More Guns, Less Crime: Understanding Crime and Gun Control Laws.

by John R. Lott Jr., Chicago: University of Chicago Press, 225 pages, $23.00

About half the U.S. population lives in one of the 31 states with relatively permissive laws regulating who may carry a concealed firearm. These states range from northern New England (Maine, New Hampshire, Vermont) to the deep South (Mississippi, Alabama, Georgia, Florida), the Piedmont (Virginia, North Carolina) to the Southwest (Oklahoma, Texas), the upper Midwest (the Dakotas) to the Pacific Northwest (Washington). They include urban states (Pennsylvania), suburban states (Connecticut), rural states (West Virginia and Montana), and everything in between. The other half of America's people live in jurisdictions like New York, where access to concealed-carry permits is limited to those who can demonstrate a specific need for potentially deadly self-protection, or Illinois, where no one other than peace officers may carry a gun.

A massive natural experiment is thus under way, one that will ultimately tell us whether liberal gun carrying laws are good or bad policy. The early results are striking. It can no longer be seriously argued that relaxing the rules against concealed carrying of handguns is an invitation to violence, to bloody shootouts over fender-benders or football games. That sort of thing, always rare, is essentially absent from crime statistics, no matter what a state's rules concerning who may carry a gun in public. What's more, it is beginning to look as though, when a state authorizes private persons to carry handguns, it takes an important step toward suppressing serious crime.

What is at issue in gun control debates is people's (mostly untutored) intuitions about which of two conflicting theories of human behavior has the upper hand in the real world. The first of these theories, sometimes called the "instrumentality theory" of lethal outcomes, holds that when firearms are more readily available, offenses such as armed robbery and murder - and impulsive homicides especially - should increase because guns make it easier to commit crimes.

The opposite theory is that of "general deterrence," which can be summed up in one phrase: more guns, less crime. That, not coincidentally, is the title of an important new book by one of America's most resourceful and fearless econometricians, John Lott, who for the last several years has been the John M. Olin Visiting Fellow in Law and Economics at the University of Chicago Law School.

Each of these theories captures a certain amount of reality. We know, for example, that x number of impulsive homicides would not occur in a gun-free world. On the other hand, we also know that the prospect of meeting armed resistance changes the calculations of human actors, whether they intend good or mischief. That is why we insist that Brinks guards, soldiers, and Secret Service agents carry guns. We recognize that if they did not, their ability to deter predators would shrink or, in some cases, altogether disappear. To know what firearms policy to pursue, one has to know which of these tendencies dominates the other. Like so many other questions with a seemingly ideological leading edge, this one, at bottom, turns out to be empirical.

For many years the public debate about which theory to credit was carried out either by a priori reasoning or, worse, through weak and often tendentious small-scale studies, many of them sponsored by openly results-oriented grantors at the U.S. Centers for Disease Control and Prevention. Almost all of these studies, if one may call them that, affirmed that guns were a "public health" hazard that was spreading by leaps and bounds, the outstanding "risk factor" for suicide and murder. The New England Journal of Medicine, in particular, has specialized in publishing such regrettable stuff, commonly complementing it with overwrought editorials calling upon Americans for God's sake at last to surrender their guns.

The instrumentality theory enjoys something of a monopoly among the news media and is a veritable litmus test for membership in Washington's intellectual establishment. One might as well believe in flying saucers as doubt the proposition that schoolyard massacres are "caused" by America's sick love affair with the gun. Congressional Democrats (to say nothing of the executive branch) are dose to unanimous on this; but Republicans - of the sort who long to be labeled "sensible" or "pragmatic" in newspaper editorials, from Richard Nixon to George Bush - think so too. Practically every gun control initiative of the last 30 years, including the Brady Act and the 1994 "assault weapons" ban, has been based on the premise that restricting the supply of firearms and thereby raising their price should reduce violent crime rates.

Every bit of this, we now know, has been wrongheaded and perverse. Since 1977, the U.S. Department of Justice has kept statistics on the most serious crimes (such as murder, rape, and robbery) in the 3,054 counties of the United States. In 1997 Lott and his collaborator, David Mustard (who was then a University of Chicago graduate student and is now a professor of economics at the University of Georgia), published an analysis of these data, the largest econometric study of crime and violence ever done, in The Journal of Legal Studies.

Their electrifying conclusion was that liberalizing concealed-carry laws drives down rates of confrontational crime, with the effect most pronounced in the counties where the problem of criminal violence is worst. Apparently, when more people are on the streets packing heat, criminals tend to redirect their predatory activities into lines where they are less apt to meet armed resistance. The results of the Lott-Mustard study, updated by several years and amplified by the analysis of additional variables, are the core of More Guns, Less Crime.

Postwar American crime rates peaked around 1979 and have been declining ever since. In most counties the decline has been steady. In the most populous cities, a short but intense murder wave, linked to crack cocaine markets and the war on drugs, complicated the picture from 1984 through 1992. Most states that have relaxed their concealed-carry laws have done so in the last 12 years, a period when violent crime was decreasing generally, though it may have been rising locally.

The resulting picture is a decidedly complicated one that makes analysis difficult. How can one know, for example, that a state's falling murder rate can be attributed to changes in its weapons laws rather than something else? Lott devotes an entire chapter to explaining the statistical techniques he used, combining cross-sectional data (comparing crime rates in different places during the same period) and time-series data (looking at how crime rates in a given place change over time).

One of the most fascinating aspects of More Guns, Less Crime is the story it tells of the firestorm that the Lott-Mustard paper set off. Lott seems bewildered by it still, though it makes as much sense as the theologians' condemnation of Galileo or Darwin. In our culture, results that come wrapped in the aura of science carry a special weight (which is why junk science is a preferred idiom of discourse among political interest groups). Lott seems to have thought that his work would merely shed some light on an issue of public policy. He underestimated the extent to which it challenged the worldview of an entrenched and self-satisfied political elite.

The organized hatred of guns coordinates an impressive amount of political energy in contemporary Washington. For ideological noncombatants like President Clinton, gun control is a low-cost program that conspicuously positions one on the right side of the crime issue. This is especially useful for Democrats, who for a generation had managed unerringly to position themselves on the wrong side.

Gun control suits liberals, who can flatter their interest groups by affirming that the locus of modern evil is not bad people but the bad objects that "society" puts in their way. It tickles feminist sensibilities because, as a Alana Bassin recently wrote in the Hastings Women's Law Journal, guns are phallic, hence patriarchal, hence evil - and ripping phallic objects away from those who cling to them is, after all, the central purpose of life. It inspires communitarians, because armed individuals are so easily portrayed as slack-jawed Archie Bunkers who, as Mario Cuomo once commented in an unguarded moment, drink too much and beat their wives. Watch MTV or Comedy Central, and you can see how contempt for guns and hunters becomes a way for hip, quick-witted city folks to celebrate their superiority over their cloddish country cousins.

Lott's work calls all this into question. It does more than advance discussion; it smashes idols.

The response to the publication of the Lott-Mustard paper was instantaneous and noisy. The critics declared not simply that their methodology was "flawed," as a staff member of the Violence Policy Center who had never laid eyes on the paper whispered to the newspapers. They insisted that the work was actually dishonest, like the pseudoscientific chaff that used to be put out by the Tobacco Institute - because, as Rep. Charles Schumer (D-N.Y.) told reporters, Lott was a hireling of firearms manufacturers. He was on the payroll of the Olin Corporation, the company that manufactures the notorious Winchester Black Talon bullet, a spokeswoman of the Violence Policy Center helpfully added.

The basis for this charge, which many newspapers repeated without comment, was that Lott's University of Chicago fellowship was endowed by the Olin Foundation. The foundation - which endows fellowships and professorships at many universities, including Oxford, Harvard, Yale, and Stanford as well as the University of Chicago - is distinct from the Olin Corporation, a chemical company that, among many other things, manufactures Winchester ammunition. The recipient of the foundation's largess was the University of Chicago, not Lott, who was selected by a law school faculty committee. Universities routinely accept money from foundations endowed by people who got rich selling tobacco, demon rum, fatty foods, even the delivery systems for thermonuclear bombs, but ideological strings on such money are seldom proposed and never accepted.

There have been a few responsible criticisms of Lott's work. In particular, a critical paper by economists Dan Black of the University of Kentucky and Daniel Nagin of Carnegie-Mellon University raised the question of whether Lott's results are driven by a selection bias in his data. One of the strengths of this book is Lott's patient refutation of every responsible criticism. His relentless counterattack on Black and Nagin is a model for academic debate. One would like to see Lott's critics respond to his refutations as specifically and in as much detail as he has answered them.

What sort of public policy do Lott's findings point to? Would the world really be better off with everybody packing heat? Well, the Lott data do not say "everybody." In fact, it appears that something like 1 percent to 2 percent of a state's population will get a concealed-carry license if given the opportunity to do so.

It is hard to be certain of the exact numbers (because, for instance, some states freely issue carry permits to nonresidents), and in any case we do not know how possessing a permit affects the actual carrying of firearms. But one cannot reason that, because a legal regime that inspires 1 or 2 percent of the population to get a carry permit does not increase but actually decreases murders, rapes, and other confrontational crimes, a legal regime that would inspire 10 or 20 percent of the population to get licenses would be similarly benign.

For all we know (and as Lott recognizes), the first 100,000 or 200,000 people who go for the permit may be significantly more law-abiding and punctilious than the next cohort of permit seekers would be. If so, it is entirely possible that the often predicted "OK Corral" scenarios that have yet to materialize may simply lie a bit further out on the experience curve. There is ample room for states to display caution and common sense before following the lead of Vermont, which allows any adult not specifically disqualified by law to pack a gun, without even getting a permit. But after a few more years of experience have accumulated - after the massive natural experiment with firearms and crime has run its course - if the Lott results are repeated, it is hard to see how the conventional wisdom about guns could possibly survive. Five or six years from now, we'll know.

Daniel D. Polsby ([e-mail]) is Kirkland & Ellis Professor of Law at Northwestern University.