Two numbers, one decision: making climate-health valuation usable across sectors

Conservatory and Botanical Garden of the city of Geneva is a museum and an institution in Geneva city in Switzerland

An OHE insight, ahead of OHE’s participation in the Health Diplomacy House panel at the 79th World Health Assembly (Geneva, 21 May 2026).

The 2025 Lancet Countdown puts the annual cost of heat-related mortality in people over 65 at $261 billion. Numbers of that magnitude command attention from finance ministries in ways that excess-death counts alone do not.

Yet when the same health case reaches formal budget appraisal, it rarely survives the journey. The obstacle is not missing evidence. It is that the two most widely used approaches to monetising climate-related health impacts, the Social Cost of Carbon (SCC) and the Value of a Statistical Life (VSL), were designed for different purposes and rest on different normative foundations. Applied without care, they can produce estimates an order of magnitude apart for identical mortality outcomes.

The core issue is one of decision context. Valuing health to set a budget calls for a different approach than valuing health to allocate within one. Conflating the two does not merely produce a different number; it answers a different question. And where different ministries apply different implicit values to the same life, the result is a systemic incoherence that cross-sector climate investment cannot afford.

The billion-dollar problem with climate-health valuation

Health is one of the strongest cases for climate action. The Intergovernmental Panel on Climate Change (IPCC) has said so, the Lancet Countdown has quantified it, and the World Health Organization (WHO) has repeated it. According to Romanello et al. (2025), the 2025 Lancet Countdown put the annual cost of heat-related mortality in people over 65 at US $261 billion. Numbers like that get attention from finance ministers in a way that excess-death counts on their own do not.

What happens after the headline is harder. When a specific climate investment is appraised against other claims on a constrained budget, the health case rarely survives the journey from communications to formal appraisal. The reason is not missing data. It is that the two most common ways to put a monetary value on a climate-related health impact, the Social Cost of Carbon (SCC) and the Value of a Statistical Life (VSL), were built for different purposes and rest on different normative foundations. They can produce numbers an order of magnitude apart for the same physical mortality, as reported by MacClancy et al. (2025) and MacClancy et al. (2026a).

The point of this piece is to make that friction visible and to offer a way of thinking about which value belongs in which decision.

The SCC: useful, partial, and easy to double-count

The Social Cost of Carbon is the standard monetary metric in climate policy appraisal. Integrated Assessment Models such as DICE (Dynamic Integrated Climate-Economy), FUND (Climate Framework for Uncertainty, Negotiation and Distribution), GIVE (Greenhouse Gas Impact Value Estimator), PAGE (Policy Analysis of the Greenhouse Effect), and the US Interagency Working Group (IWG) hybrid combine climate dynamics, physical impacts and damage functions to generate it. MacClancy et al. (2026a)reported that estimates in the recent literature range roughly between $51 and $228 per tonne of CO₂.

The complication for health is that the SCC already contains some health damages, but inconsistently. DICE embeds health implicitly inside aggregate GDP loss functions. FUND, GIVE, and the US IWG 2023 include explicit mortality and morbidity sub-modules, many of which are themselves monetised through VSL. In the ECO-CHICA review, MacClancy et al. (2026b, Recommendation 5) found these health components to be “often incomplete and embedded within functions that aggregate the damages associated with greenhouse gas emissions.”

Two implications follow, and they matter equally. Taking an SCC estimate at face value and adding a separately monetised health impact risks double-counting. Assuming the SCC has already counted a particular health channel when it has not, silently understates the case. Either way, the appearance of rigour is misleading. The constructive response is to ask what a given SCC estimate actually covers, supplement it deliberately, and integrate any additions transparently.

The VSL: powerful for visibility, contested for allocation

The Value of a Statistical Life is the dominant approach in climate, environmental and transport economics. As described by the OECD (2012) and US Environmental Protection Agency (EPA) (2020), it is derived from individuals’ willingness to pay for small reductions in the probability of sudden death, typically through wage-risk studies and stated-preference surveys. Multiplied across populations, it is what makes a figure like $261 billion possible.

For making the climate-health burden visible, the VSL has been valuable. Whether it is the right basis for choosing between investments is a separate question, and one worth naming directly:

  • VSL is derived from situations involving small, immediate changes in the probability of sudden death. Climate-related deaths are often delayed, diffuse, and cumulative. According to Hammitt (2020), whether a value derived from marginal acute risk transfers cleanly to chronic, systemic mortality is unresolved.
  • VSL scales with income. Transferring high-income estimates across borders implicitly assigns a smaller monetary value to a life lost in a low-income country.
  • Aggregate climate mortality is not marginal. The theoretical case for VSL rests on willingness-to-pay for small changes in risk, and large mortality burdens stretch the conditions of the underlying studies.
  • The same deaths produce very different totals across closely related methods. Simpson et al. (2025) estimated the social cost of mortality from London’s summer Urban Heat Island effect at £987 million using VSL and £453 million using the Value of a Statistical Life-Year. Same deaths, two-fold difference.
  • VSL-derived monetary values are typically around ten times higher than the implicit monetary values used by health-system bodies allocating fixed health budgets, as reported by MacClancy et al. (2026b).

None of this is an argument to retire the VSL. It is a reason to be precise about what it is doing in a given analysis.

The deeper question: which value, for which decision?

The friction between SCC and VSL is real, but it sits inside a larger puzzle that ECO-CHICA’s Delphi process surfaced without closing. The most useful way to frame it is as a question about decision context rather than method choice.

Within a fixed budget (a fixed health budget, or a fixed climate budget) one relevant shadow price is what is displaced at the margin: the opportunity-cost-based value of a unit of health or other benefit, conceptually derived from the binding constraint. For a climate budget, the equivalent shadow price is rarely estimated as such, and VSL is often used as a proxy for it. Whether the opportunity-cost frame is the right anchor for any given sector decision is itself a long-running debate that this piece does not try to settle.

When the question is instead how large a sector’s budget should be in the first place (whether to expand climate spending, whether to expand health spending), the relevant value is what society would give up to  obtain the gain. Willingness-to-pay-based measures (VSL, or willingness-to-pay (WTP) per quality-adjusted life year (QALY)) are the conceptually appropriate input for that question.

The gap between WTP-based and opportunity-cost-based monetary values, then, is not noise to be averaged away. It is information about the wider allocative system, though its interpretation is genuinely contested. One reading is that public budgets are set below what population preferences would support. Another is that WTP estimates over-state what populations would accept once full tax and contribution implications were made explicit. Both interpretations are defended in the literature. Either way, the gap is large enough to matter for cross-sector decisions.

This points to a more concrete answer to the cross-sector question. Within a sector’s existing budget, intra-sector shadow prices speak to displacement at the margin. Across sectors, when a budget is being set or reviewed, WTP-style values are conceptually relevant, and the gap between the two becomes information about the system as a whole. A government that allocates a cross-sector climate-action budget without making this distinction explicit is implicitly applying inconsistent values to a life or QALY depending on which department produces it. That is an incoherence worth naming explicitly.

Two further methodological threads sit alongside this one and deserve mention rather than full treatment here. Equity weighting, i.e., whether and how to adjust monetised health impacts for the distribution of who gains and who loses, across countries and across generations, is contested in both the welfare-economic and health-economic traditions and was left unresolved by the Delphi, as described by MacClancy et al. (2026a, 2026b). Discounting, particularly the choice of social rate for benefits accruing intergenerationally, has been the subject of an ongoing debate in environmental economics that health economics is beginning to engage with, according to Bray et al. (2024). Both shape final estimates in ways at least as large as the SCC and VSL choices discussed above.

Two open questions

There is no avoiding decisions about which value to use, and the choice changes the answer. Two questions follow from the analysis above:

  • When valuations within a sector and across sectors give materially different answers, what should the international system expect a finance or planning ministry to use, and what would a useful reporting standard look like, not a single method but a requirement to declare which decision context is being addressed and how alternative valuations would change the conclusion?
  • Where welfare-economic and opportunity-cost monetary values diverge by an order of magnitude, is this best treated as a methodological inconvenience or as substantive evidence that current public-budget allocations are out of step with social preferences for health?

Where this leaves us

The methods exist. What is missing is the discipline of matching each valuation to the decision it can actually answer, and being explicit about the alternatives when more than one applies. Without that discipline, climate-health evidence will continue to do more for advocacy than for allocation, and the headline numbers that prompted the conversation will keep failing to reach the appraisal stage where investment is decided. ECO-CHICA’s contribution is to make that discipline concrete: a typology of decision contexts, a transparent reporting standard, and a clear account of what each method can and cannot tell a decision-maker. The harder work, embedding it in how climate-health appraisals are commissioned, conducted and used, sits with funders, ministries, and health technology assessment (HTA) bodies. Geneva is one of the rooms where that conversation needs to happen. It is not the only one.