Viability Assessment input schedule and its assumptions

A viability assessment is only as good as its input schedule, and the input schedule is where almost every argument between an applicant and a council actually happens. Two appraisals of the same scheme can differ by several million pounds without either being dishonest, because the assumptions on sales values, build costs, benchmark land value and developer return each carry a defensible range.

So set the schedule out plainly, one input per line, each with its source and its date. An assessment that buries its assumptions in a narrative is treated as evasive, and the reviewer's first act is to rebuild it.

Gross development value and the comparable evidence behind it

For residential, gross development value comes from a schedule of units by type, size in square metres and net internal area, with a rate per square metre applied to each. Show the schedule. A single blended rate across a scheme tells the reader nothing.

Support the rates with comparable evidence: recent transactions of similar new-build units within a defined radius, adjusted for date, size, aspect, parking and specification, with each adjustment stated. Second-hand comparables need a new-build premium, and the premium has to be evidenced rather than assumed.

Commercial floorspace is valued on rent and yield with a deduction for void and incentives. State the rent per square metre, the yield, the assumed rent-free period and the letting void, and say what evidence sits behind each.

Affordable housing is valued on the transfer payment a registered provider would make, not on open market value. Where that payment is expressed as a percentage of market value, say what percentage and where it came from. This is the input that most often does the work in an assessment arguing for reduced provision, so it deserves the clearest sourcing in the document, and it should reconcile with the Affordable Housing Statement.

Build costs and where the figure comes from

Base build costs should come from a published cost index appropriate to the building type and location, or from a quantity surveyor's cost plan for the actual scheme. Say which. An index figure adjusted for location and date is acceptable on a modest scheme and thin on a complex one.

Abnormals are the contested part. Demolition, remediation, piling, retaining structures, diversions, archaeology and off-site highway works are all legitimate where they are evidenced by the site investigation and priced by someone competent. They are not legitimate as a round sum described as site abnormals.

List each abnormal separately, cross-refer it to the report that identified it, and price it. An abnormal that cannot be traced to a ground investigation, a utility response or a highway requirement will be struck out by the council's reviewer, and rightly.

Include the costs that policy imposes: biodiversity net gain delivery, the energy and carbon measures the local plan requires, fire safety provisions, and any infrastructure obligation. Exclude the ones already counted elsewhere, since double counting between contingency and abnormals is a recurring error.

Contingency should be stated as a percentage of build cost with a reason for the level chosen. A higher contingency on a deep refurbishment is arguable. The same figure on a greenfield plot is not.

Professional fees, finance and the ones that get missed

Professional fees are usually expressed as a percentage of build cost. Whatever percentage is used, justify it against the complexity of the scheme rather than stating it as a convention.

Finance needs four inputs: the interest rate, the arrangement fee, the assumed drawdown profile and the programme length. The programme is the one that moves the answer most, because interest compounds over the build and the sales period. State the assumed programme in months, with the build period and the sales rate separately, and make sure they match the programme used elsewhere in the application.

Then the items that get left out. Marketing and sales agency fees, legal fees on disposal, warranty and insurance costs, the community infrastructure levy where the scheme is liable, section 106 obligations itemised rather than lumped, and monitoring fees. A schedule missing the levy is a schedule that has to be redone.

Benchmark land value and how it is derived

National guidance is clear that benchmark land value should be established on the basis of existing use value plus a premium for the landowner, with the premium reflecting a reasonable incentive to bring the land forward while taking account of policy requirements including affordable housing.

Two things follow. The price actually paid for the site is not the benchmark. A developer who overpaid cannot pass that on through a reduced affordable housing contribution, and the guidance says so directly.

And existing use value has to be evidenced. For a working industrial building that means rental evidence and a yield. For agricultural land it means land sale comparables. For a vacant site with no use, the value is closer to a cleared site value than to the hope value of a previous offer.

Alternative use value can be relevant where an alternative use has an implementable permission or would clearly obtain one. It is not a way of importing the highest theoretical value someone can imagine for the site.

The developer return the guidance contemplates

The Planning Practice Guidance on viability states that a return to developers in the range of 15 to 20 per cent of gross development value may be considered suitable, while noting that a lower figure may be appropriate in some circumstances, and that a lower return on the affordable housing element reflects its reduced risk.

Use those figures as the reference point and explain any departure. A return argued above the range on a scheme with pre-sold affordable units and no abnormal risk will not be accepted, and asserting it damages the credibility of the inputs that were reasonable.

Where the scheme is delivered in phases, say whether the return is calculated on each phase or across the whole, because the two give different answers.

Sensitivity testing that tells the reader something

A sensitivity analysis showing plus or minus five per cent on everything is decoration. A useful one isolates the inputs that actually move the conclusion and shows the point at which the answer changes.

Run sales values, build costs, the finance rate and the programme separately, and identify the break-even level for each. Then state which affordable housing provision the scheme could support at a range of sales values. That single table is what a council's reviewer and a committee both want, and it turns an argument about assumptions into a discussion about risk.

Where the appraisal is close to the margin, expect the council to propose a review mechanism rather than accept the reduced provision outright. Drafting that mechanism is a separate discipline, covered in the note on late stage viability reviews.

Transparency and publication

Viability assessments should be prepared on the basis that they will be made publicly available, save in rare cases. Write the document accordingly. Redacting the inputs after submission is a poor look and usually fails.

Present the appraisal in a form that can be rebuilt: the input schedule, the cashflow, and the outputs, with enough detail for the reviewer to reproduce the result. Councils appoint their own advisers, and an appraisal that cannot be replicated will be replaced by one that can, which is how competing viability appraisals at appeal usually begin.

A Viability Assessment is a report councils expect to see from a suitably qualified author working to the relevant professional standards, and professional review is strongly advised before submission. For specialist technical reports, qualified professional input may be required, and the professional directory is the route to arranging that review.

PlanScope provides AI-assisted draft reports for planning applications. Reports should be reviewed before submission. We do not guarantee planning approval.