The Home Office reports that its asylum decision backlog has fallen by more than half, and Channel crossings are down sharply. Both statements are supported by official data. Neither describes the whole system.
On 27 August the Home Office published immigration statistics for the year ending June 2026. They record about 40,000 people awaiting an initial asylum decision, 56 percent fewer than a year earlier. Two weeks later, the Ministry of Justice published its quarterly tribunal statistics, which put the number of asylum appeals outstanding at 90,341. The first figure measures the queue for a Home Office decision. The second measures the queue for a judge to review that decision. Only the first has been celebrated.
This article sets the two numbers side by side, adds the latest data on small-boat crossings and the expiring UK–France returns pilot, and separates what the records establish from what remains unknown. The conclusion is not that the government has failed on every measure. It is that a system judged by its first stage alone can look healthy while the second stage fills up.
The Number the Government Leads With
The Home Office’s own summary states that there were 40,000 people awaiting an initial decision, relating to 32,000 cases, at the end of June 2026. The same release records that 16,000 supported asylum seekers, 17 percent of the total, were in hotel accommodation, half the level of a year earlier. Full Fact’s tracker puts the initial-decision backlog at 32,401 applications relating to 40,168 people, a 62 percent fall since June 2024.
These are real improvements, and a fair reading credits them. The cost of the hotel estate and the sight of unresolved claims were among the most visible failures of the previous several years. The Home Office was clearing cases faster. Summaries of the same release report asylum claims down 21 percent to 85,891 and a grant rate that fell to 38 percent from 48 percent the year before.
That last figure is where the story turns. A department that decides more cases and refuses a larger share of them will produce more appeals. This is arithmetic, not scandal, but it moves the pressure to a different institution.
Where the Cases Went
The Ministry of Justice’s Tribunal Statistics Quarterly for April to June 2026, published 10 September, covers the First-tier Tribunal (Immigration and Asylum Chamber). Its main figures, as summarised from the release, are that the chamber’s open caseload stood at 156,000, up 48 percent on the same quarter of 2025, and that the mean time to dispose of a case had risen to 65 weeks, about 13 weeks longer than a year earlier. For asylum and protection appeals specifically, the mean was 72 weeks.
The asylum-only picture, as analysed by the specialist publication Free Movement from the same MoJ data, is that outstanding asylum appeals rose from 87,450 to 90,341 over the quarter. Full Fact records that the figure was 33,227 at the end of June 2024. Across two years, the asylum appeals queue has therefore grown by a factor of about 2.7, over the same period in which the initial-decision queue shrank by roughly three-fifths.
The Migration Observatory at Oxford, in an April 2026 briefing, attributed the surge to "a combination of more initial decisions and a lower asylum grant rate." That is the mechanism described above, identified by an independent academic source rather than by this publication.
What the Figures Do and Do Not Show
Several cautions apply, and a serious reading must hold them alongside the headline.
The units differ. The Home Office counts people and cases; the tribunal counts appeals. The 40,000 and the 90,341 cannot be added together into a single "true backlog." Full Fact also notes that people awaiting an appeal may still be entitled to government support, which means they remain part of the overall asylum caseload even when they have left the initial-decision count.
The trend may be turning. Free Movement reports that new asylum appeals lodged fell from 14,749 to 11,596 in the quarter while disposals rose from 7,789 to 8,709, and that the quarterly increase in the backlog was the smallest in some time. Its author suggests the queue could begin to shrink next quarter for the first time in many years. That is a forecast, not a finding, but it is a credible one from a specialist source.
A high overturn rate cuts in more than one direction. The Observatory found that 39 percent of appeals receiving a substantive decision in 2025 succeeded, and the MoJ release summary reports a similar 39 percent for asylum and protection appeals in the latest quarter. Critics of the system will read that as evidence of weak initial decisions. Those who favour tougher enforcement will point out that most appeals fail. Both readings are available from the same number. What neither side can claim is that the first-stage figure alone tells them how well the system is working.
Some of the record is missing. Full Fact reports that the government has not published updated figures for cases pending further review, such as onward appeals or judicial review, beyond December 2022, when there were 4,051. A system that publishes the first stage monthly and the final stages rarely invites exactly the selective emphasis described here.
The Border: Real Progress, Unclear Cause
The crossings data deserve the same two-sided treatment. The Migration Observatory’s 10 September commentary reports 27,900 small-boat arrivals in the year to 9 September 2026, the lowest since 2021, and 16,300 between 1 January and 9 September, 47 percent fewer than the same period of 2025. ITV reported on 24 September a provisional year-to-date total of 18,565, down 42 percent from 32,188 at the same point in 2025, with 781 arrivals on 24 September, the busiest single day of 2026. The same report said two boats had landed undetected within two days, the first such case since September 2025. (The Observatory and ITV totals differ because they cover different periods and, in ITV’s case, a later provisional count.)
Who deserves the credit is harder to say. The Observatory lists possible policy drivers, including tighter family reunion rules, shorter temporary protection, a lower grant rate, enforcement against smuggling networks, the France returns agreement and stronger French interceptions. It also lists factors outside ministers’ control: irregular crossings into the EU fell 37 percent in the first half of 2026, and Syrian arrivals fell 67 percent after the fall of the Assad government in December 2024. Its own conclusion is that "it is difficult to isolate the effect of each factor." Any government or opposition claim that a single policy produced the decline goes further than the evidence.
The France Deal Ends Quietly
The "one in, one out" pilot, under which Britain could return small-boat arrivals to France in exchange for admitting an equal number of asylum seekers with family links there, was due to expire on 1 October. In a report carried on 30 September, Reuters quoted a French interior ministry source saying: "We are discussing the terms of withdrawal from the agreement with the British." It reported that 951 people had been returned to France under the scheme and 935 admitted to Britain, and that Interior Minister Laurent Nuñez wanted "a more European approach" and said France should not bear sole responsibility for an EU external border. The Observatory’s slightly later figure is about 1,100 returns between September 2025 and June 2026, around 4 percent of arrivals. Counts differ by date and source, and I have not been able to confirm the pilot’s final status after 1 October.
On any of these figures the scheme returned a small fraction of arrivals. Its significance was always partly symbolic: it was the first operating returns arrangement with France. Its lapse, if confirmed, removes the only such mechanism at the moment crossings are falling, which makes the durability of the decline an open question rather than a settled one.
Implications
The institutional lesson is about measurement. A department that publishes its own stage of a multi-stage process as the measure of success has an incentive to speed that stage and to leave the cost to the next one. RSN has documented similar patterns elsewhere in the Home Office’s operations: a fee regime that generates several times its processing cost, a High Court finding that immigration detention safeguards were routinely ignored, and a judicial review that forced the withdrawal of citizenship refusals under the "good character" policy. The courts, meanwhile, are strained across the board, as the Crown Court backlog shows. A tribunal that takes 72 weeks to hear an asylum appeal competes for the same scarce judicial and administrative capacity.
For readers who favour firm border control, the sober conclusion is that deterrence, fast and fair first decisions, and credible removal of those who finally lose are one pipeline. Falling arrivals and a shrinking first-stage queue are gains, but they do not demonstrate that people with no right to remain are leaving. For readers who favour a more generous system, the same data show that a 72-week wait for a decision is a hardship in its own right, and that a grant rate under 40 percent paired with a 39 percent appeal success rate raises hard questions about the quality of first-stage decisions.
What Accountability Would Look Like
Three things would let the public judge the system on its results. First, a single end-to-end series showing how many people who claim asylum in a given year are, at the end of the process, granted status or removed, published alongside the first-stage figures. Second, resumed publication of the onward-appeal and judicial review numbers that have been missing since 2022. Third, a plain statement from ministers of how they will tell whether the France pilot’s lapse affects crossings, so that the next set of figures can be read against an announced test rather than a retrospective explanation.
Until then, the fair summary is this: the front door is moving faster, fewer people are arriving by boat, and the queue has gathered in the tribunals. Both parts of that sentence are documented. Only one of them is being advertised.
Sourcing note: figures are taken from the Home Office and Ministry of Justice statistical releases and from the secondary sources linked above, which summarise those releases. Where I relied on a summary rather than the underlying spreadsheet, the text says so. Counts of UK–France returns vary by source and date.