The concept at the heart of the study is the cost of capital, or CoC: the rate of return that investors and lenders demand to compensate them for the risk of financing a project. This matters enormously because almost every low-carbon technology — wind turbines, heat pumps, electric arc furnaces, carbon capture systems — is capital-intensive. Most of the expense is paid up front, while the benefits, in the form of energy or materials, trickle in over decades. A technology whose costs are dominated by upfront investment is exquisitely sensitive to interest rates and return expectations. Raise the cost of capital by a few percentage points and a wind farm that looked cheap on paper can become uncompetitive against a gas plant, whose costs are spread across years of fuel purchases.

Despite this sensitivity, the integrated assessment models that inform reports by the Intergovernmental Panel on Climate Change have historically assumed a single, uniform cost of capital for all countries, sectors and technologies. The simplification was born of necessity: data were scarce and modelling complexity was already formidable. But it embeds a hidden assumption — that financing a solar project in Eastern Africa costs the same as financing one in the United States. The new research shows just how wrong that assumption is. Drawing on sovereign credit ratings and stock-market risk measures from the widely used NYU Damodaran database, the team calculated financing costs for eight climate-relevant sectors, including electricity, hydrogen production, cement, iron and steel, pulp and paper and food processing. The heterogeneity is striking: the cost of capital for electricity in high-risk regions such as Eastern Africa or Ukraine is roughly three times higher than in the United States. Even within a single low-risk country, sector differences matter — the cost of capital for American steel is about two percentage points higher than for American electricity, a gap equivalent to the difference between power projects in the United States and in India or Indonesia.

The team then fed these differentiated estimates into the IMAGE integrated assessment model, a well-established framework with high geographic and technological resolution that simulates energy systems and industry across 26 world regions through 2100. They ran two scenarios: one with the model’s typical uniform low-risk cost of capital of 5 percent, and one with their dynamic, region- and sector-specific estimates, which also capture how economic growth in developing countries gradually lowers sovereign risk premiums over time. Both scenarios were paired with a current-policies pathway, leading to about 2.8 degrees of warming, and an ambitious pathway that keeps warming well below 2 degrees, in line with the Paris Agreement.

The headline finding is that accounting for real-world financing conditions raises cumulative global mitigation costs through 2100 from 77 trillion to 84 trillion US dollars — an increase of roughly 9 percent. When the researchers decomposed this figure, they discovered that sectoral risk differences alone actually lower mitigation costs slightly, mainly by reducing electricity prices in some low-risk regions. The damage comes overwhelmingly from country risk, which adds about 6 percent to costs, and from the extra climate policy effort needed to compensate for the emissions that expensive capital leaves on the table, adding another 4 percent. In other words, the biggest distortion in decades of climate modelling has not been misjudging how risky the steel industry is — it has been pretending that a power plant in a country with a poor sovereign credit rating can be financed as cheaply as one in Zurich.

The emissions consequences are substantial. Without an endogenous policy reaction, introducing realistic financing costs raises cumulative global CO2 emissions through 2100 by roughly 60 to 70 gigatonnes — more than one additional year of current global emissions — driven overwhelmingly by high-risk regions. Expensive capital pushes electricity systems in those regions away from capital-intensive renewables and nuclear power and toward natural gas, which is less sensitive to financing costs than coal. It also suppresses investment in electricity grids in developing countries, a finding with major implications given that grid expansion needs for net zero are as large as, or larger than, those for power generation itself. And because energy prices ripple through the whole economy, sectors that were never directly modified in the model — transport and heating among them — also emit more, as higher power prices delay the switch from oil to electricity in passenger vehicles and trucking.

Perhaps the most striking results concern green hydrogen. Hydrogen production today is dominated by firms associated with the chemical sector, which carries above-average market risk, and electrolytic hydrogen is fully exposed to electricity prices, which themselves rise when the cost of capital rises. The result: hydrogen costs increase in nearly every region, with increases well above 20 percent in high-risk areas such as Western Africa — precisely the regions often envisaged as future hydrogen exporters. Global hydrogen production via electrolysis drops substantially in the model, while the incumbent technology, natural gas-based steam methane reforming, holds on longer. Cheaper-to-abate alternatives become more CO2-intensive and expensive, and hydrogen loses attractiveness for decarbonizing transport, industry and power, with end-of-century hydrogen usage falling accordingly under the ambitious climate scenario.

Heavy industry tells a similar story. In cement, responsible for about 7 percent of current global CO2 emissions, country and sector risks slow the transition from inefficient standard kilns to modern, efficient kilns equipped with carbon capture and storage, leaving a larger share of unabated production in place by 2100. In iron and steel, which accounts for roughly 8 percent of emissions, higher financing costs entrench the carbon-intensive blast furnace and basic oxygen furnace route, reduce the uptake of carbon capture on both that route and direct-reduced iron production, and squeeze out novel low-carbon processes such as hydrogen-based direct reduction and electrowinning. Industrial carbon capture by cement and steel falls substantially, overwhelmingly because of conditions in high-risk regions. The common thread is capital intensity: the very technologies needed to decarbonize industry are the ones most punished by expensive money.

The geographic redistribution is equally consequential. Mitigation costs rise not only in high-risk regions with persistently poor credit ratings, such as Ukraine or parts of South Asia, but also by 5 to 6 percent in major low-risk economies like the United States and China, as the model shifts cost-effective abatement toward lower-risk regions. In some high-risk parts of Eastern and Southern Africa, the effect is paradoxically reversed: abatement efforts shrink so much that overall abatement costs fall, even as global emissions rise. The study’s authors stress that economic growth in developing countries, while helpful, is unlikely on its own to close the financing gap, because sovereign risk also depends on institutional quality, regulatory stability and debt sustainability — factors that do not scale automatically with GDP.

The policy implications are pointed. Well-designed international climate finance can de-risk investments across all climate-relevant sectors in developing countries, helping to avoid what researchers have called investment traps, yet international mitigation finance remains heavily concentrated in low-carbon electricity generation, leaving grids, hydrogen and heavy industry underserved. The authors also caution that their estimates reflect market conditions rather than any judgment about whether financial markets price risk correctly, and that their model abstracts from sectors such as chemicals and transport finance. But the core message is hard to escape: the world’s climate models have been quietly assuming that money is cheaper than it is, especially where it is needed most. Correcting that assumption makes decarbonization slower, more expensive and more uneven — and makes the case for fixing global financing conditions itself a central piece of climate policy.

Subject of Research: The effect of country- and sector-specific cost of capital differences on global energy and industry decarbonization pathways

Article Title: The impact of financing cost differences on global energy and industry decarbonization

Article References: Waidelich, P., Crassier, C., de Boer, H. S., Tautorat, P., van Vuuren, D. P., & Steffen, B. (2026). The impact of financing cost differences on global energy and industry decarbonization. Nature Climate Change, 16(10), 1178-1186. https://doi.org/10.1038/s41558-026-02756-0

Image Credits: AI Generated

DOI: 10.1038/s41558-026-02756-0

Keywords: cost of capital, climate finance, integrated assessment models, decarbonization, renewable energy, green hydrogen, carbon capture and storage, developing countries, sovereign risk, steel industry, cement industry, IMAGE model