Module 6: Group Discussion

Evidence-Based Prevention of Organized Crime: Assessing a New Collaborative Approach 315

Public Administration Review,

Vol. 78, Iss. 2, pp. 315–317. © 2017 by

The American Society for Public Administration.

DOI: 10.1111/puar.12889.

Martijn Groenleer is professor of law and governance in the Tilburg School of Governance and director of the Tilburg Center for Regional Law and Governance, both at Tilburg University in the Netherlands. His research focuses on the analysis and design of regulation and governance in multiactor and multilevel settings. E-mail: [email protected]

Jorrit de Jong is lecturer in public policy and management at Harvard University ’ s Kennedy School of Government. He is faculty director of the Bloomberg Harvard City Leadership Initiative and academic director of the Innovations in Government Program at the Kennedy School. His research and teaching focus on leading innovation in the public sector. E-mail: [email protected]

Herman Bolhaar is senior fellow in the Ash Center for Democratic Governance and Innovation at Harvard University ’ s Kennedy School of Government. He was chair of the Board of Prosecutors-General, the executive board of the Dutch Public Prosecution Service, from 2011 to 2017. As a practitioner, he initiated several innovations in the fight against organized crime. As a senior fellow, he focuses on strengthening cross-sector collaboration in crime prevention and repression. E-mail: [email protected]

This manuscript was originally submitted and accepted as an Evidence in Public Administration article. The feature editors, Kimberley R. Isett, Brian W. Head, and Gary VanLandingham, are gratefully acknowledged for their work in soliciting and developing this content. Effective with Volume 78, the Evidence in Public Administration feature has been discontinued.

Maurits Waardenburg is a PhD candidate in the Tilburg Center for Regional Law and Governance, Tilburg University. He is also a research fellow in the Ash Center for Democratic Governance and Innovation at Harvard University ’ s Kennedy School of Government. His research focuses on performance management in collaborative governance settings. E-mail: [email protected]

After the publication of the U.S. Department of State ’ s 2005 Trafficking in Persons report, the Dutch Public Prosecution Service decided

to increase its efforts to fight human trafficking. The report identified the Netherlands as complying with “the minimum standards for the elimination of trafficking” but made it clear that the country could do more, especially for the 25,000 commercial sex workers in the country, of whom an estimated 50 percent to 80 percent were trafficking victims. The results of the report prompted a question at the Prosecutor ’ s Office: why had its efforts to tackle the human trafficking problem been so ineffective?

In response to this question, in 2006, the Dutch Public Prosecution Service embarked on a journey to find better ways to fight human trafficking. It developed and implemented an unorthodox approach focused on erecting barriers for all “facilitators” of sexual exploitation and forced labor, such as hotels, landlords, tax attorneys, and physicians, that might knowingly or unknowingly provide services to traffickers. Through cumulative learning from a series of high-profile cases—from Eastern European women coerced into illegal hotel prostitution to Philippine sailors exploited in inland shipping—a fundamental challenge gradually revealed itself: how to measure the effects of the new approach to fight human trafficking.

Challenges to Gauge Progress Seven years after taking the first steps toward implementing the new approach, the results were ambiguous. Key stakeholders seemed to be more aware of the human trafficking problem than before the Public Prosecution Service stepped up its efforts, with prosecutors often taking on responsibilities far beyond their traditional prosecutorial role. Politicians, policy makers, and the public at large had come to realize that, although legalized prostitution and sexual exploitation are different, they are usually connected, and that other forms of modern slavery, such as maltreatment of Philippine sailors, still occur in the Netherlands.

Besides clear improvements in public awareness, the Public Prosecution Service struggled to gauge its progress in the fight against human trafficking overall. Recent data showed that human trafficking in the Netherlands was better recognized and reported with each passing year: the number of reported victims increased from 716 to 1,222 per year between 2007 and 2011 (Nationaal Rapporteur 2012 ). It remained unclear, however, whether the increase was a result of more cases being reported thanks to the awareness-raising efforts or whether there had actually been a surge in human trafficking activity.

In a similar vein, the number of prosecutions remained constant: about 220 per year on average. Did that mean that prosecutors had maintained the same level of performance, despite the allocation of considerable resources to prevention? Or was this number disappointing in light of the higher volume of reported victims? Of the reported cases, perpetrators were indicted for trafficking about 70 percent of the time, and around 50 percent of the cases brought before the court led to convictions. Even if this was a satisfactory number, had this been accomplished as a result of or despite the new investigatory and prosecutorial practices focused on prevention?

Overall, thanks to the new approach, much had been learned about the victims, perpetrators, and enablers of human trafficking. Even more had been learned about engaging public and private partners and about the role of the news media. Prosecutors had acquired new skills and developed a different mind-set in dealing with a highly elusive problem. However, it had become less clear how the performance of prosecutors, increasingly engaged in preventive, collaborative efforts, could be measured. And while public perceptions of the problem of human trafficking seemed to have changed, performance indicators for police and prosecutors had not; they were typically evaluated in terms of number of investigations

Evidence-Based Prevention of Organized Crime: Assessing a New Collaborative Approach

Viewpoint Stephen E. Condrey,

Associate Editor

Maurits Waardenburg Martijn Groenleer

Tilburg University

Jorrit de Jong Herman Bolhaar Harvard University

15406210, 2018, 2, D ow

nloaded from https://onlinelibrary.w

iley.com /doi/10.1111/puar.12889 by Pennsylvania State U

niversity, W iley O

nline L ibrary on [21/11/2022]. See the T

erm s and C

onditions (https://onlinelibrary.w iley.com

/term s-and-conditions) on W

iley O nline L

ibrary for rules of use; O A

articles are governed by the applicable C reative C

om m

ons L icense

316 Public Administration Review • March | April 2018

completed, arrests made, successful prosecutions, and other metrics based on repression rather than prevention.

Meanwhile, new forms of transnational organized crime had come to light. The Public Prosecution Service ’ s heightened sensitivity and its novel intelligence approach, implemented in close cooperation with the National Police Force and local and regional authorities, began to uncover a reality that few people had imagined. While crime figures were improving in the Netherlands in general, it became clear that criminal networks engaged in a variety of serious crimes, including human trafficking but also production and trade of hard drugs, money laundering, and large-scale fraud, had become increasingly entrenched in society, undermining the integrity of public and private institutions.

In response, the Public Prosecution Service started to collaborate more systematically with other authorities to raise barriers for these criminal networks, similar to the approach that had been piloted for the human trafficking case. This raised similar questions regarding performance measurement: were the government ’ s efforts to prevent crime in this new way successful, and how could this be measured?

Using Outcome Proxies One of the key performance measurement challenges that prevention-oriented collaborations face is to look beyond traditional crime-fighting statistics that count outputs such as prosecuted criminals and confiscated assets (Moore and Braga 2003 ). First of all, such figures fail to capture the hidden activities of professional criminal networks, as they only relate to detected crimes (Sparrow 2015 ). Moreover, these metrics involve a post facto response—not the effects of preventive efforts. The question is how to determine whether the prevalence of organized crime has decreased and to what extent the preventive efforts have contributed to the result.

The new approach therefore required new forms of evidence, taking into account the nature and scope of the crime problems at hand, in particular their invisibility (Sparrow 2015 ). Additionally, the issue of causality had to be addressed: to what extent can success or failure to produce the desired outcomes be attributed to the collaborative preventive approach?

Despite these difficulties, practice needs metrics for accountability and performance management. One way to develop metrics to track progress without boiling the ocean is to limit the scope of the problem that a collaboration is focusing on. If a collaboration is able to define a distinct, adequately scoped “substantive problem”—for example, a series of drug trades connected to a single criminal network—it can define success more narrowly and measure it more easily. While such measures do not represent evidence of the ultimate impact on criminality as a whole, they do provide a way to track progress when relevant performance measures seem elusive otherwise (Sparrow 2008 ).

Drawing on a similar logic, some collaborations in the Netherlands started experimenting with outcome proxies—intermediate outcome measures related to the ultimate impact that one is trying to gauge that signal movements in a subproblem of the overall problem that one is trying to solve. While such proxies do not represent experimental causal evidence of impact, they are easier

to measure than the actual outcomes and may materialize faster. For example, instead of measuring criminal network activities directly, the tax authority can measure movements in the level of unaccounted wealth of individuals suspected to be involved in criminal networks to gauge criminal activities of a specific criminal network indirectly. In essence, proxies such as these form pieces of early evidence, representing “nonexperimental assessments of subgroups effects” that allow these collaborations to gauge the impact of the approaches they are piloting without resorting to, for instance, a randomized controlled trial (Crowley and Scott 2017 , 650). To develop such proxies, we observed crime-fighting collaborations combining existing data in novel ways and using targeted tests to generate new data. Four types of innovative ways to generate innovative progress metrics are described here.

Connecting existing data from different government parties in novel ways Oftentimes, governmental data sit in silos. Merely stapling these data together is not enough; they need to be connected across governmental partners to generate useful evidence. In one example, this thinking was applied to monitor the criminal activities that happened in and around a holiday park. Different government authorities had received signals of potential criminal activities in the park, but these independent signals were not sufficient to raise a red flag for any of the authorities. However, when a set of known criminals bought property in the park, the municipality, the police, the Public Prosecution Service, and the Tax Authority decided to pool their data-tracking irregularities per property (vacation rentals) to see whether they could uncover patterns. This created a map of suspected criminal activities that allowed the governmental partners to act in a more targeted and decisive manner and, simultaneously, provided a useful tool to track their success in fighting criminal activities on the park.

Monitoring data fl ows in smart networks Today ’ s smart information and communication networks record and give access to previously unavailable data flows that may reveal the nature and scope of criminal activities. Drawing on such a novel data source, a collaboration fighting domestic marijuana plantations partnered with the regional energy distributor to track where electricity was being tapped illegally, a practice often observed for the energy- intensive marijuana production process. This not only allowed the partners to identify new production sites but also gave them an active monitoring mechanism to gauge the progress of their efforts to reduce the number of home-based marijuana plantations.

Using pseudo purchases to test the scope of or exposure to criminal activities in economic subsectors Traditionally, “pseudo-purchases” have been used to test the illegal activities of criminal networks themselves. A novel evidence- generating practice that we observed was the use of similar pseudo-purchases to test the resilience of facilitators against criminal activity. In one example, the collaboration fighting human trafficking used mystery guests to test the responsiveness of the hotel sector to illegal practices. For three days, actresses posing as prostitutes sent signals of illegal activity to hotels they were staying at, such as making repeated requests for clean towels and leaving multiple used condoms in the waste bin. This provided evidence to the government partners on the responsiveness of the hotel sector

15406210, 2018, 2, D ow

nloaded from https://onlinelibrary.w

iley.com /doi/10.1111/puar.12889 by Pennsylvania State U

niversity, W iley O

nline L ibrary on [21/11/2022]. See the T

erm s and C

onditions (https://onlinelibrary.w iley.com

/term s-and-conditions) on W

iley O nline L

ibrary for rules of use; O A

articles are governed by the applicable C reative C

om m

ons L icense

Evidence-Based Prevention of Organized Crime: Assessing a New Collaborative Approach 317

to illegal human trafficking practices as well as the scope of their potential facilitation of criminal activities.

Incentivizing private parties to self-report and identify criminal activities A final way that crime-fighting collaborations generated new progress metrics was by incentivizing the private sector to self-report. In one example, the police worked together with the Dutch motorized vehicle sector organization, the BOVAG, to raise awareness among car rental companies of their facilitation of criminal networks. While helping them recognize and report signals of suspected criminal activities, such as an insistence on paying with cash or not providing adequate identification documentation, they raised awareness of the potential negative impact of facilitating criminal networks. This generated two sorts of new indicators: more consistent reports of suspected criminal activities from cooperative rental companies and the identification of uncooperative rental companies possibly engaged in conscious facilitation of criminal activities.

Concluding Thoughts Since the publication of the 2005 Trafficking in Persons report, crime-fighting collaborations have proliferated in the Netherlands. The preventive nature of these efforts and the clandestine nature of the criminal activities they are trying to fight often make it hard to track progress. As we have shown, this has driven some to develop innovative proxies for impact, going beyond organizational boundaries to combine, generate, and use data in novel ways. These proxies represent resource-efficient ways to overcome stagnation in tackling complex crime cases. They represent intermediate outcome measures that are more easily obtainable than ultimate measures of clandestine criminal activities and, oftentimes, provide immediate feedback to be able to direct collaborative activities to where they are most impactful. In this sense, they represent a potential approach to start tracking and managing the performance of collaborative crime- fighting collaborations at scale and create the opportunity to learn continuously.

Yet three remaining issues need to be explored in future research on and with crime-fighting collaborations. First, each collaborative effort is to a large extent fighting unique problems that evolve over time and in a particular context, leading to the need to reinvent specific outcome proxies for each new challenge. Finding such measures that are correlated with ultimate outcomes and are readily accessible is “rarely easy” (Behn 2014 , 129). How should collaborations foster and strengthen this creative process to ensure success? And can such a proxy redefinition process be integrated in the agencies ’ performance management practices in a structured manner?

Second, while proxies may represent more easily accessible metrics for collaborations to track progress efficiently, they are intermediate

and problem specific by their very nature. This makes it hard to judge how they are causally related to desired outcomes for the larger problem one is trying to tackle and how they interact with other alternative explanations. Can we find a more systematic way to judge the validity of these proxies in predicting ultimate public value outcomes?

Finally, while outcome proxies may represent a promising avenue to improve the performance management capabilities of specific crime- fighting collaborations individually, it is also worth considering how the impact of this new collaborative and preventive way of working can be assessed in general and across various crime-fighting collaborations. What metrics and data-gathering methodologies could be used to compare and contrast collaborations? How can effectiveness of this approach be evaluated at a meta level?

Ultimately, all these new approaches to monitoring progress, evaluating efforts, and measuring outcomes serve one purpose: to facilitate rapid and rigorous learning in practice. Crime-fighting collaborations typically have neither the resources nor the time to do comprehensive research resulting in scientific knowledge that meets the highest standards of validity. But, as the old aphorism popularized by Voltaire goes, the perfect is the enemy of the good. We need not give up on the aspiration to seek and use evidence in public policy and management if the phenomenon is too complex, the constraints too many, or the resources too few. By making intellectual curiosity a norm, data-driven decisions a standard, and continuous learning a practice, smart collaborators may advance the field of practice in the tradition of the scientific method: combining close attention to empirical facts with logical reasoning and creative experimentation and documenting results in a transparent way so they can be reviewed and discussed in a broader community of learning.

References Behn , Robert D. 2014 . The PerformanceStat Potential: A Leadership Strategy for

Producing Results . Washington, DC : Brookings Institution Press . Crowley , Daniel Max , and Taylor Scott . 2017 . Bringing Rigor to the Use of Evidence

in Policy Making: Translating Early Evidence . Public Administration Review 77 ( 5 ): 650 – 55 .

Moore , Mark H. , and Anthony A. Braga . 2003 . Measuring and Improving Police Performance: The Lessons of Compstat and Its Progeny . Policing: An International Journal of Police Strategies & Management 26 ( 3 ): 439 – 53 .

Nationaal Rapporteur Mensenhandel en Seksueel Geweld tegen Kinderen . 2012 . Mensenhandel in en uit beeld. Cijfermatige rapportage ( 2007 – 2011 ) . The Hague : BNRM .

Sparrow , Malcolm K. 2008 . The Character of Harms: Operational Challenges in Control . Cambridge : Cambridge University Press .

_____. 2015 . Measuring Performance in a Modern Police Organization . Psychosociological Issues in Human Resource Management 3 ( 2 ): 17 – 52 .

15406210, 2018, 2, D ow

nloaded from https://onlinelibrary.w

iley.com /doi/10.1111/puar.12889 by Pennsylvania State U

niversity, W iley O

nline L ibrary on [21/11/2022]. See the T

erm s and C

onditions (https://onlinelibrary.w iley.com

/term s-and-conditions) on W

iley O nline L

ibrary for rules of use; O A

articles are governed by the applicable C reative C

om m

ons L icense