Stranded assets are passé. The real risk is stranded liabilities

Everyone knows the stranded assets story by now. Coal plants that will not be run, reserves that
will not be extracted and thermal power plants that may not run for their full economic life.

What has received rather less attention is the other side of the balance sheet.

Liabilities, unlike assets, have a habit of insisting on being paid, which makes the current focus on stranded assets feel slightly one-sided. Across energy, utilities, mining and heavy industry, provisions for decommissioning, environmental remediation and litigation are all there on the balance sheet, representing obligations that were assumed to be long-dated but are now harder to defer and arriving sooner as the transition accelerates. They are disclosed but not especially interrogated.

The accounting treatment helps keep them that way. Small changes in assumptions tend to stay small on paper, which is helpful if you would prefer them not to show up elsewhere on the balance sheet.
That framing holds as long as timelines behave. The difficulty is that climate transition risk is not especially respectful of timelines.

Technologies are moving faster than expected, regulations are tightening, courts are more willing to engage, and tolerance for “eventually” is wearing thin. In turn, closure deadlines are pulled forward, decommissioning requirements become less flexible, environmental remediation is enforced earlier, and legal liabilities that were assumed to sit in the distance become harder to defer.

These are the same liabilities, just turning up earlier than expected and proving less deferrable.

Once that happens, the present value stops looking like a distant estimate. Pull a liability forward in time and it gets bigger. What sat quietly in the notes starts to look less like a provision and more like something that needs funding.

Most of these liabilities are not meaningfully pre-funded. They are backed, implicitly, by future cashflows that assume continued operations through the transition. The complication is that those cashflows are expected to do several jobs at once, and not all of those jobs are optional. If asset earnings come under pressure at the same time as liability timelines move forward, the interaction is not especially forgiving. The same forces driving the transition tend to do both at once. Carbon pricing, tighter regulation and shifting demand begin to weigh on legacy assets, while compliance costs and transition capex rise.

The result is fairly familiar. Transition capex rises, legacy assets generate less, and margins compress, just as obligations start arriving earlier. What looked like a distant liability starts competing with the business itself for cash and is generally less flexible about timing. Cash that was meant to fund transition, service debt and support operations now has an additional claimant.

That is when balance sheets start to show strain. Not because anything has failed outright, but because the assumptions that kept liabilities comfortably in the background stop holding at the same time. Provisions rarely move markets, until they start behaving like cash calls. When they do, they tend to show up in the same places markets usually notice things: earnings, ratings and
funding.

Describing these obligations as long-term does not quite solve the problem. In practice, “long-dated” often just means sensitive to assumptions that have not yet been tested. Adjust the discount rate and the number moves.

If this feels familiar, it should. Risks that are comfortably parked in the future tend to remain so until they do not, at which point they arrive with a degree of urgency that was not in the base case. The 2008 financial crisis offered a few reminders on that front.

There is also a slightly awkward interaction with the transition narrative itself. Companies can be making credible progress on decarbonisation while simultaneously carrying, or even accelerating, liabilities linked to their legacy footprint. From a distance, that looks like improvement. Up close, it can resemble a duration mismatch, where long-term obligations meet cashflows that depend on a transition that is still uncertain.

Part of the reason this sits in the background is analytical convenience. Equity narratives tend to focus on assets and growth. Liabilities sit in the notes, depend on assumptions and require scenario analysis that does not compress neatly into a multiple. Credit markets, for their part, often take provisions as given until there is a clear reason not to.

Which leaves a familiar gap. The information is disclosed, but is not obviously doing much work in the price.

None of this displaces stranded assets as a theme. It does suggest that focusing exclusively on them may miss where balance sheet stress shows up first. If transition timelines compress, even modestly, the effect on liabilities is not linear. A few years here or there can do more work than most models currently assume.

That raises a set of questions. What happens to credit metrics if liability timelines move forward by five or ten years. How sensitive are existing provisions to plausible changes in discount rates. Which sectors are effectively carrying unfunded transition obligations. And when those obligations become less deferrable, who ultimately bears them.

Markets have spent a long time thinking about assets that might not be worth what they used to be. It may be worth spending a bit more time on liabilities that might turn out to be due sooner than expected.

illuminem Voices is a democratic space presenting the thoughts and opinions of leading Sustainability & Energy writers, their opinions do not necessarily represent those of illuminem.

Track the real‑world impact behind the sustainability headlines. illuminem's Data Hub™ offers transparent performance data and climate targets of companies driving the transition.