Floorspace forecasting in an employment land assessment
Employment land forecasts are built from four methods that rarely agree. Here is how each one works, where the assumptions hide, and how a deliverable supply review changes the answer.
An employment land assessment floorspace forecast exists to answer one question: how much B2, B8 and E(g) floorspace does this area need over the plan period, and does the current supply meet it? Everything else in the study is either an input to that number or a sense check on it. Get the method wrong and the whole evidence base collapses under cross-examination, because an inspector can test each step arithmetically.
The forecast is not a single calculation. National guidance expects authorities to look at economic need through more than one lens, then exercise judgement about which answer to carry forward. An applicant arguing for release of a poor employment site, or for more land than the plan allocates, works with the same methods in reverse.
What the forecast is actually used for
Three decisions rest on it. The plan's employment land requirement in hectares or square metres. The judgement about whether a particular site should stay protected or be released. And, on appeal, whether a shortfall exists that weighs in favour of permission.
Those are different questions and they need the number presented differently. A plan-wide requirement can live with a range. A site-specific release argument needs the forecast broken down by use class and sub-area, because a surplus of large-format warehousing says nothing about whether the area is short of small workshop units.
An Employment Land Assessment that produces one headline hectarage and stops there is of little use to either side.
Labour demand against past completions
The labour demand method starts with an economic forecast of job growth by sector, converts jobs into floorspace using employment densities, then converts floorspace into land using plot ratios. It is forward looking and sensitive to the forecast house used, which is why two studies of the same district can differ by a third.
The past completions method looks backwards. It takes gross and net floorspace completions over a run of years, calculates an annual average, and rolls that rate forward across the plan period. It is grounded in what the market actually delivered, and its weakness is obvious: if the area had no available sites for a decade, low completions measure constrained supply rather than demand.
| Method | What it measures | Main weakness |
|---|---|---|
| Labour demand | Forecast job growth converted to floorspace | Depends wholly on the chosen economic forecast |
| Past completions | Delivered floorspace over a trend period | Suppressed where supply was constrained |
| Labour supply | Resident workforce change and commuting | Weak link to floorspace by type |
| Property market signals | Vacancy, rents, enquiries, time on market | Qualitative, hard to convert to a number |
Run all four and the sensible forecast sits inside the spread rather than at one extreme. Where a study presents a single method only, that is the first thing to probe.
Take-up rates and the margin of choice
Past take-up matters twice over. Once as a demand estimate, and again as the denominator for expressing supply in years. An authority holding 40 hectares of allocated employment land against an annual take-up of two hectares has 20 years of supply, which is the sort of statement an inspector can check in one line.
A margin of choice, sometimes called a buffer, is then added so that the market has a selection of sites rather than a single option. The size of that margin is a judgement rather than a policy figure, and it is usually expressed as a number of years of take-up. The arguments about it are predictable: the authority wants a margin that makes its allocations look adequate, the promoter wants one that reveals a shortfall.
Be careful about the trend period. Including the years either side of a recession, or excluding them, moves the annual average materially. State the period, say why it was chosen, and show the alternative.
Sectoral breakdown by use class
Aggregate floorspace hides the problem. The 2020 amendments to the Use Classes Order moved offices, research and development and light industry into Class E(g), while general industry stayed in B2 and storage and distribution in B8. A forecast that reports a single industrial figure cannot tell you whether the need is for E(g)(iii) light industrial units of 200m² or B8 sheds of 20,000m².
Break the forecast down by:
- Office and research and development, now within Class E(g)(i) and E(g)(ii)
- Light industrial within Class E(g)(iii)
- General industrial within Class B2
- Storage and distribution within Class B8, split between regional and local scale
- Sui generis employment uses such as waste facilities and vehicle depots
Employment densities differ sharply between those categories, and so do plot ratios. Using an office density on a distribution forecast overstates the land requirement several times over. The published employment density guidance used across the industry gives ranges, not points, so state which end of the range you have used and why.
Qualitative review of the existing supply
Quantity is the easier half. The harder half is whether the sites already allocated or in use can actually accommodate the forecast need.
A site review should record, for each allocation, its area, whether it is serviced, its access from the primary road network, its ownership and whether that ownership is fragmented, any flood or contamination constraint, whether planning permission exists, and how long it has sat undeveloped. Sites failing on access or servicing are supply on paper only.
This is where the utilities capacity evidence point bites on employment land too. An allocation with no three-phase power and no realistic connection date is not deliverable within the plan period, whatever the hectarage says.
An illustrative reconciliation
Take a district whose plan allocates 30 hectares of employment land. The labour demand method produces a need for 22 hectares, past completions rolled forward produce 34, and the authority's own monitoring shows annual take-up of 1.6 hectares. Of the 30 allocated hectares, 11 sit on a single site with no vehicular access capable of taking HGVs and no funded highway scheme.
The honest statement is that the quantitative position looks comfortable against one method and tight against another, and that deliverable supply is nearer 19 hectares than 30. That is a stronger case for a windfall employment site than any argument about the headline figure. The numbers here are illustrative and used only to show the arithmetic.
Reconciling the forecast with the plan figure
Plan requirements are policy choices informed by evidence, not outputs of a spreadsheet. A plan can legitimately set a requirement above the demand-led forecast to support an economic strategy, and it can set one below if the land simply does not exist. What it cannot do is set a figure with no evidential trail.
So when a forecast is being challenged, the question is rarely "is this number right". It is whether the authority applied its own method consistently, whether the trend period and density assumptions are stated, and whether the qualitative supply review was done at all. Those are the gaps that get found.
The same discipline applies to a retail sequential test: the evidence has to survive someone else working through it line by line.
A PlanScope Economic Impact Assessment draft can set out the policy framework, the method options and the assumption table, and it flags missing site data rather than inventing it. Councils usually expect a suitably qualified author for this kind of economic evidence, so professional review is strongly advised before anything is submitted. PlanScope provides AI-assisted draft reports for planning applications. Reports should be reviewed before submission. We do not guarantee planning approval.
This is general guidance, not legal advice; check the position with your local planning authority or a qualified professional before relying on it.
Frequently asked questions
Which forecasting method carries the most weight?
None of them automatically. National guidance expects economic need to be tested through more than one method, with judgement applied to the results. In practice a forecast that sits inside the spread produced by labour demand, past completions and market signals is far harder to dislodge than one derived from a single method, however carefully that method was run.
How is a margin of choice calculated?
It is a judgement rather than a policy figure, normally expressed as a number of years of past take-up added on top of the forecast requirement. The point is to give the market a choice of sites rather than a single option. Expect the authority and any promoter to argue for different margins, and expect an inspector to want the reasoning rather than the number.
Does Class E make employment forecasting easier?
It makes it harder. Offices, research and development and light industry all sit within Class E(g) now, so a single planning permission can cover uses with very different employment densities and plot ratios. Forecasts therefore need to be broken down below use class level, by floorspace type, rather than reported as one aggregate industrial figure.
What makes an allocated site undeliverable?
Most often access or servicing. A site with no HGV-capable access, no three-phase power, no funded highway works or deeply fragmented ownership is supply on paper only. A qualitative review recording area, servicing, access, ownership, constraints, permission status and how long the site has sat undeveloped will usually reduce headline supply considerably.
Can an applicant use the council's own evidence base?
Yes, and it is usually the strongest material available. The authority's own employment land study, annual monitoring report and take-up figures are published documents it can hardly disown. The productive line is rarely that the numbers are wrong, but that the authority has not applied its own stated method consistently across all its sites.
Sources and further reading
This article is general guidance about the planning system in England and, where stated, the rest of the UK. It is not legal or professional advice. Check your local planning authority’s validation requirements and take professional advice for your specific site.