Somewhere in the small print of a Department for Transport planning document published last week, the British government admitted that a third runway at Heathrow would add just 0.05% to UK GDP, roughly a tenth of what the airport itself had been claiming. The number should have killed the project. It didn't, because the argument for Heathrow's expansion was never really about GDP. It was about who controls the only hub-scale slot-constrained airport in the country, a piece of infrastructure so structurally irreplaceable that its owners can set terms with airlines, governments, and regulators for decades at a time. The £62.5 billion in trade-offs the DfT quietly buried in the same document tells you what building the runway costs. The silence about who captures the upside tells you everything about why the money keeps arriving anyway. This is what infrastructure investment looks like now: the returns are not in the economics. They are in the position.
The buyers have changed. That is the thing to understand before anything else. A decade ago, infrastructure assets attracted pension funds looking for inflation-linked yield, patient capital content to clip coupons on a water treatment plant or a toll road. What is arriving now is categorically different: Brookfield Asset Management, running just over $1 trillion in assets under management, has spent the last three years acquiring physical networks from Australian data centres to German port logistics. Abu Dhabi Investment Authority took a stake in a European fibre rollout at a multiple that would have been laughed out of a 2015 investment committee. The Gulf sovereign funds that once bought Mayfair hotels are now buying subsea cable landing stations. The mechanism is straightforward, even if the scale is not. Physical infrastructure, once built, has no meaningful substitute. You cannot disrupt a runway. You cannot app your way around a fibre duct running under a city street. The asset does not depreciate in the way a factory does, because the scarcity it embeds is structural rather than technological. What the new buyers have worked out, and what regulators are only beginning to notice, is that owning the physical layer of an economy is not a return-on-capital story. It is a toll story. Every airline that uses Heathrow, every ISP that leases capacity on a dominant fibre spine, every logistics company that moves through a concentrated port, is paying rent to whoever controls the bottleneck. The rent is permanent. The leverage compounds. , - The Heathrow situation is the clearest domestic example, partly because the numbers are now public. The airport is 40% owned by a consortium that includes Ferrovial, the Qatar Investment Authority, and funds managed by sovereign and quasi-sovereign vehicles out of Singapore, Canada, and China. When ministers accelerate expansion planning in the name of growth, they are accelerating the construction of a larger toll booth for a set of owners with no particular attachment to British industrial policy. The DfT's own modelling now suggests the GDP uplift is negligible. The owners' return, derived from landing fees, retail concessions, and regulated asset base expansion, is not. Heathrow's regulated asset base was valued at approximately £17 billion before the runway debate resumed. A third runway adds physical capacity and, more importantly, regulatory justification for expanding that base further. The GDP argument was always cover for a real estate transaction. This is not unique to aviation. In telecoms, XpFibre, the French full-fibre operator backed by Goldman Sachs and Altice, has been acquiring regional network operators across Europe at multiples that imply buyers expect a duopoly or monopoly outcome within a decade. The logic is simple: once fibre passes a home, the economics of a second network passing the same home are prohibitive. Whoever gets there first owns the connection. Ofcom in the UK has tried to manage this through open access obligations and wholesale price regulation, but the regulatory frame lags the consolidation by several years. By the time the rules catch up, the physical layer is already owned. , - In mining, First Quantum's copper assets in Panama illustrate a harder version of the same dynamic. The Cobre Panama mine, before the government forced its closure in 2023 following public protests, was producing roughly 350,000 tonnes of copper per year, making it one of the largest single copper sources in the world. Copper is not a commodity in the ordinary sense any longer: it is the physical substrate of electrification, and the known high-grade deposits that can be developed at scale are geographically concentrated and politically contested. Whoever controls those deposits controls a chokepoint in the energy transition supply chain. First Quantum's dispute with Panama was superficially about environmental permits. Underneath, it was about who gets to extract leverage from a finite physical asset that the global economy has committed to needing in vast quantities. The Panamanian government won the political argument. The economic argument, about where copper processing capacity and royalty flows end up, is still unresolved. The pattern across all of these cases is the same: the asset being contested is irreplaceable at the relevant scale, the buyer or owner is sovereign-adjacent or patient enough to outlast regulatory cycles, and the return is not priced in conventional cash-flow terms but in positional control. Brookfield does not buy Australian port logistics because ports have strong free cash flow. It buys them because ports are the constraint, and constraints have pricing power that is essentially permanent. , - The secondary effect is regulatory arbitrage, and it is already well underway. Sovereign and quasi-sovereign buyers are structurally harder to regulate than private ones. Governments are reluctant to compulsorily purchase or cap returns on assets owned by friendly foreign state funds, for reasons that have more to do with diplomatic relationships than economics. The Competition and Markets Authority can block a domestic merger. It has considerably less appetite for confronting a Qatari sovereign fund over its slice of Heathrow's regulated asset base. The Financial Conduct Authority can scrutinise a British fund manager. It has limited jurisdiction over the Cayman-domiciled vehicle that owns a British fibre network through three holding layers. This is the structural advantage that has gone largely unexamined. Private equity buying infrastructure is at least nominally subject to a full range of domestic regulation. Sovereign and sovereign-adjacent capital, routed through standard international structures, operates in a softer regulatory environment even when the underlying asset is a national utility. The result, over the course of a decade, is that the physical layer of the economy migrates into structures that governments find awkward to challenge directly. By the time political will forms, the legal question of whether the state can act, and at what cost, is substantially more complicated. Briefed Intelligence data recorded UK discount-seeking search behaviour at 93.3 through mid-June 2026, a level flagged as elevated. That number is a consumer signal, but it is also an infrastructure signal: when households are stretched, they compress spending on everything except the things they have no choice about. Broadband. Energy. Transport links. The inelastic services sitting behind physical monopolies. Whoever owns the pipes is, in that environment, the last landlord standing. , - The investor calculus has already shifted. The question worth asking now is whether the regulatory response can shift fast enough to matter. The UK government's Infrastructure Bank, capitalised at £22 billion, was designed in part to ensure public interest sits alongside private capital in strategic assets. In practice, it has mostly co-invested in projects where private capital was already comfortable. The harder challenge is the assets where private and sovereign capital has already consolidated, where the regulatory asset base model gives owners a legally protected return, and where the cost of unwinding the ownership structure exceeds any plausible political will to try. For operators and investors, the read is this: the premium on physical irreplaceability is not going to compress in a rising-rate environment. It widened because of low rates, and it has stayed wide because the strategic logic is now the dominant pricing input, not the yield. Assets that sit at genuine physical chokepoints, whether port capacity, slot-constrained airports, or dominant fibre spines, are being priced as perpetual options on toll collection, and the buyers are right to price them that way. The regulatory risk is real but slow. The monopoly compounds in the meantime. What would make this wrong: if governments in multiple jurisdictions moved simultaneously to cap regulated returns and impose structural separation between infrastructure ownership and service provision, the toll model breaks. Australia did something close to this with its telecommunications network through the NBN Co structure. Britain gestured at it with Openreach's partial separation from BT. Neither went far enough to dislodge the underlying dynamic. Until a government is willing to pay the political cost of genuine compulsory restructuring, the boring monopoly compounds quietly, one planning document at a time.