Lies, Damn Lies
It has been almost a month since the White House Office of National Drug Control Policy (ONDCP) issued a breathless press release trumpeting an update to research on the economic impact of drug abuse. "Illegal Drugs Drain $160 Billion a Year from American Economy," shouted the headline on the press release. "Drug Czar Details 'Direct Threat' to US Economic Security," it continued.
The press release, touting "The Economic Costs of Drug Abuse in the United States, 1992-1998," which purports to precisely calculate the impact of drug use on the economy, continued with dire warnings from newly-appointed Drug Czar John Walters. "Illegal drugs are a heavy drag on the American economy," said Walters. "This new report proves that the costs of tolerating drug use can be measured directly in dollars lost, work hours missed, and pink slips handed out. Drugs are a direct threat to the economic security of the United States," he continued. "This study provides some grim accounting, putting a specific dollar figure on the economic waste that illegal drugs represent."
No major media outlets bit at the drug czar's press release, but readers can rest assured that they will be hearing that figure in the future. Like other imaginary numbers, such as the mysterious "50,000 annual deaths from drug abuse," the $160 billion economic cost figure is sure to become another arrow in the quiver of drug war propagandists.
The study's authors break the costs down into three broad categories -- productivity losses, health costs, and "other," primarily crime-related, costs -- and then breaks down losses within each category. Productivity losses, for example, include losses from premature death, drug abuse-related illnesses, institutionalization and hospitalization, crime victims' loss of productivity, and losses due to incarceration and criminal careers. Productivity losses constitute 69percent or $110 billion of the study's calculated $160 billion total costs. "Other" costs, as noted above, are primarily law enforcement-related, and include criminal justice system funding and private costs, such as private legal defense and property damage to crime victims. These costs amounted to 22percent of the total estimated costs of drug abuse. Looking at such figures, drug abuse does indeed appear to be a detriment to the US economy.
There's only one problem: The study is bunk. It is rife with methodological problems and questionable assumptions, but one huge flaw, admitted by the authors, screams out at the serious reader: The study conflates the economic costs of "drug abuse" (which in fed-speak means any use of proscribed drugs) with the costs of enforcing prohibition. In other words, if you are arrested for smoking pot, the wages of the police officer who arrested you, the district attorney who prosecuted you, the judge who sentenced you, the guards who jailed you, the functionary who drug-tested you, and the parole officer who supervised you all become part of the cost of drug abuse. And the fact that you could not work while you were in jail? Those lost wages are also considered a cost of drug abuse, not prohibition.
The study's authors admit this up front: "In the context of this report, the phrase 'drug abuse' is used to refer to the consequences of using illicit drugs, as well as societal costs pertaining to the enforcement of drug laws."
But don't count on President Bush, Drug Czar Walters or their fellow drug warriors to make that distinction. Instead, be prepared to hear the number repeated like a mantra whenever drug policy is on the agenda.
"This is a piece of propaganda that demands to be answered," said David Duncan, chairman of the National Association for Public Health Policy's Council on Illicit Drugs. "I think the study is useful only if you assume at the outset that the current prohibitionist policy is the only reasonable policy to pursue," he told DRCNet. "On the other hand, if you understand that many of the costs are related to drug prohibition and not directly to drug use, then the study is less useful. This study fails to disaggregate drug abuse costs and drug prohibition costs, but the fact is that prohibition itself creates the law enforcement costs and increases the health care and productivity costs," said Duncan.
"That's a problem only for someone interested in what costs would look like without all that enforcement," said Dr. Peter Reuter, founder of the RAND Drug Policy Research Center and currently professor of public affair and criminology at the University of Maryland. "But the question the study poses is what, under current policy, is the cost of drugs, including drug control," Reuter told DRCNet.
Boston University economics professor Jeffrey Miron was not as sanguine as Reuter. "This study is completely useless," said the researcher on drug market economics. "It is totally tautological to say that the money we are spending on drug enforcement is a cost of drug use. That's crazy," he told DRCNet. Miron also had a harsh word for the study's authors, who were careful in the fine print to note some of the study's problems. "The authors still have to know that their work will be used for propaganda purposes, so they deserve to catch some flack, too," said Miron.
Aside from the fundamental flaw of including prohibition costs as part of drug abuse costs, the study is replete with serious methodological problems, say all three experts.
"The authors seem to assume that if illicit drugs ceased to exist, these costs would not exist," said Miron. "But people would in fact do something. They would look for substitutes. They might turn to alcohol or gambling. The authors do not do the counterfactuals that would account for all these things."
Assigning precise figures to productivity losses attributed to drug abuse also raised flags. "They are trying to estimate a number which is neither useful nor possible to estimate," said Reuter, who recently co-authored "Drug War Heresies: Learning from Other Places, Times and Vices" and who had published a critique of an earlier version of the study. "When you ask how much is productivity reduced by drug use, my answer is pick your estimate. There is a wide range of figures. The broad problem is that these estimates are very sensitive to what drug use measure is used, statistical techniques, and what years are involved."
In his earlier critique, Reuter identified a number of methodological problems in the study. There are "inherent uncertainties" in the numbers because "for almost none of the components of cost is there a strong empirical base," he wrote. There are "inconsistent findings" from studies with contradictory results, he noted. And some estimates, such as those for drug-related homicides when a large percentage of murders are unsolved, are no more than a "good guess."
Miron accused the study's authors of a "one-sided assessment" of the evidence. "Take the relationship between drug use, lower wages, and lost productivity," he said. "There are a huge number of studies that find a positive association between drug use and wage levels, not a negative one, but the study dismisses those estimates showing positive effects and only uses the ones showing negative effects."
Duncan agreed. "They are using one study that showed lower wages in persons who smoked marijuana," he said. "In fact, better studies since then have found either very weak relationships or no relationship between drug use and wages or any other measure of productivity. Whether there is any significant association between drug use and productivity is questionable."
But Reuters, for all his even-handedness, delivered the most devastating critique of the study. "Is the question important?" he asked. "One test is to see how these estimates have been cited in the past." Reuters found that "almost universally the study is cited simply as evidence that alcohol and/or drug abuse had large economic costs. In each instance, it was a prop for a policy argument, which would hardly have been affected if the number had been only half as large or twice as large."
So, the study is essentially a propaganda tool. It will be used as one. Drug reformers who wish to confront the study's findings might also find the following mental exercise useful. Let us attempt to disaggregate the numbers, separating the costs of drug abuse from the costs of drug prohibition. This will necessarily be a quick and dirty exercise using some approximations and assumptions about the roles of individual drug users and prohibition itself in generating economic costs.
The study estimates drug abuse health care costs at $14.9 billion for 2000. Assuming that all prevention costs and half of treatment costs are drug abuse-related (assuming that roughly half of all treatment slots are filled by court mandates) and that individual drug users should shoulder half the burden of drug-related diseases such as HIV/AIDS and Hepatitis C (some disease costs are generated by prohibition, i.e. lack of availability of clean needles), we see that of that $14.9 billion, about $7 billion is fairly attributable to direct consequences of drug use.
The study estimates productivity losses at $110 billion for 2000. Again, assuming that drug users should take half the responsibility for premature death, drug-related illness, and hospitalization and half the blame for costs associated with "crime careers," and assuming that 90percent of drug-related crime is prohibition-related, we find that productivity losses fairly attributed to drug use shrink dramatically. Incarceration is the single largest productivity loss factor, pegged at $35 billion, but that is clearly a prohibition-related cost. Of the $110 billion productivity loss then, approximately $40 billion comes from drug use and $70 billion from drug prohibition.
The study estimates "other" costs, primarily law enforcement, at $35 billion in 2000. Generously assuming that 10percent of these crime costs are related to the psychobiology of drug use, with the rest related to the functioning of black markets, we see that prohibition-related costs account for $31.5 billion of the $35 billion in "other" costs.
Applying this analysis, we see that costs directly associated with drug abuse account for about $50 billion in economic costs, not an insignificant figure. But it pales in comparison with the $110 billion in costs imposed by prohibition.
These figures are, of course, only rough estimates, but they illustrate the magnitude of the price society pays to maintain drug prohibition.
Finally, under its own definitions, the cost of producing "The Economic Impact of Drug Abuse in the United States, 1992-1998" is yet another drug abuse-related cost.