How many people in the UK are victims of modern slavery? At present, we don’t actually know. There is no consensus on the answer to this question, despite the wide interest in finding it, and the tools and data to do so.
Over a decade ago, before the passage of the UK Modern Slavery Act 2015, data analysts estimated that there were between 10,000 and 13,000 victims of modern slavery in the UK. Since then, there have been four further estimates between 2014 and 2023, ranging from 8,300 to 136,000.
Why such a huge range? Estimates use different indicators and definitions of modern slavery, as well as different estimation methods.
Several parliamentary inquiries have now been conducted focusing on how the UK can strengthen its response to modern slavery. Part of an effective response is understanding the nature and extent of the issue. And this understanding relies on better methods to produce statistically robust estimates of the scale of the problem.
The UK Independent Anti-Slavery Commissioner asked our team of researchers at the University of Nottingham’s Rights Lab to carry out a scoping review of modern slavery estimates.
To do this, we reviewed 46 published studies and 57 data sources, conducted a survey and held workshops with public sector leaders. We identified several methods of estimating modern slavery prevalence in the UK. These can be used depending on the type (or types) of modern slavery and populations that are of interest.
Our report also details how the UK might use existing data in novel and innovative ways, such as producing detailed modern slavery risk and vulnerability maps.
‘Hard to find’ populations
Any robust and statistically sound attempt to estimate the number of victims of modern slavery in the UK confronts what methodologists call “the fundamental problem of unobservability”.
Victims and survivors are a seldom heard, often hidden population comprised of both UK and foreign nationals. The use of force, fraud and coercion that underpin modern slavery are often difficult to detect. They are practices that, while directly experienced, are often not easily observable.
Our review examined studies from around the world that address this problem of hidden populations. Most of the studies we examined carefully follow data collection and data analysis principles to produce robust prevalence estimations for the whole world, for specific countries and cities.
The remit of our review was not to produce a new estimate, but to identify promising methods of estimation. Our findings show that the best methods to date are either “multiple systems estimation”, which analyses multiple overlapping administrative lists of victims, or some combination of sampling and carefully designed surveys. Both involve collecting and analysing data, and draw careful inferences from the data in making their estimations.
For the UK, the original estimation from 2014 adhered most closely to standard principles of data collection and analysis. But since the passage of the Modern Slavery Act in 2015, the picture has changed. The number of offences that qualify as modern slavery has expanded considerably in that time and thus a new estimate is much needed.
While sampling and survey approaches used in other parts of the world produce sound prevalence estimations, no such study using these methods has been done in the UK.
Data could be used to create maps that help governments and organisations end modern slavery.
buradaki/Shutterstock
Why counting matters
Official statistics from the National Referral Mechanism (NRM) – the UK government’s framework for identifying victims of modern slavery – show that there have been 21,285 potential victims identified between July 2024 and June 2025. However, these referrals are only for people who are known and have been formally identified. The true number, should a new estimation be produced, is likely to be much higher.
We also know that the referral process itself is highly skewed. Those who took part in our survey and workshops explained that identification varies considerably across police jurisdictions, nationalities and types of offence. These views are corroborated by a new report from the anti-slavery charity Unseen.
While providing a strong foundation, the NRM remains a “convenience sample” from which prevalence estimations and statistical inferences currently remain limited.
Our review argues that a slight reform to how data is recorded in the case management system – multiple referrals for the same person should be maintained and not merged into the same record – would enable analysts to provide an up-to-date estimate using multiple systems estimation.
This, along with other studies that focus on particular modern slavery practices for specific subpopulations in the UK, would provide strong evidence on the true number of modern slavery victims. Such analysis would be of benefit to policymakers, law enforcement, academics, charities and survivors themselves.