Starts are up, and under-19s are the only age group falling
Every lever pulled in the last two years points at young people. The data points the other way.
Provisional figures published on 16 July cover the first three quarters of 2025/26, August 2025 to April 2026. The headline is good. Starts reached 308,770, up 8.7% on 284,190. Participation reached 735,500, up 4.6%. Achievements reached 146,210, up 14%, which is the strongest of the three numbers and the one least often quoted.
The age breakdown is where it gets awkward. Starts among the under 19s fell 5.4% to 63,530. Starts among 19 to 24 year olds rose 5.1% to 83,570. Starts among those aged 25 and over rose 17.5% to 161,670. More than half of all apprenticeship starts in England are now aged 25 or over, and the only age group going backwards is the one every recent policy decision has been aimed at.
The level breakdown tells the same story from a different angle. Higher level starts reached 126,660, up 13.6%. Level 4 rose 15.7%, level 5 rose 20.6%, level 6 rose 4.9% and level 7 rose 11.6% to 28,800.
That level 7 figure needs reading carefully. Funding for level 7 was withdrawn for new starters aged 22 and over from January 2026, part way through the period these figures cover, and starts still rose. That is the pre-deadline rush rather than growth, and it will not repeat.
What it changed from: a decade in which starts among 16 to 24 year olds fell roughly 40% while spend concentrated in older, higher and more expensive provision. That is the pattern the sixteen defunded standards, the level 7 restriction, the age-based co-investment and the hiring payment were all designed to break. What it changed to, on this evidence: not much yet. The interventions landed from January 2026 onwards and most of this period predates them.
What follows: the full year release lands in November, and it is the first set of figures that will partly reflect the new policy. If it repeats this pattern, it is the strongest available argument for further intervention, and on recent form the most likely shape that takes is more defunding at the top of the system rather than more money at the bottom. If you deliver level 5 provision, that 20.6% is a number worth knowing someone is looking at.