Economic Impact Assessment: additionality and displacement

An economic impact assessment in planning is only as good as its additionality calculation. Gross figures are easy: a contractor's programme gives person-years of construction work, and a floorspace schedule gives an operational headcount. The hard part is the deduction. A planning economic impact assessment has to strip out the jobs that would have arrived anyway, the ones taken by people living outside the study area, and the ones pulled away from an existing business down the road. What survives those deductions is the net additional effect, and that is the only number a planning officer should be weighing in the balance.

Most reports that fail do so because the gross number is presented with confidence and the deductions are buried in an appendix with no justification. Officers have got better at spotting this.

Fixing the study area and the counterfactual

Two decisions made in the first hour of work shape every figure that follows. The first is the study area. A district boundary, a travel-to-work area and a combined authority footprint will produce three different answers for the same scheme, because a job that is displaced within the district may still be additional at regional level. State the area, say why it was chosen, and keep it constant throughout. If local policy asks for effects at borough level, report at borough level and offer the wider geography as a secondary case.

The second decision is the counterfactual: what happens on this site, and in this area, if permission is refused. For a cleared brownfield site with an extant consent for a smaller scheme, the counterfactual is that consent, not a blank field. For an employment allocation, it is the allocated use coming forward on a slower trajectory. Assuming a do-nothing baseline when the site has a live permission inflates every downstream figure, and it is the first thing a sceptical officer will test.

Say plainly what the counterfactual is in a sentence. Then use it.

Gross construction and operational effects

Construction effects are temporary and should be reported as person-years of employment, not as a headcount that reads like a permanent figure. The usual route is to take the construction cost, remove land and fees to get a build cost, and convert that into employment using a published cost-per-job or turnover-per-worker ratio drawn from national statistics for the construction sector. Whatever ratio you use, cite where it came from and give the year, because construction output prices move.

Operational effects come from floorspace. Employment densities differ sharply by use, so a light industrial unit, a distribution shed and an office floor cannot share a single assumption. Set out the density for each use in a table, with the source and the assumed occupancy rate, and be explicit about whether the figure is full-time equivalent or headcount. Mixing the two across a scheme is a common and avoidable error.

Deadweight, leakage, displacement and substitution

Four deductions turn gross into net, and they are applied in sequence rather than averaged together.

1. Deadweight. The outcomes that would have happened anyway, under the counterfactual you set out above. On an allocated site this is often the largest single deduction.
2. Leakage. Benefits going to people or firms outside the study area, usually measured through expected commuting patterns from census or labour market data for the relevant travel-to-work area.
3. Displacement. Activity drawn away from existing businesses inside the study area. A new gym, a new convenience store or a new distribution operator competing for the same local labour and the same local customers all displace to some degree.
4. Substitution. A firm switching activity between its own sites, which is why a relocation from two miles away is rarely additional at district level even when it looks like 80 new jobs at the front door.

Then a multiplier is applied to what is left. Applying the multiplier before the deductions, which happens more often than it should, produces a number that is wrong by a wide margin and immediately discredits the report.

Multipliers and how far to push them

Composite multipliers capture supply chain purchases and the spending of wages in the local economy. They are geography-sensitive. A borough-level multiplier is modest because most supply chain spending leaks out almost immediately, whereas a national multiplier can be considerably higher for the same activity. Using a national figure to describe a district-level benefit is the single most effective way to lose an officer's confidence in the whole document.

Pick a multiplier consistent with the study area, cite the source, and run a sensitivity case with a lower value. Sensitivity testing is not an admission of weakness. It shows that the conclusion holds even on cautious assumptions, which is exactly what a committee report needs to be able to say.

Gross value added and how it is derived

Job numbers alone tell a committee very little about economic weight, so most assessments also report gross value added. The standard derivation is net additional full-time equivalent jobs multiplied by GVA per filled job for the relevant sector and region, taken from official regional accounts. Two disciplines matter here. Use the same year for the GVA figure and the price base for any capital cost, and say whether the annual GVA total is then discounted over an assumed operational period.

Where a report quotes a headline GVA figure over 30 years without saying whether it is discounted, undiscounted or cash, the figure is not usable. An illustrative example: 100 net additional full-time equivalent jobs in a sector with regional GVA per filled job of £60,000 gives £6m a year in gross value added before any discounting. That arithmetic is transparent and easy to check. A 30-year cash total of £180m from the same inputs is arithmetically true and analytically useless, and the second version is the one that tends to appear in a press release.

Keep the annual figure in the headline and put the cumulative case in the appendix.

Where officers challenge the assumptions

Four points recur when a case officer or a consultee economist reads the assessment. The counterfactual, because a do-nothing baseline is often indefensible. The displacement rate, particularly for retail, leisure and last-mile logistics competing in a tight local labour market. The commuting assumption behind leakage, where a generic national figure has been used instead of local travel-to-work data. And the multiplier geography.

Answer all four in the body of the report rather than waiting to be asked. A short assumptions table with a source, a value and a sensitivity range for every input does more to protect the conclusion than another five pages of narrative. If a number cannot be sourced, describe the effect qualitatively and say why it has not been quantified, which is a defensible position. Inventing a percentage is not.

One honest uncertainty is worth stating openly: displacement in a local service sector is very hard to measure, and reasonable analysts will disagree by tens of percentage points. Show the range.

If the scheme also turns on whether employment land should be retained or released, that is a separate evidence base with its own method, and the two documents need to agree with each other. Trade diversion in a retail impact test is measured quite differently from employment displacement, so do not borrow assumptions between them. Where the wider social case matters, the screening logic used in health impact assessment work carries effects that never reduce to a jobs figure, and officers often find those effects more persuasive than a large gross value added total.

The Economic Impact Assessment page sets out what the report contains and who usually prepares one. Councils expect a suitably qualified author for this kind of evidence, and professional review is strongly advised before it goes in with a major application. If you are still scoping the pack, the index of 66 report types shows what else a major commercial scheme usually carries.

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.