Why are teachers leaving the profession in the UK?

Teacher retention in England improved slightly in the last two years. Fewer people left the state sector than the year before, and applications to train recovered from their low point. If you only read the headline, the problem looks like it is easing.

The rest of the data says something less comfortable. Reported wellbeing among teachers and leaders sits at about its lowest level since 2019, and the proportion saying they are considering leaving has barely moved. A profession can hold on to people and still be running on empty. That is roughly where we are.

29%

Workload is the reason most often given, but hours are only part of it

Workload comes first in almost every survey of why teachers leave, and it has done for a decade. What has changed is the shape of it. Average reported working hours have edged down. Reported wellbeing has not followed.

That gap is the interesting part. If workload were purely a question of hours, cutting hours would lift wellbeing, and it has not. Three other things travel with it:

Predictability. A heavy week that was planned is survivable. A heavy week created on Thursday afternoon by something avoidable is not.

Control. Teachers describe the same volume of work very differently depending on whether they had any say in it. Around four in ten say the demands on them are unacceptable and outside their control, and it is the second half of that sentence that does the damage.

Worth. Marking that changes what a teacher does next is tiring. Marking that exists to be seen by somebody else is corrosive.

Wellbeing and burnout

Education Support's annual index has recorded stress levels above three quarters of the workforce for several years running, with a substantial minority reporting symptoms they would describe as burnout. Leaders report worse figures than classroom teachers, which is worth saying out loud in a sector where the person responsible for staff wellbeing is often the person with the least support.

Burnout is not simply tiredness. It has three recognised parts: exhaustion, a growing distance from the work, and a sense that nothing you do is effective. The third one is what turns into a resignation. A teacher who is exhausted but still believes the job matters usually stays. A teacher who has concluded that effort changes nothing does not.

Feeling unheard

This is the one schools tend to underestimate, and the one that shows up most clearly in what teachers write in their own words.

Staff are asked their views more than they used to be. Surveys go out. What follows is frequently silence. Nothing visible changes, and nobody explains why. The reasonable conclusion a teacher draws is that the exercise was decorative.

86%

The cost is cumulative. Each unanswered survey makes the next one less likely to be answered honestly, which makes the data worse, which makes acting on it harder. Schools then conclude that staff voice work does not tell them anything useful, and they are right, because the loop was never closed.

Recognition, behaviour and progression

Three more reasons appear consistently and are worth separating out.

Recognition. Not awards. Being noticed by name when something went well, by someone whose opinion the person respects.

Pupil behaviour. Often described alongside workload rather than instead of it, because behaviour that is not backed consistently by leadership creates both stress and extra work.

Progression. Teachers who cannot see what the next five years look like are more likely to look outside the sector for the answer. This affects experienced classroom teachers who do not want to become leaders, a group with very few visible routes.

What actually holds people

The research points repeatedly to the same protective factor. Teachers who feel supported by their leadership are markedly less likely to be considering leaving, even when their workload is high. Support is not a soft finding. It is one of the strongest signals in the data.

Support is also specific. It means someone knows what your term looks like, something got taken off you when it became unreasonable, and when you raised a problem, something happened.

What a school can do about it

Almost every reason above sits inside a school's control. Not pay, not national accountability, but workload design, communication, recognition and behaviour policy all do.

Three things make the difference in practice.

Ask often and briefly. One long survey a year is a single snapshot, and by the time it is analysed the term it described is over. Short, regular, anonymous check-ins show you the shape of a year, including the weeks that predictably go wrong.

Do a small number of things. Three changes with a named owner and a date beat twenty priorities in a document. A score on its own changes nothing.

Tell people what changed. Including what is not changing yet, and why. This is the step that gets skipped, and it is the one that determines whether anyone answers next time.

The window before the letter

The thing worth holding on to is timing. A resignation is the end of a process that usually started months earlier, and for most of those months the person was still reachable. Schools rarely find out during that window, because nothing in the calendar asks.

That is the gap. Not a shortage of data about teachers, but a shortage of moments where someone is asked, in a way that is safe to answer honestly, while there is still time to do something.