The Court That Marked Its Own Homework
Britain’s asylum tribunal is drowning in cases the Home Office itself got wrong. The government’s solution is not to get more of them right. It is to abolish the people who kept noticing.
THE RATIONALS
The Home Office’s own inspectors found that nearly four-fifths of the asylum grants they examined were insufficiently evidenced and likely incorrect. The government’s response has not been to fix the casework. It has been to abolish the tribunal that keeps catching the mistakes and replace it with a body that need not employ lawyers — appointed, in the first instance, by the Home Office itself. One almost has to admire the confidence.
Here is a fact that did not make the front pages. Last autumn, Home Office officials checked their own work on asylum casework. Not the refusals, where scepticism is unremarkable. The grants — the decisions where the department had already agreed the claimant could stay.
In a sample examined by the government’s own borders inspector, 78.72 per cent were found to contain insufficient evidence to support the decision that had actually been made, and were “therefore likely to be incorrect.” Not “arguably borderline.” Not “a matter of judgement.” Insufficiently evidenced, by the department’s own reviewers, in cases where the department had already said yes.
Most people who follow this subject have absorbed one of two stories, Home Office decisions are too harsh, or not harsh enough. Almost nobody has been told the more basic problem. A comfortable majority of the department’s decisions, in either direction, cannot reliably be shown to follow from the evidence in front of the caseworker.
That is not a story about generosity or toughness. It is a story about competence. And it is the story the government would very much prefer stayed buried under the more familiar argument about numbers.
It did not stay buried. It surfaces, in bureaucratic language, across an inspector’s report, an auditor’s findings, and the department’s own internal reviews. The government’s response to finding all this out about itself has not been to fix the casework. It has been to introduce a Bill that abolishes the court currently catching the mistakes, and replace it with a body staffed by people who need not be lawyers — recruited, in the first instance, by the Home Office itself.
One almost has to admire the confidence.
This is the story of the Immigration and Asylum Bill, currently working its way through Parliament, and of the tribunal backlog it claims to be solving. It is a story with a government’s own auditors as the chief prosecution witnesses, a Law Society president delivering the driest line of the year, and a ministerial impact assessment that declines to vouch for its own policy. None of it requires embellishment. That is rather the point.
The received version of this backlog, repeated by ministers and much of the press with equal confidence, runs roughly as follows, small boat crossings and rising claim numbers have overwhelmed a court system built for a smaller caseload.
The sensible response, on this account, is faster processing, tougher enforcement, and — eventually — a leaner appeals mechanism to match the volume. It is a tidy story, and a politically convenient one. It requires nothing uncomfortable to be said about the Home Office’s own performance.
It is also, on the government’s own figures, substantially incomplete. Start with what is not in question — both sides of this argument agree on the raw numbers, even as they draw opposite conclusions from them. The tribunal that hears asylum appeals — a specialist court called the First-tier Tribunal’s Immigration and Asylum Chamber — saw its open caseload climb by 68 per cent to 152,000 over a single year, driven by a 49 per cent surge in incoming appeals.
That much sounds consistent with the volume story, too many cases, not enough courtroom. Except the tribunal’s own disposal rate — the pace at which it actually got through cases — rose by 39 per cent over the same period.
The system did not merely fail to improve. It sped up substantially and still fell further behind. New appeals were arriving faster than any plausible increase in judicial capacity could keep pace with.
That is the detail the volume narrative cannot accommodate. If the tribunal were simply under-resourced relative to arrivals, faster processing would have narrowed the gap. It didn’t.
Which raises the unglamorous but unavoidable question the received story is built to avoid, where, exactly, are all these appeals actually coming from, if not simply from more people arriving?
The answer, inconveniently for the department responsible, is the department responsible.
The government has its own watchdog for this — the Independent Chief Inspector of Borders and Immigration, whose job is to check Home Office work and report back to Parliament. That inspector published a report this summer describing the state of asylum casework in terms that would sound intemperate coming from a campaign group, and are harder to wave away coming from a government watchdog.
The finding, somewhere between 54 and 62 per cent of Home Office asylum decisions in the twelve months to October 2025 met the department’s own “satisfactory” quality bar. The gap between those two figures is a dispute between the inspectorate and the Home Office about how to do the sums.
The target, for context, was 75 per cent. It had not been met “for some years,” in the inspector’s own words.
The National Audit Office — Parliament’s independent spending watchdog — checked the same system from a different angle and arrived at a similarly unflattering figure, 42 per cent of sampled decisions in the rolling twelve months to May 2025 contained significant or outright failing errors.
The Home Office’s own 2023/24 data showed a comparable failure rate, only 52 per cent of decisions met its internal quality standards (down from 72 per cent the previous year), meaning 48 per cent did not. An organisation marking its own work, and still failing it roughly half the time.
It gets more instructive still. Recall the statistic that opened this piece, in a sample of grant decisions examined last autumn, 78.72 per cent were found to be insufficiently evidenced. That exercise deliberately targeted grants rather than refusals, on the reasonable assumption that scepticism bites hardest when a claimant is turned away. It found the opposite of reassurance. This is worth sitting with a second time, now that the wider pattern is visible.
The problem in Home Office casework is not an excess of generosity being reined in by vigilant judges. Nor is it an excess of harshness being softened by sympathetic tribunals. It is a general and fairly comprehensive inability to produce decisions — in either direction — that can be shown to follow from the evidence in front of the caseworker.
Staff surveyed for the inspector’s report put it more bluntly than any minister has, the prevailing culture rewards the quantity of decisions made over the quality of the reasoning behind them.
The downstream consequences are exactly what you would expect. Roughly half of Home Office asylum refusals that reach appeal are overturned. In 2024, the Home Office’s own Pre-Appeals Review Unit — a team that reviews cases just before they reach court, and quietly withdraws the weak ones rather than risk losing in public — pulled a large number of decisions before they ever reached a tribunal.
Of those withdrawals, 74 per cent happened because the original decision had been wrong or partially flawed. That is not an appeals system catching the occasional error at the margins. That is an appeals system functioning, in large part, as the department’s unpaid quality-control desk — correcting mistakes a competent casework operation ought never to have made in the first place.
The backlog, in other words, is not principally a story about migration volumes, however politically convenient that framing is. It is a story about an institution generating its own workload through the sheer, repeated inadequacy of its first attempt.
Faced with this diagnosis — its own diagnosis, delivered by its own inspectorate and its own quality checks — the government’s remedy is not to fix the casework. It is to abolish the court.
Put simply, the Bill treats the symptom — the swollen appeals backlog — while leaving the disease, unreliable first-instance decision-making, entirely untouched. A tribunal processing fewer cases because there is no longer a functioning tribunal to process them is not the same thing as a Home Office making better decisions.
The Immigration and Asylum Bill strips the First-tier Tribunal of its immigration and asylum jurisdiction entirely. In its place comes a new body, the Independent Immigration Appeals Authority. Its adjudicators will NOT need any legal qualification.
In its early stages, recruitment will run through the Home Office itself, rather than the Judicial Appointments Commission — the independent body that normally handles judicial appointments, precisely to keep the executive’s fingers off the scale.
The Home Secretary appoints the Authority’s chair and first chief executive. The chief executive then appoints the adjudicators who will decide whether the department’s own casework survives scrutiny.
Shabana Mahmood’s defence of this arrangement, offered to a bruised House of Commons, was that important public decisions are routinely made by non-lawyers. She cited magistrates, parole board members, planning inspectors. A person does not have to be a judge, in her words, to have good judgment.
It is a perfectly coherent point in the abstract — magistrates do indeed decide serious matters without law degrees. But magistrates are appointed through an independent process, insulated from the prosecuting authority whose cases they judge.
Asylum adjudicators, under this Bill, would be recruited by the very department whose decisions they are being asked to mark. The analogy holds right up until the moment you ask who is doing the appointing. Then it rather falls apart.
There is a further wrinkle worth noting. Claimants will be required to bring every ground of appeal at once, in a single hearing, with no second attempt if a new matter emerges afterwards. It is a one-shot process, and it raises the stakes of getting the first hearing right — precisely as the government lowers the qualification bar for who conducts it.
The Law Society’s response was the most economical demolition on offer. Understatement rarely lands this well, the Home Office, its president observed, is proposing a system to “mark its own homework.”
The Immigration Law Practitioners’ Association went further, calling the reform “an attack on justice and the rule of law” — a phrase that would read as hyperbole from most advocacy groups, and reads rather differently from a body of working immigration lawyers watching their profession’s oversight function get legislated away.
The Bar Council told MPs plainly that immigration appeals ought to be decided by legally qualified adjudicators, putting it on a direct collision course with a home secretary determined to press ahead regardless.
Here is the detail that elevates this from ordinary Westminster argy-bargy into something closer to a punchline.
When Lord Anderson of Ipswich pressed the government on the value of its own flagship reform before a House of Lords committee, he cited the government’s own impact assessment back at it. An impact assessment is the document a department publishes to justify a policy to Parliament — its own case for its own Bill. This one rates the expected value of the Independent Immigration Appeals Authority as “uncertain.”
Sit with that for a moment. A government is asking Parliament to abolish an established, judge-led tribunal and replace it with a body its own paperwork will not confidently endorse. This is not critics disputing the official assessment. This is the official assessment declining to make the case ministers are simultaneously making next door.
Mahmood has defended the reform in more combative terms, telling the same Lords committee that a system able to “frustrate legitimate removals” does not itself resemble justice.
An appeal now averaging 60 weeks is too slow. The honest response is to fund the existing court properly and improve the quality of the decisions that generate the appeals. It is not to strip out the legal qualification requirement and hope that adjudicators appointed by the department whose decisions are under challenge will somehow produce fairer, faster justice than the judges being replaced.
Slowness and unreliability are different diseases. This Bill claims to treat the first while making no serious claim — even in its own impact assessment — to have cured the second.
It would be one thing if this were an entirely novel experiment. It is not. Earlier versions of lay-adjudicator appeal systems in British immigration law were tried, found wanting, and abandoned as not fit for purpose — a history the Law Society was careful to point out. The government is not stepping into the unknown so much as revisiting a room it has already walked out of once.
None of this addresses the other half of the access-to-justice problem, which begins before anyone reaches a tribunal at all. Research cited in the House of Lords estimates that around 60 per cent of eligible asylum seekers cannot find a legal aid lawyer.
In the East of England, firms have described recruitment for legal aid immigration work as “virtually impossible.” One law centre’s contract was suspended for a full year for want of a single supervising solicitor.
A claimant without representation is less likely to assemble the evidence a caseworker needs. That feeds the poor-quality decisions already documented, which feed the appeals, which the same unrepresented claimant must then navigate alone — in a system now being redesigned to require fewer formal qualifications of the people judging them. It is a closed loop, and every point of failure is one the state controls.
There is a further point worth noting, though it is inference rather than documented fact.
The main drivers of small-boat crossings are well established and lie elsewhere entirely, smuggling economics, criminal networks, family ties already in Britain, and the shortage of safe legal routes. Nothing here argues that poor casework is why people cross the Channel. But it is one small, largely unexamined factor in the mix, and worth naming.
Small-boat arrivals are granted asylum at a materially higher rate than the asylum population as a whole — 62 per cent between 2018 and 2025, against roughly 52 to 56 per cent for applicants generally, driven by a nationality mix skewed towards Eritrean, Sudanese, Syrian and Afghan claimants, whose initial grant rates have at times exceeded 90 per cent.
Layer onto that the finding above — that around half of Home Office refusals are overturned on appeal — and the true probability of eventually being permitted to stay is higher again than the headline refusal figures suggest to anyone glancing at them from outside.
A persistently high real-world success rate, compounded by a well-documented tendency for wrong initial refusals to be corrected on appeal, does nothing to discourage a business model that depends on convincing paying customers the crossing is worth the risk.
A court that overturns half its refusals is, among the other things it is, a standing advertisement for trying again.
None of this exists in isolation. Readers of this publication will recognise the shape of it. An institution facing a well-documented failure chooses not to fix it, but to alter the mechanism that keeps exposing it.
Counting domestic hotel bills as international aid does not make Britain more generous, it makes the ledger tidier. Abolishing the tribunal that keeps overturning Home Office refusals does not make the casework more accurate, it makes it harder to check. In each case the headline metric improves while the underlying competence problem is quietly left untouched.
Suella Braverman did not invent poor-quality asylum casework. Shabana Mahmood did not invent the impulse to redesign rather than repair. Both parties, across successive governments, reach for the same manoeuvre whenever a watchdog becomes inconvenient, change who gets to be the watchdog, rather than change the behaviour it keeps flagging.
It is a bipartisan reflex, which is precisely why it survives changes of government largely unexamined.
What makes the tribunal story more damning than most is that the evidence comes almost entirely from the government’s own machinery — its inspectorate, its auditor, its internal quality checks, its own impact assessment.
The department published, more or less in plain sight, a comprehensive account of its own unreliability, and then proposed, as the fix, a body with less independent oversight than the one currently doing the noticing.
The next time a minister announces that the backlog is finally under control, do not take the falling number at face value. Ask what happened to the cases that vanished from it. Were they decided correctly, or simply decided somewhere the decision is now harder to challenge?
A tribunal that overturns half of what it reviews is not, on the evidence of the government’s own inspectors, an obstacle to justice. It is very nearly the last part of the system still doing its job. Removing it does not improve the casework. It only makes the marking easier.
And it is, after all, so much less embarrassing to fail a test that nobody is any longer allowed to check.
••••
••••
••••
••••
(UKR)