Beyond Accumulation

Capital systems designed for trust, ecology, and interdependent renewal rather than extraction.

Future Proof Intelligence. Research. No. IV. MMXXVI

Abstract

Accumulation is treated as the natural endpoint of capital. It is not. It is a design choice, encoded so deeply into ownership, fund structure, and fiduciary practice that it reads as a law of physics. This paper takes the opposite position seriously and on its own terms. It is possible to design economic and capital structures that go beyond pure accumulation, where financial return and the renewal of the asset, the ecology, and the relationships that produced the return are not opposed but engineered to reinforce each other.

The argument turns on a single distinction. Extraction and renewal are properties of the claim structure, meaning the legal and financial arrangement that defines who is entitled to demand what from a productive system, and not properties of the character of the people inside it. This explains why behavioural approaches, pledges, ratings, and disclosure overlays, fail predictably under pressure: a promise placed on top of an unchanged claim structure loses to the claim structure when the two disagree.

We examine the instruments that already constrain the claim structure rather than the behaviour: the asset lock, the separation of voting and economic rights, the perpetual horizon, designed commons governance, and universal ownership. We treat their failure modes honestly. Steward ownership, regenerative finance, patient institutional capital, and certification each have a way they fail, and naming it is the price of being taken seriously. The paper closes on the layer that is still missing: a substrate of trust and continuity that makes constrained capital legible across time and across counterparties, which is the difference between a structure that survives its founder and one that does not.

1. The Default That Looks Like Nature

1.1 What we mean by accumulation

Accumulation, in this paper, is not wealth and not profit. Both are normal features of a working economy and neither is the subject here. Accumulation is a specific and narrower thing: the systematic conversion of a productive system into a present financial claim that is then withdrawn and redeployed elsewhere, at the highest rate the system can sustain without immediately failing.

Read that definition slowly, because the word that matters is withdrawn. A firm that earns a return and reinvests it into the conditions that produced the return is not accumulating in this sense. It is compounding inside its own boundary. A firm that earns the same return and routes it out, to owners with no operational connection to the firm, to a market that will reallocate it to whatever has the highest near-term yield, is accumulating. The two firms can have identical income statements. They are different systems.

This is the first thing the conventional vocabulary obscures. We measure return as a scalar. We rarely measure where the return goes, what it does to the system that produced it, and whether the act of taking it weakened the capacity to produce it again. A scalar cannot hold that information. The information lives in the structure, and the structure is usually invisible because it is assumed.

1.2 Why the default reads as physics

Three reinforcing layers make accumulation feel inevitable rather than chosen.

The first is the legal default of the shareholder corporation. The standard joint stock company is, at its core, a machine for converting enterprise into a tradable residual claim. The residual claimant is entitled to the surplus, the claim is liquid, and liquidity means the claim can and structurally wants to move to wherever the surplus is largest. None of this is wrong. It is an extraordinary coordination technology and it built the modern economy. But it is a design, with a particular objective function, and it was selected for that objective among alternatives that also exist.

The second is the fund layer that sits above most serious capital. Closed-end fund structures have defined lives. A defined life is, mathematically, an instruction to exit. A ten-year vehicle does not ask whether a company should be sold; it asks when, because the structure has already decided that it must be. The horizon is not a market signal. It is an artefact of the wrapper, and the wrapper transmits its own deadline into every company it touches as if the deadline were a fact about the company.

The third is fiduciary interpretation. The dominant reading of fiduciary duty, narrowed over decades, treats the maximisation of the present financial claim of beneficiaries as the safe interpretation and any deviation as something requiring justification. The asymmetry is the point. Extraction is the unmarked case and renewal is the case that has to argue for itself. Defaults are powerful precisely because they do not have to win an argument.

Stack the three and you get a system in which accumulation is not chosen at any single point. It is the path of least resistance through a legal default, a structural deadline, and an interpretive asymmetry. That is exactly why it feels like nature. Nothing that requires three separate institutions to enforce it is nature.

1.2a The measurement is not anecdotal

A reader entitled to scepticism will ask whether the accumulation default is a real distortion or a rhetorical one. The honest answer is that it is measurable, and it has been measured by people with no ideological stake in the answer. Work by Andrew Haldane and Richard Davies at the Bank of England examined how capital markets discount future cash flows and found systematic excess discounting: cash flows five years out priced as if they were eight or more years out, ten-year cash flows valued as if sixteen or more years away, and cash flows beyond thirty years scarcely valued at all. Their estimate of the excess discount was on the order of five to ten per cent per year, and they found it both statistically and economically significant and rising over time. Haldane's own description of the result was that segments of the market had become, in his phrase, quarterly capitalism, with the typical owner holding for a horizon considerably shorter than a year.

This matters to the argument for one specific reason. Excess discounting is not a moral failing of investors. It is the price signal that the three structural layers above produce when they are stacked. The market is not irrationally myopic. It is rationally responding to a legal default, a structural deadline, and an interpretive asymmetry that jointly make the near claim safe and the far one speculative. The myopia is the readout of the structure, not a separate phenomenon. Any redesign that targets the myopia without touching the structure is treating a symptom that the structure will simply reproduce.

1.3 The thesis

The thesis of this paper is that the alternative is a design discipline with working precedents, not an aspiration. Capital structures exist, some of them more than a century old, in which the claim itself is constrained at the point of ownership so that renewal is not a behaviour anyone has to choose under pressure but a property of the structure that holds regardless of who is in the room. These structures are not soft. Several of them are more legally rigid than the conventional corporation, not less. They are under discussed not because they are marginal but because they sit in the gap between corporate law, finance, and governance and belong cleanly to no single profession that would otherwise champion them.

The rest of the paper builds the argument in five moves. We locate the real object of design, the claim structure. We catalogue the instruments that constrain it. We treat their failure modes without flinching. We establish the precise conditions under which return and renewal converge, and the conditions under which the convergence is counterfeit. And we identify the layer that is still missing, the one without which every structure in this paper degrades the moment its founder steps away.

2. The Claim Structure Is the Object

2.1 Behaviour is downstream of structure

The single most consequential error in the conversation about responsible capital is the belief that the lever is behaviour. It is not. Behaviour is the output. The lever is the claim structure, and behaviour is what the claim structure produces under the conditions it is exposed to.

Consider the standard intervention set: a sustainability pledge, an environmental and social rating, a disclosure framework, an impact report. Each of these acts on behaviour while leaving the claim structure untouched. The owner is still entitled to the full residual. The fund still has its deadline. The fiduciary still reads duty narrowly. Into that unchanged structure we introduce a promise. The promise is sincere. It is also, structurally, a cantilever with nothing under the far end. It holds in calm conditions because nothing is testing it. The first time the claim structure and the promise disagree, under a liquidity event, an activist position, a board change, a downturn, the claim structure wins, because the claim structure is enforceable and the promise is, at best, reputational.

This is not cynicism about people. It is the opposite. It takes the people seriously enough to notice that we have asked them to hold a load the structure was designed to shed onto them, and then we are surprised when the structure does what it was built to do.

There is a sharper way to see this. A behavioural intervention is a request to a system to act against its own gradient. Every system has a gradient: the direction it moves when no one is applying force. The conventional claim structure has a gradient that points toward extraction, because that is what the residual claim, the liquidity, and the proportional control jointly produce. A pledge is a hand held against that gradient. It works while the hand is there and the gradient is gentle. It does nothing to the gradient itself. The moment the hand tires, or the gradient steepens under a downturn or a contested event, the system resumes its descent, and it resumes from wherever the hand happened to be holding it, which is usually a worse position than if the hand had never been there, because the pledge has by then been spent as reputational cover. Structural design is the discipline of changing the gradient instead of holding a hand against it. Everything in Section 3 is a description of how to re-grade the slope so that the system, left entirely alone, rolls toward renewal rather than away from it.

2.2 Extraction lives in entitlement, not in intent

Restate the central distinction as precisely as possible. Whether a capital system is extractive or regenerative is determined by the answer to one question: what is the owner of capital entitled to demand, and from whom, and on what trigger?

If the owner is entitled to the full residual, to liquidity on demand, and to control proportional to capital, then the system is extractive by construction, irrespective of the owner's values, because the entitlement structure will eventually be exercised by someone, somewhere in the ownership chain, who holds it without the values. Values do not transfer with shares. Entitlements do. This is why a founder can run a company beautifully for thirty years and the structure can undo all of it in the eighteen months after the founder sells, dies, or is outvoted. The structure was always the thing. The founder was a temporary override on it.

If, instead, the entitlement is constrained, if there is a ceiling on the residual that may be withdrawn, if the asset cannot be sold for a capital gain, if control is held by people connected to the purpose rather than by whoever holds the most capital, then renewal is not something anyone has to choose under pressure. It is what the structure does when no one is choosing anything, which is most of the time. The objective of regenerative capital design is to move the good outcome from the category of behaviour under pressure to the category of default under neglect. A system you can neglect into the right outcome is the only kind that survives contact with time.

2.3 Figure 1, the two architectures

Figure 1. The behavioural overlay and the constrained claim.

Picture two columns. In the left column, a conventional claim structure sits at the base: full residual, liquid, control by capital. On top of it, a thin layer labelled "pledge, rating, report". An arrow of force, labelled "pressure event", strikes the structure. The thin top layer shears off; the base is unmoved. This is the behavioural overlay. It fails by design, cleanly, at the join.

In the right column, there is no top layer at all. The constraint is machined into the base itself: a capped residual, an asset lock, control held by purpose. The same arrow of force strikes. Nothing shears, because there is no join. The structure absorbs the pressure because the constraint is not sitting on the load path, it is the load path.

The figure carries the whole paper. Everything that follows is a description of how to build the right column, what it costs, and the ways it can still go wrong.

3. The Instruments That Already Exist

This section is a catalogue. Each instrument is real, has working precedent, and constrains the claim structure rather than the behaviour. For each, we state precisely what it constrains, what it leaves unconstrained, and the precedent that proves it is not theoretical.

3.1 The asset lock

The asset lock is the foundational instrument. In its strict form it states that the enterprise cannot be sold for the personal capital gain of its owners, and that on any dissolution the residual value cannot be distributed to them. The company stops being an asset that exists to be liquidated and becomes an institution that exists to operate.

This is the structural core of steward ownership, a term the German Purpose Foundation introduced in 2017 after studying older firms that had already been operating this way for generations. The reference cases are not start-up experiments. The Carl Zeiss Foundation, created by Ernst Abbe in 1889, is the sole owner of Zeiss; its constitution prevents the company from being sold and directs profits to be reinvested or applied to the common good. The Robert Bosch arrangement separates voting rights from dividend rights by placing them in distinct legal entities. These structures have survived two world wars, hyperinflation, partition, and reunification. The relevant fact is not that they are virtuous. It is that they are old. They have already passed the only test that matters for a renewal structure, which is the test of outliving the people who set it up.

What the asset lock constrains: the conversion of the enterprise into a one-time terminal payout to owners. What it does not constrain: ordinary commercial operation, fair compensation, debt finance, reinvestment, distribution of operating surplus within agreed limits. This precision matters because the most common objection to the asset lock, that it starves the firm of capital, confuses the terminal claim with the operating return. The asset lock removes the lottery ticket. It does not remove the wage, the dividend within bounds, or the loan.

There is a methodological point worth making here, because it is the strongest empirical card the regenerative case holds and it is almost never played. The nineteenth-century steward structures are not just precedents. They are something closer to a natural experiment that has been running, unfunded and unsupervised, for more than a hundred and thirty years. The control group is the entire population of conventionally owned firms founded in the same era and the same industries, the overwhelming majority of which have been sold, broken up, absorbed, or extracted out of recognisable existence. The treatment group is a small set of asset-locked firms that were exposed to the same wars, the same currency collapses, the same competitive and technological shocks, and a number of which are still operating as coherent institutions under the original constraint. No one designed this experiment. It is the unintended record of a structural choice made by a handful of founders against the default of their own time. The result does not prove that the asset lock guarantees survival; survivorship reasoning would be a fair objection if the claim were that strong. It establishes something weaker and sufficient: that a firm can be made structurally unsellable and remain a serious, competitive, technically excellent enterprise across more than a century, which is the exact proposition the conventional default treats as impossible. One counterexample retires an impossibility claim. Here there are several, and they have been audited by time rather than by anyone with an interest in the answer.

3.2 The separation of voting and economic rights

The second instrument decouples control from cash flow. In a conventional share, the right to govern and the right to the residual are bonded together and sold as one unit. Separate them and you can place control with people connected to the purpose of the enterprise while still allowing economic participation by investors, employees, or a mission entity, on defined terms.

The clearest contemporary instance is the 2022 Patagonia transfer. Ownership moved to two entities: the Patagonia Purpose Trust, holding all of the voting stock and roughly two per cent of the company, controlled by the founding family and stewards; and the Holdfast Collective, a nonprofit holding the non-voting stock and the large majority of the economic interest, directing the company's profit toward environmental work. Control and cash flow were deliberately split and pointed in different directions: control toward continuity of purpose, cash flow toward the purpose itself rather than toward a private terminal gain. By the company's own reporting, the structure has since directed on the order of one hundred and eighty million dollars to the Holdfast Collective in its first years, including roughly eighty million in 2024. We cite these as orders of magnitude and as the company's own figures, not as audited universals; the structural point does not depend on the exact number.

What this instrument constrains: the automatic coupling of money to power that lets accumulated capital purchase governance. What it does not constrain: investment itself. Economic participation remains possible. What changes is that it no longer comes bundled with the right to redirect the institution.

3.3 The perpetual horizon

The third instrument is the removal of the structural deadline. A perpetual or evergreen vehicle has no defined life, therefore no built-in instruction to exit, therefore no mechanical conversion of a productive holding into a terminal sale on a calendar that has nothing to do with the holding.

This is not a fringe construction. Evergreen and perpetual fund structures in private capital have grown sharply; industry summaries put assets under management in such structures in the order of several hundred billion dollars, having multiplied several times over roughly a decade and grown by more than thirty per cent between late 2024 and late 2025. The endowment and sovereign fund world has operated on the same logic for far longer, through the intergenerational equity principle associated with James Tobin: spend only the real return, preserve the real capital across generations. Norway's Government Pension Fund Global operationalises a version of this through a fiscal rule that limits spending to roughly the expected long-run real return, explicitly to keep the real capital intact and transfer it forward.

A caution belongs here, and the paper would be dishonest without it. A perpetual horizon removes the deadline. It does not, by itself, change the objective function inside the horizon. An evergreen vehicle optimising purely for accumulation simply accumulates forever, which can be worse, not better. The perpetual horizon is necessary for renewal and nowhere near sufficient. It only matters when combined with a constrained claim. We will return to this in Section 6, because it is one of the places the convergence of return and renewal is most often counterfeited.

3.4 Designed commons governance

The fourth instrument is older than corporate finance and was, for most of the twentieth century, assumed not to work. The presumed tragedy of the commons held that a shared resource without private ownership or central control would inevitably be depleted. Elinor Ostrom's field research, recognised with the Nobel Memorial Prize in Economic Sciences in 2009, demonstrated that this is false as a general law. Communities have governed shared resources durably for centuries, and the ones that succeed share identifiable design features: clearly defined boundaries, rules matched to local conditions, collective choice by those affected, monitoring, graduated sanctions, and accessible conflict resolution.

The lesson for capital design is precise and is usually mis-stated. The lesson is not that goodwill scales. It is the opposite. Durable shared value is held by designed institutions with monitoring and graduated sanction, not by trust in the abstract. Ostrom's principles are an engineering specification for renewal under shared ownership, and they are demanding. They make explicit that a renewal structure without monitoring and without a graduated, enforced consequence for defection is not a renewal structure. It is a hope with a logo.

3.5 The collective capital account

The fifth instrument addresses a problem the first four leave open: how a structure accumulates a buffer of capital that belongs to the institution itself rather than to any individual member or owner, so that the institution has reserves it can deploy across cycles without those reserves being a withdrawable claim.

The most studied working instance is the collective capital account in the Mondragon cooperative system, where a substantial share of capital value, in the order of roughly thirty per cent in the studied period, is held collectively rather than individually. The account is not anyone's to withdraw. It is the institution's memory of its own retained surplus, a counter-cyclical reserve and a constraint in one instrument: the members cannot vote to liquidate it into personal claims, because by construction it is not theirs to take. This is the asset lock applied not to the whole enterprise but to a defined and growing portion of its capital, which is a more graduated and often more financeable design than an all-or-nothing lock.

What this instrument constrains: the conversion of retained surplus into individual withdrawable claims, which is the slow version of the terminal-sale problem. What it does not constrain: members' fair current income, or the institution's ability to deploy the collective capital operationally. The honest qualification, which Section 4 develops, is that a collective capital account does not by itself prevent the deeper failure that the cooperative literature calls degeneration. It is a necessary buffer, not a complete answer.

3.6 Universal ownership

The sixth instrument is not a structure you build but a position some capital is already in, whether it has noticed or not. James Hawley and Andrew Williams described the universal owner: an investor so large, so diversified, and so long in horizon, pension funds, insurers, large endowments, sovereign funds, that it effectively holds a slice of the entire economy. For such an owner, the externalities one holding imposes on the rest of the economy are not external at all. They land inside the same portfolio. The cost a portfolio company exports is a cost the universal owner re-imports through its other ten thousand holdings.

This collapses, at sufficient scale and horizon, the supposed opposition between a single asset's extractive return and the health of the system. For a true universal owner, system degradation is a direct portfolio loss with a lag. The instrument here is the recognition itself: at the largest scale, the extraction-versus-renewal trade-off is partly an accounting illusion produced by drawing the boundary of the balance sheet too small. We will hold the honest limit of this argument for Section 4, because the universal owner story has a real and well documented failure that its proponents sometimes skip.

4. The Failure Modes, Stated Without Flinching

A paper that catalogues instruments and stops there is advocacy, not analysis. Every instrument in Section 3 has a way it fails. Naming the failure precisely is not a concession. It is the difference between an idea a serious reader can use and one they have to discount on sight. We take each in turn.

4.1 The signalling failure: when the trust standard is the product

The most common failure is the one that hollows the field from the inside. When a renewal structure or a responsibility standard becomes valuable as a signal, the signal becomes the thing optimised, and the optimisation runs ahead of the substance.

The clearest documented case is the trajectory of a major responsible-business certification. For years its threshold was an aggregate score across several areas, which permitted a firm to compensate weakness in one dimension with strength in another, and to hold the badge while operating in ways that drew sustained greenwashing criticism, including for multinationals certified at the subsidiary level whose parents had contested records. The certifier moved in 2025 to a performance-based standard with minimum requirements in each area, after public pressure and after at least one of its highest-regarded members declined to renew on the explicit ground that the standard was too weak for large incumbents. This is not an indictment of the certifier, which responded. It is a demonstration of a structural law: any constraint that is cheaper to display than to satisfy will, at the margin, be displayed rather than satisfied. A renewal structure is only as strong as the gap it maintains between displaying the constraint and actually bearing it.

4.2 The capture failure: the Faber problem

The second failure is governance capture, and it has a clean illustration. A large food multinational pursued near-comprehensive responsible-business certification under a chief executive who explicitly prioritised long-horizon purpose over near-term shareholder distribution. The board removed that chief executive. The certification commitments largely survived him as commitments; the strategic priority that animated them did not survive contact with the claim structure that the board still answered to.

Name this the Faber problem, after the public case. It states: where the renewal posture depends on a person occupying a control position, and the control position remains formally answerable to an unconstrained claim, the renewal posture is a tenancy, not a structure. It lasts exactly as long as the tenant. This is the single most important failure mode in the paper because it is the one that masquerades as success for years. Founder-led renewal looks identical to structural renewal right up until the founder is gone, at which point the difference is the only thing that matters and it is too late to install it.

4.3 The liquidity failure: starving the thing you locked

The third failure is real and is the strongest good-faith objection to the asset lock. If the enterprise cannot be sold for a gain and the residual is capped, the conventional equity investor's return path is closed. The capital that flows most freely is precisely the capital that wants the path you have closed. A naively constructed lock can therefore raise the cost of capital, slow growth, and in the limit protect the purity of an institution into irrelevance.

This failure is not a refutation of the instrument. It is a specification requirement. A serious steward structure has to engineer a return path for capital that does not run through the terminal sale: capped but real participation in operating surplus, redeemable instruments with defined ceilings, debt and revenue-based finance, or aligned investors who explicitly price for the constraint. The Purpose Foundation itself names aligned capital as one of the three principal hurdles for steward ownership, alongside awareness and legal complexity. That is an admission against interest from inside the movement, and it should be read as one. A renewal structure that has not solved its capital path is not finished. It is a prototype.

4.4 The ossification failure: a lock is not a guarantee of life

The fourth failure is the mirror image of the third. An asset lock and a perpetual horizon remove the forces that would otherwise discipline a failing enterprise: takeover, forced sale, exit. Those forces are brutal and often destructive, but they are also a crude form of accountability. Remove them and you must replace them, or you build an institution that cannot be killed and therefore cannot be made to change. A steward-owned firm with weak internal governance does not get renewal. It gets a protected decline, which is renewal's most plausible imitation and its exact opposite. Ostrom's principles are the antidote here precisely because they insist on monitoring and graduated, enforced consequence. A renewal structure without an internal accountability mechanism at least as demanding as the market discipline it removed is not safer than the conventional firm. It is less safe, slower, and harder to correct.

4.4a The degeneration failure: reversion under competitive pressure

There is a failure specific to member-owned and collectively capitalised structures, and the cooperative literature has named and studied it for decades under one word: degeneration. It comes in three forms, and the precision of the taxonomy is what makes it useful. Constitutional degeneration is when the structure formally reverts to a conventional form, the lock is unwound, the cooperative becomes a company. Goal degeneration is when the form is kept but the operating objective quietly becomes profit maximisation indistinguishable from the conventional firm. Organisational degeneration is when the democratic or distributed governance is captured by a managerial elite while the formal structure remains.

The studied trajectory of large cooperative groups expanding into global competition shows all three pressures operating, including the emergence of hybrid arrangements in which a cooperative core is sustained by conventionally owned subsidiaries, a structure the literature has bluntly labelled with a portmanteau of cooperative and capitalist. The lesson is not that these structures do not work. Many have worked for generations. The lesson is that competitive pressure is a continuous solvent applied to the constraint, and a renewal structure that does not have a continuously renewed mechanism for resisting that solvent will dissolve slowly enough that no single actor is ever seen to break it. Degeneration is the failure mode with no villain. It is entropy with a competitive gradient behind it, and it is precisely the failure that the missing layer of Section 7 exists to address, because the only thing that arrests slow reversion is something that re-presents the meaning of the constraint to each new cohort before the constraint has finished dissolving.

4.5 The diversification failure: the limit of the universal owner

The fifth failure is the honest limit of the universal owner argument. The theory says a sufficiently diversified, long-horizon owner re-imports the externalities it exports and is therefore structurally motivated toward system health. The documented strain, surfaced sharply in 2024 analyses of decarbonisation under a hostile engagement environment, is that the same breadth that creates the motivation removes the tool. A universal owner cannot meaningfully divest from systemic risk, because it owns the system; and engagement, its remaining lever, depends on a cooperative environment that cannot be assumed. The universal owner is structurally motivated toward renewal and structurally under-equipped to deliver it through the mechanisms it actually controls. The motivation is real. The capability gap is also real. Treating the first as if it implied the second is the field's most respectable error.

4.6 What the failure modes have in common

Read the six together and a single pattern appears. Every one of them is a case of an incomplete structure: a constraint installed without the complementary mechanism that makes the constraint hold. The signal without the cost. The purpose-holder without the structural lock. The lock without the capital path. The lock without the internal accountability. The constraint without the mechanism that renews it against the competitive solvent. The motivation without the lever. Renewal does not fail because the idea is naive. It fails because a partial implementation of a renewal structure is often worse than no implementation, and partial is the default state of anything hard. This is the bridge to the rest of the paper. The instruments are necessary. The thing that makes a set of instruments into a structure that survives is a separate layer, and most of the field does not have it.

A final observation on the pattern, because it sharpens the whole paper. Notice that none of the six failures is a failure of the idea of renewal. Each is a failure to install the second half of an instrument. The asset lock is half an instrument; the capital path is the other half. The purpose-holder is half an instrument; the structural lock that makes the purpose-holder unremovable by the claim is the other half. The collective capital account is half an instrument; the continuous re-presentation of why it exists is the other half. The field is full of first halves. First halves are visible, citable, and good for a press release. Second halves are quiet, structural, and only tested at transitions, which is to say only tested when it is too late to add them. The entire practical content of regenerative capital design reduces to a single discipline: never ship a first half without its second half, and treat any structure presenting only its first half as a structure that has not yet been built, regardless of how complete it looks while its founder is still in the room.

5. Trust as a Factor of Production

5.1 The missing input in the production function

Economics has a conventional list of the inputs that combine to produce output: labour, capital, land, and depending on the vintage of the textbook, technology or knowledge. Trust is almost never on the list, and its absence is not an oversight that can be patched by adding a fourth or fifth term. It is a category error in the other direction. Trust is not one more input alongside the others. It is the precondition that determines the transaction cost at which all the other inputs can be combined at all. Two economies with identical labour, capital, and land but different levels of durable trust are not two points on the same production function. They are running different production functions, and the lower-trust one runs a more expensive one forever, because every coordination it performs carries a verification, enforcement, and contingency tax that the higher-trust one does not pay.

This reframes the entire subject of the paper. Extraction, defined in Section 1 as the withdrawal of the present claim at the maximum sustainable rate, is not free even on its own terms. It consumes trust. Every act of extraction that a counterparty experiences as the structure choosing the present claim over the relationship is a withdrawal from a stock that priced every future transaction. The conventional accounting cannot see this because trust has no line item, which means the depletion is invisible exactly until it is catastrophic, which is the signature of every stock that is consumed without being measured.

5.2 Why trust is the input that renewal renews

Now the instruments of Section 3 read differently. The asset lock, the rights split, the perpetual horizon, the collective capital account, designed commons governance: what do they actually produce? Stated narrowly, each constrains a claim. Stated at the level that matters, each is a trust-production technology. They produce, as their real output, the credible commitment that the structure cannot do to a counterparty the thing the counterparty most fears, and that credibility lowers the transaction cost of every relationship the structure has, with employees, suppliers, regulators, communities, and capital itself.

This is the deep reason return and renewal are not opposed, and it is more fundamental than the horizon-and-boundary argument that follows in Section 6. Renewal, properly understood, is not primarily reinvestment in physical or even human capital. It is reinvestment in the trust stock that determines the transaction cost of the entire enterprise. A structure that renews its trust stock is lowering its own cost of coordination on a compounding basis. A structure that extracts is raising it. The two firms with identical income statements from Section 1 are now revealed as having radically different cost structures one decade out, and the difference is entirely invisible on the statements that the market is pricing. The market is pricing the wrong document. That is not a moral observation. It is an arbitrage, available to anyone with a structure that can credibly hold a horizon long enough for the trust stock to capitalise.

5.3 Why trust cannot be a behavioural promise

It follows, with some force, that trust as a production input cannot be created by a promise, a pledge, or a report, for exactly the reason a currency cannot be backed by a promise to back it. Trust as an economic input has value only to the extent that it is not at the discretion of the party who benefits from spending it. A structure that can extract whenever it chooses, but promises not to, has produced no trust input at all, only a forecast of its own future behaviour, which a rational counterparty discounts to approximately its enforceability, which for a promise is approximately zero under stress. The trust input is produced only when the discretion is removed, which is to say only when the constraint is structural. This is why the entire field's instinct to address renewal through commitments is not merely weak but categorically confused. It is attempting to manufacture the one production input that is defined by the absence of the discretion the commitment preserves.

This is the paper's deepest claim and it is worth stating once, plainly. The reason capital systems beyond accumulation are possible is that trust is a real factor of production, it has measurable economic value through the transaction cost it sets, it can only be produced by removing discretion rather than by promising to forgo it, and the instruments that remove discretion at the point of the claim are therefore not ethical constraints on the economy. They are economic infrastructure for producing the input the economy is most expensively short of. Renewal is not the price of virtue. It is the production of the cheapest input there is, by the only method that produces it.

6. Return and Renewal Are Not Opposed

6.1 The false trade-off

The folk model holds a hydraulic relationship: every unit of return taken is a unit of renewal foregone, and every unit of renewal is a tax on return. If that model were true, the entire project of this paper would be a moral argument for accepting lower returns, and it would be honest to say so. It is not true, and it is worth being precise about why, because the precision is also the test that separates real convergence from the counterfeit.

The hydraulic model is true under one specific condition: when the boundary of the accounting and the boundary of the time horizon are both drawn tightly around the single asset and the near term. Inside that frame, renewal genuinely is a deduction from return, because the benefit of renewal accrues outside the frame, later, or to the system, and the frame cannot see it. The trade-off is real as an artefact of where the lines are drawn. Move the lines and the trade-off moves with them, sometimes disappearing entirely.

6.2 The three conditions of genuine convergence

Return and renewal converge, genuinely and not rhetorically, under three conditions, and only when all three hold at once.

First, horizon length sufficient for the renewal to capitalise. Renewal is reinvestment into the conditions of production: the asset, the ecology, the workforce, the trust of counterparties. Reinvestment has a payback period. If the horizon is shorter than the payback period, renewal really is a loss inside that horizon, and no amount of conviction changes the arithmetic. Extend the horizon past the payback period and the same reinvestment shows up as compounding return. The perpetual horizon of Section 3 is not a moral preference. It is the mathematical precondition that allows renewal to appear on the return side of the ledger rather than the cost side.

Second, a boundary wide enough to internalise the externality. The universal owner argument is the limit case of this condition: at sufficient breadth, the cost an asset exports is re-imported and the trade-off collapses. But the condition operates well below universal scale. Any owner whose boundary includes the supplier base, the labour market it draws from, the watershed it sits in, or the regulatory trust it depends on, internalises a portion of what a narrowly drawn owner would treat as external. Boundary width is a dial, not a switch, and convergence increases monotonically as the dial turns.

Third, a constraint that prevents the horizon and the boundary from being collapsed under pressure. This is the condition the field most often forgets, and it is the decisive one. The first two conditions are unstable. A long horizon can be shortened by a single liquidity event. A wide boundary can be narrowed by a single change of control. The convergence of return and renewal is not self-stabilising; left unprotected it decays back to the trade-off the moment anyone with the entitlement to do so collapses the frame to extract the present claim. The asset lock and the rights split exist precisely to make the horizon and the boundary structurally non-collapsible. They are not the source of the convergence. They are what stops the convergence from being stolen.

6.3 How to tell the real convergence from the counterfeit

This gives a clean diagnostic, which is one of the most useful things the paper offers a practitioner. Genuine convergence of return and renewal is identifiable by a single property: the horizon and the boundary cannot be unilaterally collapsed by a holder of the claim. If any single party, an acquirer, an activist, a successor board, a fund reaching the end of its life, can shorten the horizon or narrow the boundary at will, then what looks like convergence is a temporary alignment that exists at the pleasure of whoever holds the collapse option. It is the Faber problem wearing the language of synergy.

Apply the diagnostic to any structure that claims to have transcended the trade-off and it sorts cleanly. A firm with strong values and a conventional share structure: counterfeit, because the collapse option is liquid and outstanding. A firm with an asset lock, a rights split, and weak internal accountability: incomplete, vulnerable to the ossification failure, but the collapse option is at least removed. A firm with the lock, the rights split, a solved capital path, and Ostrom-grade internal accountability: genuine, and rare, because all four are hard and the default state of hard things is partial. The diagnostic does not flatter anyone. That is what makes it worth having.

7. The Missing Layer

7.1 Why good structures still die

Everything to this point describes how to build a capital structure in which renewal is the default under neglect. Assume it is built correctly: the lock, the rights split, the capital path, the internal accountability. There remains one failure the structure cannot solve from inside itself, and it is the one that kills most of them.

A constrained capital structure is only as durable as the legibility and continuity of the constraint across time and across counterparties. The lock is written in a constitution, a trust deed, a foundation charter. Documents are inert. They are interpreted, administered, litigated, amended, and forgotten by people who arrive long after the founder and who did not feel what the founder felt. Every transition, succession, recapitalisation, jurisdictional move, merger of the holding entity, is a moment where the constraint must be re-understood by someone with the power to weaken it and no memory of why it exists. The structures from 1889 that are still intact did not survive because their documents were unusually well drafted. They survived because an institution carried the meaning of the documents forward, intact, across the gaps between the people who could have collapsed them. The carrying institution is the missing layer. Without it, a perfectly machined structure is a message in a bottle, and bottles are opened by whoever finds them, for their own reasons.

7.2 What the missing layer has to do

State the requirement abstractly, because the abstraction is the point. The missing layer is not more law and not more capital. It is a continuity-of-meaning function. It has to do four things that no document does on its own.

It has to make the constraint legible to every future counterparty, so that an investor, a successor, a regulator, or an acquirer encounters the structure already understanding what it is and what it is not, rather than discovering it adversarially at the worst possible moment.

It has to make the constraint verifiable against a standard, so that compliance with the renewal structure is something that can be assessed against an external reference rather than asserted, which is the only thing that closes the gap between displaying the constraint and bearing it that defeated the certification in Section 4.

It has to make the constraint continuous across transitions, so that the meaning of the structure is held by something with a longer memory than any individual steward and is transmitted forward at exactly the succession moments where structures historically fail.

And it has to make the constraint interdependent, connecting a single constrained structure to others of its kind, so that the renewal is not a lonely island that decays at its edges but part of a fabric where the surrounding structures hold each other legible, the way Ostrom's enduring commons were never single institutions but nested ones.

7.3 Where this layer is forming

It is tempting to treat the missing layer as a call for new regulation, and regulation will be part of it. The legislative motion in several European jurisdictions during 2025 toward a dedicated steward-ownership legal form is real and matters, because a named legal form is itself a legibility instrument: it lets a counterparty recognise the structure without re-deriving it. But statute is slow, jurisdictionally bounded, and silent on continuity of meaning. It can define the form. It cannot carry the meaning of a specific structure across a specific succession in a specific institution. Something has to operate in the space statute does not reach.

Consider what that space actually contains. It contains the moment a founder dies and a successor reads a trust deed without the context that made it make sense. It contains the negotiation in which an acquirer's counsel encounters an asset lock and treats it, by professional reflex, as a defect to be priced out rather than a property to be preserved. It contains the slow afternoon, years from now, when a board with no memory of the original purpose votes to amend a clause that no one in the room understands the load-bearing function of. None of these moments is reachable by a statute, a rating, or a pledge, because all of them are moments of interpretation under power, and interpretation under power is decided by whatever institution is present at that moment with the standing, the memory, and the legibility to make the constraint understood before it is undone. The question of regenerative capital, fully stated, is not whether the right structures can be built. Section 3 settles that. It is whether anything exists that is present at all of those moments, in all of those institutions, with a memory longer than any individual steward and a standard external enough to be trusted by a counterparty who has every incentive to disbelieve it. That, and not another financial instrument, is the actual frontier.

This is the point at which the paper arrives where it has been heading. The continuity-of-meaning layer is not a future proposal. It is the function that a small number of bodies have already been building underneath this problem, quietly, as standard, certification, and identity infrastructure rather than as another financial product. Future Proof Intelligence is one of the places this layer is being operated as a substrate: an identity layer that sits over the orchestration layer of how capital and AI systems coordinate, treating trust as a structural and verifiable property rather than a reputational one, and treating the continuity of an institution's meaning across its transitions as something that can be held, carried, and re-presented to each new counterparty rather than left to chance. We reference this the way one references an institution that is simply there: not as a thing being sold, but as a description of where, in the present tense, the missing layer is no longer entirely missing. The argument of this paper does not depend on that being so. It is, however, the difference between the structures in Section 3 being a museum of well preserved exceptions and being the early grammar of a system that holds.

8. Implications

The preceding sections are an argument about structure. This section is what the argument requires of the people who will act on it. It is divided by reader because the obligation is different for each, and it stays in the register of the idea, not of a pitch.

8.1 For institutions

For an institution, the operative finding is that your responsibility posture is only as durable as your weakest collapse option. Audit your own structure for the Faber problem before you audit anyone else's. Identify, precisely, every party who could unilaterally shorten your horizon or narrow your boundary: an acquirer, a creditor with a control covenant, a successor board, a fund in your capital stack approaching the end of its life. Each one is a place where your renewal posture is a tenancy. The work is not to add another pledge. It is to remove or constrain the collapse options, in priority order of how liquid each one is, because the liquid ones fail first and fail fastest. An institution that has done this can make commitments that survive it. An institution that has not should be honest that its commitments have an expiry date equal to the tenure of the people currently making them.

8.2 For investors

For an investor, the finding is that the trade-off you have been taught is real only inside a frame you are allowed to redraw. The discipline is to underwrite the horizon and the boundary explicitly, as terms, not as sentiment. Ask of any structure presenting itself as regenerative the single diagnostic question from Section 6: can the horizon or the boundary be unilaterally collapsed by a holder of the claim, and if so, by whom and on what trigger? If the answer is yes, you are not looking at convergence, you are looking at an alignment that lasts until someone exercises the collapse option, and you should price and document it as such. If the answer is no, you are looking at something rarer and you should understand that its capital path will not look like a conventional exit, and that this is the feature, not the defect. The investor's edge here is not virtue. It is the ability to tell a structure that will hold from one that is performing holding, and to price the difference before the market learns to.

8.3 For operators

For an operator, the founder or steward inside one of these structures, the finding is the hardest and the most personal. The structure you run well today is not the structure. You are. The system experiences your judgement as a temporary override on a claim structure that is still underneath you and still has its own objective function. The single most consequential act available to you is not a year of excellent stewardship. It is the installation, while you still hold control, of the constraint and the continuity layer that make your stewardship unnecessary. This is counterintuitive because it feels like a demotion of your own importance. It is the opposite. The founders whose structures survived from the nineteenth century are remembered not for how they ran the firm but for the fact that they made themselves removable without making the institution collapsible. That is the operator's actual legacy decision, and most operators make it implicitly, by default, and therefore make it wrong.

8.4 For the people inside these systems

For the people whose working lives, livelihoods, and trust are inside one of these structures, the finding is a standard they are entitled to apply and rarely told they may. The question to ask of any institution that describes itself as regenerative, purpose-led, or built to last is not what it intends. It is what it is entitled to do to you under pressure, and who, specifically, holds the option to do it. A structure that cannot answer that question concretely is asking for trust it has not engineered the conditions for. A structure that can, that can name its locks, its rights split, its continuity layer, and the external standard it is verifiable against, is offering something different in kind from a promise. The shift this paper argues for is, at the human level, exactly this: the right to be inside a system because of what it structurally cannot do to you, rather than because of what it currently chooses not to.

8.5 The common instruction

The four readers above receive what looks like four different instructions. They are one instruction in four registers. The instruction is: stop evaluating capital systems by their stated objective and start evaluating them by their collapse surface, the complete set of parties who can unilaterally shorten the horizon or narrow the boundary, and the trigger that activates each one. The stated objective is information about the present occupants. The collapse surface is information about the system. A serious institution, investor, operator, or participant is one that has learned to read the second and stopped being reassured by the first.

This is also why the conventional toolkit of responsible capital, the pledge, the rating, the disclosure, the impact report, is not merely insufficient but actively misdirecting. Every one of those instruments reports on the stated objective and is silent on the collapse surface. They are well-lit instruments pointed at the wrong variable, and the brightness of the light is itself the problem, because it draws attention away from the dark part of the system where the question actually lives. The discipline this paper asks for is not more measurement. It is the relocation of measurement to the thing that determines whether everything else measured will still be true after the people who are currently being measured have gone. That relocation is unglamorous, it does not produce a number that goes up and to the right, and it is the entire game.

9. Coda

There is a sentence that gets said about every well-run institution: it will be fine, because the people who run it care. It is the most dangerous sentence in capital, because it is almost always true at the moment it is said and almost always irrelevant to the question it is answering. Care is a property of the present occupants. The structure is a property of the system. The two are confused precisely because, while the right people are in the room, they are indistinguishable. They become distinguishable only at the transition, and at the transition it is too late to install the difference.

Accumulation is not the gravity of capital. It is the default of a particular design, held in place by a legal default, a structural deadline, and an interpretive asymmetry, none of which is nature and all of which can be redesigned. The alternative is not softer. It is more rigid, more demanding, more legally exact, and considerably older than the thing it is an alternative to. What it asks is not better behaviour. It asks that the good outcome be moved out of the category of behaviour entirely, into the category of what the structure does when no one is choosing anything, because that is the only category that survives the people who built it.

The structures that have survived from the nineteenth century did not survive because they were virtuous. They survived because someone made the virtue unnecessary, machined it into the claim, and then handed the meaning of that machining forward to an institution patient enough to carry it across every gap where a person could have collapsed it. That is the whole of it. Beyond accumulation is not a higher motive. It is a better structure, plus the thing that remembers what the structure is for after everyone who built it has gone. The first part is well understood and rarely built. The second part is barely named and is the layer the next decade is actually about.

References and Notes

The following are real, public, and verifiable sources. Where a figure is cited it is given as an order of magnitude attributable to the named source, not as an audited universal; the structural arguments in this paper do not depend on the precision of any single figure.

1. Elinor Ostrom, Governing the Commons: The Evolution of Institutions for Collective Action, Cambridge University Press, 1990. Nobel Memorial Prize in Economic Sciences, 2009, for the analysis of economic governance, especially the commons. Source of the design principles for enduring common-pool resource institutions referenced in Sections 3 and 4.

1. Marjorie Kelly, Owning Our Future: The Emerging Ownership Revolution, Berrett-Koehler, 2012. Source of the extractive-versus-generative ownership distinction and the five ownership design patterns (purpose, membership, governance, capital, networks).

1. James P. Hawley and Andrew T. Williams, The Rise of Fiduciary Capitalism and subsequent work on the emergence of universal owners. Foundational statement of the universal ownership concept referenced in Sections 3 and 4.

1. Ellen Quigley, Universal Ownership in Practice: A Practical Positive Investment Framework for Asset Owners, Centre for the Study of Existential Risk, University of Cambridge. Treatment of systemic stewardship and the limits of the universal owner lever.

1. Institute for Energy Economics and Financial Analysis, Universal Ownership: Decarbonisation in a Hostile Engagement Environment, 2024. Source for the documented capability limit of the universal owner discussed in Section 4.5.

1. John Fullerton, Regenerative Capitalism: How Universal Patterns and Principles Will Shape the New Economy, Capital Institute, 2015. The framing of economies as living systems, and the Capital Institute's own caution regarding greenwashing within regenerative finance.

1. Kate Raworth, Doughnut Economics: Seven Ways to Think Like a 21st-Century Economist, Random House Business, 2017. Source of the regenerative-and-distributive design framing, cited together with its standard critique regarding under-specified implementation mechanism.

1. Purpose Foundation (Purpose Economy), materials on steward ownership, including the legal guidebook and the framing of awareness, legal complexity, and aligned capital as the three principal hurdles. The term steward-ownership was introduced by the Purpose Foundation in 2017 following study of older steward-owned firms.

1. Carl Zeiss Foundation and Robert Bosch ownership structures, public corporate-history records, as the long-running precedents for the asset lock and the separation of voting and economic rights (Carl Zeiss Foundation established by Ernst Abbe, 1889).

1. Patagonia, Ownership statement and Patagonia Works press release, September 2022, on the transfer to the Patagonia Purpose Trust and the Holdfast Collective; subsequent public reporting (Fast Company, 2025) on the order of magnitude of profit directed to the Holdfast Collective through 2024. Verfassungsblog commentary, The Transfer of Ownership in the Patagonia Case, on the legal character of the structure.

1. James Tobin's intergenerational equity principle for endowments, and the Norwegian Government Pension Fund Global fiscal rule (introduced 2001) as an operational instance of preserving real capital across generations. Sources: IMF eLibrary chapter on the economics of sovereign wealth funds; AQR, The Norway Model; CFA Institute commentary.

1. Industry summaries on evergreen and perpetual private-capital fund structures and their growth (iCapital; Cleary Gottlieb private funds bulletin; Hamilton Lane evergreen strategy materials). Cited for orders of magnitude and structural characteristics only.

1. Public reporting and commentary on the major responsible-business certification discussed in Section 4 (its historical aggregate-score threshold, the 2025 move to a performance-based standard with per-area minimums, a high-profile member declining renewal, and the removal of a reformist chief executive at a large certified multinational). Sources: Raconteur; SOAS blog; ASI Central; Dr. Bronner's company statement, 2025; dairy-industry trade press, 2025.

1. Public reporting on legislative motion toward a dedicated steward-ownership legal form in European jurisdictions during 2025 (Netherlands principles for a steward-ownership corporation, January 2025; German coalition-agreement reference to a new legal form). Cited as active legislative consideration, not as enacted law.

A note on method. This paper deliberately privileges structures with multi-decade or multi-century operating histories over recent designs, because the central claim is about durability across transitions, and only time tests that claim. Recent instruments are cited for direction, not for proof. Where the field's own proponents have stated a limitation of their instrument, that admission is treated as the strongest available evidence and is cited as such.

Future Proof Intelligence . Research . No. IV . MMXXVI