Benchmarks and comparators in an economic impact assessment
Gross job numbers are a multiplication, not an assessment. Here are the four input assumptions that matter, where each is published, and how the additionality deductions work.
Economic impact assessment benchmarks are where these reports are won or lost. The arithmetic is simple enough that anyone can check it, so the only thing standing between your gross job figure and a hostile officer report is whether every input assumption can be traced to a published source and whether the additionality adjustments have actually been applied.
An assessment that reports gross construction jobs and gross operational jobs, with no deduction for leakage, displacement or deadweight, is not an economic assessment. It is a multiplication.
Where the input assumptions come from
Four numbers do most of the work: floorspace, employment density, gross value added per job, and the additionality factors. Each has a conventional published source, and each should be cited in the report rather than appearing in a spreadsheet nobody sees.
Floorspace comes from the scheme drawings, and the report should state whether it is gross internal or net internal area, because densities are published against a specific basis and mixing them produces an error of twenty per cent or more without anyone noticing.
Employment density comes from the published employment density guidance used across the industry, which gives floorspace per full-time equivalent job by use type. Those are ranges in some categories and single figures in others. State which figure you used, on which area basis, and why.
Gross value added per job is taken from national statistics, either regional gross value added per filled job or a sector-specific equivalent. Say which year's data, because the figures are revised.
And the additionality factors come from the government's own appraisal guidance, which is where leakage, displacement, substitution and multipliers are defined.
Employment densities by floorspace type
The spread between categories is large enough that the choice of category matters more than the precision of the number.
| Floorspace type | Density basis | Common error |
|---|---|---|
| General office | Floorspace per FTE, net internal | Using net density against a gross area figure |
| Light industrial | Floorspace per FTE, gross internal | Applying an office density to Class E(g)(iii) space |
| Storage and distribution | Floorspace per FTE, gross internal | Ignoring the difference between a regional shed and a local unit |
| Retail | Floorspace per FTE, net sales area | Counting back-of-house area as sales area |
| Hotel and leisure | Jobs per bedroom or per unit | Reporting headcount as full-time equivalents |
Two disciplines reduce the risk of challenge. Report in full-time equivalents and say so, because headcount inflates the figure and officers now ask. And where a use could sit in more than one category, present both and carry the lower one forward.
The same care applies on the supply side of the argument, which is why an Employment Land Assessment and an economic assessment should share one assumptions table rather than contradicting each other across two documents.
Additionality, leakage and displacement
This is the part that gets skipped. Gross effects are converted to net additional effects through a sequence of deductions, each of which needs a stated value and a justification.
- Deadweight: what would have happened anyway, including on the site itself under a fallback scenario
- Leakage: benefits accruing outside the study area, usually because the workforce commutes in from elsewhere
- Displacement: activity or jobs drawn away from existing businesses in the study area
- Substitution: a firm substituting one activity for another with no net gain
- Multiplier: supply chain and income effects within the study area, which add back
Define the study area before any of this, and define it once. A benefit that is additional at national level may be pure displacement at district level, and the study area determines which answer you get. Reports that quietly switch study area between the leakage step and the multiplier step are not uncommon, and the switch is easy to spot.
Fallback deserves particular attention on brownfield sites. Where the existing buildings could lawfully be reoccupied, the deadweight position is not zero, and an assessment that treats a vacant industrial estate as generating no jobs in the counterfactual will be challenged on it.
Construction phase against operational effects
Keep them in separate tables and use the right unit for each.
Construction employment is temporary, so express it in person-years or as an average annual full-time equivalent over the build period, never as a permanent job figure. A scheme reporting 400 construction jobs where it means 400 person-years spread over four years has overstated the annual effect fourfold.
Operational employment is ongoing and can be reported as full-time equivalents, with the ramp-up period stated. A scheme that reaches full occupation in year five should say so rather than reporting the year five figure as if it arrives on completion.
Construction gross value added and operational gross value added also need separating, because the multipliers differ.
Avoiding double counting of benefits
Three overlaps recur.
Counting both the gross value added of the new jobs and the land value uplift, which captures the same economic activity twice. Counting business rates revenue as a benefit alongside the gross value added of the activity generating it. And counting supply chain jobs through a multiplier while separately listing named local suppliers.
A short reconciliation table at the end of the benefits chapter, showing each claimed benefit once with its source step, prevents all three. It also makes the report far easier for an officer to accept, because the officer can see that you looked.
Where the assessment feeds a planning obligations negotiation, the same figures will be read against the Viability Assessment, and inconsistency between the two is a familiar line of attack.
Sourcing every figure to a published dataset
The rule to work to: no number in the report without a footnote naming the dataset, the publisher and the year.
That includes the things that feel too obvious to source, like the local unemployment rate or the district's population. It also includes anything taken from the applicant, such as expected headcount at opening, which should be labelled as operator-supplied rather than presented as analysis.
Where no published figure exists, say so and describe the effect without a number. An unsourced percentage is worse than no percentage, because it invites the whole report to be discounted.
Presenting a range rather than a point estimate
A single number invites a single counter-number. A range with stated assumptions at each end invites a discussion about the assumptions, which is a better conversation to be having.
Run a low, central and high case. Vary the employment density, the displacement rate and the occupancy assumption, and state which variable moves the answer most. If the conclusion holds across the range, say that plainly, because it is the strongest statement available.
An illustrative calculation
A 6,000m² gross internal light industrial scheme. At an assumed density of 47m² per full-time equivalent the gross figure is around 128 jobs. Apply 10 per cent deadweight for the lawful fallback use of the existing buildings, 30 per cent leakage because the catchment workforce travels in, 20 per cent displacement of activity from elsewhere in the district, and a composite multiplier of 1.3.
The net additional district-level figure lands near 84 jobs, around two thirds of the gross. Run the same calculation with a 55m² density and 40 per cent leakage and it falls to roughly 59. Both numbers are defensible; neither should be presented alone. These figures are illustrative and the density assumption in any real case must come from the published guidance.
A PlanScope Economic Impact Assessment draft sets out the policy framework, the method and the assumptions table, and flags each figure that needs a sourced input rather than inventing one. Councils expect a suitably qualified author for economic evidence of this kind, so professional review is strongly advised. The same evidential discipline applies to a retail sequential test.
PlanScope provides AI-assisted draft reports for planning applications. Reports should be reviewed before submission. We do not guarantee planning approval.
Frequently asked questions
Where do employment densities come from?
From the published employment density guidance used across the development industry, which gives floorspace per full-time equivalent job by use type. The important detail is the area basis: some categories are published against net internal area and others against gross internal area, and mixing the two produces a large error that is easy for a reviewer to find.
What is the difference between gross and net additional jobs?
Gross jobs are the total supported by the development. Net additional jobs are what remains after deducting deadweight, leakage, displacement and substitution, then adding back supply chain and income effects through a multiplier. The deductions are defined in the government's appraisal guidance and each needs a stated value with a justification.
How should construction jobs be reported?
In person-years, or as an average annual full-time equivalent over the build period, with the period stated. Reporting the total person-years as if they were permanent posts overstates the annual effect by the number of years in the programme, and it is one of the first things an officer or an objector checks.
Does the study area change the answer?
Substantially. A job that is additional at national level may be pure displacement within a district, and leakage rises as the study area narrows. Define the area once, at the start, and apply it consistently through every step. Switching area between the leakage and multiplier steps is a common and easily detected error.
Should the assessment give one figure or a range?
A range, with low, central and high cases and the assumptions stated at each end. A single figure invites a single counter-figure. A range moves the discussion to the assumptions, and if the conclusion holds across the whole range that is the strongest thing the report can say.
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.