Funding the Long Tail: How UK Permit Holders Can Defend Closure Cost Estimates

The number is not the provision

Environmental permit financial provision is sometimes managed as a compliance artefact: a closure estimate is produced, a mechanism is put in place, and the figure is revisited only when the permit or the site changes. That approach is becoming harder to defend. Regulators are entitled to ask not merely whether a provision exists, but whether it is credible, current and capable of withstanding the site conditions that will actually prevail at closure.

The risk is not confined to landfill. Waste treatment, energy-from-waste, mining and mineral processing, battery storage and some industrial installations can all carry long-tail obligations. Where a permit requires financial provision, the estimate must reflect the cost of discharging the obligations that remain. The 2026 scrutiny environment is less forgiving of percentage uplifts and stale contractor quotations than many permit holders assume.

For boards, the issue sits at the intersection of regulation and operational risk. An understated provision is not simply an accounting question. It can become an enforcement issue, a transaction obstacle, an insolvency exposure and a reputational signal. The following review agenda is intended to help permit holders test the assumptions before a regulator does.

Assumption one: contingency is doing too much work

A common weakness is a single contingency percentage applied to a base cost estimate. The percentage may be defensible in a stable project, but closure and aftercare are not stable projects. They are long-dated, uncertain and frequently interrupted by site conditions that were not visible at the outset.

A more robust approach separates contingency into identifiable components: design development, quantity uncertainty, regulatory change, market pricing and site-specific risk. Each should be linked to evidence. If the provision depends on a 15 per cent uplift to cover everything from groundwater treatment to waste characterisation, the board should ask what scenario would exhaust that allowance and what the next tranche of cost would be.

Regulators are likely to be less interested in the precision of the percentage than in the logic behind it. A provision supported by a risk register, sensitivity analysis and a clear statement of what is excluded is easier to defend than a single round number.

Assumption two: contractors will be available, willing and priced as quoted

Closure estimates often rely on assumed contractor rates and assumed availability. In specialised markets, neither is guaranteed. Landfill capping, leachate treatment, groundwater remediation, demolition of regulated plant and long-term monitoring all depend on a limited pool of competent suppliers.

A quotation obtained for a planning exercise may not survive into a regulatory review, particularly if it excludes mobilisation, welfare, permitting, waste transfer, laboratory analysis or contingency for encountering unexpected materials. The review should test whether the estimate reflects the cost of procuring the work in the market that will exist at closure, not the market that existed when the estimate was first prepared.

Where the site relies on a single contractor or a narrow supply chain, the provision should say so. A board-level review should also consider whether the estimate assumes a competitive tender that may not be realistic for a remote site or a technically complex closure.

Assumption three: monitoring ends earlier than the obligation does

Long-term monitoring is frequently the most understated element of financial provision because it is treated as a fixed period rather than a conditional obligation. Groundwater monitoring may continue until compliance thresholds are met and confirmed over a sustained period. Surface water, gas, leachate and stability monitoring may have different endpoints. Aftercare may extend beyond the period contemplated in the original estimate.

The review should identify each monitoring stream, its trigger for cessation and the evidence required to demonstrate that the trigger has been met. It should also test the cost of monitoring at the frequency the permit requires, including sampling, laboratory analysis, data management and regulatory reporting. If the estimate assumes a reduction in frequency after a set number of years, that assumption should be explicitly justified.

A useful discipline is to model a delayed endpoint. If monitoring continues for five additional years, what is the added cost? If a treatment plant must remain operational while groundwater conditions are reassessed, has that been included?

Assumption four: the site will behave as characterised

Closure estimates rest on a characterisation of the waste, soils, groundwater and infrastructure. That characterisation may be incomplete, particularly at older sites with limited records. The provision should reflect the possibility that additional contaminated material, unstable ground, unidentified voids or unexpected groundwater pathways will be encountered.

This is not an argument for unbounded pessimism. It is an argument for evidence. The review should trace each major cost line back to site investigation data, as-built records, monitoring trends or regulatory correspondence. Where the evidence is thin, the estimate should carry an explicit allowance and a stated route to reducing uncertainty.

Boards should also consider whether the estimate depends on assumptions about future waste inputs or site operations that are no longer valid. A provision built for a different operational profile may not survive scrutiny.

Assumption five: the evidence will speak for itself

Financial provision is ultimately a defended number. The regulator may ask how it was derived, what it covers, what it excludes and how it has been updated. A spreadsheet without a narrative is a weak answer.

The evidence pack should include the methodology, the cost basis, the source and date of quotations, the contingency rationale, the monitoring assumptions and the governance route by which the board approved the figure. It should record the date of the last review and the triggers for the next one. Where a parent company guarantee, bond, trust or charge is used, the mechanism should be checked against the obligation it secures.

A practical board checklist

Before the next regulatory conversation, permit holders should be able to answer five questions. First, what scenario would cause the current provision to fail? Second, which cost lines depend on contractor availability or pricing that has not been tested recently? Third, when does each monitoring obligation actually end, and what evidence is needed to end it? Fourth, which site conditions remain uncertain, and how is that uncertainty funded? Fifth, can the board produce a coherent evidence pack that explains the number from first principles?

A credible provision is not the lowest number that satisfies a template. It is the number that can be explained, stress-tested and defended when the site is closed, the contractor market has moved and the monitoring obligation outlasts the original plan. That is the standard UK permit holders should be preparing for in 2026.