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

Ambitious climate policies necessitate the phasing out of fossil fuel production, potentially leading to substantial losses for investors – so-called stranded assets. This raises concerns, particularly regarding underfunded pension plans in wealthy nations. The recent surge in fossil fuel investments due to the war in Ukraine exacerbates these anxieties. This study addresses the question of the social relevance of these potential losses, exploring whether asset stranding disproportionately impacts the affluent or a wider population. The prevailing narrative suggests significant societal consequences, potentially leading to diluted climate ambitions. However, the authors argue that this fear is unfounded because any resulting economic hardship can be compensated for at a relatively low cost. This research utilizes a novel approach, combining detailed macroeconomic and financial network data to analyze the distribution of stranded asset ownership in the US and Europe. By modeling potential profit losses from over 40,000 oil and gas fields and tracing these losses to ultimate owners via shares and funds, the study offers a comprehensive assessment of the distributional implications of asset stranding.

Existing research highlights the risks of stranded assets and their potential impact on institutional investors and pension funds (Silver, 2017; Christophers, 2019; van der Ploeg & Rezai, 2020). Studies have documented the unequal distribution of carbon emissions and footprints, showing that the wealthiest individuals and households bear the greatest responsibility (Semieniuk & Yakovenko, 2020; Chancel, 2022). However, a critical gap remains in understanding the distribution of ownership of at-risk fossil fuel assets. This paper builds on previous work on stranded assets (Semieniuk et al., 2022) and distributional national accounts (Blanchet et al., 2020; Saez & Zucman, 2016), providing a more detailed analysis of the distributional consequences of asset stranding.

The study combines detailed macroeconomic and financial network data to model and analyze the distribution of stranded asset ownership. The authors model stranded assets as the present value of potentially lost profits from over 40,000 oil and gas fields, considering varying expectations realignments towards a lower-carbon future (reflecting different global warming scenarios). Losses are traced to ultimate owners—individuals and governments—who own stocks in oil and gas companies through shares and funds. The financial asset size distribution is constructed using income tax records, wealth surveys, and national accounts data, following Distributional National Accounts (DINA) guidelines. As a first approximation, they assume that each wealth fractile is allocated stranded assets proportionally to its share of national financial assets (wealth elasticity of one). For robustness checks, the authors consider three alternative expectation realignments and different elasticities, accounting for potential biases in investment portfolios across the wealth distribution. The analysis focuses on the US and several major European countries (UK, France, Germany, Italy).

The analysis estimates losses exceeding $500 billion in a 'medium' scenario (around 2°C warming). In the US, approximately $350 billion in stranded assets are estimated, with only 3.5% impacting the poorest half of the population and one-third hitting the bottom 90%. The remaining two-thirds are concentrated among the top 1% and the next 9%. Similar skewed distributions are observed across Europe, with total losses estimated around $200 billion. Despite these substantial absolute losses for the wealthy, they represent a small fraction of their total wealth (less than 1% for the top 1% even in the most severe scenario). Conversely, for the bottom 50%, particularly in the US and UK, losses as a share of net wealth are much larger. This is linked to the structure of pension systems; market-based systems (like in the US and UK) expose a significant share of lower-wealth individuals to these losses, unlike continental European systems. The study also considers potential losses in sectors using fossil fuels as inputs, suggesting comparable magnitudes and the possibility of counterbalancing effects through increased value in other portfolio positions. Compensation for stranded assets for the bottom 50% under the 'medium' scenario would cost around $9 billion in Europe and $12 billion in the US; small compared to recent government bailouts and potential costs related to investor treaties.

The findings highlight a stark disparity in the impact of stranded assets: the wealthy bear the brunt in absolute terms, yet the losses are manageable relative to their overall wealth, while less affluent groups, particularly in the US and UK, could face severe economic hardship. This suggests a policy implication: governments could compensate for socially relevant asset devaluation at a low cost, mitigating the negative effects on vulnerable populations. The estimated compensation costs are significantly lower than other large-scale government interventions and potential costs associated with investor-state dispute settlement mechanisms. This implies that fears of substantial societal repercussions from climate policies are largely unfounded, provided that appropriate compensation measures are implemented.

The study concludes that the risk of stranded assets should not deter high-income countries from ambitious climate action. Losses are disproportionately concentrated among the wealthy, and compensating vulnerable populations would be relatively inexpensive. However, the study emphasizes the need for increased data transparency and government capacity to effectively track asset ownership to ensure fair decarbonization policies. Future research should explore the macroeconomic impacts beyond financial capital ownership, considering labor income and impacts in less affluent countries.

The study acknowledges several limitations. Data transparency and availability on asset ownership remain limited, affecting the precision of the analysis. The analysis primarily focuses on financial capital ownership in affluent countries, neglecting labor income losses and other macroeconomic effects. Further research should investigate more granular portfolio holdings and the interplay of stranded assets with other macroeconomic factors, especially in developing countries.