Bottom line
Energy security behaves like insurance: a premium paid in advance, invisible until the moment it is needed. Because no single company or politician captures the payoff, it is systematically underbought.
Across four very different countries, the binding constraint on the transition is rarely technology. It is the price signal in the UK, the cost of capital in India, domestic demand in Saudi Arabia and grid access in the US. In each case, the investment that buys security is the same one that delivers affordability and decarbonisation. Pricing security explicitly, as we already price defence and carbon, is how that investment gets financed.
Introduction: a terminal that changed its mind
In 2008, a company opened an LNG import terminal at Sabine Pass, Louisiana, betting that America would need foreign gas for decades. In 2016, the same terminal shipped its first export cargo. It took eight years to invert a multi-billion-dollar bet on the direction of global energy.
That reversal captures the problem this memo addresses: energy systems change faster than the infrastructure, and the policy, we build for them. Every energy decision is pulled by three forces. Security is who controls supply and what happens when it is cut off; affordability is the cost to households, industry and the political mandate to act; climate is the emissions arithmetic each country actually faces.
Most analysis treats climate as the driver and the other two as constraints. Climate is my end goal, but I treat security and affordability as equal partners, and my research suggests the usual ordering is often backwards. Comparing four countries that could not be more different, I find that security, priced properly, is the lever that moves the other two.
Four countries, four different problems
Geography sets the starting point, because no country chose its hand. Britain has world-class offshore wind but little else; India is rich in coal and sun but poor in oil and gas; Saudi Arabia has oil, gas and exceptional sun but no coal. The US has almost everything, which is why its problem is coordination rather than resources.
United Kingdom: decarbonising while the resource base disappears
Britain's decline is geological, not political. Around 90% of recoverable North Sea reserves are gone, production is down about 77% from its 1999 peak, and gas output is forecast to fall 99% by 2050 under current policy and 97% even with new licences. Drilling is not a security lever.
On climate, Britain is a real success: coal has fallen from two-thirds of power in 1990 to 0.1%, and the grid is about 62% low-carbon. Yet households pay roughly 33 US cents per kWh, the highest in the G7 and double the US. The cause is market design, not renewables: the price is set by the last plant switched on, usually gas, so everyone pays the gas price even when most power is clean. Affordability is the binding constraint.
India: serving growth without locking in emissions
India is the largest source of global energy-demand growth to 2035, and its transition is additive rather than substitutional: renewables and coal both expand. It crossed 50% non-fossil capacity in 2025, years ahead of its Paris target, yet coal still supplies about 75% of generation, because without storage coal covers the evening peak.
Its security exposure sits in oil, not power. Only about 18% of coal is imported, but 88.7% of crude oil is, rising towards 92% by 2035. Every electric vehicle therefore displaces imported crude, making EVs national-security policy. The deepest bottleneck is financial: the distribution companies (discoms) that buy power carry about $77 billion in accumulated losses.
Saudi Arabia: funding a transition away from what funds it
Saudi Arabia inverts the problem. Its risk is stranding, not shortage: the world ceasing to need what it sells. With about 36 million people, it exports far more per person than anyone, and that ratio drives Riyadh's choices.
Saudis pay about 5 cents per kWh, and every barrel burned cheaply at home is a barrel not sold abroad, so subsidies are foregone export revenue. With oil revenue down 45% from its 2022 peak, the kingdom is eliminating about a million barrels a day of domestic oil burn by 2030, freeing over $150 billion in export revenue by 2035. This is a revenue argument that happens to produce a climate outcome, which is why it is durable.
United States: coordinating without a decision-maker
The US has the strongest resources and the weakest ability to direct them: three largely separate grids, 50 state regimes and no single decision-maker. Shale scaled here partly because American landowners own the minerals beneath their land, and the US became a record net exporter of 11 quads in 2025.
But it is independent nationally and dependent regionally: the Gulf Coast is the only net petroleum-exporting region, and the country still imports 17 quads of petroleum. The real constraint is the grid. More than 2,300 GW waits to connect, more than total installed capacity, and transformers face 128-week lead times.
The argument: security behaves like insurance
These are not four separate cases but one system across fifty years. The 1973 embargo quadrupled oil prices, made the high-cost North Sea economic and created the British industry now in decline. US shale broke OPEC's pricing power from 2014 and helped trigger Saudi Arabia's Vision 2030. In 2022, Europe swapped Russian gas for American LNG, making the US the world's largest LNG exporter.
Security shares three features with insurance. First, the premium is paid in the good years: reserves, spare capacity and bypass pipelines cost money when nothing is wrong, which is why the IEA's 90-day oil reserve had to be mandated in 1974. Second, the payoff is a disaster that does not happen, and no politician is re-elected for a crisis they prevented. Third, under-insurance is a textbook externality: no single utility carries national supply risk, and if carbon can be priced as an externality, so can insecurity.
The 2026 Hormuz shock tested this in real time. About 13 million barrels a day, a fifth of global oil, was disrupted: the largest supply shock in the market's history. Saudi Arabia rerouted exports through its long-standing Petroline bypass, a premium paid decades earlier; India scrambled for non-Gulf barrels; the US, a net exporter, was largely insulated. The system absorbed this larger disruption better than 2022, because resilience built in between paid out.
Europe in 2022 is the counter-example. Having underinvested in spare supply, storage and grid flexibility, European governments spent about €792 billion in roughly eighteen months shielding households and firms from the price spike. That money bought no new energy, storage or grid; it is more than double the roughly $370 billion of clean-energy spending in the US Inflation Reduction Act. You either pay the premium in advance and get an asset, or pay the claim in the crisis and get nothing.
How security finances the transition
Security finances the transition through three mechanisms. In Saudi Arabia it is revenue: the $150 billion freed by cutting domestic oil burn funds the clean build, so security and transition spending are the same money. In Britain it is substitution: 15 GW of renewables avoids around 78 LNG tankers a year, so every wind farm is also a security asset. In India it is the cost of capital, the deepest lever of all.
Renewables are front-loaded: nearly all the cost is paid upfront, so financing makes up 20–50% of the final price of clean power. Solar's cost of capital is about 4.4% in Europe and 5% in the US, but around 9% in India. Raising it from 5% to 9% lifts solar's cost by about 34% and wind's by 29%, but coal's by only 10%, so expensive money tilts the field back to fossil fuels. The IEA identifies off-taker risk, the insolvent discoms, as the main driver of India's premium: a financing problem wearing a technology costume.
The engine is blended finance. A state or development bank takes the first-loss position, absorbing the first hit if a project fails, and private capital follows; one public dollar can mobilise two or three private ones. Cutting developing-economy capital costs by 200 basis points would save an estimated $15 trillion to 2050. Markets will not take first loss voluntarily, which is exactly why security must be priced: the instruments exist, but the price signal does not.
China has already answered who underwrites at scale. It invested $625 billion in clean energy in 2024, 31% of the global total, and clean technology is now about 10% of its GDP, driving roughly a third of its growth. Its overcapacity crashed global prices for everyone. The lesson for democracies is not to copy China's factories, but that someone must be the underwriter of first resort, and in the West no one yet is.
One honest limit: security can justify an LNG terminal as easily as a wind farm. Security is not inherently green; the overlap is strong in power and weaker in aviation and shipping. I am claiming the overlap, not the whole map.
Proposals: one principle, four instruments
Place the four on one map and a pattern appears. The importers, Britain and India, sit low on carbon per person; the exporters, Saudi Arabia and the US, sit high. Import-export position decides which leg of the trilemma bites first, so there is no universal answer. There is, however, one principle: find the intervention where the security return and the transition return are the same investment, then lower the cost of capital against it.
United Kingdom: reform the price, not the ambition. Expand two-way Contracts for Difference and pass their fixed strike prices of roughly £40–50/MWh through to bills, so a clean grid finally prices like one. Reframe the £40 billion-a-year Clean Power 2030 build as import substitution, and press two durable edges: North Sea carbon storage and battery storage. The tempting alternative, zonal pricing, was rightly rejected in July 2025. Government analysis found that a one-point rise in developers' cost of capital would turn its benefit into £12 billion of added cost to 2050, whereas CfDs have helped cut offshore wind strike prices by about 70% since 2015.
India: attack the cost of capital. Scale blended finance and first-loss guarantees through IREDA and the development banks, fix the discoms, and treat every EV as import substitution. The discom reform is already working. Under the roughly $37 billion Revamped Distribution Sector Scheme, which pays out only when utilities hit milestones, losses fell from 21.9% in 2021 to a record 15% in 2025, and discoms posted their first collective profit in over a decade. A creditworthy buyer means cheaper capital for every solar project selling to it. The sector has stopped bleeding but not healed, so fund the reform harder.
Saudi Arabia: decarbonise for revenue. Deliver the Liquids Displacement Program, reform electricity tariffs, and use the Public Investment Fund as the region's transition underwriter. Tariff reform sounds impossible in a petro-state, but the Citizen's Account, launched in 2016, already proved otherwise. Its targeted cash transfers to around 10 million lower-income people allowed prices to rise by up to 80% while energy subsidies fell from 5.5% to about 3.5% of GDP. A transfer protects poor households while keeping the price signal that curbs waste; a blanket subsidy destroys it.
United States: solve delivery, not production. Implement and legally defend FERC Order 1920, the biggest transmission reform in a generation; make large loads such as data centres pay their share; and use storage to route around the queue. In December 2025, PJM's capacity auction fell 6,623 MW short of its reliability target, with data centres causing about 5,100 MW of the gap. FERC then ordered data centres to pay for the capacity they cause, and in June 2026 extended this to all six regional grid operators, citing national security. With PJM capacity prices up from $29 to $329 per MW-day, who pays for the data-centre boom is the defining consumer-energy question of the next five years.
In every case the arrow runs the same way. Pull the security lever, and affordability and climate follow rather than being traded off.
Conclusion: put a price on the premium
We already price national security, through roughly $2.7 trillion a year of global defence spending, and we price carbon at around €80 a tonne in Europe. Energy security is the same kind of risk, yet it has no price, no market and no line item. The objective should not be to minimise cost, but to minimise cost while pricing in the risk, as any serious portfolio does.
This research raises more questions than it answers, and three stand out. First, build a priced security index, in which expected annual loss equals the probability of a shock times its consequence, minus a country's buffer; ministers and voters act on dollars, not diversity indices. Second, interview the energy minister, finance minister and security adviser, who make these decisions but rarely think together. Third, map the shifts in energy power, from OPEC in 1973 to US shale to China, which now refines 19 of the 20 strategic minerals the transition runs on.
In 2022, Europe paid nearly €800 billion after its energy bet failed and got nothing lasting for it. In 2026, the countries that had quietly paid their premiums in advance weathered an even larger disruption far better. Every country in this memo already runs an insurance policy it refuses to price; they differ only in the currency they pay and whether they pay before the crisis or after. The work I want to do next is to put a number on that premium, so we start paying it before the house burns down.
Sources
Figures are drawn from the sources cited in the accompanying presentation. The Saudi per-capita export figure is my own calculation (exports divided by population).
Global: IEA Oil Market Report and Global Energy Review 2026; IEA/BNEF cost-of-capital data; BloombergNEF LCOE 2026; IRENA 2026; SIPRI 2024; Our World in Data / EDGAR; World Economic Forum
United Kingdom: NSTA; DESNZ DUKES and Energy Trends 2025; Ofgem; GlobalPetrolPrices.com (Q1 2026)
India: Central Electricity Authority; CREA; S&P Global; CII-EY India Energy Security report; Rajya Sabha reply (FY2025–26)
Saudi Arabia: OPEC Annual Statistical Bulletin 2025; Enerdata; MEES; Vision2030.ai
United States: EIA Monthly Energy Review and STEO; FERC Order 1920 and PJM orders; Lawrence Berkeley National Lab; RMI; Congressional Research Service
Europe: Bruegel, national crisis support tracker (EU, UK and Norway, Sep 2021–Jan 2023)