Potential pension fund losses should not deter high-income countries from bold climate action

The paper addresses whether potential losses from stranded fossil-fuel assets should deter high-income countries from implementing ambitious climate policies. Governments may fear social repercussions if asset stranding negatively impacts pension plans and investors, especially amid recent increases in fossil-fuel valuations and expanded investments in response to energy security concerns. The core questions are: How socially relevant are stranded asset losses, and who ultimately bears them through financial ownership? The authors argue that while asset stranding is a necessary consequence of ambitious climate policy, any resulting economically meaningful hardship can be compensated at low cost. They set the context by noting the high concentration of financial asset ownership at the top of the wealth distribution and the limited analysis to date on the distribution of ownership of fossil-fuel assets at risk of stranding.

The study builds on literature showing that ambitious climate mitigation implies leaving a portion of fossil-fuel reserves unextracted, creating stranded assets and potential investor losses. Prior research has examined investor underestimation of stranded asset risks and how institutional investors consider climate risk. Work on global carbon inequality documents that the richest households account for a disproportionate share of emissions and investment-related impacts. Related modeling has quantified potential investor losses from stranded fossil-fuel assets in advanced economies and provided guidelines for harmonizing wealth and distributional statistics (Distributional National Accounts). Additional studies suggest possible cascading stranding in sectors using fossil fuels as inputs, though such sectors may have substitution opportunities that can offset aggregate portfolio losses.

• Scope and objective: Quantify and distribute potential losses from stranded oil and gas assets among ultimate owners in the United States and European countries, and assess their social relevance.
• Stranded asset modeling: Stranded assets are defined as the present value of potentially lost profits from over 40,000 oil and gas fields as investor expectations realign to a lower-carbon future. Several expectations realignment scenarios are considered, including a “medium” case consistent with roughly 2°C warming and a more “severe” case consistent with a 1.5°C pathway, as well as lower-loss scenarios. These scenarios change both the total magnitude and international distribution of losses.
• Ownership tracing: Losses are traced from oil and gas companies to ultimate owners (persons and governments) via direct equity and fund holdings using macroeconomic and financial network modeling of ownership across countries and sectors.
• Allocation across wealth fractiles: For each country, losses accruing to non-government ultimate owners are allocated across the wealth distribution in proportion to each group’s share of national financial assets (baseline wealth elasticity of stranded assets = 1). This implies that $1 of financial assets owned by any wealth group has the same probability of stranding, while recognizing that absolute holdings differ greatly due to wealth inequality.
• Data sources for wealth distribution: Financial asset and overall wealth distributions are compiled from income tax records, wealth surveys, and national accounts, harmonized using Distributional National Accounts (DINA) guidelines to ensure cross-country comparability.
• Portfolio bias robustness: To test sensitivity, alternative allocations incorporate portfolio biases: a higher fossil-fuel exposure among the bottom 90% (low-wealth bias) and, alternatively, a higher exposure among the top 10% (high-wealth bias). These yield different elasticities/distributions of stranded assets across fractiles.
• Geographic coverage and reporting: Results are reported for Europe (aggregate) and the United States, with country details for the UK, France, Germany, and Italy. Outcomes are presented both in absolute amounts and as shares of net wealth by fractile.

• Aggregate losses: In the medium scenario (≈2°C), total losses borne by individuals in the United States and Europe exceed $500 billion. Approximate regional totals: United States ≈$350 billion; Europe ≈$200 billion.
• Distribution across wealth groups (United States): Only about 3.5% of total losses accrue to the bottom 50% by wealth; roughly one-third accrues to the bottom 90%. About two-thirds of losses accrue to the top 10%, split roughly equally between the top 1% and the next 9%.
• Europe shows similarly skewed distributions, reflecting high concentration of financial assets at the top (top 10% holding 70–90% of financial assets in most countries examined).
• Relative impact: For affluent groups, stranded asset losses are small relative to their wealth. In the medium scenario, the top 1% in each country loses less than 1% of net wealth; even under the severe scenario (≈1.5°C) with a portfolio bias toward fossil fuels, top 1% losses are under 2% of net wealth.
• Lower- and middle-wealth groups generally experience smaller losses relative to net wealth because housing dominates their portfolios; however, the bottom 50% can face larger losses as a share of net wealth, especially in the United States where average per-adult net wealth is about $2,000 and negative for the bottom 20%.
• Pension exposure: In the US and UK, the bottom 50%’s financial assets are heavily concentrated in pension entitlements tied to capital market valuations, increasing exposure. Continental European pay-as-you-go systems (e.g., France, Germany, Italy) reduce such exposure.
• Compensation costs: Compensating all stranded asset losses of the bottom 50% under the medium scenario would cost about $12 billion in the United States and $9 billion in Europe—amounts modest compared to recent energy-sector support measures and potential investor-state dispute payouts.
• Funding feasibility: A US carbon price of $13/MTCO2e could raise ~$74 billion per year; compensating the bottom 50% would require about one-sixth of a single year’s revenue. A modest progressive wealth tax on the top ~0.005% (2% above $100m; 3% above $1bn) could cover total losses in ~2 years in the US and <3 years in Europe. Savings could also arise from reform/withdrawal from investor protection treaties that indemnify fossil investors.

The findings indicate that stranded fossil-fuel asset losses are highly concentrated among wealthy investors who can absorb them with minimal impact on overall net wealth. Consequently, the prospect of stranding should not deter ambitious climate action in high-income countries. The main social concern lies with less affluent groups—especially in countries where pensions are market-exposed—who may experience meaningful hardship from even small absolute losses due to low or negative net wealth. However, these socially relevant losses are small in aggregate and can be compensated at relatively low fiscal cost, especially when compared to existing energy-sector bailouts or potential treaty-based investor compensation. Potential losses in fossil-fuel-using sectors may be partially offset by gains in other sectors as portfolios adjust, and substitution options are more available outside upstream fossil fuels. Policymakers can design compensation schemes funded through carbon pricing, wealth taxation, or treaty reforms, while also using carbon dividends to neutralize regressive impacts of pricing policies. The results support the view that equitable compensation mechanisms can neutralize distributional obstacles to strong decarbonization policies.

The study contributes a first-order, cross-country assessment of who bears losses from stranded oil and gas assets and shows that these losses are disproportionately owned by the wealthiest, while the economically vulnerable can be protected at modest public cost. Therefore, fears of pension fund and low- to middle-class investor losses are not a credible deterrent to bold climate action in high-income countries. The authors call for improved governmental statistical capacity to track stranded asset ownership and note that the principal political challenge is likely to be lobbying by affluent fossil-fuel interests. Future research should integrate labor income effects, broader macroeconomic impacts, and more granular portfolio data across the wealth distribution to refine exposure estimates and policy design.

• Data transparency and availability on ultimate ownership and portfolio composition are limited; results rely on best-available macro-financial modeling and harmonized distributional datasets.
• Baseline allocation assumes stranded asset exposure proportional to financial assets (wealth elasticity = 1); although alternative portfolio bias scenarios are explored, actual exposures may vary by group and over time.
• Analysis focuses on upstream fossil-fuel assets and financial capital; indirect and downstream sector impacts are only discussed qualitatively and may offset or amplify losses.
• Labor income losses and broader macroeconomic dynamics are not modeled; results represent a first-order approximation.
• Coverage centers on affluent countries (US and Europe); findings may not generalize to lower-income, fossil-fuel-exporting countries that face different macroeconomic risks despite smaller absolute stranded assets.