The employment gap

The most-quoted figure in disability statistics: what it measures, how it is produced, and what it leaves out.

What the gap is

The disability employment gap is the difference between the employment rate of disabled people and the employment rate of non-disabled people, expressed in percentage points. It has been published regularly since the Labour Force Survey began tracking disability as a separate variable, and it is now the figure most likely to appear in a ministerial speech, a Select Committee evidence session, or a press release about workplace inclusion.

At its simplest, the calculation is straightforward: take the share of working-age disabled people who are in employment, take the same share for non-disabled people, subtract. The resulting gap has sat, through most recent years, somewhere around 30 percentage points — though the precise figure varies by quarter and by the precise population included. Any specific value printed without a source and reference period is liable to mislead, because the gap moves. The Office for National Statistics publishes quarterly and annual estimates through its Labour Force Survey outputs and the annual Family Resources Survey feeds the health-and-disability chapter of the Department for Work and Pensions' summary statistics; the right practice is to cite whichever bulletin, table and period the figure actually comes from.

What it rests on

The gap is only as stable as the definitions underneath it. Employment status comes from Labour Force Survey responses and follows International Labour Organisation conventions: a person is employed if they did at least one hour of paid work in the reference week, or were temporarily absent from a job they hold. That part is relatively consistent over time.

Disability status is a different matter. The survey question, the threshold a respondent must cross, and whether long-term conditions that do not currently limit activity are included all shape who falls on which side of the line. The wording of the survey question is not a neutral administrative act — it is the instrument that builds the denominator from which the employment rate is then derived. A change to that wording, or a change to the routing that precedes it, can shift the measured rate for disabled people without a single person's circumstances changing. This is why comparisons across years require checking whether the series is genuinely unbroken.

The non-disabled comparator group is also constructed, not observed. It consists of everyone the survey classifies as not disabled, which includes people with conditions that fall below the threshold and people who do not identify with the disability framing at all. That group is not simply "the healthy population" — it is the population the instrument fails to capture as disabled.

What the gap does not tell you

A single-figure gap conceals a great deal. Employment rates vary substantially within the disabled population by impairment type, by severity, by age, and by whether the condition is visible or fluctuating. A gap computed on all disabled people averages across these differences and can remain stable even when the position of particular groups is shifting sharply in opposite directions.

The gap is also a rate, not a count. If the disabled working-age population grows — as it has done in recent years, partly through changing question wording, partly through genuine increases in the number of people reporting long-term conditions — then a stable gap represents more people out of employment in absolute terms, not the same number. Rates and counts tell different stories, and the gap statistic tells only one.

There is a further issue the figure cannot address at all: intersecting characteristics within the disabled population. A disabled woman of working age in a deprived area faces a labour market position that the aggregate gap obscures entirely, because sample sizes at that level of granularity become too small to publish with confidence.

The employment gap is a useful summary measure precisely because it is simple. It survives that simplicity as long as the reader knows which definition built it, which survey produced it, and which reference period it covers. Without those three coordinates, the number floats free of its foundations — and a floating number, however frequently quoted, explains nothing.

Threshold is an independent publication about disability statistics and governance. It is not an advice, advocacy or support service and does not offer guidance on individual circumstances, benefits, entitlements or rights.