Unreasonable Conduct, Expert Evidence and Costs in Called‑In Planning Inquiries: Commentary on Halton BC v Secretary of State [2025] EWCA Civ 1566

1. Introduction

This Court of Appeal decision concerns when a local planning authority (“LPA”) can properly be ordered to pay costs in a called‑in planning inquiry, particularly where its case collapses because its own expert witness changes position under cross‑examination.

The appeal was brought by Halton Borough Council (“the Council”) against a decision of a Planning Inspector acting for the Secretary of State for Housing, Communities and Local Government (Mr Parsons), who had made partial awards of costs in favour of:

• the Health and Safety Executive (“HSE”), and
• Viridor, the operator of a large energy‑from‑waste plant and a Rule 6 party,

following the abortive called‑in inquiry into a residential development at Runcorn promoted by MJ Gleeson (“the developer”).

The central events were:

• The Council had resolved to approve planning permission for 139 dwellings near the Runcorn Chemicals Complex despite HSE’s long‑standing safety objections.
• The Secretary of State called in the application. A partially closed inquiry (for national security reasons) was held.
• In the closed session, the Council’s public safety expert, Mr Hopwood (of DNV), accepted in cross‑examination that, applying the relevant national policy principles (NPPG), the outcome was to advise strongly against granting permission, and that he himself would advise refusal if he were the inspector.
• In light of that evidence, the Council withdrew its support for the proposal; the developer then withdrew the application; the inquiry collapsed.
• The inspector held that the Council’s withdrawal was “unreasonable” procedural conduct and awarded partial costs to HSE and Viridor.

Fordham J in the Administrative Court upheld that costs decision. The Council appealed. The Court of Appeal (Lewison, Asplin and Coulson LJJ) allowed the appeal and quashed the costs awards.

This judgment sets an important precedent on:

• the meaning of “unreasonable behaviour” and “normal procedural requirements” in costs decisions for called‑in inquiries;
• the limits of an LPA’s duty to “stress‑test” its own expert evidence before inquiry;
• what counts as a “good reason” for late withdrawal; and
• the standard of rationality and adequacy of reasons required in such costs decisions.

2. Summary of the Judgment

The Court of Appeal allowed the Council’s appeal and set aside the Secretary of State’s partial costs awards in favour of HSE and Viridor.

Key holdings include:

1. Guidance‑based limitation in called‑in cases. Under the Planning Practice Guidance (“PPG”) para 034 (para 34 in the judgment), it is not normally envisaged that parties to a called‑in inquiry are at risk of costs for unreasonable behaviour relating to:
   • the substance of their case, or
   • actions taken before the call‑in,
   but only for unreasonable procedural behaviour in complying with “normal procedural requirements” (e.g. aborting the process without good reason).
2. Contradictory reasoning about “material change”. Mr Parsons’ decision contained an irreconcilable contradiction:
   • He accepted the Council withdrew support “in the light of evidence heard” in the closed session (para 17).
   • Yet he also said there had been no material change in the planning circumstances or evidence to justify the Council’s “volte face” (para 20).
   That was a “leap in reasoning” and a “demonstrable flaw” in the sense explained in Finch and KP.
3. Change in expert evidence is a paradigmatic “good reason” for withdrawal. The Council’s principal public safety expert conceding that he would himself advise refusal is, in principle, one of the strongest possible justifications for withdrawing support. Treating withdrawal in those circumstances as “without good reason” was untenable.
4. No “normal procedural requirement” to pre‑test expert evidence against cross‑examination. Mr Parsons’ key criticism—that the Council ought to have ensured, in advance, that its expert’s evidence could withstand cross‑examination and that failure to do so was unreasonable—“set the bar too high” (paras 58–61). There is no recognised “normal procedural requirement” to stress‑test your own expert’s evidence to see whether it will stand up to cross‑examination.
5. Need for a “trigger” before reliance on an expert becomes unreasonable. To find a party’s reliance on its expert evidence “unreasonable”, there must be some identifiable “trigger” that should have alerted the party that the expert’s view was plainly wrong or untenable (para 54). Simply losing the expert under cross‑examination is not enough.
6. Inadequate specification of “what went wrong” and when. The Council could not tell from the decision “what it did wrong or when its case changed from being ‘respectable’ to ‘unreasonable’” (para 62). That failure breached the standards of intelligible, adequate reasons required by South Bucks v Porter.

On those bases, the Court of Appeal held that the inspector’s decision was vitiated by irrationality and/or material misdirection in applying the PPG, and the costs orders had to be quashed.

3. Key Legal Issues

The case raised four central issues:

1. Scope of costs risk in called‑in inquiries. How far does PPG para 34 protect parties—in particular LPAs—from costs awards based on the substance of their case and/or pre‑call‑in conduct?
2. Meaning of “unreasonable” procedural behaviour and “normal procedural requirements”. Can it be “unreasonable” for an LPA to withdraw support mid‑inquiry following adverse expert concessions, and is there a duty to pre‑test its own expert’s evidence?
3. Standard of review in judicial review of costs decisions. Did the inspector’s reasoning contain a “demonstrable flaw” or “critical gap” amounting to public law irrationality?
4. Interaction between expert witness duties and party responsibility. When can a party be criticised, and exposed to costs, because its expert’s evidence collapses?

4. Precedents and Authorities Cited

4.1 Statutory Framework and Guidance

The Secretary of State’s power to award costs in called‑in applications derives from:

• Local Government Act 1972, s 250(4), applied to planning by
• Town and Country Planning Act 1990, s 322.

In practice this broad statutory power is constrained by published guidance, here the costs section of the Planning Practice Guidance (“PPG”), including:

• Para 28: sets out the aims of the costs regime—to encourage reasonable behaviour and avoid wasted expense.
• Para 31: “unreasonable” bears its ordinary meaning; unreasonable behaviour can be procedural or substantive.
• Para 34 (quoted at para 17 of the judgment):

  “… in a called‑in planning application … it is not envisaged that a party would be at risk of an award of costs for unreasonable behaviour relating to the substance of the case or action taken prior to the call‑in decision. However, a party’s failure to comply with the normal procedural requirements of inquiries, including aborting the process by withdrawing the application without good reason, risks an award of costs for unreasonable behaviour.”

• Para 49: gives examples of “substantive” unreasonableness in planning appeals (failure to substantiate reasons for refusal, failure to review case promptly). Its main relevance is as contrast with para 34: in called‑in inquiries, substantive unreasonableness is generally off‑limits as a basis for costs.
• Para 56: governs costs awards involving Rule 6 parties. In particular, it states:
  • awards to or against Rule 6 parties are generally confined to procedural grounds;
  • no award in their favour where the unreasonable behaviour relates to the merits of the appeal; and
  • they may be awarded costs, for example, where an appeal has been withdrawn without good reason or where an unnecessary adjournment is caused by unreasonable conduct.

The inquiry procedure itself is governed by the 2000 Inquiry Rules, with modified provisions where a national security direction is given under TCPA 1990 s 321, as happened here to facilitate a closed session on public safety matters.

4.2 North Norfolk DC and Wakefield MBC: The “Sufficient Evidential Basis” Threshold

Two leading pre‑PPG cases on costs in planning appeals (not called‑in inquiries) supply an important background test.

In R v Secretary of State for the Environment ex p North Norfolk DC [1994] 2 PLR 78, Auld J addressed guidance that authorities should “produce evidence to substantiate their reasons for refusal”. He emphasised that “substantial evidence” means:

“… the evidence upon which an authority relies to support a ground for refusal of permission must have some substance in the sense of providing some respectable basis for its stance upon a particular issue. But it need not be of such substance as to persuade the inspector to find in the authority’s favour on the issue. Otherwise every evidential failure to persuade an inspector on an issue would expose the loser to a finding of unreasonableness on an application for costs in relation to that issue.”

In R v Secretary of State for the Environment ex p Wakefield MBC (1998) 75 P & CR 78, Jowitt J endorsed this and coined the expression “sufficient evidential basis”:

“… evidence, not lacking real substance, which is capable of belief and which, if accepted, would be capable of making good the plaintiff authority’s objection.”

He added an important qualification: an authority which persists with an objection knowing the appellant can advance cogent, overwhelming evidence to refute it may still be acting unreasonably, even if its own evidence crosses the evidential threshold. That highlights two points picked up by Lewison LJ:

• there is an evidential threshold below which it is generally unreasonable to run a case at all;
• but above that threshold, unreasonableness may arise from how long a party persists in an objection in the face of overwhelming contrary material (i.e. a temporal dimension).

These concepts feed directly into Lewison LJ’s approach to whether and when reliance on expert evidence can become “unreasonable”.

4.3 Finch: The Rationality Standard

In R (Finch) v Surrey CC [2024] UKSC 20, [2024] PTSR 988 the Supreme Court articulated the modern approach to public law irrationality. Lord Leggatt explained (quoted at para 27) that courts will not interfere unless:

“… it is ‘irrational’ in the sense either that it is outside the range of reasonable decisions open to the decision‑maker or that there is a demonstrable flaw in the reasoning which led to the decision. Examples of such a flaw would be that significant reliance was placed on an irrelevant consideration, or that there was no evidence to support an important step in the reasoning, or that the reasoning involved a serious logical or methodological error…”

This passage is central. Lewison LJ explicitly uses it to characterise the inspector’s errors as “demonstrable flaws” in the reasoning: logical leaps and contradictions concerning:

• whether there had been a material change in evidence, and
• what constituted a “normal procedural requirement”.

4.4 KP: Process vs Outcome Rationality

In R (KP) v Secretary of State for Foreign, Commonwealth and Development Affairs [2025] EWHC 370 (Admin) (quoted at para 28), Chamberlain J drew a useful distinction:

• Process rationality: the logic of the reasoning—no logical error, no critical gap, no reliance on irrelevant factors, and consideration of all mandatory ones.
• Outcome rationality: whether the result lies within the range of reasonable outcomes.

He described process irrationality by asking whether the decision “does not add up” and whether there is an “unexplained evidential gap or leap in reasoning” (echoing Sedley J in Balchin and Saini J in Wells).

Lewison LJ explicitly invokes this kind of “process rationality” critique: he finds that the inspector’s decision contains:

• a gap between para 17 (acknowledging new adverse evidence from the expert) and para 20 (denying any material change in circumstances), and
• a failure to articulate what concrete “normal procedural requirement” the Council failed to meet.

4.5 South Bucks v Porter: Adequacy of Reasons

The classic statement on reasons in planning decisions is South Bucks DC v Porter (No 2) [2004] UKHL 33, [2004] 1 WLR 1953. Lord Brown held that reasons:

• must be intelligible and adequate;
• must enable the reader to understand why the decision was made and how the main issues were resolved;
• need not be lengthy but must not give rise to substantial doubt as to whether the decision‑maker has erred in law (e.g. misunderstanding policy, overlooking key matters, or failing to reach a rational decision on relevant grounds).

Lewison LJ applies these principles to the costs decision: by failing to explain what exactly the Council did wrong and when its conduct crossed from “respectable” to “unreasonable”, the inspector’s reasons fell below the Porter standard.

4.6 Criminal Costs Cases on Experts: Cornish and Aylesbury

Although involving different statutory tests (requiring “unnecessary or improper” acts), two criminal cases provided persuasive analogies for dealing with expert evidence:

• R v Cornish [2016] EWHC 779 (QB):
  • The prosecution’s manslaughter case collapsed after their expert’s evidence was effectively dismantled in cross‑examination.
  • On a costs application by the NHS Trust, Coulson J refused to find improper conduct by the Crown. He held that although, with hindsight, the expert was too dogmatic, it could not be said that what he was saying was plainly wrong or that this should have been apparent to the prosecution beforehand (para 32, 35).
  • He concluded that what happened was “a good example of the adversarial trial process in action” (para 44), not misconduct by the party who had relied on the expert.
• R (DPP) v Aylesbury Crown Court [2017] EWHC 2987 (Admin); [2018] 4 WLR 30:
  • An expert instructed by the CPS initially said indecent images were accessible to a computer user. The defence expert disagreed; on reconsideration the prosecution expert accepted he had been wrong. The CPS withdrew the case.
  • The Crown Court nonetheless made a costs order against the CPS. On judicial review, the Divisional Court (Sharp LJ) quashed that order.
  • Sharp LJ stressed that experts are independent, not part of the CPS, and owe their duty to the court, not the instructing party (para 20).
  • She stated that the appropriate test is whether the expert’s views were plainly wrong in a way that should have been obvious to the prosecuting authority (para 23). On the facts, that test was not met.

Lewison LJ acknowledges that the criminal costs threshold (“improper” conduct) is higher than mere “unreasonableness”. Nonetheless, he treats these cases as highly relevant in describing what is, and is not, fairly expected of a party relying on specialist expert evidence (paras 52, 54). In both, the key question is whether the party had reason beforehand to suspect the expert was plainly wrong.

4.7 Professional and Procedural Guidance on Experts

The judgment also references:

• Planning Inspectorate Procedural Guide, para 15.2:

  “It is the duty of an expert to help the Inspector. This duty overrides any duty the expert may have to the party that involved them in the appeal or that is paying them.”

• Royal Town Planning Institute (RTPI) Guidance for Expert Witnesses, which:
  • encourages experts to work closely with advocates to prepare and understand the weakest aspects of the case;
  • emphasises preparation for likely lines of cross‑examination; but
  • equally stresses that advocates must not tell experts what answers to give.
• BSB Handbook, rule C9 and PEBA Good Practice Guidance, affirming that barristers must not rehearse, practise with, or coach a witness “in respect of their evidence”.
• Ultraframe (UK) Ltd v Fielding [2005] EWHC 1638 (Ch), which delineates the boundary between legitimate witness familiarisation and impermissible coaching.

Lewison LJ uses these materials to show:

• it is entirely legitimate, and indeed expected, that experts and legal teams will discuss the case and test arguments in conference; but
• there is no recognised procedural obligation to “mock cross‑examine” an expert or to ensure, in some absolute sense, that their evidence will withstand cross‑examination—particularly given the constraints against coaching.

5. The Court’s Legal Reasoning

5.1 The Applicable Test: PPG para 34 and “Normal Procedural Requirements”

The inspector stated (para 16 of his decision, reproduced at para 23 of the judgment) that under PPG para 34 a party’s failure to comply with “normal procedural requirements” of inquiries, including aborting the process without good reason, may justify a costs award in a called‑in case.

Lewison LJ draws two core inferences (para 33):

• Mr Parsons did not intend to depart from the Secretary of State’s PPG—so his discretion was self‑consciously guided and constrained by it.
• He therefore had to identify a failure to comply with a “normal procedural requirement” as the gravamen of the Council’s unreasonable behaviour.

The PPG expressly excluded costs liability for:

• behaviour “relating to the substance of the case”; or
• action taken “prior to the call‑in decision”.

Accordingly, the inspector could not (and did not) rely on:

• the Council’s original decision to resolve to grant permission in the face of HSE objections; or
• the pre‑inquiry merits of its public safety stance,

as the basis for costs. Instead, he focused on the Council’s decision to withdraw support during the inquiry and its alleged failure to ensure its expert’s evidence could withstand scrutiny.

5.2 Contradiction Over “Material Change” in Evidence

The key factual trigger for the Council’s volte‑face was Mr Hopwood’s closed‑session evidence (para 10), specifically his acceptance that:

• the policy relied on by the Council did not follow the principles in the NPPG; and
• proper application of those principles led to the conclusion that permission should be strongly refused.

As Lewison LJ notes:

• HSE immediately recognised this was decisive: the GLD wrote the same day that, in light of Hopwood’s evidence, there was no basis on which the Council’s position could be maintained (para 11).
• The Council’s stated reason for withdrawing support was this change in expert evidence during cross‑examination (para 12, 17).

Mr Parsons expressly accepted (para 17) that the Council’s decision to withdraw was “in the light of evidence heard in the closed public safety session”. That is a concession that there was new, adverse evidence. Yet in para 20 he stated there had been “no material change in the planning circumstances or evidence sufficient to justify such a volte face.”

Lewison LJ finds these statements “impossible to reconcile” (para 35):

“There was a change in the evidence and its materiality was such that it caused the Council to withdraw its support; as the letter from the GLD had correctly foreshadowed.… The leap from paragraph 17 to paragraph 20 is, to my mind, a leap in reasoning which fails to justify the conclusion. It is a demonstrable flaw in the reasoning.”

He further points out that:

• In relation to Viridor’s costs against the developer, Mr Parsons treated the Council’s change of stance as a “material change in the planning circumstances” which did provide the developer with a “good reason” for withdrawing the application (para 36).
• But the Council’s change of stance was itself caused by the change in Mr Hopwood’s evidence. If that was a material change for the developer, why not also for the Council?

This internal inconsistency further undermined the rationality of the costs decision.

5.3 Withdrawal Following Expert Concessions: Is There “Good Reason”?

Both the Council and the Court of Appeal characterised the expert’s concession as the paradigmatic “good reason” for withdrawal:

• The Council argued it withdrew because its chief public safety expert no longer believed the case was tenable (para 17).
• HSE’s own costs application confirmed that Hopwood’s evidence was that he would have advised “strongly against” granting permission (para 38).

Lewison LJ states in robust terms (para 39):

“It is, with respect, difficult to conceive of a better reason for withdrawing support for the proposed development than the principal expert witness being no longer able to support the case advanced.”

To the extent Mr Parsons treated such withdrawal as being “without good reason”, Lewison LJ considers that conclusion “untenable” (para 67).

That does not, however, conclusively answer the inspector’s eventual point (as interpreted by Fordham J): that the timing of the withdrawal might itself be unreasonable if the Council should have recognised the hopelessness of its case earlier.

5.4 The Temporal Dimension: When Must a Party Withdraw?

Fordham J framed the issue (para 40) by distinguishing:

• withdrawal for good reason, which is inherently reasonable at the moment it occurs and saves future costs; and
• the question whether withdrawal occurred “at the earliest opportunity” once the party knew, or should have known, it could not maintain its case.

On that view, “withdrawal without good reason” in PPG para 34/54 extends to late withdrawal—i.e. withdrawal whose belated timing lacks good reason.

Lewison LJ assumes, in the Council’s favour, that this is indeed what Mr Parsons meant: his real criticism was not withdrawal as such, but withdrawal only after the expert’s cross‑examination. The acute question then becomes (para 41):

“… whether the Council were unreasonable in relying on Mr Hopwood’s evidence until the moment of his cross‑examination?”

That in turn required:

• considering what information and analysis the Council had at different stages (para 44); and
• avoiding hindsight bias by judging reasonableness by what reasonably appeared at the time, not by knowledge gained afterwards (para 44, borrowing the reasoning of Cornish and Aylesbury).

5.5 No Evident Flaw in the Expert’s Case Before Cross‑Examination

Lewison LJ conducts a careful step‑by‑step review (paras 56–61):

• The Council knew HSE’s objections long before the call‑in. HSE itself said its core safety concerns had been articulated over “decades” (para 57).
• The Council’s resolution to grant permission despite those objections was not found to be unreasonable by Mr Parsons (para 42).
• The HSE’s Rule 6 statement (June 2021) did not introduce materially new objections; it tracked long‑standing concerns (para 57).
• The Council’s expert proof (Hopwood) was served in August 2021. Mr Parsons did not suggest it:
  • ignored HSE’s concerns, or
  • was obviously wrong, or
  • fell below the “respectable basis”/“real substance” threshold articulated in North Norfolk and Wakefield (para 57).

Against that background, Lewison LJ rejects the suggestion that the Council had some earlier “trigger” to doubt its expert:

• There is no positive finding that the Council had information which should have alerted it that Hopwood’s views were untenable.
• It is entirely normal for experts to disagree; the mere existence of contrary expert evidence is not enough to render reliance on your own expert unreasonable (para 54).

Consequently, the Council’s reliance on Hopwood up to the point of cross‑examination was not shown to be unreasonable.

5.6 “Normal Procedural Requirements” and Stress‑Testing Expert Evidence

The only “procedural failure” positively identified by Mr Parsons was that the Council:

• should have been “satisfied with the strength” of Hopwood’s advice; and
• “crucially that they could rely on it being capable of standing up to scrutiny by any other parties through cross‑examination” (para 18 of the decision, cited at para 56 of the judgment; emphasis added).

Lewison LJ unpacks this as implying that:

• there is a “normal procedural requirement” to probe your own expert’s reasoning in advance to ensure it will withstand cross‑examination (para 58); and
• failure to do so is, of itself, unreasonable behaviour in the PPG sense.

He rejects that proposition on several grounds:

• It “sets the bar too high”. In any contested expert case, one expert will ultimately be preferred over the other. That does not mean the losing expert’s evidence was unreasonable to rely on (para 59).
• No authority or practice supports such a requirement. There is no case law or procedural rule (whether in planning or civil litigation) which demands that a party must stress‑test its expert’s evidence to see whether it will withstand cross‑examination (para 61).
• It conflicts with the nature of expert engagement. As Sharp LJ explained in Aylesbury, experts are instructed precisely because they possess specialist knowledge which parties lack (para 61). It is unrealistic, and unfair, to expect lay clients or planning authorities routinely to second‑guess complex technical judgments, save where there is some obvious flaw or external trigger.
• Professional guidance supports thorough preparation, not mock trials. The Council in fact:
  • held multiple conferences involving counsel, Hopwood and the planning officer, before and after exchange of evidence, at which both sides’ positions were “discussed extensively” (para 8, 65);
  • followed RTPI‑type good practice by reviewing proofs and testing weaker aspects; but
  • still could not have been expected to foresee the precise concessions made under adversarial questioning, especially on points (validity of policy) falling somewhat outside Hopwood’s core public safety expertise (para 66).

In short, treating the absence of an exhaustive pre‑trial stress test as a “failure to comply with a normal procedural requirement” is irrational and contrary to established understanding of how experts are used.

5.7 The Need for an Identifiable “Trigger” Before Reliance Becomes Unreasonable

Drawing on Cornish and Aylesbury, Lewison LJ holds that to condemn reliance on expert evidence as unreasonable there must be:

• some “trigger” or event which should have prompted doubts about the expert’s reliability—such as:
  • obvious internal inconsistency or error in the expert’s own analysis, or
  • overwhelming contrary evidence clearly demonstrating a fatal flaw (para 54);
• and that this must be demonstrable before the point at which the party ultimately changes course.

Without such a trigger, later collapse under cross‑examination is simply the adversarial process doing its work, not evidence of prior unreasonableness.

On the inspector’s own reasoning, no such trigger was identified. That left a “critical gap” in the logic of the costs decision and rendered it unlawful.

5.8 Inadequate Reasons and the Council’s Inability to Know What It Did Wrong

At para 62, Lewison LJ crystallises the reasons defect:

“In short, the Council cannot tell from Mr Parsons’ decisions what it did wrong or when its case changed from being ‘respectable’ to ‘unreasonable’.”

He notes that:

• the Secretary of State’s own counsel was unable to identify in argument when and why the change occurred;
• if the finding of unreasonable conduct is really based on the Council’s continuing reliance on Hopwood, then it would have been unreasonable irrespective of whether it later withdrew support or fought on to the end—the withdrawal is only tenuously connected to the alleged misconduct (para 62).

This lack of clarity breached the Porter requirement that reasons must explain what the party did wrong in terms the party can understand, especially where the consequence is a financial penalty through costs.

6. Simplifying Key Concepts

6.1 Called‑In Inquiries vs Planning Appeals

A planning appeal is brought by an applicant against an LPA’s refusal (or non‑determination) of a planning application. The LPA defends its own decision.

A called‑in application is a different creature: the Secretary of State takes the decision out of the LPA’s hands and determines the application himself, often because it raises issues of more than local importance or, as here, sensitive safety concerns. The LPA is a party, but not as “defendant” to its own decision. That structural difference underpins para 34 of the PPG: the LPA is not to be penalised for its substantive position on the application when it is called in.

6.2 Rule 6 Parties and Costs

Under the inquiry rules, “Rule 6 parties” are interested parties (such as objectors or statutory consultees) who choose to play a full role in an inquiry—submitting statements of case, evidence, and participating in cross‑examination.

PPG para 56 treats costs involving Rule 6 parties cautiously:

• they can be awarded costs, or have costs awarded against them, but only in limited circumstances;
• they are generally not to benefit from costs orders where the unreasonable behaviour relates to the merits of the appeal; procedural misconduct (e.g. causing an avoidable adjournment) is the focus.

That is why, in this case, HSE and Viridor could only obtain costs if the Council’s conduct was unreasonably procedural (late withdrawal without good reason), not merely because the Council’s substantive position on safety was eventually untenable.

6.3 “Substantive” vs “Procedural” Unreasonableness

• Substantive unreasonableness concerns the merits of a party’s planning case—e.g. advancing reasons for refusal with no evidential support, or persisting with hopeless arguments.
• Procedural unreasonableness relates to how a party conducts itself in the process—e.g. missing deadlines, failing to comply with directions, withdrawing at the last minute without good reason, or causing avoidable adjournments.

PPG para 34 essentially says that in called‑in cases, costs should be based only on the latter, not the former.

6.4 “Normal Procedural Requirements”

These are the ordinary steps all parties are expected to take in an inquiry:

• serving Rule 6 statements on time;
• serving proofs of evidence under Rule 13;
• complying with directions from the inspector;
• attending the inquiry, making witnesses available, etc.

Lewison LJ holds that there is no recognised “normal procedural requirement” that a party must:

• subject its own expert to mock cross‑examination; or
• ensure in advance that the expert’s position could withstand any conceivable line of cross‑examination.

Such stress‑testing may be good practice in some cases, but its absence is not per se “unreasonable behaviour” in the PPG sense.

6.5 Process Rationality

“Process rationality” focuses not on whether the decision was “right” on the merits, but on whether the reasoning:

• is internally consistent;
• does not omit a critical step;
• is supported by evidence at each stage; and
• properly engages with relevant guidance and facts.

The Court of Appeal found process irrationality in:

• the contradiction over whether there was a material change in evidence; and
• the failure to identify any concrete procedural requirement that the Council had breached.

6.6 The Role and Obligations of Expert Witnesses

Expert witnesses in planning inquiries:

• owe a primary duty to assist the inspector impartially, not to advocate their client’s case;
• must be willing to adjust or abandon their opinions when persuaded they are wrong; and
• are instructed precisely because they possess expertise the client and lawyers do not.

When an expert concedes under cross‑examination that their prior opinion is wrong, that is them fulfilling their duty to the tribunal. It is not inherently evidence that the party who called the expert acted unreasonably in relying on them beforehand.

6.7 “Good Reason” for Withdrawal

In costs guidance, withdrawal is “without good reason” (and so potentially unreasonable) where:

• the party has no new information or change in circumstances; and
• yet abandons the process late, causing wasted costs for others.

By contrast, a “good reason” includes:

• a material change in planning circumstances;
• a fundamental change in another party’s case; or
• a decisive change in the party’s own evidence (e.g. its expert no longer supports its position).

Lewison LJ firmly places the Council’s situation in the latter category: the expert’s concession was a textbook good reason to withdraw.

7. Impact and Implications

7.1 For Local Planning Authorities

This judgment significantly clarifies and limits LPAs’ costs exposure in called‑in inquiries:

• Protection for responsible withdrawal. Where an LPA’s expert, acting properly, changes position under cross‑examination, it is not unreasonable for the LPA to:
  • accept that advice; and
  • promptly withdraw support for a proposal it can no longer credibly defend.
  That is a good reason for withdrawal and should not ordinarily attract adverse costs.
• No duty to pre‑emptively second‑guess experts. LPAs are not required to “stress‑test” their own experts’ evidence to the point of anticipating every concession made under forensic cross‑examination, particularly where they have:
  • selected appropriately qualified experts; and
  • held reasonable conferences to discuss cases and evidence.
• Ongoing duty to review. LPAs are still expected to keep their cases under review as the inquiry progresses. If an obvious flaw or decisive contrary evidence emerges, they must reassess. Persisting in the face of such a trigger could be unreasonable in a future case.
• Clarity on “substantive” immunity. In called‑in contexts, LPAs can continue to take robust substantive positions on planning merits without fearing that a subsequent costs award will retrospectively deem their original resolution “unreasonable”, so long as their procedural conduct is sound.

7.2 For the Secretary of State and Inspectors

The judgment underscores several points for decision‑makers exercising the costs jurisdiction:

• Faithful application of guidance. Where PPG para 34 narrows the basis for costs in called‑in cases, that limitation is not merely advisory; it is a self‑imposed constraint that must be honoured or explicitly and lawfully departed from with reasons.
• Need to specify the precise mischief. Inspectors must:
  • identify with particularity what conduct was unreasonable; and
  • explain when, in the chronology, the party should have acted differently.
  Vague references to failing to be satisfied about an expert’s evidence will not suffice.
• Avoidance of hindsight bias. Inspectors must assess reasonableness based on what was reasonably apparent to the party at each stage, not by reference to knowledge only gained after cross‑examination.
• Consistency across parties. Treating a change in circumstances as “material” for one party (e.g. a developer) but not for another (e.g. an LPA) without explanation risks irrationality.

7.3 For Rule 6 and Other Interested Parties

For HSE, Viridor and other potential Rule 6 parties, the decision clarifies:

• High bar for procedural costs against LPAs. Obtaining a costs award against an LPA in a called‑in inquiry will generally require:
  • a clear procedural default (e.g. repeated non‑compliance with directions, late withdrawal with no new factor); and
  • a clear causal link between that default and wasted expense.
• Success on the merits is not enough. Even if their position is vindicated and the opposing case collapses, that does not, by itself, demonstrate “unreasonable” conduct sufficient to justify costs.

This may temper expectations among objectors and statutory bodies that costs will routinely follow a late change in an LPA’s position.

7.4 For Administrative Law More Broadly

The case is also a useful illustration of public law principles applied to a specialised tribunal context:

• Process rationality as a ground of review. Following Finch and KP, the Court of Appeal emphasises that:
  • a decision can be irrational not only because the outcome is outside the reasonable range, but because the chain of reasoning contains internal contradictions or missing links;
  • identifying such “demonstrable flaws” is sufficient to warrant intervention, particularly where guidance is misapplied.
• Guidance as a self‑binding constraint. Although policy guidance such as the PPG is not law, a decision‑maker who professes to follow it must either:
  • apply it correctly; or
  • consciously and lawfully depart from it with coherent reasons.
  Deciding a case “other than in accordance with the Guidance” is a material misdirection (para 31(i)).

7.5 For Expert Practice and Legal Teams

The judgment sends nuanced messages to those who instruct and act as experts:

• Experts should not fear proper concessions. Hopwood’s change of opinion under cross‑examination was treated as the expert fulfilling his duty to assist the tribunal, not as misconduct. Courts recognise this as a normal, if sometimes dramatic, feature of adversarial proceedings.
• Extensive preparation is encouraged but not over-specified. The Council’s multiple conferences with counsel and experts are treated as “going a long way” towards good practice (para 65). The law does not demand that those conferences cross the line into prohibited coaching.
• Boundary between preparation and coaching remains critical. By referencing RTPI guidance, the BSB Handbook and Ultraframe, the Court reminds practitioners that:
  • counsel may work with experts to refine and understand their opinions; but
  • may not rehearse or “practise” the answers to be given in evidence.

7.6 Possible Influence on Future Guidance and Decisions

Although the Court does not require formal changes to the PPG, its reasoning may influence:

• how PINS trains inspectors to approach costs applications in called‑in cases;
• how future decision letters frame “unreasonable behaviour”—with more careful attention to distinguishing:
  • legitimate expert disagreement and late concessions;
  • from genuinely avoidable procedural waste.

8. Conclusion

The Court of Appeal’s decision in Halton BC v Secretary of State establishes an important and protective principle for planning participants, especially LPAs, in called‑in inquiries:

• Costs awards in such cases must be grounded in genuinely procedural unreasonableness—identified breaches of “normal procedural requirements” such as unjustified, late withdrawal—not in retrospective criticism of the substantive merits of an LPA’s position.
• There is no general procedural obligation on a party to stress‑test an expert’s evidence to guarantee it will withstand cross‑examination; reliance on a properly instructed expert is not rendered “unreasonable” merely because the expert later concedes under questioning.
• A decisive change in an expert’s opinion during an inquiry is, in principle, a “good reason” for an LPA to withdraw support for a proposal; treating such withdrawal as inherently unreasonable is irrational.
• Costs decision‑makers must articulate clearly:
  • what the party did wrong;
  • when it became unreasonable; and
  • how this departs from established guidance.
  Failure to do so will amount to process irrationality and inadequacy of reasons under Porter.

By quashing the costs awards in favour of HSE and Viridor, the Court reinforces the integrity of both the planning process and the expert evidence regime. LPAs are encouraged to keep cases under honest review and to act on expert advice—even where that entails a late change of position—without the automatic spectre of punitive costs, provided they have not ignored clear warning signs or procedural duties along the way.