John Milne MP: speeches

Speeches

• 19 May 2025 · UK-EU Defence and Security Agreement · Hansard source

  More

  The ill-advised decision to leave the EU in 2016 saw us leave the European Defence Agency at the same time. With defence expenditure rising at record rates across Europe in response to the invasion of Ukraine, the EDA has a vital role to play in our national security. What plans do the Government have to secure a more effective working relationship with Europe’s defence procurement strategy and to form stronger links with the EDA?

• 15 May 2025 · Topical Questions · Hansard source

  More

  Airports across the country are participating in the airspace modernisation review. However, there is a clear conflict of interest between environmental imperatives and profit motives. Will the Secretary of State undertake to introduce an independent member on each airspace review panel?

• 13 May 2025 · Local Housing Need Assessment Reform · Hansard source

  More

  I beg to move, That this House has considered reform of the standard method for assessing local housing need. It is a pleasure to serve under your chairship, Mrs Hobhouse. Everyone agrees that across much of the country, homes have become far too expensive either to rent or to buy. There is less consensus on the best way to get things back under control. I will argue that throughout the history of the standard method for assessing local housing need, that method has been part of the problem, not the solution. For a long time, the free market ideology we followed was to build houses randomly until the price came down. Ever since the days of Margaret Thatcher, who single-handedly killed off the public sector contribution, we have never got anywhere near to keeping up with demand. In recent years, the strategy has been to set stiff compulsory building targets and, to that end, the Government introduced the standard method. We were told that the method would produce clear, objectively determined house building targets for every local authority. We were assured that they would be equally and fairly distributed in line with genuine local need. We can now confidently say that that failed. Many authorities got nowhere near their number. Sometimes that was through dragging their heels, but often it was because their individual targets were outright bizarre and unachievable. Meanwhile, the system has kicked up terrific public anger and opposition, which in itself gets in the way of success. At times, the Government have resorted to wielding a bigger stick or they have backed off in the face of Back-Bench pressure. Under the present Government, we are heading back towards the big-stick approach. There is almost no attempt to win consent. I will argue not only that the standard method failed to do what it says on the tin, but that the failure was inherent from the first. It never stood a chance. Far from solving the affordability crisis, the method has significantly contributed to making that crisis worse, and it will continue to do so even under the remodelled version announced before Christmas, because it is based on a false premise. To be absolutely clear, this is not about national targets. Whether we aim nationally for 200,000, 300,000 or 400,000 homes a year is a separate debate, and I hope we will not get sidetracked by that today. It is easy to tweak the standard method to meet whatever national target we want it to meet, but in practice, national targets have been not much better than slogans, such as Boris Johnson’s 40 new hospitals, which never existed in reality. Instead, it is the local target as applied to individual planning authorities that matters. Broadly speaking, the standard method compares local house prices to local wages to estimate an affordability ratio, and it adjusts targets upwards if that shows prices to be unaffordable. The sums have been fiddled with many times since the method was introduced, and I do not doubt that such a process will continue. That is where the first big failure comes in: the standard method is supposed to provide an objective assessment of local housing need but, if we were honest, we would acknowledge that it is actually designed to reflect national need. For example, in my constituency, the growth target based on existing households should now be 527 a year, but our poor affordability ratio takes us all the way up to 1,329 a year, and that is before we add on more for our neighbours. That is a whopping uplift by any stretch of the imagination. The face of Horsham district is changing at breakneck pace. Villages such as Billingshurst and Southwater are on the way to doubling in size in less than a decade. That is not because Horsham is experiencing some kind of spectacularly large birth rate; it is just an arbitrary calculation. Once again, to be clear, I wholly accept that this is a national problem and that we need national solutions. Every area, including Horsham, has its role to play, but it is insulting people’s intelligence to describe that as a local need, when we plainly have nowhere near enough locals to go around, and they mostly cannot afford the new homes anyway. If we keep telling obvious lies to people, how will we ever win public consent? This brings me to the next big failure of the standard method, which is that there is no meaningful public scrutiny. Most local councillors do not understand how it works, sadly, let alone the general public. The standard method is never an election issue, yet it has a massive impact on our communities. In this case, ignorance is not bliss. It is a big reason why Conservative councillors have, election after election, proclaimed their commitment to allocating brownfield sites over greenfield yet somehow ended up doing the exact opposite. They cannot do anything to stop the logic of their own inflexible system. The standard method is a kind of mathematical bulldozer, sweeping aside our open spaces. The single worst failing of the standard method is that it fails in the very purpose that it was supposed to be designed for. In Horsham, as in many areas, the average price of a new house is higher than that of our existing stock. Ironically, the more houses we build, the worse our affordability ratio gets, and the higher our target will be next time around. The standard method does the exact opposite of what it is supposed to do. The more housing that is built, the more the method asks to be built, with no obvious mathematical limit. I stress again that I completely agree that building many more houses than we have over the last 40 years is an essential step on the path to affordability. However an obsession with one arbitrary number, without thinking what goes into it, does not work. It is actually getting in the way of success. We have to focus attention on the type of housing we are permitting, not simply the raw total. The standard method is based on a false premise, because many things affect prices besides the house building rate.

• 13 May 2025 · Local Housing Need Assessment Reform · Hansard source

  More

  For the individuals who benefit, no one can argue with it. It is the same with the sale of a council house—if you are the family that gets it, it has clearly given you a massive uplift. What I am saying is that we have a national societal problem to solve in the housing market in general. We have a certain amount of money to put towards it. That was a subsidy. There are far better things to do with that subsidy that do not inflate prices further, as that simply eats up the subsidy. As I was saying, putting all that together, it is clear that the standard method is getting its social sums wrong. The affordability ratio is actually a lousy proxy for actual housing need. What we need to do is factor a proper analysis of local housing conditions back into the system. That should include an assessment of local homelessness rates, the need for social housing, pensioner poverty and all the other factors that make communities tick. We also need to find a clear role for neighbourhood plans. Neighbourhood plans started as a great way to bring local consent and local knowledge into housing, but from the day the standard method was introduced, they have been effectively overruled. In the latest planning reforms, they were completely marginalised and were not even mentioned. How can we change the standard method to do the job it is supposed to do? I suggest at least two inputs: a local needs calculation, which focuses on helping local people into the homes they need, and a national needs top-up. Having a separate national needs figure will help us to focus on the delivery of new towns. When our housing needs are as great as they are, new towns are essential. In contrast, the standard method spreads targets indiscriminately across every area. It leads to endless incremental add-ons to existing settlements until they begin to lose their identity altogether. In rural areas such as mine, the standard method has an inherent tendency to create low-density suburbs. Not only do they tend to be more expensive houses, but they use two or three times as much land as they strictly need to.

• 13 May 2025 · Local Housing Need Assessment Reform · Hansard source

  More

  Yes, it makes strategic planning very difficult. Provision of infrastructure, particularly in rural areas, is a major problem and not sufficiently built into the planning system to compensate for it. It is easier in urban centres where the infrastructure is already in place. This is the reason we have contrived to have perhaps as many as 1.5 million unbuilt permissions nationally, at the same time as a national housing shortage. That is because too many of them are permissions for unaffordable and, therefore, unbuildable homes. There is a degree of land banking but, for the most part, developers build as fast as they can sell. If they are serving only the top end of the market, that will be slowly. As Oliver Letwin described in his excellent 2018 report, sadly unacted on by the Government of the day, we need far greater variety in housing type. As much as 80% of housebuilding is aimed at the top 20% of the market. The fastest way to fix that is to build a guaranteed quota of social housing. My party is asking for 150,000 a year. I guarantee they would be snapped up like hot cakes, as fast as they could be built. There is a fundamental difference between permissions and actual, physical houses. If all we ever think about is permissions and alleged impediments to permissions, we will never get to grips with the problem. Wrong permissions do not increase supply, they suppress it. Wrong permissions bake high land prices into the system. Handing out more permissions like confetti simply chokes the system with unbuildable sites that will hang over the market for a generation. There are lots of ways the standard method could be reinvented, but any future form must empower local authorities to deliver social housing in significant numbers from day one. How we do that is up for grabs, but somehow it must be done.

• 13 May 2025 · Local Housing Need Assessment Reform · Hansard source

  More

  I thank the Minister for his reply and all Members for their very interesting contributions. One thing that is really striking is that we see the same problem up and down the land. It may manifest itself locally, but it is a national problem. Like many Members of this Parliament, I come from a local council background—I was the cabinet member for planning in Horsham district council—so I have personal experience of trying to get what we needed for the community out of the plan and developers. It was a battle. My reaction to the changes made to the standard method and to the Planning and Infrastructure Bill, which is currently in Committee, is to say, “Would that have given me the tools I needed to do the job? Would it have improved my chances?” I feel that the answer is, “Not really, no.” That is the standard by which I judge it. Changes to the standard method could really enhance—make or break, actually—what the Government are doing in the Planning and Infrastructure Bill. It is not just a technicality on the side, but a crucial interface. I realise that the Minister will be a bit distracted, given that the Bill is going through Parliament right now, but I hope that in the fulness of time he will take a closer look at the measure, because it can be revised at any point and does not require legislation. I again thank everybody for a very good-natured debate. Question put and agreed to . Resolved , That this House has considered reform of the standard method for assessing local housing need.

• 13 May 2025 · Local Housing Need Assessment Reform · Hansard source

  More

  I thank my hon. Friend, who makes a very good point. The system is working almost to the reverse of what was intended. In my constituency of Horsham many people either work for London businesses or perhaps have traded down from a more expensive London property. From their point of view, Horsham represents excellent value. The official affordability ratio does not reflect real working conditions in Horsham for locals, and therefore overstates local targets. Local councillors all strive to get the best for their communities, but the way we receive targets under the standard method destroys our negotiating position with developers. Developers are not stupid. They can work out as well as anyone else how many sites are needed to meet our targets. They have no need to concede on civil amenities or on affordable housing because they know that, at the end of the day, they have got the council over a barrel. I have no issue with a private developer seeking to make a profit—what else do we expect them to do?—but do not rely on them to do social planning. In areas like Horsham, years of free market ideology have turned councils into mere editors of private developer proposals. We build on greenfield sites because they are the only ones that get presented. There is literally nothing else to choose from in Horsham. The free market approach to affordability does not work for the housing market. Competition has driven prices up, not down. In Horsham we would arguably be better off if we granted a monopoly to one single developer and let them push down local land prices. To add insult to injury, we also have the standard method’s bullying friend, the housing delivery test. I am not sure whether there ever was a carrot in this process, but the HDT is definitely the stick. Failure to meet targets can ultimately result in losing local control over planning altogether. It is a Catch-22 situation: the developer controls the rate of delivery, but the council pays the price if targets slip. Heads they win, tails we lose. In fact, the single biggest factor that influences prices has nothing to do with house building. It is availability of credit. If interest rates were to double tomorrow, the price of a mortgage would soar and we would see a house price crash, yet all that would happen without a single new home being built. A succession of policies under the Conservatives only served to make the problem worse, not better. Subsidies such as Help to Buy or stamp duty holidays simply inflated prices further, like a giant Ponzi scheme. The market adjusts, and the subsidy ends up in the pockets of developers until the next upward turn in the spiral. Therefore, any analysis of UK house building must take into account the key role of finance. Since Thatcher, houses have come to be seen not simply as homes but as investments. In line with that, the explosion of the buy-to-let market in the 1990s correlates suspiciously closely with overall house price inflation. Older generations benefited from decades of property asset inflation, but today it is getting harder and harder to board that train. Putting all that together, it is clear that the standard method is getting its social sums all wrong.

• 13 May 2025 · Ukraine: Diplomacy · Hansard source

  More

  7. What diplomatic steps his Department is taking to support Ukraine.

• 13 May 2025 · Ukraine: Diplomacy · Hansard source

  More

  Last weekend the Prime Minister said that the UK would do all that it could to support Ukraine. If that is the case, why do the Government continue to prevaricate over seizing billions of pounds in frozen assets held in UK banks, which could be used to build Ukraine defences? The longer we delay, the more likely it is that those funds will become wrapped up in other negotiations and we will lose the chance altogether.

• 12 May 2025 · Personal Independence Payment: Eligibility Criteria · Hansard source

  More

  The Government say that their PIP reforms will help people into employment, but the Multiple Sclerosis Society says that 60% of sufferers believe those reforms will make it harder for them to find work, not easier. An estimate must have been made of what percentage of claimants will feasibly enter employment as a result of these reforms. Will the Minister share those figures?

• 12 May 2025 · Personal Independence Payment: Eligibility Criteria · Hansard source

  More

  10. What estimate her Department has made of the potential impact of changes to the eligibility criteria for personal independence payment on the number of people receiving that payment who will move into employment.

• 6 May 2025 · Parking Regulation · Hansard source

  More

  In Horsham town, in my constituency, we have a central car park outside a Sainsbury’s, which is operated by a third-party contractor. The number of disputed tickets is out of control. Does my hon. Friend agree that the voluntary code of practice, which was introduced last year, seems to have made absolutely no difference? I can detect no reduction in the difficulties being created.

• 6 May 2025 · Dedicated Schools Grant · Hansard source

  More

  In West Sussex, the deficit on our DSG grant is £130 million, and that will potentially double by next year. Despite that, SEND provision, which is the main driver of that deficit, is deficient across the district. So many schools approach me on this subject. Does my hon. Friend agree that we need to not only find a way to wipe out the deficit but remove the fundamental cause, which is the ballooning cost of SEND provision?

• 29 Apr 2025 · Public Authorities (Fraud, Error and Recovery) Bill · Hansard source

  More

  When it comes to public money, everyone accepts the importance of preventing fraud; there is no dispute about that. The mere thought that our benefit system could be exploited loosens the cement holding our welfare system together. However, if we look back in history, there has been a track record of fraud recovery measures not delivering what was hoped. This measure will also probably never save the £1.5 billion that is expected of it, so I ask: will the alleged rewards of this legislation ever match the scale of the imposition on our civil liberties, and are we really going after the right targets? We all want to catch deliberate and professional fraudsters, but they are precisely the people who are astute enough to change tactics, set up separate bank accounts, and avoid suspicion. Instead, it will be the innocent and the accidental claimants who fall into the trap. The implicit assumption is that we should trust in the DWP as a completely error-free organisation across the entirety of its massive operation. But the DWP does make mistakes. It makes mistakes all the time. And even when it knows that it has made a mistake, and it has been told so, it is very capable of making the same mistake all over again. In my constituency of Horsham, Anthony and his husband were accused of providing misinformation to the DWP and were overpaid £10,000 as a result. Anthony protested without success. After a long fight the case went to appeal. The tribunal wasted no time deciding in his favour—it was an open and shut case. But then, earlier this year, Anthony and his husband were migrated over to universal credit. After confirming all details were correct, the DWP overpaid them again, and then sought to claw the money back over the following months. The DWP’s mistake, but Anthony pays the penalty. The DWP has its rules, but real life does not run in straight lines. Real life is messy. How can we possibly rely on the DWP to mark its own homework when we know that there are just four fraud advisers per regional office to handle cases flagged by frontline staff? Yes, there are some checks and balances within this legislation, but what is really needed is a profound cultural change within the DWP, and that is much harder to achieve. The common experience of people who have to deal with the DWP on a daily basis is that they feel that it is always looking to catch them out. Years and years of inflammatory rhetoric under a succession of Conservative Governments have convinced people to regard the DWP as their enemy, not their friend. If anything, the Bill digs that hole a little deeper. What concerns me most about the Bill is its extreme overconfidence. It assumes that Government agencies always get things right and that individual citizens are to be automatically treated as objects of suspicion. In Committee, the Government were resistant to any amendments except their own, so I very much hope that they will reconsider today and accept the Liberal Democrat amendments.

• 28 Apr 2025 · Scientific and Regulatory Procedures: Use of Dogs · Hansard source

  More

  It is a pleasure to serve under your chairmanship today, Ms Jardine. I am grateful to speak in this very important debate. Every year, thousands of dogs are subjected to scientific experiments in the UK, often in the name of drug development. For many this raises serious ethical questions about the use of animals for human gain. Is an animal’s suffering worth the benefits to scientific discovery? For others, science and not the animals are paramount; the end justifies the means, as it were. However, animal testing does not deliver robust and useful scientific data. In fact, drug research on dogs tells us very little about whether a drug will work for humans, so campaigners argue that it is time to end animal research. Dr Jarrod Bailey, a geneticist, put it plainly: different species react differently to diseases and treatments. In other words, what works in dogs might not work in humans. In fact, drugs that pass animal tests fail in human trials 92% of the time—a staggering and costly statistic. In toxicity testing, even when dogs show no toxic response it barely improves our confidence that the drug will be safe in humans. It improves it from 70% to just 72%, which is barely noticeable. Is that really a sound basis for human medicine? Fortunately, science offers us a better path forward. Human-specific technologies such as organ chips are revolutionising drug development. Those miniaturised organs mimic how real human organs react to treatments and can be patient-specific. They have shown 87% accuracy in detecting drug-induced liver toxicity, which is a dramatic improvement over animal models. If adopted widely, these tools could create over $24 billion through research and development in the US. The US Food and Drug Administration has recognised that. Through the FDA Modernisation Act 2.0, the agency has removed the legal requirement for animal testing in drug trials. A third Act is already in the works to accelerate the validation and adoption of human-specific methods such as organ chips. In the UK we are lagging behind, not because of legal barriers, but perhaps because of entrenched industry habits, financial interests or even cultural resistance in the research community. We can change that—gradually, responsibly and strategically. I am calling for a phased approach to end the use of dogs in UK research. That means increased investment in modern human-relevant alternatives and a national commitment to shifting away from outdated animal models. When the practice of animal testing is scientifically flawed, it is also undeniably ethically indefensible. Animal suffering for unreliable and inapplicable data cannot be justified when we have the tools and knowledge to do better. Let the UK be a leader, not a follower, in creating a more humane and effective future for our scientific research.

• 23 Apr 2025 · Hospitals · Hansard source

  More

  In my constituency, there has been a popular campaign for a general hospital for many decades without success. Given that we did not even make it on to Boris’s fantasy list of 40 hospitals, we are certainly not going to succeed now. In the absence of that hospital, what we need is a strategy to move more of the treatments for which distance really matters into smaller local clinics. In Horsham, that might include wound dressing and sexual health services. Does the hon. Member agree that if we are going to continue to be denied the hospital we deserve, the kind of localised treatment strategy he is suggesting could be implemented at very low cost?

• 22 Apr 2025 · Residential Estate Management Companies · Hansard source

  More

  In my constituency of Horsham, we have many similar examples. Would my hon. Friend agree that although we certainly need legislation, the industry could act right now by introducing a voluntary code of practice? The industry does not have to wait for legislation; it should hear the call from across this Chamber.

• 22 Apr 2025 · Statutory Sick Pay · Hansard source

  More

  In my Horsham constituency office, I employ a member of staff who has ME. Fortunately, we can be very flexible with their working hours. However, under current law—where statutory sick pay is based on days worked not hours worked—an ME sufferer could easily miss out altogether on sick pay. Does the hon. Member agree that the Government should legislate to ensure that all employees are granted fair access to sick pay?

• 1 Apr 2025 · Adoption and Special Guardianship Support Fund · Hansard source

  More

  I am talking to a family in my constituency who adopted a young boy and a girl in 2021. They understood at the time that the children were likely to have special needs because of their upbringing, but they were prepared to take on the challenge. The boy in particular requires therapy to keep him in mainstream school. They have been very distressed over recent months at the possibility of losing the funding. Does the Minister not appreciate that this damages our great need to recruit more people to be adoptive parents, which does the state a great service and the children themselves the best possible service?

• 1 Apr 2025 · Eating Disorder Awareness · Hansard source

  More

  In my constituency of Horsham, we had a particularly upsetting case of a constituent whose daughter had significant mental health and behavioural issues that were very difficult to deal with and, as a consequence of those not being dealt with, she also developed an eating disorder. The only place they could send her to that could cope with that combination of factors was in Yorkshire—my constituency is in West Sussex. That was an extraordinary burden on the family. We need provision across the country to deal with the cases that present.

• 24 Mar 2025 · Planning and Infrastructure Bill · Hansard source

  More

  First, I wholly respect the intention behind the Bill; it is a serious attempt to solve a serious problem. I also recognise that what was happening under the Conservatives did not work, and never could have worked even if we had given it 1,000 years. All it achieved was to fuel house price inflation, which has now created a destructive division into a nation of haves and have-nots. But I judge this new Planning and Infrastructure Bill through the lens of my own constituency—will it work for Horsham? Will it deliver affordable homes in the right places and with the right environmental standards? I think the answer is no. The main reason is that the Bill is based on the same mistaken premise as the previous system. The problem lies with how housing targets are worked out—not the national target, which gets all the publicity, but local targets. Why are targets so hard to meet? The reason is that the Conservatives invented a catastrophically bad formula for calculating housing need, which is called the standard method. It measures the ratio of local house prices to local wages, and the bigger the gap, the higher the target goes. The idea is that communities just keep building houses until the price comes down. The only problem is that it does not work. It turns out that in Horsham—as in many places—the average price of a new house is higher than the price of the existing stock, so the more we build, the worse the ratio gets and the higher the target goes. That is the exact opposite of what the theory says should happen. Unfortunately, this new Labour Bill takes the same flawed Tory standard method and pours rocket fuel over it. Targets control planning permissions, but that is not the same thing as actual houses; Horsham already has 13,500 unbuilt permissions, including the emerging local plan. That total could double under Labour’s new targets. Does that mean that we are actually going to build tens of thousands more homes? No, it does not. We could cover every inch of Horsham district in permissions, but it is not the lack of permissions that is holding back the market. Houses do not get built faster, because developers cannot sell them any faster. Some 80% of what we build today is aimed at the top 20% of the market—all of this was described very well in Sir Oliver Letwin’s analysis back in 2017. The housing market does not behave as one market; it is like six parallel markets, and the houses we are building are largely serving the top two. I am desperate to build more affordable homes in Horsham, but clogging up the system with unbuildable permissions is not the way to do it. The best way to build more homes is to build more consent. I said that I would judge this legislation on whether it would work for Horsham, and the answer is that it will not.

• 19 Mar 2025 · National Insurance Contributions (Secondary Class 1 Contributions) Bill · Hansard source

  More

  Age UK in my constituency has told me that the employer NICs rise will cost it £50,000 a year. Does the Minister agree that it is impossible to improve the public sector by taxing the public sector?

• 18 Mar 2025 · Public Authorities (Fraud, Error and Recovery) Bill (Twelfth sitting) · Hansard source

  More

  It is a pleasure to serve under your chairmanship, Mr Western. The DWP is making extensive and growing use of algorithms for investigation purposes. Without proper oversight, these systems threaten error, unfairness and bias, which could lead to wrongful debt collection. Our amendment therefore calls for an independent audit of these systems at least every six months, to ensure accuracy and fairness. The audit must be conducted by experts in data science, ethics and social policy with no ties to the DWP or system developers. True independence is key. The audit look at issues such as accuracy, so whether the algorithms are correctly identifying overpayments; fairness, so whether they unfairly target certain groups or operate with bias; and, above all, transparency and accountability. After each audit, we suggest that a full report must be published, presented to Parliament within 14 days, and made publicly accessible. If serious flaws are found, the Secretary of State must respond within 30 days with a clear action plan to fix these issues. Overall, Liberal Democrats are positive about benefiting from new technology, but we do need to consider whether it offers help, not harm. In the wider context, what work is the use of AI generating? There are already chronic staff shortages at the DWP, with 20% vacancy rates becoming routine. Disability Rights UK has commented that operational failures now permeate every layer of welfare administration. Fraud investigation teams therefore already lack capacity to address the annual £6.4 billion of overpayments. There are only four fraud advisers per regional office to handle cases flagged by frontline staff, which has created a bottleneck, so that very often 90% of suspected fraud cases go uninvestigated. In other words, one could suggest there is already plenty of fraud to investigate without trawling for more. This amendment ensures regular scrutiny, transparency and fairness. I urge the Minister to consider it.

• 18 Mar 2025 · Public Authorities (Fraud, Error and Recovery) Bill (Twelfth sitting) · Hansard source

  More

  I am content to beg to ask leave to withdraw the motion. Clause, by leave, withdrawn. New Clause 15 Offence of encouraging or assisting others to commit fraud “(1) The Social Security Administration Act 1992 is amended as follows. (2) In section 111A (Dishonest representation for obtaining benefit etc), after subsection (1G) insert— ‘(1H) A person commits an offence if they— (a) encourage or assist another person to commit an offence under this section, or (b) provide guidance on how to commit an offence under this section.’ (3) In section 112 (False representations for obtaining benefit etc), after subsection (1F) insert— ‘(1G) A person commits an offence if they— (a) encourage or assist another person to commit an offence under this section, or (b) provide guidance on how to commit an offence under this section.’”— (Rebecca Smith.) Brought up, and read the First time. Question put, That the clause be read a Second time.

• 18 Mar 2025 · Public Authorities (Fraud, Error and Recovery) Bill (Twelfth sitting) · Hansard source

  More

  I beg to move, That the clause be read a Second time. The new clause would require that the use of algorithms, algorithmic tools and systems, and artificial intelligence, including machine learning, should be included within the algorithmic transparency reporting standard. That standard, established by the Government, is supposed to be mandatory for all Government Departments. However, last November, The Guardian reported that not a single Whitehall Department has registered the use of AI systems since it was made mandatory. Throughout debate on this issue, the Government have consistently downplayed the risk of using AI to trawl for suspect claimants, but if it really is that simple, why have so many organisations come out with concerns and opposition? That includes Age UK, ATD—All Together in Dignity—Fourth World, Amnesty International, Campaign for Disability Justice, Child Poverty Action Group, Defend Digital Me and Difference North East. I could go on: I have half a page, which I will spare the Committee from, listing organisations that have expressed concern. It is quite a roll call. Governments can and will get things wrong. History tells us that if it tells us anything. In June 2024, a Guardian investigation revealed that a DWP algorithm had wrongly flagged 200,000 people for possible fraud and error; it found that two thirds of housing benefit claims marked as high risk in the previous three years were in fact legitimate, but thousands of UK households every month had their housing benefit claims wrongly investigated. Overall, about £4.4 million was wasted on officials carrying out checks that did not save any money. We know that more mistakes will happen, no matter how hard we try to avoid them. I therefore ask the Minister to support the insertion of new clause 14 as a small measure of defence against future institutional failings.