might be mitigated by criminals’ substitution across types of guns or sources. The following framework is helpful for organizing what is known, and what we would like to know, about whether access interventions can reduce harms from criminal gun use.
There are many types of guns; the term “type” encompasses both the literal firearm type (e.g., handguns versus long guns) and the source by which it is acquired (e.g., retail purchase, private sale, theft, loan, and other types of firearm transfers).3 Furthermore, there are many types of individuals (legal possessors, juveniles, convicted felons and other persons prohibited from legal gun possession). Restrictions aim at reducing firearm possession or use by some of those groups. For analyzing the effects of these restrictions, consumer demand theory provides a useful conceptual framework, in which the use of each type of gun by each type of individual depends on the total cost that individual incurs in acquiring or retaining that gun. This generates a specific volume of use (possession or purchase) by each type of individual for each type of weapon. When the difficulty felons face in acquiring new guns rises, for example because of a targeted intervention, we assume that new gun use will decline among felons; whether that decline is substantial can be determined only empirically. Use of other kinds of guns may rise.
We use the term “cost” as broader than the money required for purchase of the item. Nonmonetary costs may be particularly important for gun acquisition by offenders, compared with purchases of unregulated legal goods; these costs include the time required to locate a reliable source or obtain information about prices, the risk of arrest by police (and sanction by a court), and the risk of violence by the seller. These are potentially important in any illicit market and have received some attention in the context of drug markets (Caulkins, 1998; Moore, 1973).
To make clear how this framework operates, consider an intervention that raises the costs criminals face to obtain new guns. The direct or “own” effect of this intervention is to reduce criminals’ demand for new guns. Yet this is not the end of the story. The total effect of the policy intervention is the sum of the “own effect” and a “cross-effect” reflecting criminals’ substitution of used guns for new ones as new guns become more costly. Even if the own effect is negative, the cross-effect might be sufficiently positive to render the overall effect close to zero.
For discussion purposes, we are dramatically simplifying the large variety of guns available to consumers. Guns vary by type (revolvers, semiautomatic pistols, derringers, rifles, and shotguns), caliber and gauge (e.g., .22, .38, 9mm, .45, 12 gauge, 20 gauge, and dozens of other bullet calibers and shotgun gauges), and manufacturers (e.g., Smith & Wesson, Sturm Ruger, Colt, Glock, Sig Sauer, Lorcin, Bryco, and hundreds of other manufacturers). There is ample evidence suggesting that criminal consumers seem to prefer certain types of guns.