2

Reform-based Training

The committee makes a distinction between training-based reform and reform-based training. It also recognizes the challenges of institutional and political barriers to the administration of training.

TRAINING-BASED REFORM

For over half a century, Global North nations including the United Kingdom and the United States have sought to reform police agencies in the Global South using what the committee considers an approach of training-based reforms. This approach views training as the primary vehicle by which the rule of law and protection of the public can be promoted. Training-based reforms also reflect a common policing philosophy in the Global North: that by providing training, the agency absolves itself of responsibility if an officer does something wrong post-training. However, training has not been clearly connected to a specific change in policing, a broader vision of agency reform, or even a wider change in local or national governance. Part of the reason for this is that police agencies have not used a framework—such as evidence-based policing—for testing whether training succeeded in producing intended reforms. Instead, the fact that training was delivered was itself the measure of success.¹

___________________

¹ A poignant example of this lack of a link between training and outcomes is the case of Derek Chauvin, the Minneapolis officer who killed George Floyd. At the time Chauvin killed Floyd, the Minneapolis Police Department (and Chauvin) had been up to date with very progressive training on implicit bias, use of force, procedural justice, and community policing. And yet, Chavin still killed Floyd while other officers stood by without stopping him (Bella et al., 2021).

One reason training-based reform has not been tested or tracked for success is that the purpose and goals of training can be numerous, highly diffuse, or not even agreed upon by police leadership or the community. For example, is the aim of training in firearms to protect the officer, to protect the community, or to satisfy the legal liability concerns of officers carrying guns? Are officers trained in diversity issues and implicit bias to improve internal diversity in the force, reduce prejudice and biased behavior toward the community, or simply to allow officers to acknowledge their own biases?

A sober assessment of this approach suggests that training alone cannot accomplish what a state has failed to achieve in other respects. There is little evidence, for example, that training alone can control corruption in government generally, get officers to take actions to prevent crime or safeguard the public, dismantle organized crime groups, or resist insurgency attacks on police stations. Nevertheless, there may be reasons to believe that police training can play an important role in accomplishing these and other worthy goals, though doing so may require a new approach of how training relates to reform.

REFORM-BASED TRAINING

The idea of reform-based training proposes a different approach. It first asks, what specific reform should be accomplished in the ways the agency delivers police services? Second, what are all the steps needed to achieve the reform, including creating new units or appointing leaders, providing new technologies, creating partnerships with local community groups, and winning political support in the national legislature or from civil servants in key ministries? Then, only after these questions are answered, the reform plan would address this question: what training is needed to support such reform efforts?

As an example, consider the problem of illegal land seizures, in which large companies clear wild forest lands for mining or growing cash crops. Local police may be legally entitled to stop them, but corruption and threats of violence against police discourage actions against such developers. Reform-based training could be a part of the solution, along with a plan for special prosecutors and investigators to seize the assets of illegal developers, with support from a national ministry of justice or interior. In this scenario, the training would teach a new procedure for local police to follow as soon as such illegal development is discovered.

Another example might be the inconsistent use of arrest or force on individuals from different racial or ethnic groups. If the reform sought is a

reduction in disparities in police actions toward the public (that is, pursuing equal protection of the laws), achieving this reform requires not only that officers be trained in how their actions contribute to bias (and ways they can mitigate those behaviors). It also requires supervisory and accountability systems to be put in place to regularly monitor officer behaviors and coach them to adjust their behavior to be aligned with the training they have received. Agencies might also have to discard certain operations that conflict with the goals of training—in this case, operations that continue to result in disparities in police action toward the public (for example, indiscriminate stop-and-frisks or traffic stops).

Evaluating such examples would thus go far beyond the question of whether training was delivered to all officers. Evaluations would include clearly describing the outcomes the agency is seeking before the training began. They would include evaluating the entire plan to achieve those reforms (with training making up only one part of the plan). Ensuring that training was effective would also require training supervisors in coaching, mentoring, and managing officers who have received training, to hold them accountable to new behaviors or activities. Finally, if the training was delivered, the evaluation would examine the effects of new police practices on the outcomes intended for those practices, such as reduced violence or racial disparity.

The committee recognizes that transitioning from training-based reforms to reform-based training could take many years and involve a deeper government and state commitment to reform. Yet there may be ready platforms for this approach in any annual reviews of the overall state of policing in a country receiving foreign donations. INL, for example, already has a yearly review process examining a full range of reform goals and concerns. That process may provide a prime opportunity to move from training-based reforms to reform-based training.

In the interim, there remain several crucial questions about the content and form of police training as part of foreign assistance efforts. Our discussion of those questions should not be seen as an endorsement of training-based reform. Instead, we address some of those questions as a kind of stopgap for doing the best that can be done with an existing approach to training, even while we would prefer to replace it.

POLITICS OF POLICE REFORM

Police reforms, including reforms of training programs, often face a range of obstacles or complicating factors. Some obstacles may come from resource constraints or technical challenges. Yet as much research suggests, the primary barriers to reform are political and institutional conditions both internal and external to the police agencies. Through quantitative studies

in both democratic and electoral authoritarian countries, scholars have shown that electoral considerations can shape leaders’ decisions regarding policing, including the enforcement of local laws (Holland, 2015) and the geographic distribution of officers (Hassan, 2017). Meanwhile, a growing literature, based mainly on qualitative case studies, has shown that political pressures, including scandals, can be a significant driver of police reforms (González, 2021; Sherman, 1978; Taylor, 2014; Ungar, 2002). At the same time, scholars have also shown that would-be police reformers may well have to contend with additional political factors that complicate reforms, including informal political rules (Sabet, 2012) and partisan competition, which can impede the adoption (Davis, 2006) and limit the durability of police reforms (Eaton, 2008; Hinton, 2006).

Although it is difficult to pinpoint the causal effect of political conditions and institutional characteristics, the social science literature suggests that these factors may play a role in shaping police reform inputs and outcomes, including training. In particular, research by Hernán Flom (2019) has shown that turnover between rival political parties can diminish the incentive to implement a predecessor’s reform policy. In his presentation to the committee about his role as a practitioner charged with overseeing training for the Argentine Federal Police, Flom described how training programs were vulnerable to partisan turnover, undermining the continuity of training programs from one administration to the next (see Box 2-1).

CONCLUSION

The relationship between police training and governmental reform is an inherently political question. An obvious source of political conflict is the macro-level politics concerning which constituencies gain the most powerful and numerous police appointments in ethnically, culturally, or tribally diverse nations. This conflict can play a central role in deciding what kind of training police will receive.

In some countries, other kinds of politics may also be paramount. For example, tensions between permanent civil servants and newly elected members of parliaments may turn police training into a political football, with control over the policy going back and forth between those two groups and—perhaps—police professionals themselves. In other contexts, police officer unions, fraternal or other secret societies, or religious groups may play equally powerful roles in shaping training decisions, often with little transparency or rationale.

All these political dynamics of police training can pose complexity for foreign assistance donors. Commitments by one government to deliver a reform-based training program (or training-based reforms) may be deemphasized by a new government, or opposed from the outset by national

civil servants who are in constant tension with police leaders. Given the dynamic and fragile nature of agreements, the specific aims of police training seem more likely to be accomplished if any politics of training itself is clear from the outset. A realistic approach to risk assessment, with a series of scenarios and fallback positions, could be informative. If the purpose of the training is to promote reform, a “pre-mortem” examination (Kahneman, 2011) of how the intervention might fail would be an appropriate precaution before any funding is transferred. Similarly, a series of compliance tests could be imposed, so that funding could halt soon after any evidence emerged of the training itself being subverted by a country’s internal politics of police training.

A final point about the politics of police training is that it remains largely unexamined. It would be of great value to donors to fund new studies on these issues, comparing successful and unsuccessful assistance from the standpoint of whether funds were spent as planned, and if not, why not. A more systematic approach to documenting and assessing training assistance efforts, including some tracking of results or even receptivity to the training, would create a written, institutional body of knowledge for future police training.