Wednesday, June 22, 2011

Related Talk

I mentioned that I was giving some talks in April that were related to the Five Drafts project. Here is one of them, "Re-Legalizing Drugs," from the TEDxUChicago conference on April 17, 2011.

Update on Draft 4.1

Well, Draft Four made it online, albeit a few days late. Once again, in certain respects, I feel that Drafts Three and Four have been more "drafty" than Draft Two. The new section Five (discussing six "complicated issues") is barebones, and the deepening of other elements, by and large, did not take place. Still, I have hopes that Draft Five will be a serious improvement. My first step in the process of improving is to print out Draft Four and read it once through, and then enter "continuous improvement" mode. I am scheduled to be away for more than two weeks, though, during which time I do not expect to be able to make a serious commitment to at least the continuous part of continuous improvement.

The beleaguered Five Drafts reader knows that I have been trying to read three relevant books for awhile now. They are:

(1) Last Call, by Daniel Okrent, 2010;

(2) Addiction: A Disorder of Choice, by Gene M. Heyman, 2009; and,

(3) Carrots and Sticks, by Ian Ayres, 2010.

The Okrent book remains on the back burner. I have now read the first 132 pages (a gain of, er, two) out of 496, as well as everything after page 310. The Heyman book has moved into the completed pile. I don't know whether it is because of or despite the fact that I am an economist that his characterization of addiction via local and global equilibria did not appeal to me. As I indicated before, it seems to be another, slightly more confusing way of contrasting myopic choices with forward-looking ones. For those who want an anti-disease view of addiction, the book is fine, good even, but I also would recommend Herbert Fingarette's old (1989) Heavy Drinking: The Myth of Alcoholism as a Disease. As for the Ayres book, some real progress: now at page 126 of 218, up from page 40 when last I reported in. The new book I will add to the pile is Drugs and Drug Policy: What Everyone Needs to Know, by Mark A.R. Kleiman, Jonathan P. Caulkins, and Angela Hawken. Seems like an easy read -- as is the Ayres book, incidentally. The Okrent book is written with great style and verve, but is more detailed, more weighty, than the books by the academics, and hence I find it harder to pick up.

Sunday, June 19, 2011

Draft Four, Composite

Draft Four, Brief Apology

Toward Drug Control: Exclusion and Buyer Licensing

Jim Leitzel

Section 1, Introduction

Section 2, Two Major Contributions to Ending Prohibition
(including 2.1, Fosdick and Scott, and 2.2, Transform)

Section 3, The Alcohol Model, Plus Exclusion
(including 3.1, Mandatory Exclusion, and 3.2, Committing to Exclusion)

Section 3.3 (Gambling Self-Exclusion)

Section 3,4 (Drug Self-Exclusion)

Section 4, The Alcohol Model, Plus Consumer Licensing (including 4.1, From Opting Out to Opting In)

Section 4.2 (Licensing Drug Users)

Section 5, A Closer Look at a Small Subset of Complicated Issues (including 5.1, Sharing and Social Norms; 5.2, The Highly-Skewed Distribution of Drug Consumption; and 5.3, The Privacy of Licensee or Exclusion Lists)

Section 5.4 (Advances in Drug-Free Probation and Bond Programs), section 5.5 (Legal Pharmaceuticals and Diversion), and section 5.6 (Advertising)

Section 6, Conclusions

References and Table

Draft Four, References and Table


Billings, John S., Charles W. Eliot, et al., The Liquor Problem: A Summary of Investigations Conducted by the Committee on Fifty, 1893-1903. Boston and New York: Houghton, Mifflin and Company, 1905.

Camerer, Colin, Samuel Issacharoff, George Loewenstein, Ted O’Donoghue, and Matthew Rabin, “Regulation for Conservatives: Behavioral Economics and the Case for ‘Asymmetric Paternalism.’” University of Pennsylvania Law Review, June 2003.

Caulkins, Jonathan P., and Robert L. Dupont, “Is 24/7 sobriety a good goal for repeat driving under the influence (DUI) offenders?” Addiction 105(4): 575–577, April 2010.

Fosdick, Raymond B., and Albert L. Scott, Toward Liquor Control. Harper Brothers, 1933.

Global Commission on Drug Policy, War on Drugs. June, 2011; report available at

Hawken, Angela, “HOPE for Probation: How Hawaii Improved Behavior with High-Probability, Low-Severity Sanctions.” Journal of Global Drug Policy and Practice 4(3), Fall 2010; available at

Hemenway, David, While We Were Sleeping: Success Stories in Injury and Violence Prevention. Berkeley: University of California Press, 2009.

Husak, Douglas, Legalize This! The Case for Decriminalizing Drugs. London: Verso, 2002.

Kleiman, Mark, “Dopey, Boozy, Smoky – and Stupid.” The American Interest Online 2(3), January – February 2007; available at www.the‑american‑ (visited February 24, 2007).

Kleiman, Mark, Against Excess: Drug Policy for Results, New York: Basic Books, 1992.

Kleiman, Mark A.R., Jonathan P. Caulkins, and Angela Hawken, Drugs and Drug Policy: What Everyone Needs to Know. Oxford University Press, 2011.

Leitzel, Jim, Regulating Vice. New York: Cambridge University Press, 2008.

Leitzel, Jim, “Self-Exclusion.” Available at SSRN:; April, 2011.

Levine, Harry G., and Craig Reinarman, “From Prohibition to Regulation: Lessons From Alcohol Policy for Drug Policy.” In Ronald Bayer and Gerald M. Oppenheimer, eds., Confronting Drug Policy: Illicit Drugs in a Free Society, Cambridge: Cambridge University Press, 1993.

Loewenstein, George, Ted O’Donoghue, and Matthew Rabin, “Projection Bias in Predicting Future Utility.” Quarterly Journal of Economics 118(4): 1209-1248, November 2003.

Long, Larry, “The 24/7 Sobriety Project.” The Public Lawyer 17(2), Summer 2009.

MacCoun, R. J., and P. Reuter, Drug War Heresies: Learning from Other Vices, Times, and Places. Cambridge: Cambridge University Press, 2001.

Mill, John Stuart, On Liberty, edited by Elizabeth Rapaport. Indianapolis: Hackett Publishing Company, 1978 [1859].

Miron, Jeffrey A., and Katherine Waldock, The Budgetary Impact of Ending Drug Prohibition. Cato Institute, 2010; available at

O’Donoghue, Ted and Matthew Rabin, “Studying Optimal Paternalism, Illustrated by a Model of Sin Taxes." American Economic Review 93(2): 186-191, May 2003.

Responsible Gambling Council, “From Enforcement to Assistance: Evolving Best Practices in Self-Exclusion.” Discussion paper, Ontario, March, 2008; available from

Substance Abuse and Mental Health Services Administration, Results from the 2008 National Survey on Drug Use and Health: National Findings. Office of Applied Studies, NSDUH Series H-36, HHS Publication No. SMA 09-4434, Rockville, MD, 2009.

Sunstein, Cass R., and Richard H. Thaler, “Libertarian Paternalism is Not an Oxymoron.” University of Chicago Law Review 70, Summer 2003.

Thaler, Richard H., and Cass R. Sunstein, Nudge: Improving Decisions About Health, Wealth, and Happiness. New Haven: Yale University Press, 2008.

Thaler, Richard H., and Cass R. Sunstein, “Libertarian Paternalism.” American Economic Review 93(2): 175-179, May 2003.

Transform Drug Policy Foundation, After the War on Drugs: Blueprint for Regulation, 2009; available at

Zinberg, Norman E., Drug, Set, and Setting: The Basis for Controlled Intoxicant Use. New Haven: Yale University Press, 1984.

Table 1: Estimated totals of top 7 arrest offenses, plus gambling and prostitution arrests, United States, 2009

Type of arrest Number of arrests

Total arrests 13,687,241

Drug abuse violations 1,663,582

Driving under the influence 1,440,409

Larceny/theft 1,334,933

Simple assaults 1,319,458

Disorderly conduct 655,322

Drunkenness 594,300

Liquor laws 570,333

Prostitution and commercialized vice 71,355

Gambling 10,360

Many disorderly conduct arrests are alcohol-related; assaults are another common arrest category that involve a substantial alcohol-related segment.

Source: FBI, Uniform Crime Reports, Crime in the United States 2009, Table 29, available at

Draft Four, Section 6

6. Conclusions
“With apprehension as well as relief the public asks: What shall replace national prohibition?” -- Fosdick and Scott, 1933, page 5.
One of the barriers to ending drug prohibition is the breadth of the possible meanings of “legalization.” The purpose of this paper is to explore two elements of regulatory systems – buyer licensing and self-exclusion – that can be helpful components of drug control and offer some assurance that drug legalization does not imply free availability. These types of regulations provide assistance to those who are worried about self-control problems with drugs, while not being significantly constraining upon those who are informed and satisfied drug consumers.
Three observations underlie my interest in licensing and self-exclusion. First, many reasonable people want to use the currently illegal drugs, and are willing to pay high prices and run not-insignificant risks to do so. Second, most use of such drugs is not personally or socially detrimental. Third, people have a strong incentive to avoid or end addictions, which are terribly costly. These three observations suggest that appropriate policy regimes can harness self-interest to do most of the work in controlling drugs, while saving coercive measures for socially harmful elements of drug consumption, and focusing treatment resources on those with the greatest medical needs.
What shall replace national drug prohibition? For softer drugs, I suggest something like the current alcohol system, along with mandated and voluntary exclusion provisions. For harder drugs, my preference is for an alcohol-style system, supplemented with buyer licensing as well as mandated exclusions. The legal regulatory structures can be used to promote, in relative terms, milder forms of drug use: opium tea instead of smoking, opium instead of heroin, and so on. License tests can help make sure that users are informed of drug dangers and procedures for dealing with problems of overdose or addiction, and license revocation can provide an additional sanction for drug-related misbehavior. In short, relative to prohibition, licensing and self-exclusion can be part of a drug regulatory structure that is much more finely tuned to the risks of harm.
My confidence in the suggestions put forth here is limited – your confidence is probably even more limited! After 50 years of global drug prohibition, it is hard to predict with assurance the consequences of various legal regulatory regimes for drug control. Unfortunately, we do have good information on the doleful consequences of prohibition, and that information is the best case for legalization experiments. While within a prohibitory regime there can be adjustments at the margin, the prison terms, arrests, violence, and corruption are endemic to prohibition, foreseeable but highly undesirable consequences of banning drugs. Though we cannot be sure of what licensing and exclusion will bring, that uncertainty, combined with the all-but-intolerable status quo, is a good reason to look at legal options – as Fosdick and Scott suggested in 1933 for the case of alcohol: “In a country as large as ours, with so great a variety of local conditions, there is room for many types of experiment. Indeed, the …states will constitute a social science laboratory in which different ideas and methods can be tested, and the exchange of experience will be infinitely valuable for the future…”[i] as we move Toward Drug Control.

[i] Fosdick and Scott (1933, p. 150); the elided words before “states” are “forty-eight.”

Draft Four, Sections 5.4, 5.5, and 5.6

5.4 Advances in Drug-Free Probation and Bond Programs

In 2004, Judge Steven Alm debuted a probation program that took advantage of the tendency of people to respond more to the speed and certainty of sanctions than to their severity. Hawaii’s Opportunity Probation with Enforcement (HOPE) probationers are randomly but frequently called in for drug tests. Failure to present for the test or to pass it results in a speedy arrest and a short period (typically 2 or 3 days) of jail time. Since its inception, HOPE has shown remarkable results in inducing a previously recalcitrant population to comply with probation requirements, including requirements to remain drug free. Angela Hawken provides a capsule summary[i]:

The Integrated Community Sanctions Unit (Honolulu’s intensive-supervision high-risk probation unit) was the first unit to pilot a HOPE program. The rate of positive drug tests fell by 93 percent for HOPE probationers during the first six months (from 53 percent to 4 percent), compared with 14 percent for comparison probationers (from 22 percent to 19 percent). These improvements in probationer behavior translated into other benefits. Recidivism fell sharply, as did arrests (arrests were more than halved), revocations, and incarceration (an average of 130 prison days were saved per probationer). Findings then were later replicated with a randomized controlled trial of high-risk, primarily methamphetamine-using probationers in a general probation unit.

South Dakota’s 24/7 Sobriety Program also indicates significant initial success at keeping alcohol offenders (as well as some other drug offenders, including methamphetamine users) abstinent, and less likely to re-offend.[ii] Started in 2005 with statewide implementation following two years later, the South Dakota approach is now being replicated in North Dakota and Montana. As with HOPE, 24/7 Sobriety employs frequent testing with high-probability, low-duration jail terms.

A small percentage of participants in HOPE and in 24-7 Sobriety repeatedly fail their tests. The swift and sure – though not strict – punishments attached to drug or alcohol use fail to deter these individuals. Within this sad outcome, however, resides at least one small compensation. Those who cannot stop using their drug of choice despite the near-immediate sanctions are the appropriate recipients of treatment resources. In the HOPE context, Hawken (2010) terms this the “behavioral triage” effect. Referrals to treatment are based on observable behavior, not unsubstantiated claims of drug usage. Further, under a re-legalized drug regime, the courts (and to some extent, treatment providers) would not be clogged with people whose only “crime” is drug consumption. Coerced treatment, independently of its effectiveness, generally is inappropriate for adults whose drug use has not been a cause of harm to others.

HOPE and 24/7 Sobriety demonstrate that exclusion is an enforceable approach to drug control. Indeed, many of the clients of these programs are methamphetamine users – and most of them choose not to use methamphetamine, one of the more reinforcing of the currently illegal drugs, given the incentives established under the programs. (Note that prohibition alone was not capable of preventing these people from establishing and maintaining addictions.) For people who some years ago believed that it was a close call as to whether prohibition or legalization was a better approach to drug policy, the success to date of HOPE and 24-7 Sobriety should tip the scales in favor of legal drug access.

5.5 Legal Pharmaceuticals and Diversion

Drugs that are available by prescription only represent an example of a licensed user regime, where the prescription constitutes the license to acquire and consume limited quantities of the drug. Some drugs currently available via prescription for medical conditions also are popular for recreational use. Recreational users either generate a legal prescription, or acquire drugs that have been diverted from the legal supply chain, often by the holder of the prescription. In 2008, some 6.1 percent of Americans 12 years of age and older used a prescription drug in a non-medical fashion, with nearly one percent of Americans taking such drugs in a dependent or abusive manner.[iii]

The significant (and largely illegal) leakage from the medical to the non-medical supply of prescription drugs poses a challenge for drug re-legalization via licensing. If the relatively stringent control of pharmaceutical drugs is so porous in practice, how will looser controls associated with the licensing of cocaine or opium fare? Does the significant abuse of prescription drugs imply that licensing does not work?

Surely the situation with prescription drugs is evidence that licensing aimed at preventing adult recreational use does not work all that well. But the buyer licensing proposed here aims to provide a legal means for adults to use drugs recreationally; it concerns facilitating, not preventing, responsible recreational drug use.[iv] It also is designed to identify those whose drug use is potentially problematic, and to focus treatment efforts on them. Transferring the bulk of recreational consumers to a regulated market will reduce the residual demand in the illicit secondary market. Further, enforcement resources will be targeted not at drug possession and use, but rather at diversion to unlicensed individuals.

5.6 Advertising

In the US, advertising for an illegal good can be suppressed without running afoul of Constitutional protections for speech. For legal goods, however, the situation is more complicated, and it is possible that advertising bans on re-legalized drugs will not be tenable.[v] For this reason, I think that there is something to be said for state-monopoly suppliers (as opposed to privately owned, state licensed retail outlets). (Fosdick and Scott likewise preferred this option, and for similar reasons – concerns that commercial interests would become too powerful if alcohol were to be sold by private businesses. Transform also supports an advertising ban for re-legalized drugs.) State monopoly suppliers can credibly commit to limited advertising, in ways that do not implicate free speech concerns. As with all of the suggestions made here for drug re-legalization, experience will indicate, over time, sets of regulations that work reasonably well.

[i] Hawken (2010).

[ii] On 24/7 Sobriety, see Caulkins and Dupont (2010), Long (2009), and “RAND Corp To Study South Dakota 24/7 Sobriety Program,” Dakota Voice, February 10, 2011, at

[iii] See Table G.4 and Table 5.2B in Substance Abuse and Mental Health Services Administration (2009); tables available at and, respectively.

[iv] Incidentally, the distinction between medical and recreational use of drugs is not sufficiently bright, to my mind, to serve as the basis for the legality or illegality of the use of a drug; see Husak (2002, pp. 37-43). One of the benefits of drug re-legalization is that the distinction between medical and recreational use takes on much less importance.

[v] See Central Hudson Gas & Electric Corporation v. Public Service Commission of New York, 447 U.S. 557 (1980), and Leitzel (2008, pp. 172-174).

Draft Four, Sections 5.1, 5.2, and 5.3

5. A Closer Look at a Small Subset of Complicated Issues

5.1 Sharing and Social Norms

Many people consume their drugs in social settings, as part of a group, and they often share drugs in these situations. Does this reality imply that exclusions or buyer licensing will prove futile? After all, even if these regulations succeed in keeping unentitled individuals away from purchasing drugs, those individuals can still use drugs provided by their legally-entitled friends.

This is a situation familiar to young adults in the United States, who are not allowed to purchase alcohol until the age of 21. Many older teens do drink, however, often alcohol purchased by slightly older friends. As with alcohol, transferring legalized drugs to an underage or unentitled person will be illegal, despite the barriers to enforcement.

Quantity controls can limit the extent to which any one licensed individual can undermine the intent of the overall program – an option currently unavailable for alcohol. Loss of a drug license is an additional sanction that can be used to dissuade illicit sharing. For drugs that often are consumed in social settings, restricting use to licensed clubs – with the clubs only available to licensed members – also can help. The possibility that drug packaging and the drugs themselves can be labeled in a traceable manner – using so called microtaggents – might also ease enforcement.[i] Nevertheless, leakage from legal users to unlicensed individuals in their social sphere will surely occur; enforcement resources are probably best aimed at diversion to teenagers and at profit-motivated secondary distribution channels.

5.2 The Highly-Skewed Distribution of Drug Consumption

Most adults do not use the currently illicit drugs. For those who do use such drugs, most are not heavy consumers. But those who are heavy consumers tend to consume the majority of the drugs, and also are responsible for many or most of the social costs associated with drug acquisition and use.

Heavy drug users typically enjoy easy access to their drug of choice. An attempt to replace prohibition with a regime that does not provide such users with similarly easy access is a recipe for maintaining illicit supply channels and the violence and corruption that accompany them. Taxes on legal sales, therefore, cannot be exorbitant, if the legal market is to out-compete the black market. Nor can license requirements be overly onerous. Pharmaceutical grade drugs of a known potency, however, typically are preferable to street drugs, so the effective price of drugs does not have to decline substantially for the legal market to dominate the black market. Further, with a large chunk of sales moved into the legal market, enforcement resources can be targeted at the relatively small number of illicit sales, driving up prices on the residual market.

The importance of heavy users to sustaining a black market implies that displacing their demand into licit channels is a key, perhaps essential component of combating illegal drug sellers. Further, their drug consumption and its related problems are sufficiently different from casual or moderate users that a separate channel of legal supply is probably appropriate, something akin to the maintenance regimes some jurisdictions have in place for opiate users. A treatment component or connection, desirable for exclusion and licensing regimes, takes on central prominence for the very heaviest users.

5.3 The Privacy of Licensee or Exclusion Lists

One of the more difficult issues with drug licensing is the extent to which the identities (and perhaps chosen limits) of license holders will be made available publicly, and whether it will be legal for employers or insurance companies to discriminate on the basis of holding a heroin license. There is something to be said for a conservative implementation of a licensing plan when drugs initially are re-legalized, so perhaps disclosure and legal discrimination might be useful as transitional devices. Nevertheless, public disclosure of the names of license holders and legal discrimination on the part of any employer against licensed consumers could easily prove so onerous that users would shun the licensing system for the black market.[ii] Therefore, I do not think these measures would be appropriate, even as temporary elements of a cautious plan for ending prohibition. Sensitive jobs, however, might be exempted: employers for these positions would then be allowed to screen employees on the basis of licenses (or perhaps by conducting drug tests, as is now done).

[i] Transform (2009) suggests the use of microtaggants to aid the enforcement of personal drug licenses. See also “Tuning in to Taggants,” by Paul Thomas,, 2006, available at

[ii] Colorado and some other states in the US that have legalized marijuana for medical purposes employ confidential registries of patients’ names; see

Draft Four, Section 4.2

4.2 Licensing Drug Users

Prohibition implies that almost every adult is unlicensed to purchase the currently illegal drugs. (Exceptions include those people who receive controlled substances via medical prescription.) The general prohibition sometimes is complemented by a further exclusion element enforced via drug testing. South Dakota’s 24/7 Sobriety Program, for instance, extends to illegal drugs for some participants. These individuals submit to frequent drug testing, either with regular on-site testing or through the use of “sweat patches,” which are worn for a week or more and then tested for evidence of drug use.[i] The state of Hawaii’s HOPE program (Hawaii’s Opportunity Probation with Enforcement) combines frequent drug testing of probationers with swift incarceration for brief periods in the event of failed or missed tests.[ii]

The legalization of the currently illegal drugs can involve buyer licensing, where the conditions under which a license is granted and maintained can be tailored to the user, the specific characteristics of the drug, and the circumstances under which it is to be consumed: drug, set, and setting, in Norman Zinberg’s celebrated triad.[iii] Consider heroin, for instance. First, it is likely that under a legal regime, the vast majority of opiate users will shun heroin and consume opium in relatively dilute preparations, perhaps in the form of tea or other beverages. Nevertheless, heroin can be made available, both for maintenance of existing addicts and for use by non-addicts. Maintenance provision of heroin (or a heroin substitute such as methadone) to addicts might be conducted along the lines of current maintenance programs in some European countries – this approach is sensible for addicts in either a prohibition or legal regime.[iv] For non-addicts, what would a workable licensing system look like?[v]

First, legal sellers of heroin will themselves be strictly controlled, either through a state monopoly regime or through a licensed dispenser model: the two alternatives that Fosdick and Scott recommended for implementing alcohol legalization. When someone turns 21 years of age, he or she would be eligible to apply for a heroin license. This could be a stand-alone license, but given the wide variety of opiates and opioids available, it might be more sensible to offer a general opiate license, with sub-categories, including one for heroin, which opiate consumers could choose to opt into. The applicant for the heroin subcategory would have to pass a short exam indicating that he or she understands the potential dangers of using heroin in different delivery modes, the steps to take in case of an accidental overdose, and the regulations concerning public use, drugged driving, and transfers to unlicensed individuals. A brief discussion with a counselor also will be mandated, aimed at determining if the user is experiencing problems (or has previously experienced problems) with her drug use, with treatment options explained and made readily available. The heroin license would last for one or two years, and renewals would necessitate further discussion, to see how the user is coping with his or her drug use.

A person who passes the exam must then choose limits on how much heroin he or she can buy in a day, a week, a month, and a year. These limits will be enforced by the seller, and an electronic data base will be necessary so that multiple dispensaries can access and update the same information. [There will be regulatory limits imposed with an eye to restricting diffusion into the black market, but those limits should not impair an ardent user from receiving the (substantial) doses that she desires. The choice of personal limits is required for license holders to help them implement their own plan for controlled usage. Recall that heavily addicted users would presumably operate under a different type of maintenance regime.] A user also might specify a mode of ingestion, so that appropriate doses can be provided; for instance, an injector can receive pre-dosed syringes. (The license might only extend to use in controlled locales, and perhaps the heroin will only be dispensed in such locales, with ingestion taking place on-site and monitored; in general I think such a rule would be too constricting a regime to place on all users, though it might be a viable option for license holders to select.)

There would be a small fee for the license, with harder forms of opium products or administration facing higher fees, perhaps. Purchases of the harder drugs might also be taxed at higher levels than purchases of the less potent varieties, on a per-unit basis. Pricing (tax inclusive) can be increasing in quantities, so that the first units of drugs each month, say, cost less than additional units. (Perhaps license holders could choose among pricing options. For instance, one price alternative would involve a constant per-unit price, whereas a second alternative would escalate charges, with lower prices for initial units but increasingly higher prices for later units.) Purchases themselves might be required to be arranged in advance, perhaps by two or three days, to restrict impulsive consumption. As with personal purchase limits and pricing schedules, license applicants also could choose advance purchase periods, no shorter than the legally-mandated delays but potentially longer than those required by statute.

Along with pricing, other elements of the licensing regime might try to prod consumers toward less potent drug forms and controlled use. New users might be subject to rather small regulatory limits on quantities. Experience with an opium license might even be required before a heroin license could be issued; such a policy would mimic graduated systems for licensing drivers, which have been shown to be effective in significantly reducing car crashes for teen drivers.[vi] For existing heavy users, however, delays and low dosages could lead to continued stimulation of the black market; for this reason, the process of obtaining a license, and its expense, cannot be too onerous. (Restriction of the black market is one important rationale for a separate maintenance system for full-blown addicts, too.)

People who are concerned about their excessive drug use (including alcohol) often intend to cut back or quit in the future, but continue as heavy users in the here and now. The regulatory regime can offer some assistance to individuals in implementing their plans for forthcoming temperance, by giving them the opportunity to commit to reduced purchase limits in the future.[vii] When a person applies for a heroin license, for example, and chooses her daily, weekly, monthly and annual limits, she might also be able to choose reductions in those limits for future years. For instance, she might agree that the licensing authority must decrease her limits by ten percent (of the original limits) per year, starting next year and continuing over the following decade, when her limit will shrink to zero. Of course, she might commit to zero consumption today by not applying for a license, but many people are not willing to give up their drug of choice entirely in the near term, even if they recognize that their relationship with the drug is damaging. Zero needn’t be the end point of such commitments, of course – a similar pace of ratcheting down could be scheduled for only five years, leaving the user with limits that are half of today’s. A currently unlicensed user might want to secure her abstinence still further. To this end, a licensing system can be complemented with a self-exclusion program, where the exclusion would bar the possibility of receiving a license for some period in the future. Licensed users could themselves adopt enforceable “no buy” phases, with their purchase rights temporarily suspended for time periods beyond those called for by the advance purchase requirements.

[i] Sweat patch requirements are controversial; see the Drug Policy Alliance’s “Drug Testing Technologies: Sweat Patch,” at

[ii] See, and section 5.4 below.

[iii] Zinberg (1984).

[iv] See section 5.2 below.

[v] Chapter 3 in Leitzel (2008) details my views on the principles underlying desirable vice controls. The licensing regimes discussed in this paper are consistent with those views. I am not suggesting that the regimes here are optimal, only (potentially) workable – indeed, I might not support all of the elements of the regime that I propose. Presumably different regimes would be tried in different places, and regulations would evolve over time as shortcomings revealed themselves and successes were replicated.

[vi] See the discussion and sources in Hemenway (2009, p. 12).

[vii] The “Save More Tomorrow” plan, which involves a formal but not binding commitment to, well, save more at a future date, has shown success in increasing savings; see Thaler and Sunstein (2008, pp. 112-115).

Draft Four, Section 4.1

4. The Alcohol Model, Plus Consumer Licensing

4.1 From Opting Out to Opting In

The default setting – the action that takes place in the absence of any positive step to override the default – can be very powerful in influencing decision-making.[i] In particular, many people will accept the default, despite having no strong commitment to the decision that the default implements, and even if overriding the default involves minimal effort. The choice of a default setting has been shown to hold significant implications for decisions in a wide variety of settings, from retirement plans to registering to be an organ donor.

For legal, non-prescription drugs such as nicotine and alcohol, the standard default rule is adult access to unlimited quantities. (The access is not entirely unhindered, however, in that restrictions of sales to licensed premises and special taxes present departures from the conditions under which regular, non-vicious goods can be acquired.) In the United States, for instance, you are not allowed to purchase alcohol if you are under 21 years of age. On your 21st birthday and dates thereafter, you are allowed to purchase as much alcohol as you or anyone else could possibly want (though on-site drinking establishments are not supposed to serve intoxicated customers). You don’t have to opt-in to being a legal alcohol purchaser – you only have to be sufficiently old.

An alternative regime can be envisioned, one that offers protection for people who are concerned that they might drink to excess. When you reach 21 years of age, you do not automatically qualify to purchase alcohol. Rather, you qualify for the opportunity to opt-in, for the opportunity to acquire a license that will allow you to buy alcoholic beverages.[ii] Alcohol sellers (who themselves are licensed) must check each purchaser’s “alcohol license,” just as they now verify the age minimum.

The introduction of drinking licenses provides new tools for alcohol control. People who misbehave under the influence can have their alcohol purchasing (and perhaps alcohol drinking) privileges rescinded. Consumers who fear the possibility of excessive drinking could bolster their self-control by choosing not to acquire a license, or by becoming licensed but specifying a legally-enforceable limit on the amount of alcohol they can purchase in a day, week, month, or year. Licenses could be tied to price regulations or taxes in such a way that prices charged to buyers could mount as the extent of purchasing goes up – the second bottle of spirits during a week could cost more than the first, for instance. The strictness of the overall alcohol regulatory regime can be adjusted by tweaking the ease of acquiring a license or the circumstances in which a license could be revoked.

It is possible, therefore, to use buyer licensing to convert alcohol purchasing from an opt-out default to an opt-in default. One of the difficulties of the current opt-out system, for those who face alcohol self-control issues, is that maintaining abstinence or temperate consumption requires near continuous re-iteration of the decision to opt out (at least in the absence of an exclusion system as detailed in section 3).[iii] A buyer’s licensing system might make it much easier to pre-commit to avoiding or restricting drinking – buyer licensing subsumes a self-exclusion system, as failure to obtain a license results in ineligibility to purchase alcohol. Assuming that the process and cost of acquiring an alcohol buyer’s license are not very onerous, this pre-commitment option can be provided with little imposition upon those who place no value on limiting their alcohol purchases.[iv]

Licensing imposes upon all would-be consumers, whereas self-exclusion (at least in some enforcement modalities) exempts the bulk of consumers from interacting with the regulatory regime. This disadvantage of licensing is offset to some extent by the possibility that at the time of licensing, all consumers can be assessed for alcohol-related problems, and put into contact with counseling and treatment options. Alcohol difficulties, therefore, might be identified and addressed before they escalate to the point where the drinker would choose exclusion. Further, shortfalls (including unavoidable shortfalls) in the enforcement of exclusion might militate in favor of a buyer licensing system that offers more enforcement options.

[i] Thaler and Sunstein (2009, pp. 83-87).

[ii] Kleiman (2007) and Leitzel (2008, pp. 163-165).

[iii] Kleiman (1992, pp. 98-101) distinguishes between positive and negative licenses. A positive license is one in which the default condition is unlicensed, where a would-be consumer must take some positive steps to opt in, to become licensed. A negative license shifts the default, so that adults automatically have the right to purchase alcohol, say, but that right can be revoked in the face of alcohol-related misbehavior or for other reasons (perhaps even voluntarily, as with self-exclusion).

[iv] Policies that assist people who might be less than rational in their drug related choices, while not imposing in a significant way upon other users, meet the criteria of vice policy robustness, libertarian paternalism, asymmetric paternalism, or related concepts. See Leitzel (2008), O’Donoghue and Rabin (2003), Thaler and Sunstein (2003, 2008), Sunstein and Thaler (2003), Camerer et al. (2003), and Loewenstein, O’Donoghue, and Rabin (2003).

Draft Four, Section 3.4

3.4 Drug self-exclusion

As with mandated exclusion, self-exclusion can be employed with the currently illegal drugs as part of a legal regulatory system. To be specific, imagine a self-exclusion system with respect to legal marijuana. Upon reaching the age of majority, all individuals would be eligible to purchase and consume (perhaps limited) quantities of marijuana. Users who are concerned that their marijuana use has become habitual and is detracting from their lives could choose to put themselves on the list of ineligible buyers. Presumably the method of enforcing the voluntary bans would involve all marijuana customers producing identification, which then would be compared with the names on the exclusion list.

The detailed provisions of exclusion that arise in the instance of gambling exclusions have precise parallels with drug exclusions. How long should an exclusion last? When an exclusion expires, is any positive step required for reinstatement of the privilege to purchase marijuana? Would it be sensible to allow enforceable self-limits, partial exclusions that permit the purchase of restricted quantities of marijuana? Could family members initiate an intervention that could result in an enforceable exclusion? Different jurisdictions could experiment with different exclusion options, and a set of best practices could emerge: a dynamic that is ongoing in the case of gambling self-exclusion programs.[i]

An alcohol-style regulatory system, enhanced with both mandatory and voluntary exclusion mechanisms, might be a good option for the legalization of relatively soft or safe drugs like marijuana. For harder drugs, however, a more stringent regulatory system might be a better route for legal provision, at least in the initial post-prohibition situation.

[i] Responsible Gambling Council (2008).

Draft Four, Section 3.3

3.3 Gambling self-exclusion

Gambling jurisdictions around the globe have demonstrated the viability (and popularity) of a commitment system that can aid voluntary abstainers and recruit sellers into the enforcement mechanism: self-exclusion programs. In the US, adults are eligible to gamble by default: no positive steps need be taken to acquire the privilege to gamble. Nor are time or money limits generally imposed. Self-exclusion programs, however, allow individuals not only to opt out, but to do so in a manner that is binding for some time in the future. A person can voluntarily place himself on a self-exclusion list, and then he will not be permitted into casinos, nor keep any winnings if he evades the exclusion and manages to gamble anyway. Many jurisdictions allow trespassing charges to be brought against gamblers who attempt to enter a casino from which they have excluded. For people who face severe self-control problems, a self-exclusion program can be a significant boon.

Self-exclusion presents a physical barrier to entry into a casino, and by confiscating winnings, renders less alluring any gambling that takes place in defiance of the ban. Self-excluders are removed from any promotional or frequent bettor plans, too, in an effort to limit temptations.

Without self-exclusion (or some sort of licensing of gamblers), opting out of gambling requires those ongoing decisions to desist. The willpower to continue to abstain often proves wanting in those whose rationality with respect to wagering is most suspect. With a self-exclusion program, however, the opt-out can occur at any moment, and maintain a degree of enforceability for long periods of time. The new, voluntarily chosen default economizes on willpower, and has shown itself to be a valuable tool in curbing pathological gambling and the problems induced by pathological gambling.[i]

Gambling self-exclusion programs differ across a wide variety of dimensions, such as the duration of an exclusion. Most venues offer multiple choices for exclusion length, from six months, say, to five years. Other locales offer only one exclusion length; the state of Missouri provides a single option, though it is of variable duration: self-excluders sign up for a lifetime ban from the state’s thirteen casinos. Another important issue concerns what happens when a self-exclusion order (short of a lifetime ban) comes to an end. In some cases, the person can immediately resume gambling. In Illinois, however, reinstatement requires not only that the period of exclusion be expired, but also that a counseling professional certify that the excluded individual is now capable of controlled gambling.

One method of enforcing exclusions is to try to visually recognize excluded gamblers and to remove them from the gambling premises. The Canadian province of Ontario currently is implementing a computer and camera-based face-recognition system to help enforce exclusion orders.[ii] An arguably more thorough enforcement regime involves requiring identification from all would-be gamblers, though such a regime would impose somewhat on the non-excluded. (Identification checks for relatively youthful gamblers might be necessitated by age restrictions in any case.[iii]) The collection of large winnings from electronic gaming machines presents another enforcement opportunity. Jackpots in modern slots are not paid out directly by the machines; instead, receipts for credits are delivered to the lucky player. US federal tax laws necessitate that forms be filled out before winnings of $1,200 or more are disbursed; identification is part of this process, so casinos can check the names of winners against the excluded list. As noted, jackpots will be confiscated from excluded gamblers, with the proceeds in many jurisdictions earmarked for treatment of problem gaming. Electronic tracking devices (sometimes employed for casino loyalty programs) can be used to enforce exclusions, too; furthermore, they can help implement partial exclusions, whereby gamblers set time or loss limits on their betting. Australia is considering mandating that all players of electronic gaming machines establish monetary limits before gambling.[iv]

The exclusion system can be extended to involuntarily excluded gamblers.[v] Casinos do this on their own, barring individuals because of, for instance, prior cheating or criminal activity. But the more interesting extension concerns individuals who, like typical self-excluders, are problem gamblers or at-risk of becoming problem gamblers. For these people, family members might be empowered to initiate an intervention, one that could result in the involuntary exclusion of a gambler from betting locales. Singapore and some jurisdictions within Australia already make provision for such family exclusions; Singapore also automatically excludes from its casinos residents who are receiving public assistance or who are in bankruptcy.[vi]

In between voluntary and involuntary exclusions are exclusions that result from some active prodding. Gambling providers can track bettor behavior, and have personnel discreetly contact someone whose frequency or pace of wagering suggests potential problems. The casino employees (or officials empowered by the regulatory agency) might suggest counseling or self-exclusion, and provide information about these and other options. Casinos in the Netherlands check, upon entry, the identification of all gamblers. These ID checks facilitate the enforcement of exclusion orders, but they also are used to track the frequency of entry; gamblers who enter a casino more than eight times in a month will be approached by casino personnel to discuss the possibility of control issues and self-exclusion.

[i] See the relevant posts at the blog Self-exclusion, at


[iii] The Canadian province of British Columbia is setting up electronic identification checks in casinos in response to an investigative media report on underage gambling. There are plans to link the identification checks with the self-exclusion database; see


[v] The state of Missouri’s casino self-exclusion program developed when a problem gambler learned of the existence of mandatory exclusions, and asked if he could become excluded, too; see


Draft Four, Sections 3.1 and 3.2

3. The Alcohol Model, Plus Exclusion

Alcohol control offers one known model upon which to build for the regulation of legal, psychoactive drugs. Standard approaches to alcohol control start with a ban on purchases by minors. Sellers are licensed by the state, or the state maintains its own retail stores – these are the two options closely examined by Fosdick and Scott. The sellers operate under a panoply of rules governing location, opening hours, advertising, and so on, and these controls differ depending on whether the sales are for on-premises consumption or for carry-out. Special excise taxes are applied to alcohol; these taxes typically are calibrated in such a way as to disfavor relatively potent (distilled) alcoholic beverages.

3.1 Mandatory Exclusion

The alcohol regulatory model probably is too permissive to serve as a comprehensive template for a system of control aimed at the currently illegal drugs, at least for the harder drugs within the prohibited class. The model can be augmented, however, by continuing prohibitions on users whose consumption is particularly problematic: mandatory exclusions. Adults can lose, for cause, their privilege to purchase or consume the otherwise legal drug. In the case of alcohol, some of the infractions that would lead to a revocation of purchasing privileges would be transfer of liquor to an underaged or otherwise excluded consumer, violent behavior under the influence, drunk driving, and use of alcohol in forbidden areas (such as drinking in public, for instance). Loss of a purchase privilege (for a limited time, or, in severe circumstances, permanently) might not exhaust the penalties imposed for these infractions, of course.

How can such an individual-specific exclusion be enforced? One possibility is that all potential alcohol buyers would have to be screened at the time of purchase, to ensure they had not lost their purchasing privilege. A version of this approach currently is applied to enforce the minimum age requirement: all sufficiently young-looking individuals are supposed to be asked for proof of age when attempting to purchase alcohol. An alternative (or complementary) exclusion enforcement method involves frequent testing for alcohol consumption. One advantage to this type of enforcement is that only the proscribed individuals need be involved, by submitting to the requisite tests. Everyone else, including sellers and entitled adults, can take no notice of the banned users.

Exclusion from the privilege to purchase or consume traditionally has not played a significant role in the regulation of alcohol within the US: convicted drunk drivers might lose their driver’s license, but not their drinker’s “license.”[i] Variations on the theme of alcohol exclusion currently are spreading, however. People who are in legal trouble connected with alcohol can be monitored to ensure that they are not drinking. Sometimes people adopt these measures voluntarily, though perhaps also with an eye to benefiting in ongoing legal proceedings. In other instances, courts impose alcohol monitoring as part of pretrial release or probation. South Dakota’s 24/7 Sobriety Program, for instance, aimed at repeat drunk driving offenders, seeks to keep participants away from drinking.[ii] People enrolled in the program either check in twice a day for an alcohol breath test, or wear an electronic alcohol monitor. These monitors, which can come in the form of small ankle bracelets, allow frequent testing without the necessity to travel to a testing location; they have been adopted by many court systems in the US.[iii]

3.2 Committing to Exclusion[iv]

Enforcement via testing is a key component of court-supervised exclusions – people who drink or take drugs in violation of the order face negative consequences imposed from without. People who voluntarily choose to be abstinent from alcohol or drugs do not face the same degree of testing and enforcement – though the illegality of some drugs typically does raise a barrier to use, relative to what would exist in a legal, lightly regulated market.

Voluntary abstainers might try to enforce their commitment. They might make a public pronouncement of their abstinence intention, so that any deviations from their plan might be noted by others and cause embarrassment. They might enlist their friends and family into serving as watchdogs and supporters of their pledge. They have the option of setting up enforceable contracts that would result in “fines” or other negative consequences if they stray from their declared path: see Even with these measures, the voluntarily abstemious generally lack the participation of sellers in preventing their alcohol or drug consumption. Indeed, sellers presumably have a pecuniary incentive to overcome the reluctance of wavering abstainers.

[i] Mandated ignition interlock devices are a sort of targeted driving exclusion, temporarily suspending the driving (though not the drinking) privileges of someone who fails the requisite in-car sobriety test.

[ii] See, and section 5.4, below.

[iii] For the website of one leading alcohol monitoring company, see

[iv] The remainder of section 3 draws upon Leitzel (2011). The notion of commitment is central to the current paper, which involves, in part, a sort of self-referential experiment; see A presentation related to this paper is available at