Executive summary
Create a National Violence Prevention & Justice Delivery Unit and Safety Observatory.
The first shocking security statistic is not a homicide count. It is institutional invisibility. Saint Lucia’s official National Crime Victimization Survey estimated that 18% of adults experienced at least one of fourteen measured crimes during September 2018 to August 2019, yet only 34.2% of crimes were reported to any competent authority. A reactive system is therefore working from a partial picture.
Safety also determines whether people use streets, public transport, businesses and community life. In the same baseline survey, 38% of adults said they felt unsafe or very unsafe walking alone in their neighbourhood at night. In Castries and Gros Islet the combined figure was 42.7%. Public safety is therefore a mobility, health and economic-participation issue as well as a policing issue.
Saint Lucia already has useful building blocks: the Swift Justice initiative, a Criminal Backlog Reduction Court, virtual hearings, youth-diversion work and a national crime-prevention policy under development. SLPA proposes joining prevention, lawful enforcement, victim support, justice flow and public measurement in one national safety operating system.
Developed from a Saint Lucia security, justice and resilience research draft and implementation annex. Crime-survey, court-flow and reform claims were checked against official CSO, ECSC, Government of Saint Lucia and CARICOM–UN records in July 2026.
Key findings
Low reporting leaves most victimisation outside the system.
When roughly two-thirds of measured victimisation is invisible to competent authorities, patrol, investigation and prevention priorities inherit a selection bias. Safe reporting routes, victim liaison, response standards and repeated trust measurement are operational capabilities—not public-relations extras.
Evidence-led findingAnnual court data show case-flow pressure.
ECSC’s 2023 tables recorded 405 new criminal High Court filings and 268 dispositions, and 1,536 Magistrates’ Court filings against 1,099 dispositions. These are annual flows, not the same case cohorts or conviction rates, but they support an end-to-end caseflow response with better disclosure, triage and age-of-case data.
Evidence-led findingMicro-place data need strict privacy safeguards.
Violence concentrates in networks, moments and micro-places. A small-island strategy can focus resources precisely, but public hotspot data can stigmatise communities or identify people. Operational precision must be matched by privacy, proportionality and independent rights review.
Evidence-led findingTrack firearm cases from recovery to final outcome.
The relevant system runs from lawful recovery and digital chain of custody through ballistic examination, regional tracing, customs intelligence, police–prosecutor case conferencing and fair adjudication. Counting seizures without measuring trace completion and case progression leaves the mechanism unseen.
Evidence-led findingWhat the system could see
About two-thirds of measured crimes were not reported.
% of measured crimesShare of crimes in the 2020 National Crime Victimization Survey, covering incidents from September 2018 to August 2019. This is a historical official baseline—not a 2026 condition report.
SLPA policy proposal
National Safety Outcomes Compact
Judge public safety by harm prevented, trust earned, evidence converted into fair consequences and cases resolved without unnecessary delay—not activity counts alone.
One accountable national safety team
Give a small Violence Prevention & Justice Delivery Unit a Cabinet-approved outcomes mandate spanning police, justice, health, education, equity, youth, CSO and community partners, with an independence-preserving judiciary liaison.
National Safety Observatory
Join privacy-protected police, trauma, court, corrections, school and social-service data; repeat the victimisation survey and publish definitions, revision notes and a small annual trust pulse.
Local plans for high-harm micro-areas
In a few transparently selected high-harm micro-areas, combine focused lawful enforcement with outreach, mediation, hospital referrals, school-retention work, treatment and paid routes to income.
Track firearm evidence through the justice system
Standardise recovery, digital chain of custody, ballistics, regional tracing, customs intelligence and police–DPP case conferencing while publishing safe aggregate turnaround measures.
Permanent case-flow reform
Build on Swift Justice with early triage, disclosure protocols, procedural virtual hearings, witness support and public age-of-case, adjournment and pre-trial-time measures.
Delivery sequence
Set the outcomes, owners and baseline
- Designate the accountable delivery lead and publish a narrow national outcomes mandate.
- Sign privacy and data-sharing protocols, then issue a baseline using existing CSO and ECSC evidence.
- Map pending criminal matters by age and status and adopt a common crime-gun evidence protocol.
Run and evaluate the first safety pilots
- Publish the first quarterly safety scorecard with methods and revision notes.
- Run live referral pathways in selected micro-areas and commission an independent process-and-rights review.
- Prepare the field plan for a repeat national victimisation survey and audit backlog-court results by case age and outcome.
Expand proven safety pilots
- Expand place-based compacts only after harm, trust and rights outcomes are evaluated.
- Institutionalise the observatory and delivery architecture in law if the model proves useful.
- Publish prevention expenditure alongside enforcement and justice performance so resource choices are inspectable.
Public accountability
Measures for public accountability
Recommended publication: quarterly operating signals and one independently reviewed annual outcome report.Tests harm and whether residents believe the system is usable.
Shows whether recovered evidence becomes actionable intelligence.
Measures speed without confusing activity with justice.
Tests whether prevention changes life trajectories.
Limits, uncertainty & sources
Limits of this analysis
- Victimisation, reporting, safety-perception and firearm estimates describe 2018–19. They are essential baselines but not estimates of conditions in 2026.
- The survey marks some estimates as having moderate or low precision. A repeated survey should publish clear confidence information with every headline.
- Court filings and dispositions can concern different cohorts. The displayed ratios are flow signals, not backlog size, conviction rates or same-year completion rates.
- Small-area public data can re-identify or stigmatise people. The Observatory should publish safe aggregates and prohibit black-box predictive-policing labels.
Primary and institutional sources
01