Certain substances are harmful to health but, at least after a while, become very attractive to their users. We have banned some of these and allowed others to be sold, subject to regulation and/or taxation.

Our current mix of prohibition and regulation is largely determined by historic factors. Certainly there is no principled defense for simultaneously allowing alcohol (indeed, even the promotion of that addictive and life-threatening substance) and prohibiting marijuana. It is simply a consequence of the almost accidental evolution of our society's habits and we should not be surprised to find other societies (predominantly Muslim) in which the opposite pattern is found.

But if there is no principled defense there is a pragmatic one. We should at least contemplate the prohibition of all seriously health-threatening substances that are not already in mass use. Reversing history by law, as was done during Prohibition, may simply pose too great a risk to the social consensus necessary for continuation of the nation. Reversing it by deliberately attempting to alter tastes seems much more acceptable, though it still remains to be seen whether that can be done on a large-scale basis.

But that of course begs the question as to when the threat posed by the substance is so great and the extent of use sufficiently small that we should incur all the costs of creating illegal markets in order to prevent more use. Our knowledge and our use patterns change in ways that can affect the decision. We have now learned enough about the consequences of cigarette use that a prohibition on its sale could be justified. On the other hand, marijuana use has expanded so rapidly in the last generation that one might reasonably ask whether discrimination between the legal status of alcohol and marijuana use can be maintained.

Debates about changing the mix of prohibition and legal availability are divisive and rarely conducted. Discussions about drug policy are normally confined then to the appropriate level of enforcement (including the severity of sanctions) of prohibitions and stringency of regulation of those that are permitted. Since most of my own recent research is concerned with the first of these narrower issues (namely, the consequences of various kinds of drug enforcement) rather than the large questions of principle,' I shall focus on the implementation of our current legal policy, just occasionally sniping at the prohibition of marijuana use and the costs of our heroin policy.

Given a prohibition on the use of a drug, what constitutes success for the prohibition? With enough of a police state, we can reduce undesired drug use almost to zero; clearly we do not seriously consider that a desirable situation. Enforcement of prohibitions must not threaten the basic civil liberties of the nation. When drug enforcement agents started battering down the doors of innocent neighbors of suspected drug dealers in the early 1970's,2 there was an immediate backlash against aggressive enforcement.

Beyond that we count among the costs of drug enforcement the direct expense of carrying it out (still a fleabite of public expenditures, at $1.7 billion federally); the creation of criminal incomes and gangs; and the labeling of individuals as criminals simply as the result of their consumption of a prohibited substance. Success for drug enforcement then is a net rather than gross measure; it is the reduction in use of the prohibited substance less the other costs of achieving that reduction.

In the following discussion I largely ignore the costs of enforcing drug prohibitions. Success here is essentially the gross concept, the reduction in drug use from what it might otherwise have been. That is admittedly a very partial way of analyzing the problem. My defenses are the standard ones; I lack the time to deal with the larger issue and almost everyone else contents themselves with a partial (though usually different) analysis as well.

The starting point of all the symposium's contributors, with the notable exception of Leon Kellner, the one government official in the group, is that the War on Drugs (hereafter WOD) is at an impasse. That in fact is a great simplification, leading many to believe that law enforcement (the central component of WOD) has had, and can have, little impact. Its success has been highly variable.

WOD has been quite successful at restricting the spread of heroin, and may have had substantial impact on some varieties of synthetic drugs, such as methaqualone. It is currently failing to prevent increased use of cocaine and PCP, which may turn out to be the most dangerous drugs yet in popular demand. We will do a better job of deciding to what extent major changes in policy deserve serious consideration if we analyze WOD's successes and failures.

This article has three themes. The first is that policy and debate should make clearer distinctions among drugs. The current federal allocation of drug enforcement resources probably overemphasizes the least serious of drugs, marijuana, and fails to give adequate resources to the most effective levels of enforcement against heroin, namely street enforcement. Similarly, there is a good argument for legalizing marijuana and a very good argument for keeping cocaine illegal. Heroin presents the most serious analytical problem; recent history provides good evidence for both sides of the legalization debate, though political rhetoric has ensured that this debate is very muted.

Second, the current debate, with its resurrection of "demand side" measures (rhetorically at least), is largely irrelevant to drug use in the next five years. Current patterns of use are not likely to be changed by prevention/treatment efforts launched today. Treatment currently has significant impact on the use of only one drug, heroin. Though there are numerous patients supposedly in treatment for marijuana, the evidence is that they are in fact being treated for abuse of other drugs.3 For cocaine and PCP, the treatment literature provides little basis for optimism.s•'

Prevention is almost a complete gamble. No one has demonstrated that prevention programs have any significant impact on drug use. It is certainly a sensible gamble to invest more in prevention than has been done previously, but we should not expect to see any significant impact on usage for many years, even if good prevention programs are developed.

But the third argument provides some reassurance after this pessimism. Drug use patterns change quite rapidly for reasons that we understand poorly but which appear not to be dominated by government policy. We should not assume that current growth patterns will continue.

I. Distinguishing Drugs

Current rhetoric blurs distinctions among illegal drugs. Indeed the Reagan administration has been extremely explicit about this; the difference between "soft" drugs like marijuana (perhaps exclusively marijuana) and hard drugs like heroin is mostly timing. In other words, marijuana is simply the first drug on the path to use of much more dangerous drugs and that is a justification for taking enforcement against marijuana use and trafficking seriously. In fact the justifications for intense enforcement of prohibitions are very different for marijuana and the other two drugs we shall consider, cocaine and heroin."

A. Marijuana

The President's Commission on Organized Crime recently recommended that the federal government take active steps to reverse the permissive legislation of the 1970's with respect to marijuana use. Eleven states passed legislation decriminalizing the possession of small amounts of marijuana;* all of them still allow for civil fines, comparable to parking tickets. This decriminalization was probably the work of middle class parents, outraged that their children could acquire a criminal record for consuming a drug that seemed to be no worse than the alcohol they then drank, in pre-MADD days, with so little conscience. With 400,000 marijuana possession arrests each year, such stigmatization seemed a serious issue.

It is not only scolds such as the President's Commission that have turned against decriminalization. In the last decade, the percentage reporting in Gallup polls that they favor criminal penalties for possession of small amounts of marijuana for personal use has risen for all age cohorts; the most marked increase came from individuals age eighteen to twenty-nine, where the percentage favoring criminal penalties rose from 29 percent in 1977 to 44 percent in 1985.5 One would expect these figures to continue to have risen in 1986 in light of recent events, such as the death of Len Bias, increased media coverage, and general WOD rhetoric from highly visible individuals. No state, to my knowledge, has yet reversed a prior decriminalization statute but the threat is undoubtedly there, given the political attraction of anti-drug stands nowadays.

One can argue that this reversal of attitudes is an extension of the growing American concern with personal health, rather than with a decreasing tolerance for differences in taste. The genie of alcohol has been let out of the bottle (no pun intended), but we can still do something toward stuffing marijuana back in. Given the growing evidence that long periods of heavy use of marijuana raise a variety of health risks,6 that is clearly not an irrational position.

I think, though, that this interpretation is overly generous in its assessment of changes in public attitudes. The reversal is more reasonably attributed to the drumbeat of warnings of a drug "epidemic," the standard term of high level political rhetoric on the subject in the last few years. Indeed, what is somewhat alarming is the ubiquity of this phrase in political speeches during 1986, despite the clear evidence that use of most drugs is actually declining. The result of the rhetoric has been to lessen the willingness of the public to tolerate a policy which makes distinctions among drugs on the basis of their likely harms. The proposition seems to be that if we fail to act harshly against marijuana use (expulsion now being almost the minimum school penalty), then we will face a rising tide of use of more dangerous drugs.

There is little doubt about the etiology of cocaine and heroin use; they are preceded by marijuana use. That does not imply that major reductions in marijuana use will have major impacts on the use of these other drugs, for at least two reasons. First, those most likely to be deterred from marijuana use by heavy enforcement against marijuana are probably the large majority of users who do not go on to the "heavier" drugs; later cocaine users tend to be more committed and less deterrable in their marijuana use. Second, the existing etiology is partly determined by current availability. If marijuana did not exist, we would still see cocaine and heroin use but the path to their use would be different.

For the moment let us ignore the dynamic effects, so to speak, of marijuana use and focus on the harms arising directly from the drug. The harms are a function not only of its pharmacology but also of the modes and intensity of use. We do not have much evidence that marijuana users maintain heavy use for extended periods. Heavy marijuana use seems to be, for most persons, a relatively brief phase. Three times as many high school seniors report having been daily marijuana users in some previous three month period as report currently being in that state.' The heavy user population is very dependent on new recruits because marijuana use is not addictive and young adults move out of the heavy user pool fairly rapidly, probably because they move, after school, into a world that is not so full of "infected" people.

This is not to say that marijuana use is harmless. It only implies that we do not know whether current use patterns are sufficiently extensive and intensive to present significant risks.

The federal agencies have responded to the clarion call for enforcement against marijuana. A very significant share of the rapidly increasing federal enforcement budget now goes to investigating, prosecuting, and incarcerating marijuana dealers, as well as interdicting shipments of foreign marijuana on the way to the U.S. For example, the Coast Guard interdiction budget, which is predominantly for marijuana enforcement, rose from $194 million in 1982 to $326 million in 1986.8

What is the result of this focus on marijuana by federal enforcement agencies? The price of marijuana has gone up and its use has declined, but I shall argue below that the decline in use seems to come from factors other than enforcement. More plausibly the federal marijuana effort, most heavily focused on foreign sources, has helped foster the growth of the domestic marijuana industry. It is tempting to note the irony of this; the free trade oriented Reagan Administration has managed to provide effective protection to the domestic industry to which it is most hostile, marijuana growers.

The domestic industry is probably more adaptive and enforcement resistant than the importing industry. With the spread of production to all fifty states there is no longer a need to assemble large shipments for cross-country sales; instead we probably have lots of small producers selling into local markets. This is just the kind of market against which federal efforts are least likely to be successful since there are no concentrations of power to serve as attractive targets for sophisticated enforcement efforts.

In fairness, it should be pointed out that the administration has tried not to discriminate in favor of domestic growers. The Attorney General has personally launched domestic eradication efforts each year, with highly publicized raids on domestic crops. The predictable result of that effort has been to force growers to adopt more technology intensive (and less obtrusive) growing techniques, which probably increase the potency of marijuana smoked in this country. A University of Mississippi project has monitored the amount of psychoactive ingredient (THC) present in samples of marijuana seized by law enforcement officials. They estimate that the THC level has risen from less than .5 percent in 1973 to close to 4 percent in 1984.9 Indeed, that may be, as Mark Kleiman conjectures," the major result of the marijuana interdiction effort, since foreign producers grow less potent marijuana than do their domestic counterparts.

So for marijuana, as for heroin, enforcement has heightened the health risks of drug users. Users now consume marijuana with higher THC content and are forced to smoke it rolled in cigarette papers, which is the form most dangerous to their lungs. Enforcement against head shops and the effort to make usage more clandestine have reduced the attractiveness of water pipes, though these would reduce the health risks of use, the principal marijuana carcinogens being water soluble.

B. Cocaine

We can now write with a clear conscience about cocaine enforcement. There is no longer any doubt that the drug is seriously addictive for a significant number of regular users, particularly in its newer forms of administration." It poses very serious health hazards to its regular users, quite apart from its effects on their ability to continue normal social and work lives.

The spread of cocaine has been fueled by its declining retail price, disguised in the official figures by lack of data on the rising purity of retailed cocaine. It is also fueled by the availability of these new modes of administration which enable the user to get more of the drug more quickly and hence at lower cost.

Enforcement has been singularly unsuccessful. The explanation is not difficult to find. It is produced in countries relatively close to the United States, which have large numbers of nationals here; that simplifies both transportation (small planes and boats) and distribution to U.S. wholesalers. It is very compact per dosage unit.

At the retail level enforcement prospects look even bleaker. The markets for cocaine are predominantly private. Regular users are not to be found searching in street bazaars, where low quality and high priced drugs are to be found. That is for young suburban neophytes. The heavy user will purchase his drugs in the privacy of his office or his - apartment, quite secure from police surveillance. Though large quantities are seized, they are seized at levels of the market at which their replacement cost is quite modest.

The decline in the price of cocaine since 1981 is a genuine mystery. The enforcement pressure against cocaine, even when one adjusts for growth in the market, has certainly increased in the last five years. The price should be rising, unless there has been a major erosion of some monopoly position in the distribution system or there are hitherto unrealized economies of scale in distribution. Neither of these seems very likely.

Mark Kleiman conjectures that we are observing the results of learning in the importing sector; smugglers have acquired experience that enables them to be more efficient in reducing risk. The same may be true in domestic wholesale markets as well. It may be a one-time phenomenon, with prices soon reaching a new equilibrium level and staying there. But there is little reason to think that cocaine enforcement is likely to be much more successful in the future than it has been in the past.

For cocaine the term impasse may be appropriate. What can we do in the face of it? Certainly there is little to be said for taxed and regulated legal sale; we might indeed reproduce the disaster of our alcohol policies that way. I defer to the final section a discussion of the alternatives before us.

C. Heroin

John Kaplan's title for his book on heroin, The Hardest Drug," provides a nice treble entendre. It is the hardest drug to obtain, the hardest in terms of the damage that our policy wreaks, and the hardest drug for which to make policy. The three are linked.

The central problem of heroin policy is easily summed up. Some significant share of urban crime is related to heroin use;12.' the very high price of heroin, and the particular effects of its use, ensure that regular heroin users must commit numerous crimes in order to support their habit.

Note that we need both conditions. High price alone is not enough. Cocaine was as highly priced as heroin in the early 1970's (the change in relative prices since then is itself an interesting phenomenon), but was never associated with street and property crime as heroin was.

The explanation probably lies in who was using cocaine then and, relatedly, the effects of that use. The fact is that those who can keep their use of cocaine moderate (and that seems to have been true of many regular nasal users) were able to perform adequately in their lives and jobs. Indeed there is alarming evidence that part of cocaine's attraction is precisely the ability that the drug gives some users to improve their job performance in the short run. I note that it was widely reported that George Rogers ran for the National Football League rookie rushing record while regularly using cocaine. Some of the National Basketball Association players expelled for cocaine use were performing near their career bests at the time of their expulsion.

Heroin is a narcotic, cocaine a stimulant. It is no surprise that they differ substantially in terms of the social consequences of their use. Those who use heroin, even in the debased form that it is available to American addicts, are not induced by the drug to perform at their peak. Rather, they are able to escape the cares of life; it is a drug particularly attractive to those near the bottom of legitimate American society, for whom the allowed alternatives are rather bleak.

It has often been noted that the cost of heroin control is absurdly high. The social cost of crime — if one includes (as is proper) the decay of communities — is enormous, and a non-trivial fraction of that (at least in large cities) can be ascribed to heroin addicts. To many that is an argument for the legalization of heroin; to others it suggests the danger of the drug. For the drug itself does not contribute to crime; quite the opposite. It is only the very high price (roughly $2,000 per gram) that leads to addict crime.".2

Regulated legalization, or even less enforcement, would probably reduce the social cost of heroin consumption. But it would dramatically change the distribution of that cost. Inner city minority populations would have greatly expanded heroin user populations, while the rest of the community would probably incur a much lower crime rate. Much of the resistance to the discussion of legalization seems to come from leaders of those communities which would in fact bear the cost of expanded use.

Heroin is drug enforcement's success story. It is absurdly expensive to obtain, and can be bought only in nasty parts of town from very dangerous people; it takes a long time to find the drug. Alas, successful drug enforcement, at least for an addictive drug that appeals to a criminally active population, looks like a socially expensive proposition.

II. Demand Side Measures

In face of the consensus that drug enforcement has failed, there has been a large shift in the rhetoric of WOD and a modest shift in its budget allocations. For the first time in a decade, serious attention is being given to "demand side" programs, prevention and treatment.

But there is little reason for optimism about these programs. We have turned to them mostly as the result of despair about drug enforcement, not because we have proof positive that such programs hold any significant promise to deal with our current problems. I shall refer heavily here to a RAND study which I co-authored.13

The treatment literature makes clear that heroin addicts can be substantially affected by treatment, particularly methadone. One cannot say "cured"; there is growing evidence that heroin addiction is a near permanent state. But those who are in treatment use very much less heroin and commit many fewer crimes. That is a substantial gain both for us and for the addicts. It is important to assure that there are enough treatment positions available for heroin addicts seeking treatment, since one of the major paths by which heroin enforcement benefits the community is through driving addicts to seek treatment.

So far cocaine treatment does not seem to offer much. We know a great deal less about such treatment for two reasons. First, large scale cocaine treatment has been operating only a relatively short period. Second, most cocaine addicts have been treated in private clinics rather than through publicly funded agencies of the kind that treat heroin addicts. Thus we have less access to the experience of such treatment programs.

But what evidence we have is not encouraging. Cocaine addiction has different roots than heroin addiction; a different group of persons is involved and the users seek something different from the drug. As yet there is no cocaine counterpart to methadone — i.e., a substitute drug which reduces the craving for the initial drug and does not produce similar anti-social behavior on the part of the user.

Prevention may well be the "magic bullet" of the 1980's, the supposed panacea that everyone endorses prior to evidence that it works. RAND is involved in a major experiment aimed at testing the efficacy of some innovative cigarette prevention programs in drug prevention. The programs are school-based, offered to children in seventh and eighth grades. They emphasize developing resistance to social pressures, focus on short-term harms from drug use; some also involve older children to deliver the messages."' The results of the experiment will not be available for at least another year. Other large scale demonstrations are also being carried out at this time.

These experiments are the first serious tests of prevention efforts. While school-based prevention programs have been around for at least a decade, there is no evidence to suggest that any of the earlier generation were particularly successful and even a little evidence to suggest that some were counter productive." The newer demonstrations are generally based on sounder behavioral models and make better use of what we know about learning. But there is no denying that at this stage we simply do not know what works.

This is not a mere scholarly quibble; researchers can always be relied on to assert that "too little is known about . . . ." In this case we genuinely do not know. The focus on early socialization, which can take many operational forms, is plausible. I certainly find it so, having to put up with my nine year old son's protests about my (harmless) after dinner cigar. But we might be entirely wrong about this, just as we were wrong a generation ago in the effort to base prevention efforts on frightening (and exaggerated) statements about the consequences of marijuana use.

Certainly the federal commitment of $250 million in fiscal year 1987 for drug prevention is a large gamble. The money is to be distrib-
uted to schools which have few effective models from which to work. They will probably do no harm with the money. Alas, we are also unlikely to know whether they have done any good with it.

III. Of Times and Tastes

One does not have to be entirely pessimistic about drug abuse in this country. Use patterns change quite sharply without new government policies. The decline in reported heavy marijuana use among high school seniors is the most recent example of a positive change that appears to be unrelated to any government initiative. Pessimism about the prospects for government interventions should not be taken to be pessimism about the future use of illicit drugs in America.
In 1978, eleven percent of high school seniors reported daily use of marijuana, almost double the figure found in the first of the surveys in 1972.'5 This rapid increase has never been given a systematic explanation, but roughly speaking it could be seen as an epidemic phenomenon.

In other words, more seniors used marijuana in 1975 than in 1974 because the number in 1974 was higher than in 1973, so that each "uninfected" person was more likely to come into contact with a user in the later year and thus become a user him/herself.
But what then explained the sudden reverse? The downturn was as sudden as the increase. Daily marijuana use had declined by 1983 to almost the same figure as in 1974, about 5.5 percent. Analysis of the survey data suggests that the decline was the result of changes in the youth population's attitudes towards their own health and the health consequences of marijuana use. Certainly the percentage reporting that marijuana was readily available remained high (about 88 percent) throughout the entire period," suggesting that enforcement was not the prime cause of the decline.

One could argue that this was a consequence of government policies, namely the increased emphasis in school curricula on personal health. But it would be stretching matters to call this a form of drug control policy in the late 1970's. The explicit integration of drug education into health curricula, still not completely accepted, did not become a mass phenomenon until the 1980's.

Nor is this the only instance of a rapid downturn in the rate of new infections. Heroin use extended rapidly in the late 1960's and early - 1970's." A rise in price and generally adverse market conditions, the products of policy, accounts for this initial cessation of growth. Heroin became more expensive and of lower quality, the time needed to find a dealer may have significantly increased, and methadone treatment provided for the first time a reasonable alternative for addicted users.

But when market conditions in terms of price and "availability" returned to their 1972 state, there was no further growth. The population of regular users seems to have stagnated with very little new recruitment into the user population. Indeed, what is striking is the evidence that we are dealing with exactly the same population that was being treated in the early 1970's. The proportion of heroin users in New York City treatment centers over the age of 30 was 31 percent in 1977; in 1985, that figure had increased to 65 percent." Lester Grinspoon suggests that the decline of recruitment into heroin use in the last decade is the result of non-users in the same communities as users observing the degradation in the quality of life of users; of course one can argue that the degradation is itself primarily a function of effective enforcement.

The two changes we have discussed represent very different phenomena. The end of the heroin epidemic meant that the pool did not continue to grow but that it did not shrink much either; this is precisely because heavy heroin use appears to be a permanent condition. The figures on the aging of the population entering treatment is very persuasive on this point." The numbers of heavy marijuana users may actually have declined because there has always been substantial turnover in this population.

IV. Conclusion

American drug use has been subject to unexpected shifts over time, shifts that we can only rather weakly relate to drug policy. In light of this we should be very careful about extending present trends into the future. The surge of cocaine use, even with falling prices for the drug, may not continue very long. If it is indeed as addictive in its new forms as the literature suggests, the recruits of the past five years may serve as effective warnings for the potential next generation of users. There is a silver lining to be found in even a phial of crack.

Heroin is clearly a drug in decline, perhaps a testimony to the eventual success of law enforcement, albeit at great social cost. Marijuana has probably found its place in American society as one of the ways in which a significant chunk of the nation's more rebellious adolescents establish their independence from their parents and other adult controllers. It will not make them better but, if I am correct about the short duration of intensive use, it won't leave a very large mark on them.

Some level of illicit drug use will always be with us and we should not measure success of a policy against an unattainable perfection. In that connection it is worth making some clearer connection to Prohibition. For though Prohibition has always been rated one of the great failures of American law, it did in fact accomplish a great deal in terms of its policy goals. The best evidence, which is not too bad, is that it at least halved the consumption of alcohol in the United States.

A comparison of per capita drinking levels in the area immediately prior to Prohibition and that immediately following Repeal shows the latter figure to be only half of the former.2° It seems reasonable to assume that drinking at the end of Prohibition was less than that following Repeal, so that this comparison understates the extent to which Prohibition reduced alcohol consumption.

Despite this success, Repeal came easily. In contrast the debate about drug policy, which is generally seen to have failed just as substantially as Prohibition, is almost non-existent; it is a quibble among politicians about how much more to spend on different elements of the enforcement program. I believe that there are three factors which explain the difference in the response to the failure of drug and alcohol prohibitions.

First, the use of illicit drugs is still concentrated among the young and the poor. Most Prohibition era judges were probably well acquainted with alcohol use prior to 1919. Though Baby Boomers are now moving to judicial and congressional ages, their marijuana use is well behind them. Cocaine use on Wall Street notwithstanding, it seems unlikely that the senior partners of law firms and other contemporary dignitaries (as opposed to celebrities) number cocaine among their standard recreational outlets. They sentence from a distance. Probably a majority of the adult population sees drug use, particularly cocaine use, as someone else's vice.

Second, the corruption surrounding bootlegging was much more systemic and broad-based than that around drug dealing, Miami Vice notwithstanding. We do not worry that the city of Chicago is in the hands of the drug counterparts of Al Capone, if any exist. The privacy of most drug transactions makes it much less necessary to obtain comprehensive protection from law enforcement agencies and political authority.2'

Third, the American people take health much more seriously than they did in the 1920's. The campaign for Prohibition emphasized abandoned children and labor absenteeism rather than cirrhosis of the liver. Though the drug crusaders of today are still moral entrepreneurs, the growing evidence that none of these drugs is good for your health is an important fall-back.

The effort to control drug use in America is, like all moral campaigns, flawed by exaggerations and simplifications on both sides. Some prohibited drugs represent relatively slight dangers to their users in the quantities commonly used; others are dangerous mostly because of the conditions of use that society has created.

But the critics of our current policy who ignore the fact that the War on Drugs has significantly affected the extent of use of substances hazardous to the users' health do their cause no good either. Drug use is a permanent feature of American society. It is not a desirable one. But if we are realistic about what can be accomplished with the various tools that are available to the institutions of social control and we realize that cries of epidemic are counter-productive, a good deal more can be done to lower the costs in money and civil liberties of both drug use and drug enforcement.

* B.A., Univ. of New South Wales, 1966; M. Phil., 1971, Ph.D., 1980, Yale. Mr. Reuter is a Senior Economist at the RAND Corporation in Washington, D.C. The views expressed are solely the responsibility of the author and do not necessarily reflect the opinions of the RAND Corporation or its sponsors.
Valuable research assistance was provided by Patrick J. Murphy.
© Copyright 1987 by Peter Reuter.

1. These questions are admirably addressed in Kleiman, Liberalism and Vice Control, 6 J. POLICY ANALYSIS & MGMT. 242 (1987).

2. E. EPSTEIN, AGENCY OF FEAR (1977).

3. WISH, DEREM & RAINONE, AN OVERVIEW OF PROGRAMS FOR CLIENTS WHO ENTER TREATMENT WITH MARIJUANA AS THE PRIMARY DRUG OF ABUSE (1983) (publication of Narcotic and Drug Research, Inc.).

3.1. Siegel, Cocaine Smoking, 14 J. PSYCHOACTIVE DRUGS 271, 359 (1982); Gorelick, Wilkins, & Wong, Diagnosis and Treatment of Chronic Phencyclidine (PCP) Abuse, in PHENCYCLIDINE: AN UPDATE (D. Clovet ed. 1986) (publication of National Institute on Drug Abuse).

3.2 I focus on these largely because they are the drugs about which most is known.

4. UNITED STATES DEP'T OF JUSTICE, BUREAU OF JUSTICE STATISTICS, SOURCE BOOK OF CRIMINAL JUSTICE STATISTICS (1985) (NCJ-93682).

5. G. GALLUP, THE GALLUP POLL: PUBLIC OPINION 1985 (1986).

6. ADDICTION RESEARCH FOUNDATION, REPORT OF AN ARF/WHO SCIENTIFIC MEETING ON ADVERSE HEALTH AND BEHAVIORAL CONSEQUENCES OF CANNABIS USE (1981); ADVISORY COUNCIL ON THE MISUSE OF DRUGS, REPORT OF THE EXPERT GROUP ON THE EFFECTS OF CANNABIS USE (1982); NATIONAL ACADEMY OF SCIENCES, COMM. ON SUBSTANCE ABUSE AND HABITUAL BEHAVIOR, AN ANALYSIS OF MARIJUANA POLICY (1982).

7. L.D. JOHNSON, DRUG USE AMONG AMERICAN HIGH SCHOOL STUDENTS, COLLEGE STUDENTS, AND OTHER YOUNG ADULTS: NATIONAL TRENDS THROUGH 1985 (1986) (publication of National Institute on Drug Abuse).

8. PRESIDENT'S COMM. ON ORGANIZED CRIME, AMERICA'S HABIT: DRUG ABUSE, DRUG TRAFFICKING, AND ORGANIZED CRIME (1985).

9. UNIVERSITY OF MISSISSIPPI POTENCY MONITORING REPORT (1983 and 1986).

10. Personal communication to the author.

11. R.M. Post, Chronic Cocaine Administration: Sensitization and Kindling Effects, in COCAINE: CLINICAL AND BIOLOGICAL ASPECTS 107-68 (E. Uhlenhuth ed. 1986).

12. J. KAPLAN, THE HARDEST DRUG: HEROIN AND PUBLIC POLICY (1983).

12.1 For a summary of the evidence concerning this see Wish & Johnson, The Impact of a Substance Abuse on Criminal Careers, in CRIMINAL CAREERS AND "CAREER CRIMINALS" 52, 86 (A. Blumstein, J. Cohen, J. Roth & C. Visher ed. 1986).

12.2 The relationship of cocaine use to crime remains relatively unexplored. Some studies have found such a relationship in non-random samples. See, e.g., Collins, Hubbard, & Rachal, Expensive Drug Use and Criminal Income: A Test of Explanatory Hypotheses, 23 CRIMINOLOGY 743, 764 (1985). Crack, which attracts younger users because of its low dosage cost and high potency, may lead to a heroin-like relationship to crime.

13. M.J. POLICH, P. ELLICKSON, P. REUTER, & J. KAHAN, STRATEGIES FOR CONTROLLING ADOLESCENT DRUG USE (1984) (publication of the RAND Corp.).

13.1 P. ELLICKSON, PROJECT ALERT: A SMOKING AND DRUG PREVENTION EXPERIMENT FIRST YEAR PROGRESS REPORT (1984) (publication of the RAND Corp., N2184-CHF); P. ELLICKSON, DESIGNING AN EFFECTIVE PREVENTION PROGRAM: PRINCIPLES UNDERLYING THE RAND SMOKING AND DRUG PREVENTION EXPERIMENT (1984) (publication of the RAND Corp., P-7068).

14. Schaps, A Review of 127 Drug Abuse Prevention Program Evaluations, 11 J. DRUG ISSUES 17 (1981).

15. JOHNSON, supra note 7.

16. Id.

17. DIVISION OF EPIDEMIOLOGY AND STATISTICAL ANALYSIS, EPIDEMIOLOGY OF HEROIN, 1964-1984 (1985) (publication of National Institute on Drug Abuse).

18. COMMUNITY EPIDEMIOLOGY WORK GROUP RECORDINGS, EPIDEMIOLOGY OF DRUG ABUSE: RESEARCH, CLINICAL, AND SOCIAL PERSPECTIVES (1985) (publication of National Institute on Drug Abuse).

19. DIVISION OF EPIDEMIOLOGY AND STATISTICAL ANALYSIS, supra note 17.

20. D. KYVIG, REPEALING NATIONAL PROHIBITION (1979).

21. Reuter, Police Regulation of Gambling: Frustrations of Symbolic Enforcement, THE ANNALS, July 1984, at 36.