Police reform and the use of performance indicators in South Africa


During the transition to democracy in South Africa, the nature of police institutions and the process of policing itself were dramatically reformed. At an institutional level, the South African Police Service (SAPS) was created as a result of the integration of the former South African Police and the ten “homeland” police agencies. Considerable work was required to align and integrate these structures and their various procedures, ranks and administrative systems. In terms of delivery of policing itself, there were also significant reforms, including a focus on improving relations between police and community; redesigning systems for selection and training; preventing and actively responding to torture or other human rights abuses; and inculcating a culture of service delivery into police work. The complexity and multi-faceted nature of the reform process, along with the requirement for police, policy makers and the newly created civilian secretariats at national and provincial level to manage and monitor the process more effectively, led to the establishment of a performance management system based on indicators.

Entry point

The critical entry point for establishing the system was the desire of the new political leadership to track the impact of the police reform process, including on the level of security experienced by local people. On an annual basis, the police, the civilian secretariat and elected policy makers set the overall performance targets. Detailed performance indicators were agreed with civilian policy makers (at both national and provincial level) and then published. The police established a data management system to record and report on performance. Finally, levels of performance were evaluated by the civilian secretariats.

Lessons learned

Too many indicators — Given the number of issues requiring monitoring, the early years of the process focused on an excessive quantity of indicators, numbering well over 100. This required a significant bureaucratic effort to collect data and resulted in the presentation of detailed tables of numbers that were difficult to interpret. The indicators also held data on widely varying topics, from levels of reported crime to the percentage of formally disadvantaged officers receiving promotion. That made it difficult to provide clear answers as to how well overall reform efforts were proceeding. Many of the indicators were also difficult to verify independently (for example the number of patrols conducted or meetings held), leading to suspicions from civilians and operational police alike that the reporting system lacked integrity. Fewer, more focused and clearly verifiable indicators would have simplified and increased the influence of the system.

Difficulty of analysis — The civilian secretariats responsible for the analysis of the data at both national and provincial level were short of skilled human resources and ill-equipped for the task. While some progress was made at national level in producing reports on police performance, this was seldom replicated further down the system at the level of provincial secretariats. The analysis produced was also heavily data-driven (reflecting the statistical inputs provided by the police), and given the complexity of the task it was difficult to produce information that policy makers could use effectively. Indicators that measured activities (patrols or meetings) were impossible to reconcile with those that measured outputs (reducing crime) in a way that produced useful analytical outcomes. In combination with a simplified set of indicators, greater investment in analytical capacity would have significantly enhanced the quality of the reporting, and thereby the impact.

Weakness of the feedback loop to policy makers — Analysis of police performance in relation to the indicators was seldom made public, reducing the pressure to use the system as a management tool. While the minister for safety and security was briefed quarterly on the results by the national secretariat, given the challenges spelled out above it was difficult to integrate findings directly into a set of formal policy directives. Nevertheless such briefings, as well as those to parliament and provincial
legislatures, contributed to an understanding of the complexity, challenges and overall progress of the reform effort. A more clearly established and widely recognised reporting process in which the findings were presented and debated and decisions taken could have strengthened the link to national policy making.


While the system suffered from clear weaknesses, it provided the only country-wide data-driven system to determine overall progress with regard to police reform. Indeed, the system of indicators was often taken much more seriously at the level of local and area police, several of whom established a dedicated and often very sophisticated monitoring capacity. In the longer term the use of indicators during the transition inculcated a much stronger culture of measuring performance in the SAPS than
would otherwise have been the case.

An overview of the current process of using indicators for measuring police performance in South Africa, as well as its links to the system of police oversight, can be found in “The Police We Want: A Handbook for Oversight of the Police in South Africa”, Open Society Justice Initiative, 2005, http://www.justiceinitiative.org.