The price of prevention
A cost-benefit analysis of the UK’s Prevent strategy
8 June 2025
In my previous article on Medium, I applied a standard metric used in clinical diagnostics — the Diagnostic Odds Ratio (DOR) — to assess Prevent’s capacity to distinguish between individuals who pose a genuine threat and those who do not. Drawing on publicly available data, and making conservative assumptions where necessary, the analysis revealed that Prevent performs only marginally better than chance. It identifies very few actual terrorists, while wrongly flagging hundreds of innocent people. The evidence suggests not only that Prevent is an ineffective predictive tool, but that its design may be fundamentally flawed — raising concerns about its impact on civil liberties, public trust, and the rational allocation of security resources.
To fully evaluate the effectiveness of the Prevent programme, however, it is essential to look beyond diagnostic accuracy and consider whether the benefits it delivers justify the substantial costs it incurs. A cost-benefit analysis (CBA) provides a structured way to assess this, by quantifying both the direct and indirect financial and social costs of the programme, and comparing these to the estimated benefits. Although data is limited and many assumptions are necessarily speculative, a fictional worked example — using conservative and reasonable estimates — can illustrate the scale and direction of these effects. A full breakdown of all the assumptions and workings in this analysis is available online here.
Costs
Let us assume a three-year period, from 2022 to 2024, for which we attempt to estimate total costs and benefits. In terms of direct public expenditure, Prevent receives significant central government funding, estimated at approximately £160 million over this three-year window. This includes operational budgets for referrals, Channel interventions, and the maintenance of regional Prevent coordinators. Added to this are staffing and training costs, conservatively estimated at £60 million, and administrative overheads of around £20 million. This brings the direct financial cost of the programme to £240 million.
However, the true cost of Prevent is not merely a matter of public expenditure. Its indirect costs are more difficult to quantify but arguably more significant. Over the same period, some 900 individuals may have been wrongly flagged under Prevent — a figure supported by publicly available data — incurring reputational damage, emotional distress, and in some cases legal consequences. If we assign a modest value of £10,000 per individual to account for the personal, educational, and professional disruption caused by such referrals, the total cost of false positives amounts to £9 million. Additional institutional costs — such as time spent by educators, police officers, and social workers on these false referrals — might plausibly add another £5 million.
Get Dylan Evans’s stories in your inbox
Join Medium for free to get updates from this writer.
Beyond these immediate consequences lie deeper social harms. Prevent is widely perceived by many communities, particularly among British Muslims, as stigmatising and discriminatory. This erosion of trust in public services may be conservatively valued at £20 million, while the long-term damage to community cohesion and feelings of political alienation might justifiably be placed at a further £15 million. The chilling effect Prevent may have on public discourse, particularly in educational settings, where students and teachers self-censor for fear of triggering referrals, adds an additional estimated cost of £10 million. Taken together, these indirect costs total £59 million. Combined with the direct financial cost of £240 million, the overall cost of the Prevent programme over three years may be estimated at £299 million.
Benefits
The question, then, is whether the programme delivers benefits sufficient to justify this expenditure. Here, assumptions must be clearly stated. Let us assume that Prevent successfully disrupted two serious terrorist plots over this period. Based on academic and governmental estimates of the economic cost of a typical large-scale terrorist attack in the UK — which includes emergency services, long-term medical care, property damage, and loss of productivity — a conservative figure of £75 million per attack is plausible. Preventing two such attacks would therefore save £150 million. If we assume that each of these attacks would have resulted in five fatalities, we may assign a further £20 million in benefit using the UK government’s own Value of a Prevented Fatality (VPF), currently set at £2 million per life.
Other benefits, while more difficult to price, should not be ignored. Early intervention in the case of vulnerable individuals — particularly young people at risk of radicalisation — might yield substantial lifetime benefits in terms of desistance from crime, social reintegration, and avoidance of imprisonment. Assuming twenty such successful interventions at a value of £500,000 each, this would amount to £10 million. To this we might add a modest figure of £5 million for general deterrence and public reassurance, and a further £5 million for any intelligence value Prevent might provide to the wider counter-terrorism infrastructure. These more diffuse or second-order benefits bring the estimated total to £190 million.
On these assumptions, the cost-benefit analysis produces a negative net present value of −£109 million. The benefit-cost ratio is 0.64: for every £1 spent on Prevent, society receives only 64p of benefit. These figures suggest that the programme is not cost-effective in its current form. Even when fairly generous assumptions are made about its successes in preventing attacks, the social and institutional costs it imposes are sufficiently high to outweigh the measurable benefits.
Sensitivity analysis
Of course, any such model is highly sensitive to assumptions. Were Prevent shown to have disrupted five attacks instead of two, the total benefits would rise substantially: £375 million saved in attack prevention, £50 million in the value of lives saved, and a further £20 million in broader social benefits. In this scenario, the total benefits would rise to £445 million, and the benefit-cost ratio would climb to 1.49. Prevent would then be cost-effective. However, there is currently no robust public evidence to support such optimistic assumptions. Conversely, if we underestimated the social harms — particularly the deep alienation and distrust the programme causes in minority communities — the costs could rise still further, deepening the negative value of the programme.
This worked example is illustrative, not definitive. But it highlights a serious concern: even on relatively conservative figures, the Prevent programme appears to impose substantial costs while delivering only modest and uncertain benefits. Without a clear and transparent demonstration that Prevent regularly thwarts attacks or successfully deradicalises at-risk individuals, its continuation as a publicly funded programme is difficult to justify on cost-benefit grounds alone.
When taken together with the earlier diagnostic analysis, which showed that Prevent’s Diagnostic Odds Ratio was only 4.44 — a value that reflects weak discriminative power — the cost-benefit perspective reinforces the conclusion that the programme is both inefficient and ineffective. Not only does Prevent fail to reliably distinguish between individuals who pose a genuine threat and those who do not, but the resources it consumes and the collateral damage it causes far outweigh the benefits it appears to deliver. Any serious defence of Prevent must therefore go beyond vague affirmations of its importance and instead engage with robust evidence demonstrating that its harms are justified by meaningful gains. At present, such a defence remains unmade.
