The BATF study found that 43 percent of the trafficking investigations involved the illegal diversion of 10 guns or fewer but confirmed the existence of large trafficking operations, including two cases involving the diversion of over 10,000 guns. Corrupt FFLs accounted for only 9 percent of the trafficking investigations but more than half of the guns diverted in the pool of investigations. Violations by FFLs in these investigations included “off paper” sales, false entries in record books, and transfers to prohibited persons. Nearly half of the investigations involved firearms trafficked by straw purchasers, either directly or indirectly. Trafficking investigations involving straw purchasers averaged a relatively small number of firearms per investigation but collectively accounted for 26,000 guns. Firearms were diverted by traffickers at gun shows and flea markets in 14 percent of the investigations, and firearms stolen from FFLs, residences, and common carriers were involved in more than a quarter of the investigations.
Braga et al. (2002) suggest that the three sources of data on illegal gun markets are not directly comparable but broadly compatible. Each data source has its own inherent limitations and, as such, it is difficult to credit the insights provided by one source over another source.
None of the three sources of data contradicts the hypothesis that stolen guns and informal transfers (as opposed to transfers from legitimate sources) predominate in supplying criminals and juveniles with guns. However, they also clearly suggest that licensed dealers play an important role and that the illegal diversion of firearms from legitimate commerce is a problem. In their review of these three sources of data, Braga and his colleagues (2002) suggest that, in the parlance of environmental regulation, illegal gun markets consist of both “point sources”—ongoing diversions through scofflaw dealers and trafficking rings—and “diffuse sources”—acquisitions through theft and informal voluntary sales. As in the case of pollution, both point sources and diffuse sources are important (see also Cook and Braga, 2001). Braga and his colleagues (2002) also speculate that the mix of point and diffuse sources differs across jurisdictions depending on the density of gun ownership and the strictness of gun controls.
Real interventions in gun markets tend to target particular types of firearms or sources. If policy raises the difficulty (cost, time, risk) of obtaining a particular type of gun or using a particular type of source, the effect