What Should a Development Viability Sensitivity Analysis Test?
A practical guide to sensitivity analysis in development appraisals, including which variables to test and how to identify the assumptions that can change a site decision.
7 min read
A development viability sensitivity analysis should test the assumptions most capable of changing the decision on a site. In practice, that usually means development value, development quantum and mix, build costs, abnormal costs, affordable housing and planning obligations, programme, sales rate, finance and developer return. The point is not to create a large grid of numbers. It is to show which assumptions move profit or residual land value enough to change what the team should do next.
How This Works
Start with one clearly labelled base case. The base case should contain the current best view of the scheme, values, costs, obligations, programme, finance and target return.
Then identify the assumptions with either high uncertainty, high financial impact, or both.
Typical sensitivity variables include:
private sales values or other GDV inputs
development quantum, unit mix and tenure
base build cost
external works and infrastructure
abnormal costs
affordable housing assumptions
Section 106 and CIL
construction and sales programme
interest rate and finance assumptions
developer return
Homes England notes that development appraisal tools are commonly used to analyse multiple development scenarios and run sensitivity analysis to explore the effect of changes in key variables on viability. The useful word is key. Testing every cell equally can hide the variables that actually matter.
A simple one-way sensitivity changes one assumption while holding the others constant. A two-way sensitivity changes two important assumptions together, such as sales values and build costs. More complex scenario testing can be useful, but only when the assumptions remain understandable and evidence-led.
Worked Example
Take an illustrative residential appraisal with:
GDV: £12.0 million
development costs excluding land and target profit: £7.0 million
target developer profit: £2.1 million
base residual land value: £2.9 million
Now test sales values.
If GDV falls by 5%, GDV reduces by £600,000. All else equal, the residual falls from £2.9 million to £2.3 million.
Now test build cost. If the base build allowance within the appraisal is £4.5 million and that allowance rises by 5%, cost increases by £225,000. All else equal, the residual falls from £2.9 million to £2.675 million.
If both happen together, the simplified residual falls to £2.075 million.
Nothing about the physical site changed. The decision may still change because the commercial assumptions did.
That is the practical value of sensitivity analysis. It tells the team which questions deserve more evidence before money or negotiating position is committed.
Review Method
A useful sensitivity review should answer six questions.
What is the base case?
The team should be able to see the exact assumptions being sensitised and the date and source behind them.
Which inputs are genuinely uncertain?
Do not spend time sensitising a confirmed minor fee while ignoring an untested sales value, unit count or abnormal cost.
Which inputs have the biggest effect on the decision?
The most important variable is not always the one with the widest percentage range. It is the one capable of changing the acquisition, pricing or investment conclusion.
Are the sensitivity ranges reasonable?
Ranges should be based on evidence, professional judgement or a clearly labelled stress case. An arbitrary plus or minus percentage can create false precision.
Does timing need to be tested?
A slower build or sales rate can increase finance costs and delay receipts even if the headline GDV and cost totals are unchanged.
What is the decision trigger?
Record the point at which the team would pursue, renegotiate, reappraise, hold or reject the site.
Sources Used
Limitations And Professional Review
Sensitivity analysis does not tell you which assumption is correct. It shows what happens if an assumption changes.
The ranges used can be as misleading as the base case if they are not supported by evidence. A sensitivity table should therefore sit next to the source, date and status of the important inputs.
Where the appraisal is being used for formal valuation, planning viability, lending, tax, cost advice or another professional purpose, the relevant standards and qualified review should be applied.
Where Harold Viability Fits
Harold Viability is designed to keep sensitivity next to the appraisal assumptions and evidence that produced the base case.
The current workflow can structure the development schedule, GDV, affordable value, CIL, Section 106, build costs, externals, abnormals, professional fees, contingency, finance, target profit, residual land value, cash flow and sensitivity in one reviewable place.
That makes the useful question easier to answer: which assumption is driving this result, and would changing it alter the decision on the site?
Get your site appraised with Harold Viability.
For the wider land workflow, see Harold Land.
Article written by
