TCS Daily

More Drugs, Less Crime?

By Iain Murray - July 22, 2002 12:00 AM

Drug czar John P. Walters, writing in last Friday's Wall Street Journal (link for Journal subscribers only), argued that legalizing drugs would not reduce America's crime problem, and would add a public health problem on top. Some academics, on the other hand, have recently argued that a hard line stance on drug enforcement directly contributes to a nation's problem with violent crime. Their case, however, is far from proven.

The argument is analogous to a strong theory about gun ownership. The suggestion that fewer restrictions on gun ownership decrease rather than increase crime has looked increasingly valid in recent years (John Lott of the American Enterprise Institute's groundbreaking book "More Guns, Less Crime" is the classic in this area). Academic attention is now turning to ask whether the same can be true about drugs.

The economic hypothesis is fairly simple. Completely suppressing sale and distribution of a product for which there is a market forces the market to go underground. In such a market, legitimate competition by means such as advertising is unavailable and so competition proceeds by unconventional means. These means may often take the form of violence. Therefore, we would expect violence rates to be higher in countries that more aggressively prohibit drugs. Others counter with the argument that violent crime is much more likely to be driven by firearms availability, and the drug issue is secondary.

One recent paper, presented at a conference organized by John Lott among others, looked at the connection between violent crime, drug prohibition and restrictions on firearm ownership in 66 different industrialized or semi-industrialized countries. It found that weaker firearms prohibitions were associated with lower crime, while strong drugs prohibition was associated with higher crime. The author of the paper, Jeffrey Miron of Boston University, points out that over three-quarters of the homicides in a sample of precincts in New York City in 1988 were due in some fashion to drug-related causes: disputes over drug territory, drug debts and so on as well as murders committed when either victim or perpetrator was under the influence of drugs. He concludes that "differences in drug prohibition enforcement explain differences in violence, which in turn explain differences in gun ownership that correlate positively with violence but do not cause that violence."

The hypothesis certainly sounds convincing, but there are significant problems with Miron's methodology. To begin with, his categorization of nations' approaches to firearms regulation is far too crude to be a sensitive measure of their effect on violence. This leaps off the page when one reads that the United States, with its generally liberal approach to gun control and the United Kingdom, with its highly restrictive approach that bans private ownership of virtually all sorts of firearm, receive the same score in the measure of gun prohibition. This immediately weakens the credibility of the study in measuring the effect of gun laws on violence.

Second, and as important, the study is weakened by its reliance on the use of the homicide rate between 1993 and 1996 as its measure of violence levels. In many ways the homicide rate, especially in the case of the United States, is not really a useful measure of the violence of a society. To begin with, murder is the rarest violent crime, and so forms a very small subset of the phenomenon being measured. Second, murder rates do not correlate particularly well with overall violent crime rates. For instance, according to the new British crime figures released last week, the USA continues to have a murder rate three times as high as the UK's. Britain, however, has twice as many rapes, 40 percent more robberies and over 30 percent more assaults per head of population than the US. Which is the more violent country? By sheer volume of crime, the UK wins by some way, but if we use the murder rate as our measure the US remains well ahead.

Moreover, using such fixed, snapshot data as Miron did in his analysis ignores the way crime changes over time. In the years since 1993, crime in the US has plummeted, even though drug prohibition has remained strict. In the UK, to take another example, crime has increased or remained steady, while drug laws have been weakened to the extent that cannabis was effectively decriminalized last week. The fact that many gun control laws have been liberalized in the US while they have been strengthened in the UK adds a further dimension to this analysis that has not fully been taken into account.

Overall, then, the case that guns don't cause crime but drug laws do has not been proven. Herman Goering famously remarked in 1936 that he would rather have guns than butter. That famous old morphine addict would probably have been delighted to hear an argument that he should not have to choose between guns and drugs. The rest of us, however, will have to wait a while before "More Drugs, Less Crime" appears on the nation's bookshelves.



TCS Daily Archives