Something Changed at Kent Police: Black Arrests Soared 79% in One Year — Part Two

In Part One, we examined what happened on the streets of Dover, where hundreds of anti-migration protesters descended on the town, dressed in black and wearing balaclavas, roads around the Port of Dover were blocked and no arrests were made on the day.

That was one day, one protest and one set of operational decisions. It raised questions, but it couldn’t by itself tell us whether what happened in Dover formed part of any wider pattern.

Part Two therefore widens the lens.

We’ve interrogated five years of Home Office arrest data for Kent Police to ask a broader question: when ethnicity is brought into the equation, what does the force’s wider arrest record actually show? The Home Office publishes both its underlying arrest open data and a dedicated ethnic-disparity time series alongside its annual police powers statistics.

The answer isn’t neat.

The figures don’t prove Kent Police discriminates against Black people. Arrest statistics record the use of the power of arrest; they don’t, on their own, establish criminality, guilt or what ultimately happened to the person arrested. The Home Office itself treats arrest figures separately from subsequent criminal-justice outcomes.

But one change in the latest Kent figures plainly deserves an explanation.

Between 2023/24 and 2024/25, Black arrests recorded by Kent Police rose from 764 to 1,367 — an increase of 603 arrests, or 78.9%. White arrests rose by 9.7%, while total arrests across all ethnicity categories increased by just 5.1%. These figures come from our analysis of the Home Office arrest open-data and ethnic-disparity datasets.

At the same time, Kent’s Black-to-White population arrest-rate disparity jumped from 1.45 times to 2.37 times.

That is where Part Two begins.

Five years of Kent arrest data

When somebody is arrested, the Home Office says they are asked to define their ethnicity using the 2021 Census categories. For its headline analysis, these are grouped into White, Black or Black British, Asian or Asian British, Mixed, Other ethnic group and Not Stated.

The Home Office’s ethnic-disparity time series uses population data to express arrests as a rate per 1,000 people. For the latest five-year period, the population basis is the 2021 Census, allowing the same underlying population denominator to be used when comparing the force-area rates over those years.

A disparity rate then compares another ethnic group’s population arrest rate with the White rate. So, for example, a disparity of 2 means the arrest rate for that group is twice the White population rate.

Here is Kent, based on our analysis of the Home Office’s Arrests – ethnic disparity time series:

Something changed in 2024/25

Between 2023/24 and 2024/25, Black arrests went from 764 to 1,367. That is an increase of 603 arrests — 78.9% in a single year.

White arrests also increased, but nowhere near as quickly: from 15,244 to 16,720, a rise of 1,476 arrests, or 9.7%.

Total Kent arrests across all recorded ethnicity categories rose from 20,892 to 21,959 — an increase of just 5.1%.

So whatever produced the change in 2024/25, it wasn’t simply Kent Police arresting everybody at roughly the same increased rate. That is the first important feature of the data.

The population-based rates moved in the same direction. The Black arrest rate rose from 13.46 per 1,000 to 24.09, while the White rate increased much more gently from 9.27 to 10.17.

That pushed Kent’s Black-to-White disparity from 1.45 times to 2.37 times. Calculating the change from the underlying rates — 13.46/9.27 compared with 24.09/10.17 — gives a 63.1% increase in the disparity ratio.

It is the highest Black/White arrest disparity recorded in Kent during this five-year period.

But Kent’s national position tells another story

If we stopped there, it would be easy to paint Kent as a national outlier.

It isn’t.

We compared Kent with all 43 territorial police forces contained in the Home Office ethnic-disparity time series. The Home Office confirms there are 43 territorial forces in the England and Wales comparison.

Our comparison of those 43 forces shows that in 2024/25 Kent’s Black/White disparity of 2.37 was actually the ninth-lowest. More strikingly, in 2023/24 Kent had the lowest Black/White arrest disparity of all 43 forces. It was fifth-lowest in 2022/23, eighth-lowest in 2021/22 and fifth-lowest in 2020/21.

So the figures emphatically don’t support a claim that Kent is routinely one of England and Wales’s worst forces for Black/White arrest disparity. What they show is something rather different: a sharp deterioration in Kent’s own position during the of the latest year dataset available.

The national figures reinforce that point. The underlying Home Office time series gives the Black/White arrest disparity across England and Wales in 2024/25 as approximately 1.87 times, which the published statistical commentary rounds to 1.9. Kent’s 2.37 was therefore higher.

But remove the Metropolitan Police and the England and Wales figure rises to approximately 2.59 times, rounded by the Home Office to 2.6. Kent’s 2.37 sits below that measure.

There is a good reason for treating the Met separately. The Home Office says that in 2024/25 the Metropolitan Police recorded self-defined ethnicity as unknown in 61% of arrests, compared with 8% across the rest of England and Wales. Because the Met accounts for such a large share of arrests, that missing data materially affects the national comparison.

That makes the latest Kent movement more interesting, not less.

A force which had the lowest Black/White arrest disparity of all 43 territorial forces in 2023/24 moved to 2.37 just one year later.

There is an important caveat hiding in the figures

There is another issue which has to be taken seriously before drawing conclusions: self-defined ethnicity wasn’t recorded for every arrest. The Home Office specifically warns about missing self-defined ethnicity data when interpreting arrest figures.

In Kent, the number of arrests categorised as Not Stated was 14,357 in 2020/21, 5,401 in 2021/22, 2,772 in 2022/23, 3,561 in 2023/24 and 2,135 in 2024/25, according to our extraction of the Home Office arrest dataset.

That means around 62.0% of all Kent arrests in 2020/21 had no self-defined ethnicity stated. By 2024/25, that had fallen to 9.7%.

Better recording is obviously welcome. But it also means comparisons across the whole five-year period need caution. Some apparent movement between ethnic categories could reflect improvements in recording as well as genuine changes in arrest patterns. That problem is precisely why the Home Office discusses missing self-defined ethnicity separately in its statistical release.

Even the latest year requires that caveat. The Home Office says 8% of arrest records outside the Metropolitan Police were missing self-defined ethnicity in 2024/25. Kent’s 9.7% was therefore slightly higher.

More importantly, between 2023/24 and 2024/25 — the same year in which Black arrests jumped sharply — Kent’s Not Stated category fell from 3,561 to 2,135, a reduction of 40.0%.

We cannot simply decide that those better-recorded cases would otherwise have belonged to one particular ethnic category. There is nothing in the published data which allows us legitimately to make that allocation.

But Kent Police can reasonably be asked how much of the 2024/25 movement it believes is attributable to better ethnicity recording, and how much represents an actual change in arrest patterns.

The disparity isn’t evenly spread across offences

The Home Office also publishes arrest-level open data which allows the headline figures to be broken down by the primary offence associated with the arrest.

For drug offences in Kent during 2024/25, the dataset records 170 Black arrests. Using the same 2021 Census population denominator as the disparity series, that equates to about 3.00 arrests per 1,000 Black residents.

There were 1,037 White drug arrests, equivalent to roughly 0.63 per 1,000 White residents. On that measure, the Black drug-arrest rate was approximately 4.75 times the White rate.

Robbery produced an even larger relative difference. The data records 50 Black arrests and 220 White arrests, equivalent to approximately 0.88 and 0.13 per 1,000 respectively — producing a Black/White rate ratio of about 6.59 times.

But there is an important warning attached to that number. Fifty Black arrests is a comparatively small total, and ratios based on relatively small numbers can move substantially with relatively few additional cases.

Violence against the person gives us a considerably larger numerical base. Kent recorded 542 Black arrests and 7,496 White arrests. Those equate to approximately 9.55 and 4.56 arrests per 1,000, putting the Black population rate at about 2.10 times the White rate.

Those numbers tell us who was arrested for the recorded primary offence. They don’t tell us who committed more crime, why particular people came to police attention, or why the disparities exist.

One “arrest rate”, two very different things

There is another potential statistical trap.

Kent Police publishes what it calls an arrest rate within its stop-and-search figures. That is a completely different measure from the population-based arrest rates above.

Kent Police’s own 2024/25 statutory-duty report records 1,275 searches involving Black people, of which 289 resulted in arrest — an arrest rate of 22.7%. For White people it records 11,181 searches and 2,063 arrests — an arrest rate of 18.5%.

That doesn’t mean 22.7% of Kent’s Black population was arrested. It means 22.7% of those stop-and-search encounters involving Black people resulted in an arrest. Kent Police publishes these statistics specifically as outcomes following stop and search.

Two different denominators.

Two different questions.

What these statistics cannot tell us

Do these figures prove Kent Police discriminates against Black people?

No.

Saying that would go considerably further than the data permits. Arrest figures can be affected by reported crime, the locations where offences occur, proactive policing, operational priorities, deployment, demographic profiles and repeat arrests. The Home Office data describes police activity; it doesn’t by itself establish the cause of an ethnic disparity.

Nor does an arrest tell us the eventual criminal-justice outcome. The Home Office’s stop-and-search data, for example, expressly records a range of possible outcomes including arrest, caution, summons, community resolution and no further action, while the arrest dataset itself isn’t a conviction dataset.

An arrest is not a conviction.

But neither does that mean a Black/White disparity of 2.37 times, following such a rapid one-year increase, should simply be shrugged off.

The questions the data leaves behind

Why did Black arrests rise by 78.9% in one year when total Kent arrests increased by only 5.1%?

Why did the Black/White disparity increase by 63.1%, from around 1.45 times to 2.37 times, when our comparison of the Home Office time series shows Kent had recorded the lowest disparity of all 43 territorial police forces only one year earlier?

What explains a Black drug-arrest rate approximately 4.75 times the White rate?

How much of the latest movement can be explained by the 40% fall in arrests where self-defined ethnicity was Not Stated, and how much represents an actual change in arrest patterns?

Does Kent Police routinely analyse these disparities internally? If it does, what did that analysis conclude when the 2024/25 figures arrived?

Those aren’t accusations.

They are questions arising directly from official Home Office data.

And they matter because arrest is one of the most significant coercive powers the state can exercise against an individual.

There is also one final methodological reason why we’ve started this comparison in 2020/21 rather than reaching further backwards. The Home Office says Kent Police has only been able to identify the primary offence associated with an arrest since November 2018. Before that, where somebody was arrested for several offences, Kent could count every offence separately.

So comparing the present five-year series with much older Kent figures ris

ks comparing two different counting methods.

Part One began with what Kent Police did — and didn’t do — on the streets of Dover. The question it left behind was whether hundreds of Black protesters, similarly dressed, masked and causing the same disruption, would have been policed in precisely the same way.

Part Two shows that the wider arrest data doesn’t provide a simple answer. Kent isn’t routinely a national outlier. Indeed, our comparison of the Home Office figures shows that just one year before the latest figures Kent had the lowest Black/White arrest disparity of all 43 territorial forces. But something changed sharply in 2024/25, and the published statistics alone don’t tell us why.

The numbers don’t deliver a verdict on Kent Police. They give us questions which the force now needs to answer.

And that takes us to Part Three.

Because statistics can tell us what happened to Kent’s arrest rates. They can’t tell us whether anything inside the organisation might help explain the broader questions raised by this investigation.

In Part Three, we move from the numbers to Kent Police itself: racism and alleged racist conduct, officers suspended or dismissed, misconduct findings and what official inspections and other evidence have said about the force’s culture.

The Shepway Vox Team

Dissent is NOT a Crime