---
title: what the tragedy of the horizon costs a universal owner
canonical_url: https://ensurance.app/guide/what-the-tragedy-of-the-horizon-costs-a-universal-owner
markdown_url: https://ensurance.app/guide/what-the-tragedy-of-the-horizon-costs-a-universal-owner.md
subtitle: the mandate is fifty years. the scorecard is three
category: nature-finance
---

# what the tragedy of the horizon costs a universal owner

*the mandate is fifty years. the scorecard is three*

The **tragedy of the horizon** is the gap between when the costs of climate change arrive and when the people who could act are measured. Mark Carney named it in a speech at Lloyd's of London on 29 September 2015: the catastrophic impacts fall beyond the horizon of most actors, including central banks, and by the time they define financial stability it may be too late to act.

For a universal owner, the tragedy is internal. Under every holding in its book are living systems: the watersheds, soils, forests, pollinators, wetlands and coasts that make the conditions companies earn in. They exist whether or not anyone holds a certificate. They change over decades. Almost everything the owner uses to judge its own work runs on a shorter clock. [**Ensurance**](/?from=guide) is how an owner can fund one of those systems, at a named place, now. It is not the place.

:::johnson
**the mandate is fifty years. the scorecard is three.** — A universal owner promises income across generations and is graded in windows far shorter than the systems it depends on. The living system keeps its own clock.

[see the flows a place produces each year →](/natural-capital/flows?from=guide)
:::

## what carney actually said

Carney's point was not that finance is careless. It was that each actor is doing its job on its own timetable, and the timetables do not reach far enough. Business cycles, political cycles and the mandates of authorities such as central banks all run shorter than the period over which climate costs build. Nobody in the chain is paid to price what lands after their window closes.

The speech was about climate. The same structure holds for the other living systems under the market. The figures below are about climate; for soil and water the pattern is the mechanism, not a measured cost. In many places soils lose organic matter and aquifers draw down over decades, long before a well runs dry. A forest that burns this summer can stop holding its slope for years. These are slow costs with sudden edges, and a scorecard built for quarters sees the edge, not the slope that led to it.

## we are long-term already

This is the objection, and it deserves its full weight.

A universal owner is the most patient capital in the market. A pension fund pays members who have not yet been born. A sovereign fund like NBIM holds wealth for future generations by design, and in 2026 it set eight core nature expectations for the companies it owns, from board oversight to targets. ABP names biodiversity loss among its systemic risks. Railpen launched a 2026–2030 systems-stewardship strategy as a universal owner and added nature to it. The pattern is wider than these funds: smaller and public schemes in other regions may face the same chain of clocks.

All of that is real. If any investor can see past the horizon, it is this one.

The trouble is not the owner's intent. It is the machinery the intent runs through.

## the horizon is inside the agency chain

A universal owner does not act as one mind. It acts through a chain, and each link is measured on its own clock.

| link | answers to | measured against |
|---|---|---|
| Members and beneficiaries | A lifetime of income | Whether the money is there when it is owed |
| Trustees or board | Members, regulators, sponsors | Funding level, contributions, peer comparison |
| CIO and the internal team | The board | The policy benchmark, reviewed on rolling windows |
| External managers | The CIO | Return relative to a market index, against peers |
| Portfolio companies | Shareholders | Earnings, guidance, the share price |
| The living system | No one | Its own condition, over years to decades |

Every row is doing what it was asked to do. A manager who beats its benchmark has served the client; a CIO close to policy has served the board. Nobody in the chain is careless, and nobody in it is assigned the last row.

That is the tragedy in its owner-sized form. The fund's promise runs for fifty years. Its review cycle runs in quarters and years. The conditions that decide whether the fifty-year promise holds sit below every review, because they are not in any benchmark.

## why a long mandate still misses the living system

For most listed holdings the benchmark is the market, and many managers are judged against it or against peers. A manager's job is to beat peers who hold roughly the same market. If the conditions under the whole market erode, every manager in the peer group loses together, and relative performance does not move. The scorecard says everyone did fine. The beneficiary is poorer. Schemes that measure against liabilities face a sibling problem: the liability clock is long, but the review windows are not.

A fifty-year owner that grades its managers on relative return is, in effect, the owner of a short judgment repeated. Each window is sound. The sum of the windows never looks down.

One modeled case suggests the slow cost can reach an owner's own numbers. In a UK defined-benefit case study for the Institute and Faculty of Actuaries, Ortec Finance and Cambridge Econometrics found that climate pathways delayed the scheme's full funding by three to nine years, with no change in its strategy. That is a model of one scheme, not a forecast for every fund. It shows the shape: the horizon cost arrives as years added to the time it takes to keep a promise, and no single quarter shows it.

Insurers show a sharper edge: they reprice on short contracts, so a slow cost can turn up at renewal. Not every loss below is attributable to climate or nature loss, and State Farm's decision was its own. State Farm stopped taking new California property applications from 27 May 2023, citing catastrophe exposure. Swiss Re Institute put 2025 natural catastrophe losses at about US$220 billion economic and US$107 billion insured, according to its *sigma* report of 2026. A universal owner holds the insurers, the homebuilders, the lenders and the utilities in those places. When cover leaves, the cost does not leave with it. [The protection gap is a condition gap](/guide/the-protection-gap-is-a-condition-gap?from=guide) covers what happens to the uninsured part.

The owner's tools have not failed. Disclosure, engagement and voting have moved companies; [a disclosure is not a transaction](/guide/a-disclosure-is-not-a-transaction?from=guide) sets out what disclosure does well. The claim is narrow: those tools work on companies, on the company's clock. Water funds, easements and land trusts already fund named places, and they work. They sit outside the benchmarked book.

## a hold that runs on the living system's clock

If every scorecard is shorter than the system, one answer is a position whose clock is the system's own: not a longer review cycle, but a different thing being reviewed.

That is the shape ensurance is built around. An account is addressed to a named place or purpose: a watershed, a forest, a stretch of coast. A certificate against that account is something an owner can hold under its own rules. Proceeds fund the parcel first, the land, water, species and flows on and from it, then the people who keep it. And the condition of the place is read every period, now, from what it is doing this season. The read does not ask what would have happened without the money. It asks whether the place is working.

A manager's report says how the book did against the market. A place's report says how the ground under the market did against its own last reading. The second is slow, and honest about it.

The place also has an end state, which most horizon problems lack. Each account moves through stages:

| stage | what proceeds fund | when it ends |
|---|---|---|
| Unensured | Nothing yet; the parcel carries cost, debt, claims and risk | When funding starts |
| Ensured | The parcel's capital cost: acquisition, debt retirement, clearing claims, the restoration that gets it secure | When the place is debt-free, claim-free and no longer at risk |
| Entrust | Ongoing stewardship: the people and work that keep the parcel, its species and its flows | It does not need to end |

The ensured stage is the part with a finish line. Once the place is secure, its cost is out of the picture and what arrives goes to care. [A long-horizon book still needs a living clock](/guide/a-long-horizon-book-still-needs-a-living-clock?from=guide) works through how to match a long mandate to a place that can last; this post does not repeat it.

A certificate is not a deed and not a claim on the land. It does not lengthen a manager's mandate or change how a fund is benchmarked. It gives the owner one funding commitment, beside its stewardship, that is judged on the condition of a place rather than on a window. [Risk Resilience](/risk-resilience.ensurance?from=guide) and [Water Abundance](/water-abundance.ensurance?from=guide) are two live accounts that show the shape.

It is early. On 24 October 2026 the protocol held 2,085 active accounts, 86 with a wallet deployed, 26 policies and 191 coins, with seven-day volume of about US$8. That is not a size a universal owner allocates to. It is a working clock at a named place.

For a fiduciary reader the case is pecuniary: beneficiaries are paid over decades, and the conditions under the market are part of the risk to their money over that span. Whether funding a named place is prudent is for trustees and counsel to decide. This post is not investment or legal advice.

## frequently asked questions

### what is the tragedy of the horizon?

It is the mismatch between when climate and nature costs arrive and when the people who could act are measured. Mark Carney used the phrase at Lloyd's of London on 29 September 2015: the worst impacts fall beyond the horizon of most actors, so by the time they are urgent, it may be too late to act.

### who coined the tragedy of the horizon?

Mark Carney, then Governor of the Bank of England, used the phrase in that 2015 speech at Lloyd's of London.

### aren't pension funds already long-term investors?

Their liabilities are long-term and many of their beliefs are too. Their machinery is shorter. Trustees, the CIO and external managers are each measured on their own windows, and most of those windows judge return against the market. When the conditions under the whole market erode, relative performance does not show it.

### how does the tragedy of the horizon affect a pension fund?

Slowly, then in numbers. In one UK defined-benefit case study for the Institute and Faculty of Actuaries, climate pathways delayed full funding by three to nine years with no change in strategy. It is a model of one scheme, but it shows the cost arriving as time added to the promise.

### does ensurance solve short-termism?

No. It does not change how funds benchmark managers or how boards review them. It offers one position whose clock is a living system's: an account addressed to a named place, with condition read every period, proceeds that fund the parcel first, and a public ledger. It is small and early.

## next

The next post asks why owners who see the same risk still act one at a time: [why universal owners act alone and lose together](/guide/why-universal-owners-act-alone-and-lose-together?from=guide).

If your organization already funds places on a generational clock, [foundations](/solutions/foundations?from=guide&topic=universal-owners) shows how we work with endowed capital on named places. If you set the rules a fund's horizon runs on, [governments](/solutions/governments?from=guide&topic=universal-owners) is the page for you. If you can name the place your book leans on, [tell us which one](/contact?from=guide&topic=universal-owners).

## sources

[Carney, M. 2015. *Breaking the tragedy of the horizon: climate change and financial stability*. Speech at Lloyd's of London, 29 September 2015. Bank of England.](https://www.bankofengland.co.uk/speech/2015/breaking-the-tragedy-of-the-horizon-climate-change-and-financial-stability) — The definition of the tragedy of the horizon and the horizons of business, political and technocratic actors.

Ortec Finance & Cambridge Econometrics, for the Institute and Faculty of Actuaries. UK defined-benefit pension scheme climate case study (2020/2022). — Climate pathways delayed full funding by three to nine years with no change in strategy.

Swiss Re Institute. 2026. *sigma* 1/2026, natural catastrophes in 2025. — About US$220 billion of economic and US$107 billion of insured natural catastrophe losses in 2025.

State Farm General statement on California new business, May 2023, as reported by CBS News. — New California property applications no longer accepted from 27 May 2023, citing catastrophe exposure.

[Norges Bank Investment Management](https://www.nbim.no/) — Eight core nature expectations of portfolio companies.

ABP, nature and biodiversity policy. — Biodiversity loss named among ABP's systemic risks.

Railpen. 2026. Systems stewardship strategy 2026–2030. — Railpen's universal-owner stewardship strategy, adding nature.

## the series

1. [what is a universal owner](/guide/what-is-a-universal-owner?from=guide)
2. [the risk a universal owner cannot diversify away](/guide/the-risk-a-universal-owner-cannot-diversify-away?from=guide)
3. what the tragedy of the horizon costs a universal owner (this post)
4. [why universal owners act alone and lose together](/guide/why-universal-owners-act-alone-and-lose-together?from=guide)
5. [fiduciary duty and esg for a universal owner](/guide/fiduciary-duty-and-esg-for-a-universal-owner?from=guide)
6. [why pension funds don't invest in nature](/guide/why-pension-funds-dont-invest-in-nature?from=guide)
7. [what system-level investing looks like at a named place](/guide/what-system-level-investing-looks-like-at-a-named-place?from=guide)