Are Small Boat Migrants Really 24x More Likely to Go to Prison?

Spoiler: No. But here’s how the media got suckered by bad data and worse assumptions.

Jul 05, 2025

The Conservatives have claimed migrants from small boats are “24 times more likely to end up in prison”, but this figure is wrong on so many levels!

On Thursday, I was sent this headline and asked if it was true…

It took a lot of digging to get to the bottom of it!

Alberto Brandolini wasn’t kidding when he said:

“The amount of energy needed to refute bullshit is an order of magnitude bigger than that needed to produce it.”

The claim

The Times were the first to report on this story, shortly followed by the Express, The Daily Mail, and The Telegraph, before it became a talking point on GB News and Talk TV. And the only source for the whole story was… the Conservative party.

There was no link to the analysis or any kind of report, just “Analysis by the Conservative Party”. From listening to the GB news interview, it would seem that Chris Philp, Shadow Home Secretary, might have been responsible.

The claim being that nationalities that are found in small boat crossings are massively over-represented in the UK prison population, and therefore those arriving on small boats will be 24 times more likely to end up in prison.

We’ll dig into why this analysis doesn’t hold up later on, but first, we’ll try and fathom the numbers used!

With no report to read and no sources given, I had my work cut out trying to establish where the numbers even came from, in order to work out if they were sensible.

The data

Eventually, I worked out that they used three sources of data:

The government’s Offender Management Statistics for March 2025, which breaks down the England and Wales prison population by nationality
The 2021 Census data on how many people held passports from different countries
The government’s immigration statistics on irregular arrivals (mainly small boat crossings)

Let’s start with the prison stats.

The numbers used in the article are correct:

76,663 British nationals
10,838 foreign nationals

They then picked 4 of the top 15 nationalities found in small boat arrivals - highlighted in blue.

You’ll be able to decide why you think they chose Somalia, despite it representing only 1% of crossings, a bit later on!

Data from The Migration Observatory

They found the correct prison numbers for:

Albania1,246
Iraq 281
Somalia 258
Iran 257

So, the raw prison stats were correct.

Where their analysis went haywire was in what they used as the base population for each country.

Because to determine that “12% of all Somalians in the UK are currently in prison” (Yup, that was the claim!), we need to establish what the total population of Somalians in the UK is.

I got lost down many rabbit holes trying to work out how on Earth they calculated that 12% figure, because that would suggest there were only just over 2,000 Somalians in the UK, and that is clearly not the case.

The clue came in the article when it said:

The analysis is based on the 10,838 foreign criminals — those living in the UK who, according to the latest census in 2021, hold a foreign passport and do not have a British passport — in prisons in England and Wales at the end of March.

Now it’s worth noting that the government prisoner stats do NOT use passport holders to determine nationality, they use self-reports from the prisoners.

So the article is incorrect in its definition of ‘foreign criminal’.

It is also worth noting that the prison population statistics include all those in prison, whether they’ve been convicted or not. For British prisoners, 82% have been convicted, but for foreign prisoners, only 61% have been convicted. The rest are on remand or not criminal.

So again, these numbers are not all ‘foreign criminals’.

Onto the passport data

During the 2021 Census, they collected data on what passports people held on that day, and THIS was the baseline data used by the Conservatives in their analysis

‘United Kingdom Passport’ PLUS ‘No passport held’ formed the baseline number for British nationals (53,708,125)
All other passports formed the baseline for Foreign nationals (5,889,415)
Iraq 10,237
Somalia 2,130
Iran 16,462

However, the ONS data only gives data on around 40 individual countries, and Albania isn’t one of them, so I can’t work out what they used as their baseline for Albanians.

Now, the eagle-eyed amongst you might have noticed that 2,130 Somalian passports is a really low number compared to the others.

In fact, of all the data we have on passport holders from individual countries, it is the lowest number!

If you have a really small baseline, you can make a really high percentage.

Do we think there’s a reason they chose to include Somalia now, given it makes up only 1% of small boat arrivals?

So, that gave us:

British 76,663 / 53,708,125 = 0.14%
Foreign 10,838 / 5,889,415 = 0.18%
Albania 1,246 / ??? = 6%
Iraq 281 / 10237 = 2.7%
Somalia 258 / 2,130 = 12%
Iran 257 /16,462 = 1.6%

What’s wrong with that baseline?

As we know, the right-wing media are usually very keen on overinflating immigrant numbers, generally including anyone born in another country, sometimes including people whose parents were born elsewhere, and occasionally including entire households that contain a foreigner.

So it might come as a shock to discover that for this ‘analysis’, they were only interested in counting those who held a foreign passport in 2021!

This excludes:

Everyone born abroad but who has lived here for 5 years and been able to gain British citizenship
Anyone with dual citizenship
Anyone who arrived AFTER the 21st March 2021, when the census took place - that’s the section in blue!
From the Migration Observatory

Remember, the smaller the baseline population, the bigger the percentage number you can achieve for your headlines….

What do the foreign-born stats show?

There are several methods for measuring foreign populations living in the UK. The census collects national identity, passports held, ethnicity, and country of birth. I believe most people’s instinctive go-to when they think ‘Kenyan’ or ‘French’ would be the country of birth. And it’s certainly the most used one in the media.

So, for consistency, let’s use the 2021 census data for country of birth:

British 76,663 / 49,579,570 = 0.15%
Foreign 10,838 / 10,017,980 = 0.11% (not 18%)
Albania 1,246 / 68,672 = 1.8% (not 6%)
Iraq 281 / 89,393 = 0.31% (not 2.7%)
Somalia 258 / 108,921 = 0.24% (not 12%)
Iran 257 /109,168 = 0.24% (not 1.6%)

Yes, there do appear to be higher percentages of people in prison from those 4 countries, but closer to 2x rather than 24x!

Please note, that while the 2021 census showed 108,921 people were born in Somalia, it estimated a Somalian population in the UK of 176,645 based on ethnic group, national identity or both. Using this figure would drop the percentage further.

And if we were to look at best estimates for those populations now, in 2025, those percentages would drop even further, but my brain is fried and I assume yours is too at this point! You get the idea.

So we’ve established that the 24x figure is ludicrous on multiple fronts:

It compares a prison population that self-reports their nationality with a base population based solely on passport ownership
It ignores all foreign-born people who have gained a British passport, or who have dual nationality
It ignores the couple of million immigrants who have arrived here since 2021!

By using the smallest possible numbers for their baseline, they’ve managed to create crazy sounding percentages for these chosen countries.

OK, but would the analysis be sound if they’d used more accurate numbers?

In short, no, not really.

You cannot assume that the 258 Somalians in prison in 2025, from an estimated 2021 UK Somalian population of 176,000, must have come from the 1,356 Somalians that have arrived by small boat in the last 7 years!
No attempt has been made here to actually find out the criminality rate of small boat migrants by studying how many people in prison arrived that way.
No attempt has been made to adjust for the fact that those populations tend to be much younger (a key driver of crime) when compared to the British population, which spans 120+ years!
No attempt has been made to compare the crime types, sentence types/lengths, and whether they’ve even been convicted yet between the British prisoners and the prisoners from those specified countries.
Nor do we know whether those prisoners were long-term UK residents, here on a visa, or students as opposed to asylum seekers or refugees.
Just because a nationality is overrepresented in both boat crossings and prisons doesn’t mean they are the same individuals. This is a classic example of what’s known as an ecological fallacy: assuming that patterns seen at the group level (e.g., nationality) apply to specific individuals (e.g., asylum seekers arriving by boat).

As for whether this analysis shows that:

“…some 3.4 per cent of small boat migrants could go to prison, a rate that is 24 times higher than the average for British citizens and 18 times more likely than the average migrant in the UK. It would mean that about 700 migrants who arrived on small boats this year could end up in prison if the rate is applied to the 20,422 people who have crossed the Channel so far this year.”

No, it doesn’t.

The ‘3.4%’ claim is not based on any actual tracking of small boat migrants. It’s a reverse-engineered guess using the prison stats, with a deeply flawed denominator, projected onto migrants arriving in small boats.

The prison stats are a snapshot of people in prison on the 31st March 2025 and cannot be used as a way of tracking what percentage of any cohort will later be imprisoned.

To conclude

As you can see, the claim that migrants arriving on small boats are 24x more likely to end up in prison was an easy manipulation of available statistics. It wasn’t entirely fabricated - they can point to ‘official statistics’ to claim credibility - but it’s clearly misinformation all the same.

A Tory MP briefed it to a right-wing newspaper, which published it unquestioningly. That was picked up by the rest of the right-wing media ecosystem, and now the 24x figure is firmly in the minds of those who want to believe it, and is being repeated all over the internet, and across dinner tables and garden fences.

Meanwhile, it’s taken me many hours to debunk the claim, write it up, and share my workings across social media, and not even a fraction of those who’ve seen the original claim will ever see the debunk.

The newspapers won’t cover it, and they certainly won’t correct themselves, and so the lie sticks.

And all the journalists had to ask themselves was, “If 258 Somalian prisoners account for 12% of the UK Somalian population, then that means there are only ~ 2,000 Somalians in the UK. Does that seem plausible?”

That one sniff test would have told them they were being fed a flawed analysis.

But they chose not to/didn’t bother.

And now there is yet another deeply flawed statistic floating around that demonises asylum seekers in an already divisive media and political ecosystem.

It is so frustrating and tiring……

As ever, when you see a stat that sounds implausible or that is clearly designed to make you feel something - anger, rage, fear - take a minute to question it. Ask who is trying to manipulate you and how they might have done that?

And then, come and ask me to debunk it! 😆

Although I might have to start charging - personal debunker, anyone?!

Having said that, thank you so much to those who pay for their subscriptions, or who buy me coffees☕️.

These debunks take a lot of time and effort, and while I primarily do them because I can’t help myself when I see a misleading headline(!), knowing I have a small amount of income coming in to cover some of that time really helps!