Should the police be assessed using crime statistics?

This policy brief by the Institute for Security Studies argues that crime statistics are an unreliable measure of police performance due to variables beyond police control, such as public reporting habits and the nature of contact crimes. It highlights systemic fragmentation within South Africa's criminal justice system (CJS) and recommends shifting toward alternative performance indicators and the implementation of the Integrated Justice System to identify operational bottlenecks.

• Using crime statistics to assess police performance is problematic because crime rates are influenced by variables that cannot be attributed to police actions. For example, a decrease in reported sexual offences may reflect a lack of public trust in police ability to act rather than a reduction in crime, while an increase in reporting due to improved trust could paradoxically make crime levels appear to rise.
• The South African Police Service (SAPS) cannot effectively prevent many 'contact crimes' because they often occur in private spaces or between people who know each other, frequently involving alcohol or drug abuse. A 2015 analysis of 2,912 murder cases found that 88% of offenders and 57% of victims had consumed alcohol before or during the crime.
• The Medium Term Strategic Framework: 2014–2019 (MTSF) suffers from a lack of linked performance measures across criminal justice system (CJS) departments. This compartmentalised reporting makes it difficult to track cases from the initial report by SAPS through to prosecution by the National Prosecuting Authority (NPA) and incarceration by the Department of Correctional Services.
• There are significant discrepancies and 'bottlenecks' in the transition of cases from police to prosecutors. In 2016/17, the NPA declined to prosecute 430,363 dockets in district courts (55%) and referred 253,710 back to SAPS for further investigation, while only prosecuting 12% of dockets received in those courts.
• The authors recommend that the 'reduction of contact crime' indicator be narrowed to include only crimes preventable by sound policing, such as carjacking and bank robbery, rather than all contact crimes. They suggest SAPS use crime statistics as a management tool for strategic planning rather than as a primary performance measure.
• To resolve systemic fragmentation, the brief advocates for the speedy implementation of the Integrated Justice System (IJS), a modernisation initiative designed to map cases across the CJS and identify bottlenecks through shared information and database technology.

APA

Schnetler, J., & Lancaster, L. (2018). Should the police be assessed using crime statistics? Institute for Security Studies. https://issafrica.s3.amazonaws.com/site/uploads/pb116.pdf

Chicago

Schnetler, Johann, and Lizette Lancaster. Should the police be assessed using crime statistics? Institute for Security Studies, 2018. https://issafrica.s3.amazonaws.com/site/uploads/pb116.pdf.

Wikipedia

{{cite report |last1=Schnetler |first1=Johann |last2=Lancaster |first2=Lizette |title=Should the police be assessed using crime statistics? |publisher=Institute for Security Studies |date=2018 |url=https://issafrica.s3.amazonaws.com/site/uploads/pb116.pdf |access-date=17 August 2026 |via=Climate Insights Directory}}

BibTeX

@techreport{schnetler2018should, author = {Schnetler, Johann and Lancaster, Lizette}, title = {{Should the police be assessed using crime statistics?}}, institution = {Institute for Security Studies}, year = {2018}, url = {https://issafrica.s3.amazonaws.com/site/uploads/pb116.pdf}, urldate = {2026-08-17}, note = {Indexed by Climate Insights Directory} }