Immigration in NI: Does Belfast provide homes to more “illegal migrants” than every other UK city except Glasgow?

• This list relies on a misreading of new Home Office data. The main problem is that the list is wrong. London should be at the top.
• The relevant figures concern asylum seekers, not “illegal migrants”. Claiming asylum is not illegal. Other issues include that the data does not count the number of “homes” provided to people, it counts the number of people allocated a home.
• The claim does not make clear that the data only counts certain asylum seekers.
• The figures underpinning this list exclude everyone using the major schemes to resettle people from Ukraine and Afghanisation, as well as all asylum seekers currently being put up in hotels.

On 21 August, a social media user calling themselves “Basil the Great” claimed in a post on X (formerly Twitter):

This isn’t accurate, for several reasons.

First of all, London should be at the top.

While the claim refers to cities, the shared list matches Home Office data for local authority areas with the highest number of asylum seekers currently housed in “dispersal accommodation” (see below for an explanation) and excluding those using the dedicated Ukraine or Afghanisation asylum schemes.

However, London is made up of 33 different local authority areas and the figures for these need to be added together to get a figure for the UK’s biggest city.

Beyond this, the original post has a few mistakes and misconceptions and could use some further context:

• The underlying data does not count the number of houses allocated to asylum seekers, it counts the number of asylum seekers provided with somewhere to live. Given that it is possible for several people seeking asylum to share a house or flat or hotel room, these two figures can be different.
• These numbers, as noted above, do not include anyone claiming asylum via the Homes for Ukraine or Afghan Resettlement schemes.
• Furthermore, these numbers only include asylum seekers staying in “dispersal accommodation”, such as private rental accommodation, and excludes all asylum seekers in “contingency accommodation” such as hotels. While one might argue that a house is a home in a way that a hotel is not, on the other hand someone’s home is simply the place they live. The original post does not explain this clearly.

For a deeper explanation of all of this, and a closer look at asylum figures for Northern Ireland specifically, read on.

This claim also gives us a good example of how an AI tool – in this case Grok – won’t necessarily provide the full picture in response to a query.

• Source

FactCheckNI made efforts to contact Basil the Great but, at the time of writing, had received no reply.

• Misconceptions and misinterpretations

Figures from the list posted by Basil can be found amid Home Office data published on 21 August as part of the UK Government’s quarterly immigration system statistics, which includes data correct at the end of June 2025.

However, the names listed in the claim are in fact the ten local authority areas with the most current asylum seekers (discounting those involved in either the Homes for Ukraine or Afghan Resettlement schemes) placed in dispersal accommodation (more on that later).

Figure 1 – UK local authority areas with the most asylum seekers (excluding those on the Ukrainian and Afghan schemes) housed in dispersal accommodation [source: Home Office]

Before we get onto some of the further problems with the claimant’s interpretation of this data, we should first point out that asylum seekers are not “illegal migrants” – claiming asylum is, of course, not against the law (we’ve covered this previously). 

Secondly, this list in the claim does not count the total number of homes or houses or dwellings (again, more on that later) that are in use by asylum seekers, instead it shows the number of asylum seekers (excluding those using the dedicated Ukraine or Afghanistan programmes) in those houses.

As several asylum seekers could be sharing accommodation, there is a risk of counting the same dwelling on more than one occasion.

• A closer look at the detail

Another problem with the list is that it is wrong.

The city with the most asylum seekers in dispersal accommodation is not Glasgow, it’s London.

Figure 1, above, shows that Glasgow City is the single local authority area with the most (non-Ukraine scheme, non-Afghan scheme) asylum seekers in dispersal accommodation, and by quite a distance, with 3,523 in total.

However, as London comprises 33 separate local authority areas, to calculate the number of asylum seekers (excluding the Ukrainian and Afghan schemes) in dispersal accommodation in the whole city requires adding together the figures for those 33 local authorities. In total, London has 4,350 asylum seekers in such circumstances.

Using per capita figures would be a much more appropriate way to compare statistics across different sized local authority areas.

Glasgow City has a much smaller population than London – which, according to the ONS’ data portal Nomis, is home to almost nine million people – which means that per capita it has a much higher number of asylum seekers in the circumstances described above. 

The definition of a ‘city’ is a rabbit hole that FactCheckNI has jumped into in a previous article: we’ll not repeat that exploration here, but even simple terms like ‘city’ can have complicated definitions.

• Context: what is dispersal accommodation?

The data used for this list only looks at a certain subset of current asylum seekers.

Firstly, it excludes those using either the Homes for Ukraine or Afghan Resettlement schemes – large scale programmes that were established due to the 2022 Russian invasion of Ukraine and the Taliban’s return to power in Afghanistan in 2021 respectively.

Secondly, it only considers a certain type of accommodation.

While awaiting the results of their application, asylum seekers can ask for asylum support. If they are deemed eligible for asylum support, they may be provided with somewhere to live, a cash allowance, or both.

Any asylum seekers who receive this support and are allocated a place to live will find themselves in one of three types of accommodation:

• Initial accommodation – according to the Home Office notes accompanying the data, this “is provided to asylum seekers who have indicated that they are unable to support themselves or their families and are therefore at risk of destitution. It is made available in order to provide shelter while a request for asylum support is being assessed. Occupants of Initial accommodation who receive a positive outcome to their support request generally move to Dispersal (longer term) accommodation when suitable property becomes available.”
• Dispersal accommodation – “[this] is provided to Asylum Seekers whose claim for Asylum Support has been agreed. Occupants of Dispersal accommodation are permitted to stay there whilst they remain eligible for Asylum Support.”
• Contingency accommodation – “[this] is temporary accommodation (including hotels) used when there is insufficient Initial or Dispersal accommodation available. People housed in contingency accommodation generally move to Dispersal Accommodation when suitable property becomes available.”

The data also mentions a fourth “Other” category of accommodation. The only local authority area where this is relevant is Braintree in Essex, where 653 people currently live in such accommodation. Braintree is home to Wethersfield Airfield (formerly RAF Wethersfield) which is used to house asylum seekers.

Given the nature and focus of ongoing debates about immigration, it is useful to note that the list in the claim looks exclusively at dispersal accommodation and so excludes all asylum seekers currently placed in hotels.

The number of people staying in initial accommodation is generally low but, depending on the area, the number of people staying in dispersal accommodation (such as private rental housing) can either be much higher or much lower than those in contingency accommodation (such as hotels).

Regardless, to get the full number of asylum seekers staying in a local area you need to add together the numbers staying in each type of accommodation and also include those involved with Homes for Ukraine or Afghan Resettlement.

Of the 20 local authority areas with the most asylum seekers housed in hotels (discounting the Ukraine and Afghan schemes), ten are boroughs of London.

The Birmingham council area has the fifth highest number of asylum seekers staying in dispersal accommodation (1,389) and also the fourth most in contingency accommodation (1,226).

Home Office figures show that as of 30 June 2025, Belfast currently has no asylum seekers supported to stay in hotels (although some other Northern Ireland council areas do: Causeway Coast and Glens has 124, Antrim and Newtownabbey has 78, Ards and North Down has 41, Derry City and Strabane has three).

This important context is left unexplained in the claim. Although it could be argued that “homes” do not include hotels, it could also be said that someone’s home is simply where they live at a given time.

On the other hand, it could be argued that all these types of accommodation lack permanence so none are really “homes”. All of the above represents temporary accommodation that is only provided for the duration of the asylum claim. According to Home Office advice for asylum seekers: 

“Once you have received your decision, you will have 28 days to leave the accommodation if you have been granted Leave to Remain (LTR), and 21 days if you have had a negative decision.”

Use of the word “homes” in the claim is unclear, rather than incorrect – explaining exactly what this list includes and excludes would allow anyone who reads it to understand it much more clearly.

• Context: relative populations

Some local authority areas have bigger populations than others. The original post does not suggest in any way that its list is on a per capita basis, which is fine, but that does not mean that a per capita analysis would not be useful.

The Home Office’s data tables include, for each local authority area, the percentage of the total population that is currently claiming asylum.

This includes all asylum seekers – those on the Ukraine or Afghanistan schemes, and all further supported asylum seekers including all accommodation types and those receiving subsistence help only.

When considering all asylum seekers as a percentage of the total population, the top ten local authority areas include seven London boroughs.

Figure 2 – source: Home Office

Out of the 361 local authority areas in the Home Office data, Belfast has the highest percentage of asylum seekers as part of its population in Northern Ireland and has the 32nd highest such percentage across the UK.

Figure 3 – source: Home Office

The rest of Northern Ireland’s local authority areas have relatively low proportions of asylum seekers.

In Derry City and Strabane, 0.26% of the population (roughly one in every 400 people) is an asylum seeker. This is 260th out of 361 total UK local authority areas.

Mid and East Antrim, where 0.13% of people are claiming asylum – just over one in every 800 people – is 349 out of 361. In other words, it has the 13th lowest proportion of asylum seekers out of all local authority areas in the UK.

• A word on AI

This claim was posted on X, formerly Twitter, and one of the replies to the post asked Grok – X’s own in-house AI – to fact check the original post. Grok’s first reply stated:

“The latest Home Office data (June 2025) on supported asylum seekers in accommodation shows top local authorities as: Glasgow City (3,844), Birmingham (2,755), Hillingdon (2,481), Liverpool (2,361), Manchester (1,997). The listed cities don’t fully match this; no specific data found on “homes given to illegal migrants.” Sources: http://GOV.UK immigration stats.”

A different user asked Grok to explain what it meant by “don’t fully match this”, to which Grok responded:

“I meant the claimed list doesn’t fully align with official Home Office data on supported asylum seekers as of June 2025. Official top 5: Glasgow (3,844), Birmingham (2,755), Hillingdon (2,481), Liverpool (2,361), Manchester (1,997). The post’s list has mismatches like Belfast  (not in top 5) and omits others. Source: http://GOV.UK stats.”

That same user responded again, asking for the “actual figures”, but received no reply.

Say what you like about the efforts of Basil the Great to handle this Home Office data – and this article has said plenty – but ultimately the claim does originate from a real list featuring real data and hasn’t just been made up.

Grok has not managed to use data as wisely. Instead, it appears to have pulled information from a different column in the same spreadsheet (or, if it didn’t, it found that same data elsewhere) and announced that the lists don’t match (that list, by the way, is the ten local authority areas with the highest total number of supported asylum seekers in any form of accommodation – but still excluding the Ukraine and Afghanistan schemes…).

Ultimately the ability to critically think remains useful for any form of analysis and thinking is something large language models simply do not have the capacity to do.