The severe test: when highway impact justifies refusal

An application can only be refused on highway grounds where the residual cumulative impact on the road network would be severe, or where there would be an unacceptable impact on highway safety. Those are two separate tests, and confusing them is the most common weakness in both objections and appeal statements. Safety is not qualified by the word severe. Capacity is.

That distinction decides how a highway case should be built and where the evidence has to be strongest.

Where the test comes from and what it means

The severe test entered national policy in 2012 and has survived every revision since, including the current edition of the National Planning Policy Framework published on 17 August 2026. It replaced an earlier approach under which any measurable worsening of junction performance could support refusal.

Two words carry the load. "Residual" means the impact remaining after mitigation, planning obligations and the effects of sustainable transport measures have been taken into account, so the comparison is never with the do-nothing case. "Cumulative" brings in committed development, which is why a modest scheme can be refused in a growth area while a larger one elsewhere is not.

Severe itself is undefined, deliberately. Inspectors have consistently treated it as a high bar and have declined to convert it into a queue length or a degree of saturation. What they respond to is a characterisation of the effect on real users: how long the delay is, when it occurs, how many movements experience it, and what it means for people trying to leave a side road or cross on foot.

Safety is a different test

Highway safety sits separately in policy and is not subject to the severe qualifier. An unacceptable impact on highway safety is a sufficient reason for refusal on its own, and inspectors apply it more readily because the consequence is injury rather than delay.

The evidence base differs too. A safety case rests on visibility splays measured on site, on the geometry of the access, on speed surveys where the recorded speed differs from the limit, and on recorded personal injury collision data across a defined study area and period. A capacity case rests on trip generation, distribution and junction modelling.

Mixing them weakens both. An objector who argues that a junction will be busy and therefore dangerous has made a capacity argument and will be answered with a capacity analysis.

Residual impact after mitigation

The mitigation package is where most of the argument happens. A Transport Assessment should present the network with the development and the agreed mitigation in place, and the residual figure is what is then tested against the policy.

Mitigation counted towards the residual case has to be deliverable. Three questions decide that.

1. Is the work within highway land or land the applicant controls, and if not, how is it secured?
2. Is it funded, and through what mechanism: a section 278 agreement, a section 106 contribution, or a condition?
3. Has the highway authority confirmed it is acceptable in principle, including any technical approval or safety audit stage?

Measures that fail those questions are aspirations. A junction improvement requiring third party land with no option agreement will not be counted by an inspector, and a scheme relying on it is exposed.

Sustainable transport measures count as mitigation, but only where they change behaviour in a way the evidence supports. A Travel Plan with monitoring, targets and a funded coordinator carries weight. A travel plan promising a 10 per cent mode shift on a site with no bus service does not, and claiming it invites the highway authority to discount the whole package.

Cumulative impact and committed development

Agreeing the committed development list early is worth more than any modelling refinement. The list should cover permissions not yet built, applications resolved to grant, and allocations where delivery is programmed within the assessment year.

Disputes usually turn on two things: whether a large permission will actually come forward in the assessment period, and whether its own mitigation has been assumed. Double counting mitigation, where two schemes both rely on the same junction improvement, is a specific point worth checking in an objection, because it is common and it is decisive.

Where the network is already congested, the base position matters. An inspector looking at a junction operating over capacity before the development will ask what the scheme adds rather than what the total is. A small percentage increase on an already failing arm is the classic shape of a case that fails the severe test despite genuine local frustration.

The evidence that persuades an inspector

Modelling outputs on their own rarely decide these appeals. What works is translating the output into experience.

Present peak hour delay in seconds per vehicle and as queue length in vehicles, with and without the development. Say when the peak occurs and how long it lasts. Identify the arms that carry the change rather than quoting a junction-wide average, because a two second worsening overall can hide a 40 second worsening on a minor arm.

Then deal with the users who are not in cars. Pedestrian crossing opportunities on the minor arm, severance, cycle movements through the junction and bus journey time reliability all carry weight, and they are frequently absent from both the assessment and the objection.

Recorded collision data belongs in the safety case rather than the capacity case, and it should be presented with the study area, the period and the severity classification stated, along with an honest note on what the data does not record. Appeal decisions are indicative of how inspectors have approached an issue rather than binding precedent, so cite them for reasoning rather than as authority.

Conditions and obligations instead of refusal

National policy expects decision makers to consider whether an acceptable outcome can be secured before refusing. In transport terms that usually means a phased trigger condition, a section 278 agreement for works on the highway, a section 106 contribution towards a wider scheme, or a travel plan condition with monitoring and review.

Highway authorities sometimes resist this because they would rather have the improvement built than monitored. The planning authority still has to ask whether the objection could be met by condition, and an applicant who offers a properly drafted trigger condition puts the burden back on the authority to explain why it is insufficient.

Where a Transport Statement rather than a full assessment was submitted, expect the highway authority to argue that the evidence base is inadequate for the mitigation proposed. Scale the document to the scheme at the outset, using the local thresholds and the guidance on transport assessments, and agree the scope in writing.

Running the argument at appeal

Get the statement of common ground to do real work. Trip rates, the committed development list, the survey dates and the modelling software and calibration should all be agreed, leaving the inquiry or the written representations to deal with the judgement rather than with the arithmetic.

Then make the severe case in ordinary language. An inspector reading a proof of evidence wants to know what a driver, a pedestrian and a bus passenger will experience at the worst moment of the worst day, and whether that is materially different from today. If the honest answer is that it is worse but tolerable, the highway objection will not carry the decision, whatever the local strength of feeling. That is uncomfortable to say to a parish council, and it is usually right.

Browse all 66 PlanScope report types and start a draft to see which transport documents a scheme of a given scale normally needs.

This is general guidance, not legal advice; check the position with your local planning authority or a qualified professional before relying on it. PlanScope provides AI-assisted draft reports for planning applications. Reports should be reviewed before submission. We do not guarantee planning approval.