A thirty-year mandate underwrites two things. It underwrites that the cash flows arrive — and it assumes, silently, in every model, that the physical world those cash flows come out of still works the way it did when the spreadsheet was built.
Investment committees spend months on the first assumption. The second one rarely reaches the page.
what long-term investing actually is
Long-term investing is the practice of allocating capital on a multi-year to multi-decade horizon, accepting illiquidity and interim volatility in exchange for returns that shorter money cannot reach. When the OECD worked with the G20 on why institutional money struggles to fund long-lived assets, it gave that practice three qualities. Their words, not ours:
| quality | the OECD's description |
|---|---|
| patient capital | "allowing investors to access illiquidity premia, lowers turnover, encourages less pro-cyclical investment strategies and therefore higher net investment rate of returns and greater financial stability" |
| productive capital | "providing support for infrastructure development, green growth initiatives, SME finance etc., leading to sustainable growth" |
| engaged capital | "which encourages active voting policies, leading to better corporate governance" |
That is a good definition, and the G20 endorsed the principles built on it in 2013. Read it once more and notice its shape. All three qualities describe the behavior of the capital — how long it waits, what it finances, whether it votes. None of them describes the condition of the physical systems the productive assets sit inside.
That is not a flaw in the OECD's work. It is the boundary of the discipline. Long-term investing has a precise vocabulary for how patiently money should behave, and almost none for whether the world it is being patient about is holding still.
the one input nobody stress-tests
A serious long-horizon book gets tested hard, and then it gets tested again by a regulator.
| what a long-horizon book stress-tests | what it quietly holds constant |
|---|---|
| rate paths, curve shape, inflation | rainfall, and where it falls |
| credit spreads, default correlation | soil that holds water and grows food |
| longevity and mortality tables | rivers arriving roughly on schedule |
| currency, sovereign, and policy risk | a climate the actuarial past still describes |
| liquidity under stress | pollination, fire behavior, aquifer levels |
The left column gets hundreds of pages, a scenario deck, and someone senior asking hard questions. The right column gets treated as a constant, inherited from a century in which it more or less was one.
Three findings, from sources a fiduciary already accepts, say that century has ended.
The Dasgupta Review — commissioned by HM Treasury, the first full economics-of-biodiversity review sponsored by a finance ministry — found that between 1992 and 2014 produced capital per person roughly doubled and human capital per person rose about 13%, while the stock of natural capital per person declined by nearly 40%. Its framing claim is blunter than most investment memos: economies are embedded within nature, not external to it.
Swiss Re Institute, building an index for its own underwriting rather than for a brochure, found 55% of global GDP — roughly $42 trillion — dependent on high-functioning biodiversity and ecosystem services. That is an insurer measuring exposure, not an NGO making an appeal.
The Global Commission on the Economics of Water reported in 2024 that the hydrological cycle is out of balance for the first time in human history, with more than half of global food production at risk by 2050 on the current path.
Every long-duration valuation contains one input nobody is ever asked to defend: that the biosphere will keep behaving the way it did during the sample period.
That is the term this post is about. Not the discount rate. The thing the discount rate is applied to.
short-termism is the same error at a shorter tenor
John Kay's 2012 review of UK equity markets reached a conclusion worth quoting exactly: "short-termism is a problem in UK equity markets, and that the principal causes are the decline of trust and the misalignment of incentives throughout the equity investment chain." Not villains — architecture. A culture of transactions and trading had displaced relationships of stewardship, all the way along the chain from saver to company.
The diagnosis still holds, and it explains something otherwise strange about the institutions with the longest liabilities on Earth: they frequently behave like the shortest money in the market. Quarterly relative performance. Career risk measured in one-year windows. Funding rules that mark decades-long liabilities to today's curve and force action in a stress. And a thin supply of genuinely long claims to buy even when the mandate wants them. The measurement mechanics, and the duration math underneath, are worked through in match your liabilities to living systems — this post will not redo them.
Put the two failures side by side and they are one failure at two tenors. Short-termism discounts the next decade because the scoreboard resets in ninety days. The constant-biosphere assumption discounts the next century because the model has no cell for it.
a thirty-year book on a failing water cycle is a story
Take two portfolios. On paper they are the same: same tenor, same target return, same sponsor, same underwriting file.
The first holds farmland, a hydro concession, municipal water revenue bonds, timberland, insured coastal property, and a data center campus. Every one of those positions is priced off a service — water arriving, soil holding, fire staying survivable, heat staying manageable — that no one has contracted to deliver and no one is being paid to maintain.
The second holds the same assets, and somebody, somewhere in the structure, is paying to keep the addressable parts of those systems in working condition — conversion, drainage, fuel load, diversion, stocking rate, the soils and forests the water and snow land on.
That second book is not weather-proof. Funding condition does not make snow, and it is not a climate-term hedge. It is the part of the position a contract can actually reach. The first book treats even that part as a free input.
You can hold an asset for thirty years. You cannot hold it against a basin that stops delivering, because there is no counterparty to call. Nothing in a capital stack replaces a snowpack. A hold can still fund the ground the snowpack lands on.
This is the gap in the market, and it is a plumbing problem rather than a values problem. Insurance pays after the loss — that is its business model. Carbon and biodiversity credits price a counterfactual, arguing about what would have happened otherwise. Neither instrument pays for a functioning basin now, which is exactly the thing a long book is silently relying on.
the part of the position that can actually be funded
Ensurance is the mechanism for that missing line item: funding the protection, restoration, and stewardship of natural assets before loss, rather than compensating after it. The natural asset — the basin, the wetland, the working forest — is the thing that matters. Ensurance is how it gets paid for.
Two instruments, in plain terms:
- general ensurance — protocol-wide coins. Indirect funding: market activity across the whole system routes proceeds into protection. Early and thin — a way to observe proceeds, not a position you can size a mandate against.
- specific ensurance — certificates tied one-to-one to a named agent and a named place. Direct funding, with condition and proceeds attached to that place rather than to an average.
Three features matter to a long-horizon allocator specifically. First, the payment is present-tense: it funds a system that is generating ecosystem services now, so there is no counterfactual baseline to defend and no loss event required. Second, condition is the credit: ecosystem service value against cost, plus measured condition, is the analog of a rating — the thing an underwriter reads. Third, the claim is designed to end. A hold can run to an entrust pathway in which the place is permanently protected and the financial claim is retired, rather than refinanced forever. Sovereign debt rolls. This finishes.
the honest limits
Three things this is not, stated before your investment committee says them for you.
It is not a gilt substitute available today. A duration-clear coupon on pure protection still has to be manufactured: an identified payor, a payment schedule that produces a stated duration, a residual position agreed in writing. Working nature — timber, farmland, operating water — already competes for institutional dollars because harvest and operations generate the cash. Pure protection does not yet do that at scale. That gap is the work, and anyone describing the instrument as finished is selling you the story again.
It is not a low-volatility claim. What is on offer is duration and the purpose of capital, not a flatter line on a chart. Assets that look smooth because they are marked infrequently are delayed information, not diversification.
It is not a valuation of nature. Price is a bridge — it exists so capital can act on a living system's behalf. It is not a claim that the dollar figure is what a basin is worth.
where this fits in a mandate
Three doors, by the question you are actually holding.
- "Who can hold this, and for how long?" — the money that can wait is the hub: buyer profiles, ecological duration, and the gap between long liabilities and short behavior.
- "Where does it sit on my curve?" — match your liabilities to living systems has the duration math, the key-rate shape, and the honest constraint.
- "This is family or endowment money, not a hedge." — the forever asset covers perpetual capital that never needs a harvest.
- "Is this even an asset class?" — is nature an asset class yet? gives the honest answer: demand is real, supply is the hard part.
If you would rather see the live instruments than read another framework, start with general ensurance and follow the proceeds. If you have a mandate with a real horizon and want to know whether any of this is underwritable inside it, talk to someone who can answer in your terms — coupon, payor, duration, residual, evidence.
frequently asked questions
what is long-term investing?
Long-term investing is allocating capital on a multi-year to multi-decade horizon, accepting illiquidity and interim volatility in exchange for returns and stability that shorter money cannot access. The OECD's working definition gives it three qualities: patient capital that lowers turnover and harvests illiquidity premia, productive capital that finances real long-lived capacity such as infrastructure, and engaged capital that exercises stewardship instead of exiting on a bad quarter.
is long-term investing the same as longtermism?
No. Long-term investing is a mandate question: can the book wait, and what physical world is it waiting on? Longtermism is a moral argument about how much future people count — associated with William MacAskill and Toby Ord, and already covered in longtermism still needs a living present. Same time-word. Different job. This post stays on the mandate.
why does long-term investing depend on nature?
Because ecological services are inputs to nearly every long-dated cash flow, and they are not contracted. Farmland needs rain and soil; hydro needs snowmelt timing; property insurance needs survivable fire and flood behavior; municipal water revenue needs a functioning basin; data centers need water and manageable heat. Swiss Re Institute found 55% of global GDP dependent on high-functioning biodiversity and ecosystem services, and the Dasgupta Review found natural capital per person down nearly 40% between 1992 and 2014 while produced capital doubled. A thirty-year model that treats those inputs as fixed is not a conservative model — it is an unpriced short position on the physical world.
None of this is investment, tax, or legal advice.
the series
Five posts on the same fault line: long horizons that assume a static planet.
- longtermism still needs a living present — the philosophical pillar, for readers who arrived through effective altruism
- long-term investing still needs a living present — you are here
- short-termism is why long money still acts short — in draft
- future generations inherit a living system or they inherit a story — in draft
- a long-horizon book still needs a living clock — in draft
For duration mechanics, buyer profiles, and the instruments themselves, the patient-capital hub is the money that can wait.