Infrastructure Funds

Infrastructure investing buys the things a country runs on. Toll roads and airports. Power grids and water networks. Mobile towers, data centers and pipelines. Not shares in the company that owns them — the assets themselves.

What makes them unusual is how safe the income is. Most of these assets are the only one of their kind in the area, so there is nowhere else for customers to go. People cannot easily stop using water, power or roads. And the money that comes in is usually fixed by a contract or by a regulator for the next 20 to 50 years, often with a clause that raises prices in line with inflation.

Pension funds and insurers need income like that, because they have to pay pensions and claims decades from now. They rarely buy an airport directly. Instead they put money into an infrastructure fund: a pool run by a manager — Macquarie, Brookfield, GIP and KKR are the big names — that buys the assets, runs them, improves them and collects the cash. The fund is built much like a private equity fund, only it usually lives longer, and some never close at all.

The idea started with Australian pension funds in the 1990s. It is now worth several trillion dollars, and three things are driving it hard right now. Countries are rebuilding how they make and move electricity. Everything is going digital, which needs towers, fibre and data centers — the AI buildout is really an infrastructure story. And governments would rather private money paid for all of it than their own budgets.

Why the profile is unusual

Stated in one line: cash-flow stability that reads like a bond, ownership that behaves like equity, and an inflation link written into the contract. It is the closest private markets come to an inflation-linked perpetual bond with an operating upside — and the risks are correspondingly un-bond-like.

The risk spectrum

• Core: operating assets, contracted or regulated revenue (a regulated water utility, a fully leased tower portfolio). Target: 7–10% net, mostly cash yield. The "bond-plus" end.
• Core-plus: mostly contracted with some market exposure or growth capex (an airport with retail revenue).
• Value-add / opportunistic: development, construction, repowering, platform build-ups (build the data-center campus, then sell it as core). Target: 12–18%, mostly appreciation — private equity in a hard hat.

Revenue models — the analysis that matters

An availability-payment hospital earns the same whether full or empty; a regulated grid earns an allowed return on its asset base (reset every few years by the regulator); a contracted pipeline earns what the take-or-pay says; a toll road or merchant power plant earns whatever traffic and markets deliver. Two assets in the same fund can sit at opposite ends — "infrastructure" is a label, the revenue model is the risk.

Leverage and structure

Stable cash flows carry serious debt — 50–80% loan-to-value at the asset level, typically non-recourse project finance: each asset's debt is trapped at that asset, so one failure doesn't sink the fund. Returns decompose as levered cash yield plus growth plus (for value-add) the de-risking re-rating from build to core.

Valuation: DCF in a discount-rate cage

Private infra marks come from long-horizon DCF at appraisal-set discount rates — which move slower and less than markets. The 2022 rate shock exposed the machinery: listed infrastructure de-rated 20–30% while private marks barely moved, reviving the "volatility laundering" debate from private equity with higher stakes (longer duration = more rate sensitivity, not less). The honest analysis re-underwrites: terminal-value assumptions, the equity discount rate versus current risk-free plus spread, and — for regulated assets — the gap between the allowed return and the market cost of capital, which regulators eventually close in the investor's disfavour.

The risks with teeth

• Regulatory reset: allowed returns follow rates with a lag both ways; UK water (Thames Water's spiral) shows the full failure mode — leverage loaded against a regulatory bargain that tightened, ending in quasi-renationalisation talk and equity marked toward zero.
• Political/contract risk: retroactive renegotiations are a genre — Spain's solar-tariff cuts (2013), Australian toll-road bankruptcies (traffic forecasts, not politics), airport concessions through Covid: "essential" did not mean "immune", it meant merchant exposure was hiding in the retail and volume lines.
• Transition-asset stranding: gas networks and pipelines price a terminal-value question the DCF must answer explicitly — infinite-horizon assumptions on finite-horizon assets are where infra marks hide their optimism.

Structure evolution and the listed mirror

The asset class is migrating from 10-year closed funds toward open-end core vehicles (matching perpetual assets with perpetual capital, at the price of NAV-based entry/exit and queue mechanics — the semi-liquid tensions private credit knows) and super-core yield vehicles. Listed infrastructure (utilities, tower REITs, listed funds) offers the same assets with daily pricing — the private/listed valuation gap is the asset class's own version of the CEF discount, arbitraged slowly by take-privates whenever it widens far enough (the 2022–24 listed-infra privatisation wave, KKR/GIP buying listed vehicles, priced it in public).

The formulas above are standard textbook formulations, simplified for teaching. They explain the mechanism — they are not a valuation tool, and they will not reproduce a dealer’s price.