TEARDOWN Published 29 September 2026 at 06:24. Evidence-based. Source-cited. No sponsored content.

DWP's new employment fund promises £10 million testing what helps disabled people into work, then a further £50 million guided by the results. Its own timetable opens that bigger round before any funded project has delivered a single month of work.

4 out of 5 stars4/52 documented mistakes in this teardownHow ratings work

Estimated reading time: 6 minutes

Telford Jobcentre Plus building, a modern low-rise office block with a car park in front.
Jobcentre Plus, Telford, the kind of frontline DWP employment support building not directly part of this story. Photo: Ian S / Wikimedia Commons, CC BY-SA 2.0.

In short. The Pathways to Work Innovation Fund's applicant bid pack lists two objectives: getting disabled people and people with long-term health conditions into work, and "to build evidence about what works and for whom, so that successful approaches can be adopted more widely" (page 7) [2]. The same document's own milestone table shows award decisions for the first, smaller round only starting in January 2027 and project mobilisation only beginning "From February 2027" (page 16) [2]. DWP's own guidance page says the second round, worth five times as much, "is expected to open to bids early in 2027" [3]. Separately, nothing in either document names who aggregates the fund's cross-project learning, publishes it, or by when. Two documented mistakes: four stars.

Pat McFadden, the Work and Pensions Secretary, launched the Pathways to Work Innovation Fund on 14 July 2026, citing 2.8 million people out of work due to ill health and the Keep Britain Working review's estimate that economic inactivity linked to health conditions costs the country £212 billion a year [1]. That review's own final report puts the figure slightly differently: "the state faces an unsustainable cost from economic inactivity due to ill-health of £212 billion per year, equivalent to 7% of GDP... through lost output, increased welfare payments and additional burdens on the NHS" [4] - a cost to the state and the health system, not only to "the economy" as the press release frames it, though the figure itself checks out. The fund itself is worth up to £60 million over two years and sits inside Pathways to Work, described on DWP's own guidance page as "the government's £3.5 billion investment to support disabled people and people with health conditions to move into, stay in and progress in work" [3]. Applications for the first round opened 28 September 2026 and close 26 October 2026 [2].

A fund built to test first, but its own clock rules that out

The applicant bid pack is explicit about what the fund is for. "The Pathways to Work Innovation Fund has two objectives: To support disabled people and people with long-term health conditions to move towards, enter, stay and progress in work" and "To build evidence about what works and for whom, so that successful approaches can be adopted more widely" (page 7) [2]. The logic is a standard test-then-scale model: fund a first round of ideas, learn what works, then use that evidence to steer where the far larger sum goes next.

The document's own timetable does not leave room for that sequence to happen. The first round's application window closes 26 October 2026, shortlisting outcomes follow on 9 November, and a refinement stage runs into early December. Then: "Award decisions communicated From January 2027" and "Project mobilisation begins From February 2027" (page 16) [2]. Funded projects then have up to twelve months to deliver: "Projects must be delivered within 12 months of signing the Grant Funding Agreement" (page 10) [2], with funding for the whole fund running "up to the end of March 2028" (page 10) [2].

Against that, DWP's own guidance page for the fund states plainly: "Up to £10 million is available for the first funding round which is opening September 2026" and "Up to a further £50 million will be available for the second funding round, which is expected to open to bids early in 2027" [3]. Round one's projects are only beginning mobilisation in February 2027, by the fund's own table; round two, carrying five times the money, is expected to be taking bids in that same window. No first-round project will have delivered a day of funded work, let alone generated evidence about what helped someone into a job, before the fund starts allocating the larger sum. The document that promises to "build evidence... so that successful approaches can be adopted more widely" has published a schedule in which the bigger round cannot be informed by anything the smaller one produces.

No one named to turn "learning" into a verdict

The fund's second objective rests on a word that recurs constantly through the bid pack without ever being pinned to a process. "We recognise that innovation involves uncertainty. Not every idea will succeed, but every project should generate learning and evidence that can improve future services" (page 10) [2]. Applicants themselves are assessed partly on "the value in the learning that could be generated" (page 10) [2], and are asked directly whether they "can participate in monitoring, evaluation and learning activities... and comply with Fund reporting requirements if funded" (page 32) [2].

What the document never states is what DWP itself will do with that learning once collected: who reviews it across projects, what counts as a "successful approach" worth adopting more widely, or when any aggregate finding will be published. The bid pack's own glossary defines evaluation only in the abstract: "Assessing how well a project worked and what can be learned from it. Information on how the government approaches evaluation can be found in the Magenta Book available on GOV.UK" (page 28) [2] - a pointer to the Treasury's general evaluation guidance, not a commitment specific to this fund. The Ministerial Foreword promises "a duty to ensure our investment in this has the maximum effect" (page 4) [2], but nothing in either document names an owner, a synthesis process or a publication date for the evidence the fund exists to generate.

Credit where due

Several parts of this are handled well. The bid pack is genuinely candid that failure is expected and useful: "not every idea will succeed", and the Fund says it is "interested in learning from challenges and adapting approaches during delivery as well as capturing success" (page 32) [2], rather than quietly dropping projects that do not deliver. The one financial safeguard the document does spell out is specific and sensible: Mobilisation Payments of up to 30% of a grant are discretionary, and "unspent funds, or funds not used for the agreed purpose, may need to be repaid" (page 17) [2]. The bid pack is also published in Easy Read, large print, British Sign Language and audio formats alongside Welsh, and an expert panel including the Paralympian Tanni, Baroness Grey-Thompson, is named as advising on which bids get funded, "ensuring the voices and experience of disabled people are placed at the very heart of the process" [1]. None of that changes the sequencing problem above; it means the fund is well run in the parts it has actually specified, which makes the two gaps more conspicuous rather than less.

The claims, tested

The claim What the evidence shows Verdict
The Fund's second objective is "to build evidence about what works and for whom, so that successful approaches can be adopted more widely" (page 7) [2] The bid pack's own milestone table has round-one award decisions only "From January 2027" and mobilisation only "From February 2027" (page 16) [2]; DWP's guidance page has the five-times-larger second round "expected to open to bids early in 2027" [3] The bigger round is timetabled to open before the smaller round has produced anything to learn from
Projects are assessed on "the value in the learning that could be generated" and must "comply with Fund reporting requirements" (pages 10, 32) [2] Evaluation is defined only generically, pointing to the Magenta Book, with no owner, synthesis process or publication date named for this fund's own findings (page 28) [2] Individual projects must report learning; nothing says how or when that learning becomes a fund-level verdict

Two documented mistakes. (1) The Fund's stated purpose is to test ideas and use the evidence to guide "successful approaches" that can be "adopted more widely", but its own milestone table has the first round's projects only beginning mobilisation in February 2027, the same window its guidance page names for the second, five-times-larger round to open for bids - leaving no time for evidence to flow from one to the other. (2) Applicants must report on "learning" as a condition of funding, but neither the bid pack nor the guidance names who reviews that learning across projects, what counts as a "successful approach", or when any findings will be published. Rated four stars out of five.

Sources

  1. £60 million fund launched to transform employment support for disabled people (GOV.UK news story, Department for Work and Pensions, 14 July 2026)
  2. Pathways to Work Innovation Fund: applicant bid pack (PDF, 65 pages, Department for Work and Pensions)
  3. Pathways to Work Innovation Fund guidance (GOV.UK, Department for Work and Pensions)
  4. Keep Britain Working Review: final report (GOV.UK, independent review commissioned by the Department for Work and Pensions)
SHARE THIS ARTICLEXBlueskyMastodonLinkedInRedditEmail