Latest news from FE Week

Watchdogs on the lookout for AI experts

Ofqual is searching for an “exceptional” senior leader to head its artificial intelligence (AI) plans and protect the integrity of assessment.

The qualifications regulator is on the lookout for a director of assessment integrity and innovation to provide “increased leadership capacity” on technology.

In the advert for the £81,000 a year job, Ofqual said: “While technology can deliver meaningful benefits for learners, the standards and public confidence underpinning England’s world-leading qualifications must be protected.

“Balancing these priorities is central to the role.”

The appointee would be responsible for “protecting assessment integrity and test security” from technological risks, including malpractice.

They would also shape Ofqual’s regulatory approach to AI and other new technology, making sure any innovation was “safe and beneficial”.

Ian Bauckham, the chief regulator, told the Festival of Education this month that written coursework would face “far, far more scrutiny” once reformed qualifications were rolled out to ensure that AI-generated output did not replace learners’ own coursework.

But he acknowledged that AI could “improve efficiency and both cut costs and increase accuracy” in marking.

Ofsted has also launched a recruitment round for a lead AI security specialist.

‘Thought leader’ on AI

It said the new director must be an “exceptional senior leader” who would help the regulator “navigate this changing landscape and shape how regulation responds”.

It involved anticipating and addressing threats to validity, security and trust across qualifications, and setting up effective ways of preventing and detecting malpractice enabled by AI.

The director would also be a “authoritative voice and thought leader” to influence government, regulators and the sector and “shape national thinking” on AI and assessment integrity.

Successful candidates should understand the potential and use of technology in assessment, and have experience leading “through ambiguity and making high-stakes decisions with incomplete evidence”.

The £70,000-a-year Ofsted AI lead would act as a “subject-matter expert on AI security”, making sure the systems used were “secure, resilient, and compliant”.

They would also provide expert cyber security guidance across projects, working with different teams to identify risk and to make sure that AI use was compliant with policy and standards.

Ofsted said it was committed to making sure its AI use “complies with legal obligations, mitigates risk to the organisation, providers, the public and employees” and fits civil service guidance.

Ofsted bins £3.9m IT project paused to fund pay rises

Ofsted has written off the entire £3.9 million it spent on software meant to replace its scheduling and case management system, after an external review found the stalled project was not worth reviving.

The watchdog recorded a £3.1 million impairment of a software asset in its 2025-26 annual report and accounts, published this week. It followed an earlier £800,000 impairment of the same asset in 2024-25, which Ofsted confirmed covered work that would have needed redoing.

Ofsted told FE Week the two write-offs account for the full £3.9 million spent on the project before it was paused.

The inspectorate paused the work in 2023-24, when it reallocated funding to support civil servant pay in the organisation after the Conservative government announced unfunded public sector pay rises.

Ofsted confirmed the system was being built to replace the software it uses to schedule, case manage and support workflow across all its inspection and regulatory work.

Its latest accounts for the financial year ending March 31, 2026, said an independent review had confirmed that restarting the work would not represent value for money, because of “technology changes and evolved business requirements”.

“We resumed work on the replacement system last year, starting with an external review to decide whether we would be best re-start the paused work or begin again with a different approach,” an Ofsted spokesperson said.

“That review recommended we should go down a different route, leading to us needing to write this work off.”

Ofsted’s accounts state that “elements of the work completed will inform future digital solutions”, though the watchdog did not specify what elements have been kept. A further £125,000 impairment was identified after the year end, which Ofsted judged immaterial and did not adjust for.

Ofsted is required to report write-offs of more than £300,000. Its latest accounts also reported a £400,000 “fruitless payment” for three months of rent and running costs on a vacant London office, incurred after Ofsted relocated as part of the Government Property Agency’s programme to reduce Civil Service office space.

Total losses reported by the inspectorate rose to £3.67 million across 68 cases in 2025-26, up from £54,000 across 50 cases the year before.

The scrapped project is unrelated to Ofsted’s electronic evidence-gathering (EEG) system, the software inspectors use to record notes during inspections.

Ofsted reiterated that it was considering replacing the EEG system, after long-running technical issues of lost data and freezing glitches when inspectors attempt to type their notes into the application.

A spokesperson added: “We’re currently undertaking a programme of digital modernisation aimed at ensuring long-term and sustainable digital inspection tools, including evidence gathering tools. We don’t have any further update at this stage.”

Ofsted’s digital and IT costs have climbed steadily since the pause. The organisation spent £6.4 million in 2023-24, rising to £7.2 million in 2024-25 and £9.6 million in 2025-26 — an increase of nearly a third in a year.

To fix apprenticeships we must back employers, not just fund them

The government’s new ‘deal for young people’ and renewed focus on youth apprenticeships is a welcome step towards reversing the long-term decline in under-25 apprenticeship starts.

But if this ambition is to succeed, we must recognise a fundamental truth: the problem is not a lack of interest from young people.

Every month, Lifetime receives more than 5,000 applications from young adults looking for apprenticeship opportunities, yet vacancies fall well short of demand.

That tells us something important. Young people are motivated, ambitious and eager to learn. The challenge is creating enough opportunities for them to take that next step.

Apprenticeships remain one of the most effective ways for employers to build a skilled and loyal workforce, develop talent aligned to business needs and create a pipeline for long-term growth.

Yet for many employers, particularly in hospitality, retail, care and early years, recruiting a young apprentice can still feel like a significant commitment. Many are taking on someone entering the workforce for the first time, requiring additional time, support and investment.

Increased funding has an important role to play, especially during the early stages of employment. But funding alone will not unlock employer demand.

Employers consistently tell us they need practical support: clear guidance, simple processes and confidence that they have the right structures in place to help a young person succeed. When that confidence exists, they are far more likely to invest.

This is where policy needs to go further. We should focus on genuinely de-risking the decision to recruit, whether through targeted incentives such as salary offsets or practical support that helps employers manage the realities of bringing new talent into the workplace.

The government’s recent apprenticeship announcements reinforce both the opportunity and the challenge.

The review of funding bands has the potential to better reflect the true cost of delivering high-quality apprenticeships. However, without additional funding entering the system, difficult trade-offs remain and there is a risk that some areas of provision will be squeezed.

The renewed focus on younger learners is welcome and overdue. Rebuilding entry-level pathways is essential for social mobility, productivity and economic growth.

However, entry-level does not always mean young. In sectors such as adult social care and early years, many new entrants are over 25. A strong apprenticeship system must support young people starting their careers, older career changers, existing employees looking to upskill and those returning to work. Solving one challenge should not create another.

Alongside employer demand, there is another equally important priority: preparing young people for the workplace.

Employers consistently tell us that communication, teamwork and other employability skills matter just as much as technical knowledge. Stronger collaboration between schools, employers and training providers can make a real difference.

Better work experience, mentoring and opportunities to develop workplace confidence can help young people understand employer expectations before they apply for an apprenticeship.

Too often, employers engage too late or have limited relationships with schools and local communities. Our recent research found that only 37 per cent of parents had received any information from employers about apprenticeships. Earlier engagement would raise awareness of apprenticeship pathways, shape expectations and strengthen future talent pipelines.

The government’s ambition to create 50,000 additional youth apprenticeship starts by 2030 is achievable. But it will require more than additional funding. It means giving employers the confidence to recruit, helping young people become work-ready and maintaining a balanced apprenticeship system that supports young entrants, career changers and workforce development alike.

The talent is already there. Now we need a system that gives more young people the opportunity to fulfil their potential while helping employers build the workforce they need.

The career ladder is missing its first rung

Alan Milburn’s recently published Young People and Work Review has drawn significant headlines, but one of Milburn’s most striking observations – that “the first rungs on the old career ladder have weakened” – deserved greater focus.

This is not simply about whether young people have the right qualifications, or access to careers advice and employment support. It is about whether the labour market still provides enough opportunities for young people to get started, gain experience and build careers. The deterioration of established pathways from education into entry-level roles and stable career-track employment may be one of the most important structural drivers of declining social mobility and young people’s growing disillusionment with the economic and political system.

While many young people are affected by these changes, those from disadvantaged backgrounds are likely to face the greatest challenges because they are often most reliant on accessible entry-level opportunities to gain a foothold in the labour market. As these opportunities decline, it may become harder for them to transition into independent adulthood.

Before the mid-noughties, we became used to the labour market offering large numbers of jobs that acted as stepping stones into sustainable careers for young people. These roles did not always pay well, but they provided accessible entry points for young people to gain experience, develop workplace skills and build a platform from which to progress.

However, opportunities to access low-skilled, accessible jobs have long been waning and recent labour market changes suggest this trend has accelerated. NFER research, The Skills Imperative 2035, showed that AI, automation, and wider demographic, environmental and economic changes are reshaping the labour market, with employment growth concentrated in higher-skilled, higher-paid occupations, while most lower- and middle-skilled occupations decline. Actual employment changes in recent years suggest the pace of change has been faster than previously projected, by as much as three times for some groups.

Many of the jobs that historically provided young people with a foothold in the labour market – such as administrative and routine customer service roles – are shrinking. These roles were important not necessarily as destinations, but as first steps; jobs where young people built skills and experience, helping them later progress into more stable career-track jobs.

At the same time, the requirements of the remaining entry-level jobs are rising. Employers increasingly seek experience and a combination of technical and transferable skills, creating a challenge for young people who need employment in order to gain the very experience employers are asking for. This risks creating a cycle where young people face higher barriers to accessing opportunities that were previously more accessible.

Consequently, young workers now start further down the career ladder than they used to and progress more slowly after that point . Even graduates are experiencing a ‘triple whammy’ of challenges due to a general labour market slowdown, a sharper decline in graduate-level job openings, and reduced demand for lower-skilled ‘stepping stones’ roles. NFER research shows that the sharp slowdown in hiring in the tech sector, for example, has hit entry level roles disproportionately.

The result is that the labour market is shifting from one that facilitates efficient education-to-employment transitions to one with fewer entry points into occupations offering good career and salary progression, especially outside London; rising experience requirements for roles that historically provided that experience; and a growing risk of becoming trapped in forms of underemployment.

As is often the case when opportunity structures weaken, disadvantaged young people are likely to bear the brunt particularly those with fewer qualifications and family connections or financial support that can help others navigate a more competitive labour market.

Milburn’s interim report rightly acknowledges occupational changes in the structure of the labour market and the erosion of early career entry points. But these changes should be treated as more than just background context. They are central to understanding why so many young people are struggling to establish themselves in the labour market.

Policy focus is on improving the education, welfare, health, and careers systems around young people, without adequately acknowledging that the structure of opportunity has changed and is likely to change further still. This risks helping young people compete for existing opportunities, rather than addressing the shrinking structure of opportunities themselves.

The erosion of entry-level opportunities is not just a temporary disruption but the consequence of deeper structural changes that are reshaping the demand for labour. Milburn’s final report needs to start from this premise. Young people don’t just need help to climb the rungs of the career ladder; we also need to rebuild the ladder itself.

Our new vocational model could shape future ESOL

Up until a few years ago, we had a problem that we knew wasn’t unique to us at The City of Liverpool College: many learners of English for Speakers of Other Languages (ESOL) were getting stuck when they reached entry level 3 ESOL and then dropping out. Normally, this is the point at which students progress onto vocational programmes. However, these courses do not always provide the right language support or learning environment that ESOL learners need.

So, about two years ago we introduced a new suite of ‘ESOL into’ courses designed to break the cycle. They combine ESOL and vocational pathways for our learners who have reached Level 1 ESOL, where they then have the option to join a specific vocational pathway. Following the success of the first year, we now have eight vocational pathways for students to choose from, ranging from Plastering to Art and Design.

Working with the local authority

We first started to offer vocational taster sessions to ESOL students in 2023. That same year we successfully applied for and received Test and Learn funding from the Liverpool City Region Combined Authority as part of the local devolved adult education budget (Adult Skills Fund), helping to make our concept become a reality.

We would encourage others to consider supporting specific learner groups and/or applying for the funding, which if available, enables organisations to design and support innovative pilot projects that help to engage learners, meet employer needs, address skill shortages. They are specifically aimed to facilitate innovative practise where there’s a safe space and flexibility for experimentation.

For us, the programme has enabled us to improve ESOL learners’ participation in vocational education significantly. But it’s far from a separate project – it’s now part of our curriculum plan and embedded across many different subjects.

Unlocking progression opportunities

It’s been brilliant to see the students progressing on to the next level within the vocational courses. Two thirds of our current ‘ESOL into plastering’ students are continuing to the level 2 plastering course, while six of our ESOL into teaching assistant graduates have already taken up positions in local schools. For others, it’s simply about building their confidence as they work and study in a new language.

Replicating the model

There are many providers and colleges who are already running a vocational ESOL model in some shape or form. But we do truly believe that this comprehensive framework could help shape future ESOL programmes across the country, and it shouldn’t be difficult for others to replicate the model.

The first step is to work with local authorities to secure funding and to sell this model as a way to benefit students and the local economy. This can take time: liaising with different strategic authorities and navigating complex funding structures are potential barriers that providers might face when implementing a similar programme.

The next step is getting buy-in from all staff involved. In our experience, we’ve had nothing but positivity from both the ESOL and vocational teaching teams when it comes to teaching the new courses. They’ve all been heavily involved in the curriculum design and have told us that the new model is exactly what they’ve needed over the last few years.

We have found that our “ESOL Into” model works particularly well where there are particular skillsets or terminology that the students need to master, for example, taking shorthand notes in health and social care, or describing pictures in art and design. In short, making sure that the students have the language to be able to follow a set of instructions clearly. This means that most of the pathways that we offer are vocational as opposed to academic subjects, with the exception of our ESOL in science course, which can act as an entry point into further academic study.

The next challenge for us, and others, is to ensure that there are clear progression routes for all our ESOL pathways. It is also important that learners who may have other work or family pressures are aware that these vocational courses are a bigger time commitment than a standalone ESOL course.

On reflection, the model has been a big hit with everyone in the college, students and staff included. We love it when ESOL students start to view themselves as vocational students – ultimately, it’s about building their confidence.

AI might accelerate processes, but it can’t sit with ambiguity

One of the most quietly dangerous phrases in modern workplaces is: “Can we just get a quick answer?”

It is dangerous because what people often actually mean is: “Can somebody make the discomfort of uncertainty disappear?”

Modern organisations are obsessed with speed – faster responses, faster systems, faster delivery, faster decisions. Everyone is “circling back”, “touching base” and escalating things marked urgent that absolutely were not urgent five minutes earlier. Meanwhile, half the workforce is running on caffeine, cortisol and Microsoft Teams notifications, so naturally we’ve collectively decided AI will fix this.

To be clear, AI is incredibly useful. I often use it myself. It can reduce administrative friction, process huge amounts of information quickly and automate repetitive tasks that human beings probably should not have been doing manually in the first place. But there is a significant difference between accelerating processes and exercising judgement, and organisations are increasingly starting to confuse the two.

AI performs brilliantly in structured environments: clear rules, defined outputs, historical patterns and predictable systems. The trouble is that human organisations are almost never like that. Most genuinely difficult workplace decisions happen inside ambiguity – not spreadsheet ambiguity, but human ambiguity.

That is the kind where nobody has the full picture yet, where the data technically says one thing but your operational instincts are screaming another, where a safeguarding concern first appears as “something feels slightly off”, or where culture problems emerge through strange tension in meetings long before they appear in engagement surveys. Risk rarely arrives neatly labelled. More often, it develops quietly through accumulated workarounds created by exhausted people trying to survive impossible workloads.

A lot of governance work – real governance work – is essentially professional pattern recognition under conditions of uncertainty. That is also why I am increasingly unconvinced by the notion that the future belongs entirely to the people who can produce answers fastest. Sometimes the most valuable person in the room is the one saying: “Hang on. I don’t think we fully understand what’s happening yet.”

Unfortunately, many workplaces still reward the opposite. Confidence is often treated as competence, while fast responses get mistaken for good judgement. Reflective people can be perceived as hesitant simply because they insist on sitting with complexity for longer than is socially comfortable. Corporate culture still has a strange tendency to reward performance signalling over actual thinking, and AI may accidentally make some of this worse rather than better.

Because AI tends to mirror the logic of the systems around it, if an organisation already prioritises speed over reflection, visibility over substance and certainty over nuance, then introducing AI into that environment may simply accelerate existing dysfunction more efficiently – which is not quite the futuristic utopia everyone put in the PowerPoint.

I also find this conversation intersects interestingly with neurodivergence. Many neurodivergent professionals, particularly those used to navigating unpredictable or cognitively demanding environments, often develop strong systems-thinking and pattern-recognition abilities. Sometimes, we identify operational tensions long before those tensions become formally visible to everybody else.

At the same time, many workplaces still assess professionalism through communication style rather than decision quality: who sounds polished, who speaks confidently in meetings, who performs calmness convincingly enough, who understands the unwritten social choreography of corporate environments. Those things are not always the same as good judgement.

And in increasingly AI-enabled workplaces, that distinction matters enormously because eventually somebody still has to sit in the room where the information is incomplete, the politics are unspoken, the risks are emerging, everybody is uncomfortable and there is no clean answer yet.

AI is excellent at generating outputs. It is still remarkably bad at sitting with uncertainty without hallucinating confidence. And frankly, quite a lot of humans are bad at that too.

I suspect ambiguity tolerance – the ability to remain thoughtful, reflective and operationally calm without forcing premature certainty – is quietly becoming one of the most valuable workplace skills of the next decade. The irony is that many organisations still do not recruit, reward or promote for it nearly as much as they should.

Burnham could create a new dawn for further education and lifelong learning

The vision that Andy Burnham has laid out of the future is refreshing, particularly his assertion that education is a key to growth, hope and opportunity.

Central to this vision I would like to see the revitalisation of further education. I have many reasons to believe in the life-changing importance of FE. Here are some of them.

Given Burnham’s emphasis on opportunity and hope, I would like to tell my own story. I failed my 11 plus and left school in 1965, aged fifteen. Ten years of dead-end jobs followed.

At the age of twenty-five I went to my local FE college to enquire about studying for some evening class O-Levels. They had just started a new full-time course for mature students and encouraged me to apply. After a year I had gained four O-levels; another year, three A-levels. My local authority gave me a grant for these two years.

I was then accepted to study social sciences at the University of Bath. FE, and then university, completely changed my life – it gave me hope and opportunity. Since those times in the 1970’s the opportunities for older people to follow such a path have diminished considerably.

After graduation I started as a part-time lecturer in FE teaching something called liberal studies to day and block release students, including plumbers, electricians, carpenters and joiners, motor vehicle engineers, and many others. These people were on proper, well-recognised and funded apprenticeships. I soon became a full-time lecturer and my local authority funded day release to study for a Certificate in Education

I taught in FE for 22 years. In 2002, after a tasty redundancy offer, I joined the FE teacher training team at my local university. I gained an MA in education studies and, before retirement became a senior teaching fellow in the university’s department for education, health and sciences. I have published several books on learning and teaching in further and higher education. All this might sound like bragging but I am proud of what I did and the way an FE college set me on the path.

I would very much like to see FE providing opportunities for people of all ages to study, train and retrain. This will become especially important when the idea of a job for life has gone, with technological shifts causing the economy to change more rapidly than ever before. In short, lifelong learning will become crucial. Former education secretary David Blunkett had a vision for this in his 1998 green paper The Learning Age.

In 1993, FE colleges were taken out of local authority control and forced to become independent entities. Reductions in lecturers’ pay, reduction in courses and general decline in the status of further education colleges followed. Perhaps Andy Burnham will consider giving local mayors more control of funding, curriculum and management.

I would also like to comment about his desire to put an end to the academic and vocational divide. This is a socially constructed divide reinforced by class prejudices; Professor Alison Wolfe said that vocational education was often regarded as ‘a great idea for other people’s children’.

Clearly skills are vital to personal and economic growth, but the phrase ‘learning and skills’ seems to imply that skills are for people who can’t manage learning. We should only talk about ‘education’ and ‘learning’. In the future people will need to learn, unlearn and re-learn throughout their lives.

My final wish for FE is a revitalisation of its role in building communities and bringing people together. I remember the times when non-vocational courses such as painting, woodworking and yoga were well attended, and relatively inexpensive. People learned, made friends and enjoyed themselves; they didn’t even have to undertake a qualification!

Let’s make further education central to life and learning again.

Hybrid assessment models are becoming essential for modern learners

As higher and further education institutions across the UK enter peak assessment periods, long-standing cracks in traditional assessment models are becoming harder to ignore. Ageing exam hall infrastructure, limited campus space and tightening budgets are colliding with rising student demand and forcing institutions to rethink how assessments are delivered.

During exam season, campus space is often stretched to capacity, as sports halls, classrooms and other venues are repurposed to accommodate students sitting their exams. This creates logistical challenges and limits teaching flexibility.

With 89 per cent of universities planning to scale back repairs and maintenance by 2029 and a £5.6 billion backlog in restoring research facilities, the reliance on paper-based exams is becoming increasingly difficult to sustain. At the same time, students expect seamless digital experiences while institutions must safeguard academic integrity in the age of generative AI.

In further education, the Institute for Fiscal Studies has reported that funding would need to rise by £150 million by 2028-29 to maintain the current spend per student. Meanwhile, demand for further education courses is rising sharply, placing additional strain on resources and forcing some institutions to either turn away or enrol more students with little to no additional funding.

Hybrid assessments are continuing to rise

Amid these pressures, hybrid assessment models that combine paper and secure online exams are continuing to gain traction across different education settings. Rather than replacing traditional, in-person exams entirely, this approach gives institutions the means to align assessment methods with capacity, student needs and peak demand periods.

Beyond flexibility, hybrid models can also enhance accessibility when implemented with the right oversight and governance. Providing students with a wider mix of assessment methods can help ease logistical pressures and packed exam timetables as well. For students balancing studies with employment, training, commuting or caregiving, and particularly for SEND students, the option to take certain exams remotely makes participation more manageable and ensures exam delivery isn’t limited by physical campus restraints.

Futureproofing assessment models

While online exams raise valid concerns about cheating and this often acts as a barrier for adoption, addressing these concerns requires more than replicating in-person exam conditions digitally.

Preventing cheating in online environments requires a layered approach. Randomised question banks and clever question design, like open-ended or scenario-based questions, encourage students to think critically. These act as a soft barrier that makes AI-assisted cheating more difficult by design. A technical layer, including browser lockdowns or copy-paste restrictions, provides an additional hard barrier, while the proctoring layer adds oversight by monitoring students during an exam.

Cheating definitely needs to be managed in online exams, but applying maximum levels of surveillance is not the shortcut to prevent it. The goal should be to create conditions where students can focus on the exam itself, not the feeling of being watched.

By spreading risk across these layers and embedding integrity across the entire exam structure, institutions can demonstrate they are not reliant on surveillance alone to validate results. This is also critical for GDPR compliance and addressing student privacy concerns, as it reduces the need for excessive data collection.

As demand for physical space and student numbers continue to rise, a one-size-fits-all approach to assessment is becoming less viable. Hybrid models allow institutions to tailor assessment strategies accordingly, aligning delivery methods with learning outcomes and practical constraints.

Instead of viewing hybrid models as an integrity trade-off, institutions should think of it as a design challenge that needs to be addressed not just through technology but also with governance and strategic intent.

Modern learning requires a flexible approach

Learning and assessment cannot be thought of as a binary choice between paper and digital. It’s not about replacing one form of assessment with another, but rather an opportunity to create exam systems that are capable of scaling with physical campus demands, accessible and fair for students, and able to withstand the financial pressures many higher and further institutions are currently navigating.

Institutions that embrace hybrid assessment models will be better positioned to balance academic demands with operational realities, while meeting evolving student expectations. In a sector shaped by growing student numbers and limited resources, flexibility is no longer a nice-to-have but an operational necessity.

Whitehall targets limit flex, mayors tell spending watchdog

Mayors with new devolution deals are complaining that targets dictated by Whitehall are limiting their local flexibility, according to a report by the national spending watchdog.

Seven mayor-led combined authorities now have integrated settlements, or “single pot”, deals that give “greater funding flexibility” to address local needs.

The deals mean that the combined authorities are handed a single multi-year funding pot running from this financial year to 2028-29, with spending organised into six “themes” such as skills and employment support.

This replaces a system of grants for specific funding streams, such as the adult skills fund and skills bootcamps, that are ring-fenced and renewed annually.

But according to the National Audit Office (NAO), agreement about what targets mayors should work towards was delayed until March this year due to a “tension” between mayors and government departments that previously controlled the funding.

In a report ‘Devolution in England: funding and accountability’, the watchdog said mayors felt the targets, set out in pre-agreed “outcome frameworks”, were too strongly shaped by national priorities, contained too many output measures, and were “limiting their ability to reflect local need”.

Mayors told the watchdog some departments were “initially reluctant to flex” on the number of targets, with the Department for Work and Pensions demanding specific outcomes, and the Department for Transport initially proposing 44 separate indicators.

Some mayors told the NAO they felt government departments treated them as “delivery arms of central government”.

But one department told the spending watchdog their approach was “deliberate” and aimed at managing delivery risk due to mayors proposing targets that “lacked a clear logic or baseline”.

Defending their approach, government departments told the NAO that the targets were a necessary part of accountability to ministers and Parliament, which ultimately approves all public spending.

But performance reporting burdens on mayors are currently “greater than initially envisaged” in the English Devolution white paper, which promised to replace “top-down micromanagement” of local leaders with “locally accountable autonomy”, the watchdog concluded.

Due to the tensions, the NAO said negotiations leading up to March were “protracted and resource intensive” and called on mayors, government departments and the Ministry for Housing, Local Government, and Communities, to ensure targets in future deals are “proportionate” and reflect the minimum assurance needed.

It also called on the government to strengthen local scrutiny arrangements, minimise the amount of performance information departments ask for, and to agree outcomes frameworks in a “timely manner”.

Reports from combined authorities show that outcome frameworks were supposed to have been agreed by January this year, but slipped until mid-March, shortly before the start of the new financial year.

In an update in March, South Yorkshire Combined Authority said the delays were due to demands from the Department for Transport and “relatively late” inclusion of targets linked to the government’s construction skills package, which is overseen by the Department for Education and Department for Work and Pensions.

It added that the targets only reflect funding streams in the integrated settlement, rather than “everything we care about in South Yorkshire”.

What are the targets?

Skills-related targets agreed by West Yorkshire include numbers of residents enrolling on education courses, starting a supported employment programme, or completing an industry placement.

While some annual targets such as numbers of learners progressing to a level 3 qualification for the first time have “baseline” figures, several do not.

The Greater London Authority has agreed to a target of 33,200 adults being trained to progress in or towards key sectors over three years, against a baseline of 13,000 in 2023-24.

Combined authorities are held to account for their performance by “programme boards” of government officials who meet with mayoral teams every six months to discuss progress.

The English Devolution and Community Empowerment Act, which received royal assent in April, will also give extra scrutiny powers to committees of local councillors from April next year.

This includes the power to carry out thematic inquiries, to require key people from a combined authority to attend meetings and answer questions, and a duty to assess value for money of decisions mayors take.

The West Midlands Combined Authority, which has a trailblazer integrated settlement which launched in 2025-26, a year earlier than most areas, has already set up a new local scrutiny committee.

At its first meeting yesterday, officials told reported that three out of ten outcomes were rated ‘amber’, with the remaining seven rated ‘green’.

Its target of increasing the number of residents receiving level 3 qualifications was rated amber, as only 9,812 learners started a such a programme against a target of 14,200.

Gareth Thomas, a skills policy consultant, said: “The use of nationally set targets does support management of allocations across the country, as well as national statistical analysis and statistical publications, however it does potentially restrain the impact budgets can have locally.

“For example, if we want to compare the performance of England (or the UK) to other OECD countries we need a common measure.

“National measures and headlines are important for national statements and political headlines. This may also have something to do with why we still have such measures.”

He added that some of the adult skills related targets related to enrolments and achievements appeared to be based on DfE “thinking” and could evolve following skills policy moving to the DWP and Andy Burnham becoming prime minister.

An MHCLG spokesperson said: “We welcome the National Audit Office’s report’s recognition of the significant progress made to move power and funding into the regions, empowering mayors to make decisions that work best for their communities.

“Our English Devolution and Community Empowerment Act gives mayors unprecedented powers over planning, transport and local growth, making it easier and quicker to devolve powers out of Whitehall to regions across England.”