Gun Control: Arm Yourself with Evidence

Who's right-the gun lobby or the gun control lobby? Most of what we hear is rhetoric, but there's now a whole body of data on the effects of controlling gun ownership.


March 30, 1981, was a typical day in our nation's capital. Rain fell, panhandlers panhandled, Congress debated without result, and a psychotic tried to kill the president. Nations seem to have traditions in crime, in their social life. Italy has its terrorists, South America its political massacres; we have our media-hungry basket cases who always "act alone."

Nations also have traditions in their reaction to major crimes. Our television media had barely stopped competing for the best close-up and most authentic blood tones when they began publicizing demands for measures against gunmakers who "profit from violence." Soon the print media took up the call, and a debate raged on the best form of gun control, whether waiting periods (presumably future presidential assassins would be in a hurry) or bans on cheap guns (perhaps they would be short of cash or have left their American Express at home), or total handgun bans (although one might wonder whether Brady would have recovered from his head wound, or Reagan from a shot in the chest, if the assassin had used a sawed-off shotgun rather than a .22 pistol). The debate then moved into the direct-mail business, as various gun-control groups sought to get money while the getting was good.

In all this, no one seemed to ask the question that should have been asked first: Why, in 1982, are we talking about gun control at all?

Gun control is hardly a novelty in America. Our first laws controlling concealed weapons date from 1813; handgun permit systems were extensively applied to free blacks in the antebellum South; our first handgun ban came in Georgia in 1837. By 1920, when England adopted its strict handgun controls, much of our Southeast and Southwest banned most or all handgun carrying.

Not that it did either England or America much good. The southern laws were repealed or amended out of existence in the years since, and homicide rates fell nonetheless—by 65 percent in Atlanta and 78 percent in Memphis, for instance. The English controls have been steadily tightened, and a few years ago a Cambridge University criminology scholar said of his findings: "One is forced to the rather startling conclusion that the use of firearms in crime was very much less when there were no controls of any sort and when anyone, convicted criminal or lunatic, could buy any sort of firearm without restriction."

Far from being novel, virtually all modern gun control contentions ("crimes of passion," self-defense, mandatory sentencing, and others) can be found in articles written during the 1920s. Even the sinister "gun lobby" appeared as early as 1911, when a New York politician slapped that label on hardware store owners testifying against a gun law.

Given the brevity of human memory, rhetorical reruns are perhaps to be expected. That the arguments recur is not so astonishing as the refusal to look to their results in the past.

A History of Failure

Probably the clearest example of unfulfilled hope is New York. Back in 1911, "Big Tim" Sullivan had promised that his handgun permit law would do more to stop killing "and save more souls than all the priests and ministers in the state." New York City's murder rate proceeded to shoot up 18 percent in 12 months. Rather than admit the possibility that there was something to be learned here, the legislature concluded that the law was too weak: it was amended 68 times in the following 70 years. The state's murder rate persisted in rising at twice the national rate.

Still, the legislators piled on more restrictions: New York City adopted its own, tighter permit laws and refused to honor state-issued permits. State permit requirements were extended to rifles and shotguns as well as handguns. Issuance requirements were subdivided and narrowed. Applicants were required to submit not only fingerprints and photographs but medical certifications and testimonial letters (police investigated the signers of the letters as well as the applicant).

Then there were the extra-legal barriers. Class action suits had to be brought to force police to hand out application forms, to issue permits in compliance with the law, and to stop denying permits on subjective judgments that a person who changed jobs recently or cleared his throat too often was unstable or too nervous.

Of course, the police and judges were consequently less available for dealing with murderers, rapists, and robbers. By 1980, only one felon arrest in a hundred actually ended in a prison sentence. The New York City chief of detectives was forced to announce that his office would henceforth not even investigate burglaries involving under $5,000 in stolen property. The district attorney for the Bronx estimated that in his borough alone there were over 250,000 unserved felony bench warrants, issued because the defendant had not bothered to appear voluntarily for trial: he estimated about 25,000 of these involved persons charged with violent felonies.

The solution—another gun law. Things were a bit desperate: New York had very nearly run out of gun laws to try. Handgun permits (the fashion of the 1920s), rifle permits (de rigeur for a gun controller in the 1960s), waiting periods (the finishing touch for the 1970s) all had been enacted. What was left?

New York's politicians at length found a new fad, one invented in Massachusetts. In April 1975, that state had imposed a one-year mandatory sentence (no probation or parole) for carrying a firearm without a permit. The main impact of the law was upon the judiciary system, which suffered the appeal rate in gun cases jumping from 20 percent of convictions to 95 percent. The conviction rate in appealed cases, meanwhile, plummeted from 91 percent to 35 percent as judges and juries refused to convict under the draconian law.

The Massachusetts law did start with some advantages. April 1975 was a good time to impose any criminal statute, since crime rates had already begun falling, both nationwide and in Boston. From January through March, for instance, Boston had only 11 gun murders, down 35 percent from the same period a year before. In March, gun assaults began a similar drop, falling to barely over half the levels of preceding months. By April, it was a safe bet that the legislature could enact a law mandating good will among men and see it followed by declines in violent crime rates.

Yet normally logical people attributed the decline to the new firearm law, which only formed the basis of 40 convictions in Boston its first year—compared to 2,800 violent firearms crimes and over 11,000 violent crimes overall. A federal research grant soon was underwriting a study at local universities (commonly referred to as the "Pierce-Bowers" study, from two of its authors). Before the gun law was imposed in April, gun murder rates had already begun their drop (in January) and gun assault rates were beginning to drop (starting in March), whereas gun robbery rates reached new highs after April and in fact remained high well into the following year. The expected conclusion would be that an April firearm law hardly caused changes occurring in January, February, and the following year. But the study managed to suggest causation: the January murder drop must have been due to an attempted enactment of the law at that time; the March assault drop must have been due to the publicity campaign that began in February; the 1976 robbery decline must have been due to robbers having "adopted a 'wait and see' attitude on the gun law as to how it would be applied."

The authors did generously allow that "we presently lack empirical evidence to estimate" the veracity of these explanations—although the supposed purpose of the entire study was to determine whether the law caused the decline. More to the point, after 173 pages of statistical ruminations, the researchers conceded that Boston's experience may be equally consistent with the effects of removing all restrictions on gun ownership and simply penalizing use of a gun in crime—in short, with the National Rifle Association's utopia! "We have not reached the point of knowing whether it is changes in punishment imposed for committing assault or robbery with a gun or simply for carrying a gun without a license which is responsible for the altered crime pattern. This is, of course, critical for evaluation of the relative advantages in terms of crime control of felony firearms laws which mandate additional punishment for crimes committed with a gun as compared to new felony firearm laws aimed at the ownership, possession, and/or carrying of firearms." It goes without saying that this conclusion went unmentioned in all news reports of the study.

Within months the New York legislature was being pressed for such a law by a media campaign, ultimately successful, organized by Governor Carey and by Mayor Koch of New York City. But the latest in firearm laws had the same effects as earlier statutes. In 1980, New York City moved from its position as our fourth most violent large city up to our second most violent—led only by Boston. In 1981, it broke its all-time record for homicides. The following year, the New York Times reported: "Crimes committed with handguns have risen at an alarming rate, according to city police figures that compared 1981, the first full year the law was in effect, with 1979, the last full year before the gun law. Slayings committed with handguns increased 25%, robberies rose 56%, and assaults with handguns rose 20%."

After 9,900 arrests of violators in New York City during the law's first year, the city's police commissioner could still say, "The law needs to be enforced and tightened, but even so, we need to give it more time to work.…Every improvement will come slowly, incrementally." This was not, of course, the theme of the campaign for the law in the first place: his estimate that "it will take 15 or 20 years to get us back to a peaceful society" was not heard in those halcyon days. Nor are such estimates heard when the mayor today proposes ammunition controls as the finishing touch to New York's 71 years of gun control failure.

Spiraling Crime Rates

New York at least has company in its failure—the company of every state that has experimented with similar legislation. New Jersey in 1966 adopted a law described by its sponsors as "the most stringent gun law" in the nation. It required registration, waiting periods, a police permit to buy and a separate permit to carry a handgun, plus a permit to purchase a rifle or shotgun. Two years later the state's murder rate was up 46 percent, and its robbery rate had nearly doubled. The following year Hawaii imposed a similar measure. On an isolated island, with handgun registration for over half a century, such a law had a test under optimum conditions. Its murder rate was, moreover, but 2.4 per 100,000—miniscule compared to most states. After four years under the new restrictive gun statute, it was 5.3, a 220 percent increase. The legislature then imposed a truly draconian two-year mandatory sentence for carrying without a permit. Four years later, the rate was 7.7. The legislature imposed a "Saturday Night Special" ban; today, the murder rate is 8.7 per 100,000 population.

Nor was this the only disappointment. The District of Columbia in 1976 banned all civilian handgun sales, required registration for all other guns, and mandated that all firearms kept in the home be disassembled or trigger-locked. Violent crime rates had gone into a slump just before its enactment—a rare enough event in the District—but they promptly began spiraling after the new gun law was imposed. In the two years before its enforcement, for instance, murder rates had fallen 30 percent; in the two years after they rose 18 percent. The law was a disaster.

The only resource was creative accounting. A special project had been set up within the US Conference of Mayors, funded by a $600,000 foundation grant, specifically to propagandize the expansion of gun laws. The raw number of murders had declined—because people were moving out of crime-ridden D.C. faster than the murder rate per population unit was increasing. Taking only raw numbers and not rates, the project proclaimed the gun law a success. Rather than admit that rates had fallen over the two years before the law, they averaged those years to get a "before enactment" figure. Nor did they mention that Baltimore, the nearest major city to the District, had over the same period seen a 46 percent drop in handgun homicides without a new gun law. The National Rifle Association was, of course, not slow to point out these errors.

The end result was that by 1981 advocates of gun controls were back where they had been in 1961—citing the examples of England and Japan, simply because there were no American jurisdictions with any favorable results. The obvious distinctions, both cultural and judicial (for one, those two nations solve 81 percent and 90 percent of their violent crimes; we solve about 45 percent), simply had to be glossed over. And never mind the possibility that English and Japanese citizens kill each other at a reduced rate compared to the US citizens simply because they attack each other at a reduced rate—not because those attacks are made with less-lethal weaponry. In 1975, for instance, England had but 1,141 assaults; we had 484,710; the current Japanese assault rate is but one-ninth ours, their current murder rate about one-fifth—meaning that when an assault does occur in Japan, it is nearly twice as likely to result in a killing.

Occasionally, the desperation of gun law advocates in their search for a jurisdiction, somewhere, that has reduced crime with a gun law becomes almost humorous. When South Carolina tightened its laws a bit (requiring, for example, better identification of a purchaser), it did so a year before crime rates entered a temporary drop, not just in that state but throughout the East Coast. After the statistics showing the drop were released, Sam Fields of the National Coalition to Ban Handguns drafted an entire article asserting that the gun law had caused the decline. It was a masterpiece of chutzpah.

No one was ever going to argue that South Carolina now had a "strict" gun law—least of all NCBH. So Fields argued that the decline showed what even a very modest change would do. The annoying fact that violence rates had declined throughout the East Coast was taken to prove that South Carolina's gun law had obviously interrupted illegal gun markets all up and down the coast, so that its modest change had influenced crime rates in a quarter of the United States. There were other annoyances: the homicide rate had begun to fall a year before the law was enacted, and in the period following its enactment, blunt-object homicides fell 33 percent, showing that the drop in killing was not due to the firearm law.

Gun Law Data

Nor has NCBH and its sister organizations been able to find much solace in the social sciences. The three earliest studies of firearm laws had all concluded that no significant relationship between firearm laws and crime rates could be shown. These early studies were rather crude: for the most part they treated states as either having "gun laws" or not, whereas firearm laws come in wide varieties and degrees; and they failed to consider other social conditions.

In the mid-1970s a second generation of studies became available. These divided state gun laws into different classes—permit to own, permit to carry, dealer licensing, waiting periods, and so on—and thus could test specific laws as well as groupings of those. Moreover, they used statistical techniques to determine the effects of social variables such as income, education, sex, race, age, and urbanization. The 1975 study of Douglas Murray, an economist at the University of Wisconsin, set the stage. Taking into account all social variables and seven forms of handgun laws, he looked at state rates for handgun homicide, suicide, assault, robbery, and accidental deaths. Murray found no significant relationship between any one handgun restriction, or all of them combined, and the rates for criminal or harmful use.

Murray's study did not go uncriticized. Matthew DeZee of Florida State University pointed out several shortcomings and reran the study using corrected data and format. DeZee found no significant relationship when laws were taken individually, none when they were taken in various combinations, and none when all were considered as a mass. "Gun laws do not appear to affect gun crimes"—an unusual conclusion from a social scientist who advocated handgun prohibition.

Joseph Magaddino, a professor of economics at California State University at Long Beach, went Murray one better. Magaddino formulated an elaborate econometric model that could take into account not only gun laws and social variables but the deterrent impact of probabilities of arrest and punishment. His findings: (1) the probability of arrest has a strong effect on violent crime rates; (2) increasing probability of imprisonment has a strong impact on homicide and a lesser impact on assault rates; (3) the only firearm restriction not already incorporated into federal law that appeared to have any impact on crime was a ban on carrying loaded firearms in vehicles, and all that could be said even of this was that it appeared to have "some level of deterrence" for robbery and assault but not for homicide.

With Magaddino's study, the development of the "second generation" statistical work was complete. The results had been universally and strongly negative to the concept of gun control as a means of crime control.

The "third generation" studies brought no more solace to the proponents of such restrictions. These studies zeroed in on individual states, sometimes in great detail. The first was conducted by the California attorney general and involved computerizing all violent deaths in the state for the first half of 1975, then breaking them down by means, location, county, and cause. Among its relevant conclusions, the study found no significant difference between the percentage of gun and nongun robberies that resulted in murder of the victim (0.8 percent and 0.8 percent) and found compelling evidence that concealability of weapon was not a factor in homicide. The large amount of data available enabled the study to focus in on firearm homicides alone and to break these down between residential and nonresidential killings and handgun and long-arm use. The finding was that a murderer who attacked away from his home was no more likely than one who attacked in his own home (where concealed carry is obviously not vital) to choose a handgun over a rifle or shotgun.

A year later, sociologists David Bordua and Alan Lizotte completed a still more intensive study of the 102 counties in Illinois. They obtained data on gun ownership in each county, augmented this with extensive statewide surveys on reasons for and nature of gun ownership, and considered economic and social data for each county. The results: (1) considered in isolation, gun ownership rates had an almost perfect negative correlation with violence rates; (2) when social variables—largely urbanization—were considered, there was no relationship between violence rates and gun ownership; (3) the segment of the population with the highest percentage of handgun ownership from self-protection motives was not "rednecks" but single black women living in high-crime urban areas.

Thus the case for gun control went from bad to worse as statistical studies progressed into steadily more refined and detailed formats. Yet do we still hear cries for more firearm control?

Criminal Results

The main effect of restrictive gun laws is to artificially inflate the crime rate by criminalizing conduct, on purely regulatory grounds, that is engaged in by tens of millions of citizens. And how are such laws to be enforced? Barely 0.02 percent of handguns are involved in a homicide annually. We can safely assume this would be the last 0.02 percent to be registered or turned in; street thugs are not known for reverence for the law. Since few proponents of any criminal law can claim 99.98 percent enforceability, the problems are obvious. Enforcing officials can always start kicking in doors, but with guns in half the households in the nation…The only alternative is draconian punishments sufficient to cause compliance by deterring violation. But where handguns are kept for self-defense, citizens will give them up only when they become more fearful of their government than they are of street thugs.

Even attempted enforcement would quickly overload the criminal justice system with these artificial criminals. That system is already stretched beyond the limit. Gary Kleck, in an advanced time-series analysis, concluded that homicide rates are a self-perpetuating cycle: more killings overload the system, reducing probabilities of arrest and conviction, and reduced probability of arrest and conviction results in more killings. Between 1960 and 1970, the homicide conviction probability dropped steeply, and the homicide rate doubled; in the mid-1970s the conviction probability jumped upward by one-fourth, from 30 percent to 39 percent, and the homicide rate fell over 8 percent (despite, by the way, several million new handguns entering the market). The main limits are simply manpower and money. A given police officer can investigate and solve only so many crimes per year, a court try only so many cases, and a prison hold only so many offenders.

Chicago's gun control law has required the dedication of two courtrooms (which thereby are unable to deal with overloads of other criminal cases) solely to trying gun law offenses. One of the judges testified a few years ago:

The most striking experience I can take away from the Gun Court…is with regard to the kinds of people that appear there as defendants. For most, this is their first arrest of any kind. I don't mean now that this is their first conviction, but I mean this is their very first arrest of any kind. Many of them are old people, many of them are shopkeepers, persons who have been previous victims of violent crime.

Naturally, most of these "criminals" get probation, which is why the advocates of stricter gun laws currently consider it imperative to press for mandatory sentencing for gun law violations. Presumably, overloading the courts is an insufficient remedy; now we should push out of prisons a few hundred or thousand convicted robbers or rapists to make room for the new artificial criminals. Considering that weapons arrests even under existing gun laws total well over 100,000 per year, we are talking about no small additional load for the criminal justice system and no small number of genuine criminals removed from confinement to make space for gun law violators.

And so we return to the original question: why are we even debating gun control as a crime solution? To have discussed it in the 1920s, or even the 1960s, would have been a reflection on the lack of hard data and detailed studies then available. To have discussed it in the 1970s might have been evidence of emotional commitment. To discuss it today as an abstraction, in the sense that we might discuss a return to liquor prohibition, prior restraint on the press, or what if Longstreet had come up at Gettysburg, might be a source of diversion for depressing moments of quiet that cannot be otherwise disrupted. But to now discuss gun control as a serious approach to crime, and to ignore other solid proposals to deter crime or decrease interpersonal violence in the first place, shows either ignorance of the subject matter or a willful disregard of experience.

David Hardy is a contributor to Restricting Handguns: The Liberal Skeptics Speak Out. Formerly an attorney in private practice in Tucson, he recently took a position in the office of the general counsel for the National Rifle Association.


Virtually the only books on the American firearms issue that are well researched and readable are Lee Kennett and James Anderson, The Gun in America (Greenwood Press, 1975), and Don Kates, ed., Restricting Handguns: The Liberal Skeptics Speak Out (North River Press, 1979). (A second Kates book, Firearms and Violence: Issues of Public Policy, is scheduled to be published soon by Ballinger and the Pacific Institute.)

The firearm regulation issue is capably discussed from an English standpoint in Colin Greenwood, Firearms Control (Routledge & Kegan Paul, 1972). My discussion of British crime rates before and after adoption of their strict law, which dates only from 1920, is based on information provided in Greenwood, p. 234, and Kates, p. 31.

Because of the difficulty in compiling crime statistics in turn-of-the-century America, I relied on secondary sources for information on early New York crime rates. These were obtained from Kennett and Anderson, p. 185; a most interesting article in the New York Times for May 24, 1913; and LaMar Beman, ed., Outlawing the Pistol, (Reference Shelf, 1926), p. 33. Beman is also a fascinating source of pro-and antigun rhetoric of the 1910–25 period.

All post-1930 statistics are taken from the FBI's Uniform Crime Reports for the relevant years, except that figures for the District of Columbia were computed using the number of crimes reported by the FBI and population figures within the limits of that jurisdiction. FBI rate figures for the District are premised upon its Standard Metropolitan Statistical Area, which takes in several Virginia and Maryland counties not subject to the District gun law.

Those interested in additional studies on firearms laws and civil liberties should examine Hearings on Gun Control and Constitutional Rights, Senate Subcommittee on the Constitution, 96th Cong., 2d Sess., serial 96-83 (Sept. 15, 1980).

Major statistical studies cited may be found at: Krug, The Relationship between Firearms Ownership and Crime Rates, reprinted in 113 Cong. Rec. 20060, 20064 (1967) and 114 Cong. Rec. 1496 (1968); Murray, Handguns, Gun Control Laws, and Firearms Violence, Social Problems (1975); Magaddino, "Comparative Cross-Cultural Statistics, "in Kates, p. 31. These and other studies are summarized in Hardy, Firearms Ownership and Regulation, William and Mary Law Review (1978). The Pierce-Bowers study may be found at Rossman et al., The Impact of the Mandatory Sentence Gun Law in Massachusetts (unpublished, Boston University School of Law, 1979).