British pensions are about to bankroll the American tech revolution
From SpaceX to Anthropic, America’s next industrial revolution is being powered by British pensions, writes Paul Armstrong
Anthropic’s IPO filing matters less because of Anthropic itself and more because of who may follow. OpenAI is widely expected to pursue a public listing at some stage. Reports suggest SpaceX has already confidentially filed paperwork of its own. Taken together, three of the most important private companies in the world will become investable within a relatively short period, creating a capital markets event unlike anything seen since the emergence of the commercial internet.
Few investors under 40 have experienced anything quite like it. Financial markets occasionally encounter moments when entirely new industries become available to public investors. Railways created one such moment. Electricity created another. Cars, aviation and the internet followed. OpenAI, Anthropic and SpaceX increasingly look less like technology companies and more like the infrastructure layer for the next economic era. Artificial intelligence, satellite communications, autonomous systems, defence technology and advanced computing increasingly meet inside a handful of businesses whose influence already stretches far beyond their balance sheets and increasingly getting them into courtrooms.
A lot of the discussion around Britain’s pension reforms has focused on encouraging greater investment into growth assets. Policymakers want pension capital working harder, pension funds need returns and savers need growth. The logic behind those reforms is difficult to dispute. Less attention has been paid to where much of that capital is likely to end up.
Britain participates mainly as an investor rather than as an owner of the underlying ecosystem, and that distinction matters more than it might appear. British investors financed vast amounts of American industrial expansion during the nineteenth century, backing railways, ports, mining operations and infrastructure projects that generated returns for London while helping economic power accumulate elsewhere. Similar patterns emerged during the internet era as British institutions benefitted from owning Microsoft, Amazon, Apple, Google and Meta while the highest-value jobs, infrastructure and strategic influence remained firmly rooted in the United States. OpenAI, Anthropic and SpaceX (and others) represent the next version of that story.
Supporters will argue that none of this matters. Pension funds exist to maximise returns rather than pursue industrial strategy.
A retiree in Leeds should not receive lower returns simply because an investment happens to be based in California rather than Cambridge. Pension managers owe their members returns, not national loyalty. Any pension manager who ignored potentially transformational businesses purely because they were American would rightly face scrutiny, but commercial logic isn’t the same thing as national advantage.
Every pound invested into OpenAI, Anthropic or SpaceX helps finance American infrastructure, American energy projects, American research facilities and American jobs. Data centres not built in Birmingham, or launch facilities not appearing in Bristol. Most of the highest-value supply chains, talent clusters and secondary industries surrounding these businesses are developing elsewhere and there lies the contradiction sitting at the heart of Britain’s technology ambitions. Political leaders on both sides seem to endlessly talk about creating the industries of the future while market incentives increasingly direct British capital towards financing someone else’s. The contradiction has existed for decades but the scale of what is now arriving makes the likely impact much harder to ignore.
OpenAI has received a valuation of around $850bn, after closing a record $122bn funding round in March. SpaceX is targeting a $1.5 trillion valuation ahead of an anticipated IPO later this year. Anthropic this week raised $65bn at a valuation approaching $1 trillion, briefly overtaking its rival. Combined, those three businesses are already worth more than $3 trillion in private markets. Public markets have rarely been asked to absorb companies of such scale operating at the centre of technologies that governments increasingly view as strategically important.
Markets don’t possess infinite capacity and pension funds, sovereign wealth funds and institutional investors all have to make choices that carry an opportunity cost. Railway mania distorted capital allocation across Victorian Britain. Internet enthusiasm shaped investment behaviour during the late 1990s. Investors organised portfolios around a small number of companies perceived to be defining the future, often at the expense of everything else. A similar dynamic looks likely to emerge again. British pension funds seeking growth may find it difficult to justify not owning businesses sitting at the centre of artificial intelligence and space infrastructure. Every allocation towards those companies, however rational in isolation, represents capital unavailable to smaller listed firms, domestic growth companies and emerging challengers elsewhere.
The more consequential question isn’t whether artificial intelligence represents a bubble but what happens if the technology succeeds exactly the way supporters predict. Success requires unprecedented amounts of energy, computing infrastructure, specialist talent and financing. Economic value would become increasingly concentrated among a small number of companies operating at extraordinary scale. Public markets are then deeply exposed to their performance. Pension funds would become deeply exposed to their performance. Governments would become increasingly dependent upon infrastructure and services controlled by businesses whose influence already extends into defence, communications, education, healthcare and enterprise software.
There’s a further risk that rarely features in the commentary; extreme concentration of pension exposure to a handful of AI companies creates retirement security risks that have nothing to do with geopolitics. If these businesses stumble, the consequences go beyond portfolio statements and ripple through the savings of millions of people who never consciously chose to bet their retirements on the outcome of an artificial intelligence race.
OpenAI, Anthropic and SpaceX will likely prove to be exceptional investments for various reasons, and Britain’s pension funds will be hard pressed to ignore them. A stranger possibility sits underneath the excitement. Future generations could retire on gains generated by American technological dominance while wondering why so little of that dominance was built at home. A once-in-a-generation investment opportunity may be approaching, and a permanent transfer of economic power might just arrive alongside.
Paul Armstrong is the founder of emerging tech advisory, TBD Group, and its intelligence community, TBD+