412 268 8787.E-mail address: [email protected] (G. Loewenstein).0167-6296/$ – see front matter © 2007 Elsevier B.V. All rights reserved.doi:10.1016/j.jhealeco.2007.01.002Author's personal copy866 G.J. Badger et al. / Journal of Health Economics 26 (2007) 865–876One of the central questions in the literature on drug abuse and addiction is why peopletake addictive drugs in the first place. As Goldstein (1994, p. 12) queries in his seminal bookAddiction,“If you know that a certain addictive drug may give you temporary pleasure but will, in thelong run, kill you, damage your health seriously, cause harm to others, and bring you intoconflict with the law, the rational response would be to avoid that drug. Why then, do wehave a drug addiction problem at all? In our information-rich society, no addict can claimignorance of the consequences.”Several lines of research in economics have addressed this riddle. Perhaps most famously,Becker and Murphy (1988) argue that the addict’s decision is, in fact, rational—based on anunbiased appraisal of alternative current and future options. Objecting to the implicit characteri-zation of “happy addicts” in this perspective, Orphanides and Zervos (1995) modified the modelto allow for uncertainty on the part of potential addicts about their own susceptibility to addiction.Although they assume that estimates are unbiased, on average, those who underestimate their sus-ceptibility are most likely to end up addicted. They assume, however, that individuals learn fromexperience about their own susceptibility, albeit sometimes only after they have already becomeaddicted when they no longer have an incentive to stop taking the drug. Gruber and Kosczegi(2001) further relax the assumption of rationality, modifying Becker and Murphy’s model byintroducing hyperbolic time discounting, which leads to systematic time inconsistency, as well asmisprediction of future behavior if addicts are naive about their own preferences (on this point,see e.g., O’Donoghue and Rabin, 1999).Researchers outside of the field of economics have also proposed explanations for theparadox of initial drug use. Some have posited that addicts do not notice that they arebecoming addicted because the development of drug-dependence is so gradual (Herrnstein andPrelec, 1992; Heyman, 1996). Others have found positive support for a variety of additionalcausal factors, such as steep discounting of future consequences (Bickel and Marsch, 2001;Giordano et al., 2002), genetics (Crabbe, 2002), and peer influences (Niaura, 2000; Westermeyer,1999).While not denying that drug use has some rational component, that people may exhibit time-inconsistency but be na¨ıve about it, or the importance of the other factors discussed in the literature,we test for the operation of an additional possible factor that may play a role in initial decisions totake addictive drugs. We hypothesize that people systematically underestimate the motivationalforce of drug craving on their own future behavior. As a result, they may begin taking drugs withan unrealistic view of how easy it will subsequently be to quit.The concept of craving plays a central role in the psychological literature on drug addiction,(e.g., Marlatt, 1987). Craving refers to a “strong desire or intense longing” for a drug whichproduces a powerful, often overwhelming, urge to consume the drug (Kozlowski and Wilkinson,1987, p. 31). Almost all drugs, including cocaine (which at one time seemed to present theanomalous case of an addictive drug that did not produce withdrawal or craving) have beenshown to produce craving (Gawin, 1991). While the economic models of addiction discussedabove do not deal explicitly