Sign the Resolution
Contents | Feedback | Search
DRCNet Home | Join DRCNet
DRCNet Library | Schaffer Library | Special Collections
Milton Friedman on the Drug War
The War We Are Losing
by Milton Friedman
From: Searching for Alternatives: Drug-Control Policy in the United States, pp. 53-67. Edited and with an Introduction by Melvyn B. Krauss and Edward P. Lazear. Stanford, Calif.: Hoover Institution Press, 1991.
After everything that has been said on all sides of this issue, there is little new that is left to be said. I was going to say that one thing on which everybody has agreed is a need for more money for research-especially for the research we ourselves do--but I like to be contrary so I will express a disagreement with that. If on any subject whatsoever we waited until all the research we wanted to do was done, we would never do anything. If we are going to act, we have to act on the basis of the evidence that there is. I do not agree with those like my good friend, Ed Meese, who say that you need a detailed and well-reasoned alternative before you do anything about the present system, that the burden of proof is upon those who want to change the system. If the system is making a mess, it is a good thing to do something to change it even though you may not have a fully detailed alternative.
One thing that needs to be kept in mind is that we are fundamentally all on the same side. We all have the same objectives. We all recognize that drugs are currently doing a great deal of harm. What divides us is our judgment about the best means to minimize the harm done by drugs. We must not let ourselves get diverted from trying to reach reasonable, sensible conclusions by attributing bad motives to those who disagree with us. There is a famous statement, which I have used many times, made by Pierre S. du Pont almost precisely two centuries ago (September 25, 1790) to the National Assembly in revolutionary France in which he said, "Gentlemen, it is a disagreeable custom to which one is too easily led by the harshness of the discussions, to assume evil intentions. It is necessary to be gracious as to intentions; one should believe them good, and apparently they are; but we do not have to be gracious at all to inconsistent logic or to absurd reasoning. Bad logicians have committed more involuntary crimes than bad men have done intentionally.'
I am obviously not going to add any arguments to the large number already presented. I want to use my limited time simply to try to bring a little order out of the discussion and to add a little evidence.
People tend to discuss the issue of drugs on two levels. One level was well described by one of the speakers as Plato versus John Stuart Mill: The philosophical disagreement between Plato's view that it is right for some of us ("philosopher kings") to tell others of us what they must do because it is good for them, and the doctrine of John Stuart Mill that the role of government is simply to prevent people from doing harm to others and that it is not right for government to try to force people to do anything simply for their own good. The philosopher-king perspective and the libertarian perspective, if you will. No doubt there is a wide disagreement on that level, and as many of you know my own sympathies are on the side of John Stuart Mill. That consideration is not decisive in this issue, however, as it is not in many. Nonetheless, it does affect people's attitudes and the way in which they look at things. I think that it is worth recognizing.
Why is it not decisive? Because even the libertarians justify interference to prevent harm to others. In my opinion, the most basic distinction that needs to be kept in mind in this discussion is between innocent victims and self-chosen victims. That has come out again and again in many discussions. As everyone recognizes, self-chosen victims may and do harm others as well. Even if there were no laws against drugs whatsoever, if they were completely legal, there would still be innocent victims. The most obvious, of course, are the crack babies. I don't know how many there are-that is for you medical people to decide-but insofar as there are any they are obviously innocent victims of their mothers. So legalizing drugs would not eliminate all innocent victims. Even a strict libertarian might argue for prohibiting certain drugs, or putting strict limits on them, on the ground that interference with individual behavior is more than offset by the prevention of harm to innocent victims.
That brings the real issue to the second level-the level of expediency. We now have a system to control drugs. Is it working? Is it doing more good or more harm? If it is doing more harm, let's stop doing that harm and let's not wait until we have a fully worked-out, detailed plan for exactly what we are going to put in its place. Let's eliminate those features of it that are clearly and obviously doing the most harm. Again, everybody agrees on this level that the present methods are doing a great deal of harm. Dr. Clarke movingly and effectively presents one of the most important components of that harm (in Chapter 25).
The attempt to enforce the prohibition of the use of drugs is destroying our poorer neighborhoods in city after city, creating a climate that is destructive to the people who live there. This phenomenon is perhaps the greatest disgrace in the United States at the moment. I say "perhaps" because an alternative is what we are doing to other countries-a subject discussed in Chapter 20. Can anybody tell me that the United States of America is justified in destroying Colombia because the United States cannot enforce its own laws? If we enforced our laws, there would be no, problem.
I don't mean to say we could not enforce our laws. In principle, there is no doubt that we could completely eliminate drugs if we were willing to use the methods that Saudi Arabia is willing to use: If we were willing to cut off the hands of a drug offender; if we were willing to impose capital punishment on drug dealers. We are not, and all of us without exception are proud of the fact that we are not willing to use those methods. Those are cures that are clearly worse than the disease. Given that we cannot enforce our own laws, I believe that there is no way to justify behavior by the United States that leads to the destruction of other countries.
We are destroying the poorer neighborhoods in central cities, but at least we are doing that to ourselves. I don't justify it-don't misunderstand me-but I see even less justification for destroying other people's countries. I have asked this question of many people who are in favor of our present policies. I have never had what I regard as a halfway satisfactory answer.
In discussing the issue of drugs, I, like many others, have cited the prohibition of alcohol in the 1920s as an obvious example of the evil effects of prohibition, as does Dr. Morgan (Chapter 24). In response, I have received a good deal of correspondence. Those who object to my conclusions tend to make two arguments in response, and the little bit of data that I would like to add is in response to these arguments.
Everyone recognizes that the prohibition of drugs makes drugs into a profitable illegal activity and creates a class of criminals. However, the proponents of prohibition answer, if you legalized or decriminalized drugs, or in any other way changed the situation, these people would still be criminals; they would just go on to other crimes. Look, they say, what happened after prohibition. You had Al Capone and the gangs, and after prohibition ended they just shifted over to other sectors. Unquestionably, there is some truth in that. The building-up of a criminal class is going to leave a hangover, and the hangover is going to mean more criminality.
How serious is it? I have a graph (Figure 3. 1). The series that goes all the way back to 1910 is the homicide rate. Its scale is on the left. It goes to 1987, which is as far as the data were readily available. From 1910 on, there is an almost explosive growth in the number of homicides. The first part of the explosion is during World War 1, and one phenomenon you observe over and over is that wars tend to lead to a rise in crime. What happened after the end of the war? The homicide rate kept on going up very rapidly and reached a peak in precisely the year in which prohibition was ended, 1933. It then fell drastically, and it stayed down throughout the forties and the fifties, except for a rise during World War II. Since 1933 was also the end of the great contraction, it can be argued that the Great Depression was also a stimulus to crime and to homicide. Throughout the prosperous 1920s, however, homicides per hundred thousand persons were very much higher than throughout the prosperous forties and fifties, let alone in the late thirties which were not so prosperous. I believe that no one who looks at the evidence can doubt that ending prohibition had a significant and prompt effect on the homicide rate.
Homicides started to go up early in the 1960s and rose very rapidly after Nixon introduced his drug war. More recently, the rate has come down a little, but it is still at the same level as in 1933. I believe one can have great confidence that if drugs were decriminalized the homicide rate would fall sharply, most likely back to the level that it maintained throughout the fifties. That is no small matter: A reduction in the homicide rate from its average during the eighties to its average during the fifties would, with our current population, mean a saving in excess of 10,000 lives a year!
As another bit of evidence I have plotted the number of prisoners (per 10,000 population) received into all prisons-federal, state, and local-year by year. Those data, at least in the sources readily available to me, only went back to 1926. From then on, the number of prisoners received went up very sharply until 1931. It then went down, then rose again to 1940, went down sharply during the war, rose thereafter to a peak in 1961 and came down sharply to 1969. From 1970 on, the number of prisoners received rose dramatically, to a level in 1987 more than twice as high as in 1931. The increase in the number of prisoners received coincides with the beginning of Nixon's drug war, and received an additional boost when the Reagan drug war started.
To say the least, those are disheartening figures. Most discussions of innocent victims, including those I have heard here, leave out what I regard as one of the most important classes of innocent victims, those of us who are not protected by the police because the police are too busy trying to do something about drugs and are being corrupted by the drug industry. The destruction of the atmosphere of law enforcement, of the whole climate of law obedience, adds greatly to the list of innocent victims. Personally, I find it hard to see how anyone can deny the enormous importance of the innocent victims who have been produced by making possession of specified drugs and dealing in them a crime. Few persons do deny the importance of such innocent victims. Those who nonetheless defend drug prohibition reply that decriminalizing drugs may well reduce the number of such innocent victims, but the price society pays for that gain will be a large increase in the number of addicts. Again they go back to alcohol and its prohibition for evidence. They claim that the end of alcohol prohibition was followed by a tremendous increase in the fraction of the population consuming alcohol and in the number of alcoholics.
The next two charts from my trusty computer are designed to answer that claim. The first (Figure 3.2) shows the fraction of total consumption expenditures spent on alcoholic beverages. It is available only for legal alcoholic beverages; that is why it starts in 1933. Unfortunately, all estimates of alcohol consumption during the prohibition era are necessarily highly indirect and uncertain, so I have chosen to stick only to the figures for legal beverages. Dr. Morgan refers to some data on consumption during prohibition, and it is clear that consumption did not disappear. Incidentally, among the innocent victims of prohibition are the addicts themselves, because of the factors that Dr. Morgan brings out. In an illegal market, there is bound to be adulteration and impure substances, which shows up in people dying. Indeed, it has always seemed to me that the greatest beneficiaries from the decriminalization of drugs would be the present addicts. They are made to become criminals. They can't ask for help without admitting that they are criminals. The argument in favor of the present method, thus, has to be that if drugs were decriminalized, you would have a vast increase in the number of addicts.
What does our experience after alcohol prohibition tell us? In the first three years, as legal beverages were being substituted for illegal beverages, it is not surprising that the reported percentage of all consumption expenditures spent on alcoholic beverages went up sharply. It peaked in 1937, then went down to 1940, then rose during the war until 1945. Thereafter it went down gradually but persistently.
I am old enough to be a veteran of that period myself. I remember a few months after prohibition had been repealed going to a Swedish restaurant in New York City with a Swedish friend of mine, a fellow graduate student at Columbia. It was a restaurant in which he had been able to buy Aquavit all during prohibition and he tried to get Aquavit for us. I had never tasted the stuff, and he thought that I ought to have that experience. They said, oh no, they couldn't serve it now because they hadn't received their license yet. He talked Swedish to them and finally was able to persuade them to take us back into the kitchen and give us a little taste of Aquavit. Anyone who believes that during prohibition there was any difficulty in getting alcohol in most of the United States should look at the evidence. I wasn't very old and was not much of a drinker but there was no difficulty in finding speakeasies.
To return to the chart, the skeptic may reply, and correctly, that it is a percentage. Total consumption is going up. Perhaps the smaller percentage of a larger total conceals a very large increase in the amount of alcohol consumed.
Figure 3.3 shows the expenditure on alcoholic beverages expressed in constant 1982 prices between the same dates. As you will see, absolute expenditures, like the percentage spent, went up to 1937 and then fell briefly. During the war, expenditures went up sharply, peaking this time in 1946. Expenditures then fell and remained fairly constant during the forties and fifties and then, beginning in 1961, there was a sharp increase in expenditures on alcoholic beverages. For our purposes, however, the important lesson from the chart is that the legalization of alcohol clearly did not stimulate alcoholism. The legalization of alcohol was followed by a plateau in the consumption of alcohol. The kinds of things that many people have talked about as occurring during the sixties produced the sharp increase in expenditures on alcoholic beverages from 1961 to 1980. Since then expenditures have been falling in absolute terms and not only as a percentage of total consumption.
The obvious implication is that if currently illicit drugs were decriminalized and handled exactly the way alcohol is now handled, there is no reason to suppose that there would be a vast increase in the number of addicts. That is by no means a certainty, but every statement that I have seen asserting the contrary is based on pure conjecture and hypothesis. I have seen no hard evidence. The closest to it that I have come across is reference to the opium craze in China. Given the evidence we have-not only from alcohol prohibition but also from Holland, Alaska, and others-the burden of proof, it seems to me, is on those who maintain that there would be a completely unacceptable increase in the number of addicts.
One thing we really do know for certain is that what we are now doing is not working. There is a wider measure of agreement on that proposition than appears on the surface. It is natural for people to exaggerate their differences. It is hard to impress people without overstating one's case. I suspect, for example, that on the issue of marijuana that Dr. Grinspoon addresses so movingly (Chapter 21), few people believe that dealing in marijuana ought to be a capital offense. I suspect that almost everybody would agree that there is no case whatsoever for treating marijuana the way we do.
It seems to me that we ought to recognize the harm that we are now doing, and not let the tyranny of the status quo prevent us from making some changes that can stop the killing in the slums, and ghettos of our cities. We can stop destroying the possibility of a decent family life among the underprivileged in this country. I do not agree with many people who would agree with me on that point about the role that government ought to play in the treatment of addiction. I do not agree either with those who say that the tragedy of the slums is really a social problem, that the underprivileged do not have enough jobs and therefore government has to provide them with jobs. I want to tell those people that government performance is no better in creating jobs and solving other social problems than it is in drug prohibition. just as a very large fraction of our crime is, in my opinion, caused by government measures, a very large fraction of our poverty is caused by government measures. If those of you who have studied the drug situation were to study as carefully the effects of government measures in the areas of welfare, social services, housing, and so on, you would not have any difficulty in recognizing that there is at least a little bit of sense in what I am saying. That is a different subject, however, and we ought to separate those subjects. Let's not draw ideological lines on this issue because, although there is no doubt an ideological element, the expediential considerations are so strong and so overwhelming that it seems to me they really dominate the situation.
Q. I just wanted you to comment on some of the precedent that has been set by the fact that nicotine is legal and a great amount of education has gone into it recently and I believe nicotine consumption is going down on a per capita basis.
Friedman: I know nicotine consumption is going down on a per capita basis just as alcohol consumption is going down on a per capita basis.
Q. And the second part of that is what about the drug enforcement agencies. If we do legalize it, how do we respond to their losing their income or their financing from the government. If we legalize it we have to have something for the enforcement agencies ...
Friedman: Nonsense. Why? The taxpayers are wasting their money now. Why should they continue to waste it?
Q. I am in 100 percent agreement, but how do we convince them not to? How do we turn the funds off?
Friedman: We can't convince them. We cannot get a change in procedure by convincing the people who have a strong interest in continuing the present methods. There is no area in our government in which we can do that. We cannot get rid of tariffs by trying to convince the people who are being Protected by tariffs. We can't get rid of farm subsidies by persuading the farmers. It's the rest of us as citizens who have to be convinced. I don't believe that maintaining a full-employment policy for bureaucrats ought to deter us from taking the right measures.
Q. (Kildare Clarke) The question you state that it is more of a medical problem, which I do agree with because if we legalize or decriminalize drugs, even if we increased the number of addicts, it is a medical problem. Right now as a physician I think God has given me the privilege to cure someone who is an addict, but he has not given me the power to raise Lazarus from the dead. When a man is dead with a gunshot wound to his head ...
Friedman: I couldn't agree with you more.
Q. (Kildare Clarke) Therefore, based on that factor, would you comment a little about whether we should be looking at this as strictly a medical model as I state in my paper?
Friedman: I am not sure what that means and I don't think we want to draw hard and fast lines. For example, I believe that you would agree that we should make the sale of all drugs to youngsters under the age of 18 illegal, as we now do for cigarettes and alcohol. The question of advertising is a very difficult question. I must confess that my libertarian instincts lead me not to want to prohibit advertising, and yet I am repelled at the notion of seeing a pretty young lady on the television screen saying, smoke my brand of cocaine instead of his, or my brand of smack or crack. So I am very much torn on the question of advertising, but again we don't have to decide all those issues. We now prohibit much advertising of hard liquor, and that is why I tried to say let's treat drugs the way we treat alcohol now, which is a mixture of regarding it primarily as a medical problem, but on the margins, on the fringes, as a criminal problem.
Q. (Robert Millman) I would like you to belabor a point. I don't think that legalization of drugs will cure the problem of the inner city.
Friedman: It won't.
Q. (Robert Millman) And you act as if there is nothing that government can do, or that anyone can do, and I disagree. I think we can do enormous amounts to change the educational system. I think we have to start sooner, doing more along the way. My point is that there is a tendency to think that a technologic answer like legalization will alter suicides or alter crime or change the American dream in those areas, and that worries me.
Friedman: I don't call it a technological change. But whatever you call it, legalization of drugs will improve the conditions of the inner cities enormously, even if it does not completely cure the problem. In my opinion, the most important thing we can do to improve the inner city beyond that is to get the government out of doing most of what it is doing and into something else, which is privatizing the school system. The most destructive area of government intervention in the inner cities, other than the prohibition of drugs, is in the school system. What we have to do is to privatize the school system. This is something that I suggested doing thirty years ago through unlimited vouchers for schooling which could be used in any governmental or private school. In New York City it is also easy to see that you need to eliminate rent control, which has been a major factor in destroying housing in the inner cities.
Q. (Joseph McNamara) Professor Friedman, what do you think realistically is the political potential for decriminalizing drugs, given what we have seen for seventy years?
Friedman: Unfortunately, very small, but you know that doesn't mean we shouldn't keep trying. For example, in education, I have been working on the voucher system for thirty years, even longer, and I have participated in one unsuccessful attempt to get one experiment started after another-in New Hampshire, in Connecticut, in Michigan, in Oregon-but that does not mean we should give up hope. I have also seen cases in which things that were said to be politically unfeasible became politically feasible almost overnight. They are not in this area, but I could cite you examples.
Q. (Joseph McNamara) I have a personal example. I have been trying to get some reasonable firearms control for twenty years.
Friedman: I understand we're all in that position, and the one thing we must not do is to give up the fight. If we do, we are through.
Q. Professor Friedman, you described it as mere conjecture that consumption of the currently illicit drugs would go up under legalization. If cocaine were made legally available and taxed at the same rate per dose as alcohol is now taxed, the price of cocaine would fall to approximately 10 percent of its current value. Do you think it is mere conjecture that the price elasticity for demand of cocaine is likely to be unity, more or less?
Friedman: No, no. Don't misunderstand me. I did not deny that consumption might go up. There is a high likelihood that it would. But there are offsetting forces that I will come to in a moment that might keep that from happening. What I designate as conjecture is the notion that there would be a manifold multiplication of the number of addicts, that the number of drug addicts would become comparable to the number of people consuming alcohol. That I believe is pure conjecture. If you make something lower in price, it is likely that more will be purchased. There is price elasticity. But there are two offsetting forces. One is the effect of the forbidden-fruit kind of thing, especially on the young to whom doing something illegal has a certain value in and of itself. But, second ...
Q. But under your proposal it would remain illegal. So the forbidden fruit wouldn't change.
Friedman: I beg your pardon. It would remain illegal for the children, but consuming drugs would not be doing something that in a sense was sneering at social values. That is what they want to do. Second, and more important, under present circumstances drug dealers have a financial incentive to create addicts. If they can create an addict, they have a captive customer. Under a legalized system, you don't see anybody on the street corners giving away bread. It doesn't pay you to give away bread because if you create somebody who loves bread he can buy it at the cheapest place. It would not pay anybody, for financial reasons, to create addicts.
Q. Unless people create brand-name drugs as, for example, R. J. Reynolds does hand out free examples of their cigarettes.
Friedman: They do hand out cigarettes, that is right. If there were brand-name drugs of that kind, the same thing would happen. But it would be on a much smaller scale. I am not questioning for a moment that the number of drug addicts is likely to go up. What I am saying is that the innocent victims created by even a much higher level of addiction would be vastly smaller than the number now created by drug prohibition. There have been cost-benefit analyses such as Rich Dennis published in the Atlantic Monthly in which he tried to estimate the financial costs and benefits from decriminalizing drugs. He establishes a highly persuasive case that the benefits are far greater than the costs.
However, I am inclined to rest the case much more on the human innocentvictim aspect than on the financial consequences. With respect to the notion that drugs should be taxed, I believe that we overtax alcohol now, in the sense that we tend to generate an illegal black market in alcohol. Very likely, the same thing would occur with drugs. We probably would end up overtaxing them. Everyone of us has a different law he would write, and no one of us is going to be able to write the law. What we can do is to agree that the present system is a mess and isn't working and that we should at least move to some extent in a direction that will reduce the major harm now being done.
On the political level, I want to cite one example that offers a bit more hope. Back in the 1940s and 1950s we had a military draft. And I may say, I was just as opposed to the military draft as I now am to the prohibition of drugs. It looked as if you couldn't get rid of it. It was politically unfeasible to get rid of the draft. We had a conference like this at the University of Chicago; I have forgotten the exact date--sometime in the fifties or early sixties. It was one of the few conferences in which opinions were changed. I hope this will be another. We took a poll at the beginning of the draft conference. We had, just as here, people in favor of the draft, people opposed to the draft-a much wider group than here, including politicians, academicians, and so on. At the beginning of that conference the vote was one-third in favor of the volunteer army and two-thirds in favor of the draft. After three days of the conference, the vote was precisely reversed. Two-thirds expressed themselves in favor of the volunteer army and one-third still in favor of the draft.
I believe that was a major factor in starting the ball rolling, which ultimately got rid of the draft in 1973. 1 believe that this is the same kind of an issue. The evidence is highly persuasive to those who are willing to look at it from the point of view not of one extreme or the other, but of the sensible middle that everybody is looking for. We must change the present policy. I am not without hope that something will happen. At least, the vigor of the attempt at enforcement will lessen.
Q. The thing that concerns me, of course, is exactly the issue that you don't say much about, except to say it is speculative, how much the user pool would increase. But let me ask you this. just make the assumption now that the use of cocaine, particularly crack cocaine which is smoked, that is a socially acceptable way to take things. Just suppose we ended up with as many crack smokers as we have nicotine users today. So that the expansion would be that great. Would that change your views at all about the issue of legalization?
Friedman: It probably would, not about the legalization of marijuana or some other drugs, but about crack. However, I believe that the likelihood of that happening is extremely small. I cannot claim to be an expert on this, but I have looked at a great many studies. I find the evidence of anything like that explosion in usage occurring very underwhelming.
Q. But that is the central issue, isn't it?
Friedman: No, it is not the central ...
Q. Since you would change your position if that were true, then it is important to get a better estimate of whether that is true.
Friedman: Yes, and let me ask you a question. In the meantime, should we allow the killing to go on in the ghettos? Perhaps 10,000 additional murders a year? In the meantime, should we continue to destroy Colombia? Let me ask you a question. Suppose we succeeded in legalizing drugs and the smoking of crack went up 5 percent. Would you change your position?