Patient Capital, Public Risk

At the Financial Times' Global Insurance Summit in late June, Amanda Blanc, chief executive of Aviva, made an offer that sounded like generosity. Insurers, she said, hold patient, long-term capital and stand ready to pour it into Britain's new towns, its social housing, its schools, hospitals and high streets — to "fall in behind" a government willing to commit. Build the pipeline, structure the projects appropriately, and "investment from private capital will follow." She called it a flywheel of opportunity.

It is an attractive picture, and the honest place to begin is by conceding how much of it is true. Britain genuinely needs houses and infrastructure. Long-dated, patient institutional money genuinely is the natural holder of long-dated assets — a pension annuity owes its members for thirty or forty years, and an asset that pays out slowly over decades is a better match for that than the quarterly impatience of public markets. Aligning the two is, in principle, good policy. Anyone who pretends otherwise is not arguing in good faith.

But an offer should be read for what it actually asks, not for how it is dressed, and Blanc's asks — taken precisely — describe something rather different from a benefactor with money to spare. They describe an industry asking the state to manufacture a margin for it, and to carry the risk while it does.

The capital is not lying around

Start with the phrase doing the most work: "patient capital ready to invest." It conjures an image of war chests — great reservoirs of insurer cash searching for a home. There are no such reservoirs. An annuity writer is, by construction, fully invested at all times. The moment it takes a pension scheme's premium it must put that money to work earning a return, because it owes those pensions for decades and idle cash is a liability, not a luxury. There is no vault.

So what is the "capital ready to invest"? Two things. The first is the ordinary forward flow of money the insurer is constantly reinvesting as old assets mature and new business arrives — money it is going to deploy somewhere regardless. The second, and more telling, is the capital that the Solvency UK reforms released by thinning the buffer held behind the annuities. The headline "£100bn of patient capital for Britain" was never spare wealth. It was, in large part, the safety margin that used to stand behind your pension, reclassified as investable when the rules decided less needed to be held in reserve. "We have capital ready to support Britain" means, decoded, "you let us hold a thinner cushion, and we would like to put the difference to work." That is not a gift looking for a cause. It is redirection, on terms.

The margin is made by the state, not the market

Why, then, must insurers lobby for these projects? If new towns and infrastructure were simply good investments, the money could buy them today. The answer reveals the real mechanism. The return an insurer cares about is not the headline yield but the margin after regulatory treatment — yield relative to the capital it must hold and the accounting benefit it can claim. And that margin is not a property of the asset. It is conferred by how the asset is structured, classified and backstopped.

An infrastructure loan pays a superior margin to an annuity writer only if its cashflows are predictable enough to qualify for the matching adjustment — the mechanism that lets the insurer book decades of expected spread as value today and hold less capital against it — and only if an implicit state backstop lets it be treated as low-risk. Strip away that treatment and the same asset reverts to ordinary illiquid credit: higher capital charge, no early recognition, something the insurer could already buy and largely chooses not to.

So when Blanc says government must "create the projects," she is not only asking for a pipeline to exist. She is asking for the projects to be built to qualify — long-dated, contracted, predictable, quasi-sovereign — so that ordinary yield becomes extraordinary margin. The state is not being asked to provide opportunities. It is being asked to provide the treatment that makes the opportunities pay better than what is already on the shelf.

The assumption that carries everything

All of which rests on a single load-bearing assumption: that the projects get delivered — built, profitable, and on budget. Predictable cashflows are the foundation of the whole edifice. They are what make the assets matching-adjustment-eligible, lightly capitalised, and worth the premium. Remove delivery-to-plan and the investment case does not wobble; it collapses.

And delivery-to-plan is precisely what Britain has most reliably failed to achieve. HS2 ran from an early estimate of around £33bn to well over £100bn before its northern leg was cancelled outright — several times the money for a fraction of the railway. Crossrail arrived years late and billions over. Hinkley Point C slips in cost and timetable with grim regularity. This is not misfortune; it is the pattern. The scholar Bent Flyvbjerg, who has assembled the largest database of major projects in the world, calls it the iron law of megaprojects: over budget, over time, under benefit, over and over again. "Predictable cashflows" is the one thing the country's record most dependably refuses to supply.

Here patient capital meets its limit, and it is worth being exact about where. Holding an asset to maturity protects an investor against a liquidity shock — the temporary, panic-driven markdown that reverses if you simply wait. It offers no protection whatever against an asset that fundamentally underperforms. You cannot hold a half-built new town to maturity and have it heal, any more than you can a defaulted bond. The long horizon the industry waves as its great strength is irrelevant to delivery risk. Patience cures the wrong disease.

Manufactured value, on the sovereign balance sheet

Now the cleverest ask, and the one that should detain us longest. Blanc — and Aviva, Legal & General and the USS pension fund, who wrote jointly to the Chancellor on this earlier in the year — want development-corporation debt treated the way it is in parts of Europe: as the debt of a separate, commercial, long-term body, and therefore not counted as government debt. The appeal is obvious. It lets the state build without the borrowing showing up against its own fiscal rules.

Strip the framing and what is requested is that a category of debt be reclassified so that it no longer scores. The economics do not change. The town costs what it costs; the money is still borrowed; the state's involvement is still real. Only the recognition moves. This is the matching adjustment applied to public finance — fiscal headroom manufactured not by reducing a liability but by relocating where it is recorded.

There is a legitimate version of this. Under the statistical rules, a genuinely commercial public corporation that covers its costs from its own revenues and bears its own risk can properly sit outside government debt. That is economic substance, not a trick. But the legitimacy turns entirely on whether the risk has truly left the state — and for a politically backstopped new town, it has not. No government allows a flagship new town's development corporation to default; the political cost is unpayable. The implicit guarantee is real whether or not it is written down. So to classify that debt as non-government is to recognise a smaller liability than the one that actually exists — to manufacture fiscal space by declining to count a contingent obligation that is plainly there.

Whoever is holding it when the overrun lands

Put the two together — projects that reliably overrun, and debt structured to sit off the public books — and the destination of the risk becomes clear. When, not if, a project runs over, the loss has only two places to go. Either the state steps in to absorb it, in which case the "non-government" classification was a fiction all along and the liability crystallises onto the public balance sheet at the worst possible moment; or the state does not, and the loss falls on the investors — the annuity portfolios, and behind them the pensioners whose security was the entire justification for the exercise. There is no third door.

This is the shape of the thing once the dressing is removed. The day-one margin is booked privately, on an optimistic assumption of delivery. The downside, when the assumption fails, is exported — not offshore to a Bermudan reinsurer this time, but to a fiscal footnote, to land on the taxpayer or the pensioner. Manufactured value, exported risk, now on the sovereign balance sheet.

And note the final, quietest ask. Blanc's plea for "stability and consistency," her warning against "lurches in any direction," delivered into a moment of political flux, is patient capital asking the state to pre-commit to the conditions that keep its returns safe. Reasonable from an investor's chair — you cannot underwrite thirty-year money into uncertainty. But read structurally it is the beneficiary asking government to guarantee the environment in which its margin is earned: the soft form of capture, not the buying of a rule but the securing of the stability the rule depends upon.

The honest charge

None of this is an argument that insurers should not fund infrastructure, or that the match between long-dated money and long-dated assets is bogus. It is genuine, and a country starved of investment should welcome it. The objection is narrower, and therefore harder to wave away.

It is this: the favourable treatment is being claimed on an assumption of deliverability that Britain's own record flatly contradicts; the margin that treatment confers is manufactured by the state rather than earned in the market; and the structure proposed assigns the consequences of failure to parties — taxpayers, pensioners — who are not the ones booking the day-one return. If the public is to stand behind the risk, the public should capture the upside, not merely underwrite the downside while a private margin is taken off the top. "Patient capital, ready to support Britain" is a fine slogan. The version that survives the small print reads: patient with the upside, impatient to socialise the loss.

Build the new towns, by all means. But build them on an honest accounting of who is bearing the risk — and on a sober memory of how rarely, in this country, the diggers stop on budget.