Bilal Khan (Department of Mathematics & Computer Science, John Jay College, CUNY)
Mitch Downey (Department of Economics, University of San Diego)
Meredith Dank (Urban Institute, Washington D.C.)
Kirk Dombrowski (Department of Sociology, University of Nebraska-Lincoln)
In 2010, the National Institute of Justice funded the Urban Institute’s Justice Policy Center to measure the size and structure of the underground commercial sex economy in eight major US cities. The goals of this study were to: (1) derive a more rigorous estimate of the underground commercial sex economy (UCSE) in eight major US cities; and (2) provide an understanding of the structure of this underground economy. Over the course of the study, however, it became clear that to estimate the size of the UCSE accurately, we had to produce simultaneous estimates in each city for the size of the cash-based trade in both illegal drugs and illegal firearms. Ordinarily, estimates for each of these are made singly, and often for a single locale at a single time. Our approach goes in the opposite direction. The operating assumption of the estimation process that follows is that estimates of the size of various domains of the underground economy (UE) are more accurate when comparative data across time and across different locations are taken into account, and when estimates of the size of one domain of the UE are forced to balance estimates of other domains in the UE with which they coincide.
Underground economies are pervasive, but the cash trade in commercial sex, illegal guns, and illegal drugs are concentrated and most readily observable in urban settings. As discussed below, including multiple urban sites into a single estimation process provided the opportunity to leverage relative estimates of the size of each UE domain across cities as elements in a large system of linear constraints. Here we take up the size of illegal gun, commercial sex, and illegal drug economies in Atlanta, Denver, Dallas, Miami, San Diego, Seattle, and Washington D.C. Further, by making use of multiple proxy measures within each UE domain (illegal gun, commercial sex, and drugs), we mitigate the influence of errors in any single proxy, and distribute the dependence of the final estimates for all cities across a wide range of measured sources.
In seeking to estimate the relative sizes of UE domains, the viability of using arrest data as a proxy measure proved problematic. Law enforcement strategies across US cities vary considerably, and even more so when evaluated across time. Using arrest data as the basis for proxy measurement of various UE domains across a wide range of cities requires making the erroneous assumption of uniform arrest rates across geographical location and time. Fortunately, viable non-arrest-based data is available for use as proxy measures for both the illegal drug and illegal gun trades. As few non-arrest-based sources of data are available for estimation of the UCSE, the project used interviews with arrestees to estimate the size and income levels of pimps/traffickers, in addition to sex workers, in each city to collect the data necessary to complete this estimation process. Because these data came from arrestee sources, special means had to be developed to compensate for the data collection source. These corrections draw from what is normally considered a major challenge to single market size specification: the fact that UCSE participants (especially pimps/traffickers) exhibit high levels of geographical mobility. As described in this chapter, pimp mobility data among the study cities became a key element in the construction of the UCSE proxy.
This project was supported by Award No. 2010-IJ-CX-1674, awarded by the National Institute of Justice, Office of Justice Programs, U.S. Department of Justice. The opinions, findings, and conclusions and recommendations expressed in this document are those of the authors and do not necessarily reflect those of the Department of Justice, or of the Urban Institute, its trustees, or its funders. Elements of this paper were reported to the National Institute of Justice in the form of a final technical report as per grant obligations. The authors would like to thank: (1) final technical report co-authors: Cybele Kotonias, Debbie Mayer, Colleen Owens, Laura Pacifici and Lilly Yu (2) the National Institute of Justice and Janine Zweig, Senior Fellow at the Urban Institute’s Justice Policy Center, for their careful review of project findings.