I. Introduction: The Morning After Rebellion
The previous two essays have traversed the following path: the first essay diagnosed the injustice of excess wealth: the functional phase transition of wealth turns it from a resource for living into a control variable; this phase transition spans both the natural-person and legal-entity levels; institutional capture skews the rules; deprivation of survival escalates injustice into a violation of basic rights. The second essay inquired into the logic of action for the deprived: individual reclaiming is a dead end; liberal action theory loses efficacy under deep capture; solidarity is structurally the most promising response; class struggle and rebellion provide a re-founding of legitimacy.
But the endpoint of this path (collective action breaking the old structure) is only another starting point. What comes after the breaking?
This is the question that no revolutionary theory can evade, and it is the most difficult question in history. Breaking an unjust structure is relatively easy; once historical tensions accumulate to a certain degree, the structure will crack on its own. What is difficult is: after it cracks, how to ensure that the new structure will not repeat the same mistakes?
This question is not hypothetical. It has a heavy historical record.
II. The Paradox of Revolution: New Masters, Old Structures
2.1 The Lesson of the Soviet Union
The largest-scale revolutionary experiment of the twentieth century ended with the Soviet Union. The capitalists and landlords of Tsarist Russia were overthrown, and property was "nationalized." But what was the result?
The party bureaucracy (a nascent oligarchic group) replaced the old capitalists and gained control over the allocation of resources. They did not need nominal ownership, because they controlled the power to allocate resources itself. They frequented special supply stores and exclusive sanatoriums, enjoying material conditions unimaginable to ordinary Soviet citizens. Their children entered elite schools, inheriting not bank accounts but social networks and political capital, a more concealed but equally effective form of intergenerational transmission.
The Soviet Union did not eliminate excess holdings. It merely converted excess holdings from the form of private capital to the form of bureaucratic power. The function of wealth did not change; it remained a variable controlling others' feasible sets, now exercised by party bureaucrats in place of private capitalists.
2.2 The Risk of Structure
The Soviet case reveals a structural risk of revolution: without institutional design, the risk of structural replication is inherent in revolution itself.
The French Revolution did dismantle feudalism in the long run, and South Korea's 1987 democratization did open space for sustained reform; revolution or large-scale collective action does not always end in structural replication. But the lesson of the Soviet case is a more precise proposition rather than the claim that "revolution inevitably fails": Revolution replaces persons, but does not automatically replace structures. The core of the old structure — resource concentration → power concentration → rule skewing → further resource concentration — this cycle can restart in new forms under a new regime. The entry point of the Soviet cycle changed from "capital" to "party membership," but those who controlled resource allocation remained the few, and those who were controlled remained the many. Excess holdings still existed, institutional capture still existed, deprivation of survival still existed; the labels alone were new.
If the first essay's diagnosis holds (the problem lies in the phase transition mechanism of wealth itself, not in the specific individuals who hold wealth), then replacing individuals cannot automatically solve the problem. You could send every excess holder to the guillotine, but if the new institution allows resources to concentrate again above the phase transition threshold, new excess holders will emerge within a generation. The disease of the structure will not be cured merely by changing the patients.
2.3 Revolution Is Not the Endpoint
This means: revolution is not the endpoint. Breaking the old structure is only the first step, and not the most difficult one. The most difficult step is: how to design a new structure that makes wealth's re-phase-transition difficult, makes the cost of institutional capture high, and enables the deprived to retain the capacity for sustained intervention.
The second essay argued that solidarity and collective action are necessary conditions for changing the structure. They are also necessary conditions for maintaining the new structure. Institutional design is a structure requiring sustained maintenance rather than a one-time engineering project, and the force of maintenance is, again, solidarity.
III. Preventing Re-phase-transition: The Core Problem of Institutional Design
3.1 Defining the Objective
The goal of the new institution is not to eliminate wealth. This is neither possible (any complex society requires mechanisms for the storage and accumulation of resources) nor desirable (wealth as a resource for living is not in itself unjust; the injustice argued in the first essay occurs after the phase transition, not before it).
The goal is: to prevent wealth from phase-transitioning from a "resource for living" into a "control variable."
More specifically, the new institution needs to ensure:
- The holding of wealth does not cross the phase transition threshold, that is, does not shift from a means of direct use into a variable that alters others' feasible sets.
- Even if someone attempts to cross the threshold, the institution has sufficient anti-capture mechanisms to make success difficult.
- Even if capture partially occurs, the institution has the capacity for self-correction.
These three progressive layers (prevention, resistance, correction) constitute the basic framework of institutional design.
The first essay has already argued that phase transition occurs at both the natural-person level and the legal-entity level: after corporate wealth crosses the threshold, it likewise shifts from an operational resource into a control variable, and the independent personality of the legal entity becomes a cover for the phase transition. Therefore, the constraints of the new institution must simultaneously cover both natural persons and legal entities. For natural persons, what is constrained is the upper limit of the individual wealth stock; for legal entities, the constraint targets the control function of corporate wealth rather than the elimination of companies (companies as a form of productive organization have their efficiency-based legitimacy): ensuring that the company's benefit rights are not exclusively attributed to a small number of shareholders, that the company's decision-making rights are not unilaterally monopolized by capital, and that the company's wealth cannot be converted into a force of capture over rule-making.
3.2 Why Taxation Alone Is Not Enough
The first essay deliberately avoided discussion of policy instruments. But at the level of institutional design, one question must be confronted directly: why is the traditional progressive tax system insufficient to prevent re-phase-transition?
The progressive tax system is an ex post adjustment of stock wealth; it intervenes only after wealth has already concentrated. But the first essay argued the self-reinforcing cycle of institutional capture: once wealth concentrates to the degree that it can influence rule-making, holders can rewrite the tax law itself. Piketty in Capital in the Twenty-First Century documented the empirical manifestation of this cycle: the effective tax rate at the top continued to decline in the late twentieth century, while wealth concentration continued to rise. The force of taxation depends on the will of the legislature, and the legislature can be captured.
This is not to say that taxation is useless; it is one tool in the toolbox. But it is a downstream tool, not an upstream tool. Upstream tools alter the generative structure of wealth (who obtains what share of resources under what conditions), not merely ex post redistribution.
3.3 Upstream Tools: Ownership Dispersion and Universalization of Benefit Rights
If the root of the problem is resource concentration, then the most fundamental preventive mechanism is: to keep resources less concentrated from the very beginning. This means institutional design needs to proceed from two levels.
The first level is ownership dispersion: embedding dispersion mechanisms at the point of wealth generation to prevent capital from over-concentrating in the hands of a few. Specific instruments include the progressive structure of the estate tax, employee stock ownership plans, and strict enforcement of antitrust law.
The second level is universalization of benefit rights: even if ownership dispersion cannot be achieved in one step, institutional arrangements can be made so that part of the returns created by companies is shared by all the people rather than exclusively attributed to shareholders. The key insight of this approach is: the generation of corporate returns depends on multiple inputs: beyond shareholder capital, these include the infrastructure, public education, legal framework, and data contributions provided by the whole society. Attributing part of the benefit rights to all the people is an institutional recognition of the multiplicity of sources of returns.
The two levels form a progressive relationship: ownership dispersion is the fundamental solution, but it requires surgery at the property-rights level, and the political cost is extremely high (the lesson of the Meidner Plan); universalization of benefit rights is a prerequisite scheme: it preserves the existing ownership framework but embeds a universal distribution mechanism at the level of return flows, with lower political cost, and can operate in advance when the conditions for ownership dispersion are not yet ripe.
But universalization of benefit rights has a prerequisite that must be supplied: benefit rights without the guarantee of control rights are a tree without roots. If the people enjoy only dividend rights without any right of intervention in corporate governance, companies can use accounting means to transfer "profits" into "internal pricing" to shrink the dividend base, or they can lobby the legislature to lower the proportion of universal returns. Therefore, universalization of benefit rights must be paired with certain constraints on corporate control rights. These need not amount to a complete governance restructuring; at minimum they ensure that the proportion of universal benefit rights is not unilaterally determined by the company, that the company's political expenditures are subject to restrictions symmetric with the individual wealth ceiling, and that the recipient of universal benefit rights has a supervision mechanism independent of corporate governance.
IV. Historical Cases and Institutional Synthesis
Three historical cases provide the most important empirical material for institutional design. The following analyzes each one's design logic, strengths and limitations, actual implementation, and the resistance encountered, and on this basis discusses whether a more robust scheme can be synthesized from them.
4.1 Rawls: Property-Owning Democracy
Design logic. Rawls in Justice as Fairness: A Restatement (2001) distinguished between two institutional arrangements: welfare-state capitalism and property-owning democracy. Welfare-state capitalism allows wealth to concentrate freely in the market and then corrects it ex post through taxation and transfer payments. Property-owning democracy, by contrast, disperses the broad holding of the means of production from the very beginning, ensuring through education, estate taxes, antitrust, and competition policy that capital does not over-concentrate in the hands of a few. Ownership remains private, but the distribution is sufficiently broad that no one can dominate others' political choices solely by economic power.
Strengths. Rawls's key insight is that the center of gravity of institutions should shift from redistribution to pre-distribution: preventing over-concentration at the point of wealth generation rather than correcting after wealth has concentrated. This is consistent with the first essay's argument. Rawls's scheme is also not limited to ownership dispersion; it also includes complementary institutions such as broad public education, basic health care, and public financing of political campaigns. He explicitly noted that ownership dispersion alone is insufficient; complementary reform of political institutions is also needed to prevent wealth from eroding political equality. This holistic perspective is consistent with the idea of multiple redundancy: anti-concentration cannot rely on a single instrument.
Limitations and implementation. Property-owning democracy has never been fully implemented in any country. Rawls himself positioned it as an ideal type, a normative benchmark for evaluating real institutions, rather than a concrete policy blueprint. Its limitation is: it assumes that the political process is itself sufficient to formulate and sustain broad ownership dispersion policies, but the mechanism of institutional capture argued in the first essay precisely erodes this premise. If the legislature is captured by wealth, who enforces the progressive structure of the estate tax? Who enforces antitrust law? Rawls's scheme is self-consistent under the presupposition of institutional neutrality, but under conditions of deep capture, its enforcement mechanism is vacant.
4.2 Sweden: The Meidner Plan
Design logic. The Swedish labor movement proposed a structural scheme in the 1970s. The "wage-earner funds" (löntagarfonder) plan designed by Rudolf Meidner required all large enterprises to inject 20% of profits each year in the form of new shares into funds managed by trade unions. Over time, the funds would gradually accumulate corporate shares, ultimately achieving workers' collective holding of productive capital.
Strengths. The subtlety of the Meidner Plan lies in its being a self-executing structural scheme: once launched, it does not require sustained legislative maneuvering to maintain, nor does it depend on a potentially captured legislature to make decisions year after year. It alters the distribution of capital through an automatic mechanism of ownership transfer, a point that to this day remains the design with the greatest structural insight among all institutional conceptions.
Limitations, implementation, and resistance. The Meidner Plan was ultimately substantially diluted. Capital owners launched a massive media offensive, depicting it as "the Swedish Gulag." Under pressure, the Social Democrats reduced the 20% to a lower proportion, removed the unions' management authority, and the final scheme became a modest supplementary pension plan. By the early 1990s, the funds were abolished entirely.
The failure of the Meidner Plan is a textbook case of institutional capture. Its core weakness is: converting profits to shares directly transfers ownership, which gave capital its most effective mobilizing slogan: "defend property rights." At the same time, the plan had no anti-capture protection layer: once the political winds changed, the scheme could be repealed by simple majority legislation. The 20% extraction rate was also too aggressive; the one-step design triggered capital's maximum counteroffensive.
But it also left a key lesson: self-executing automatic mechanisms are the right direction, but the automatic mechanism itself also needs to be protected.
4.3 Germany: Co-Determination
Design logic. Germany's co-determination (Mitbestimmung) took another path. Since 1951, German law has required worker representatives on the supervisory boards of large enterprises. The 1976 Co-Determination Act (Mitbestimmungsgesetz) extended this requirement to all enterprises with more than 2,000 employees: labor and capital each occupy half the seats on the supervisory board.
Strengths. Co-determination does not alter the ownership structure, but it alters the decision-making structure. Workers have an institutional seat at the highest decision-making level of the enterprise, making it impossible for the enterprise to unilaterally channel profits in directions that harm workers' interests. It is a practical form of power-dispersion mechanism: by embedding itself within the corporate governance structure, once legislated it is difficult to simply repeal.
Limitations and implementation. Co-determination did limit capital's unilateral power to a certain extent, making Germany's labor-capital relations more stable than those of the US and UK. But its efficacy has clear boundaries: the parity seats on the supervisory board are broken by a capital-appointed chair in the event of a deadlock; co-determination applies only to large enterprises; it does not alter the degree of ownership concentration, only the degree of participation in decision-making. More critically, as German manufacturing globalized and financialized, capital circumvented the constraints of co-determination through cross-border restructuring.
The synthetic lesson of the three cases: the direction of structural solutions is correct (starting from the dispersion of ownership and decision-making rights, rather than relying solely on ex post redistribution), but every concrete scheme is weakened by capital's counteroffensive. Rawls's scheme lacks an enforcement mechanism, the Meidner Plan lacks an anti-capture protection layer, and co-determination can be circumvented by cross-border restructuring. Institutional design requires not only the correct direction but also sufficient capacity to resist capture.
4.4 A Scheme Synthesized from Historical Lessons: Universalization of Benefit Rights
Can a more robust scheme be synthesized from the lessons of the three cases? What follows is an improved combination of the above design logics rather than a "new invention."
Lessons absorbed. From the Meidner Plan comes the self-executing automatic mechanism: profits are automatically extracted at a legally fixed proportion, not depending on annual legislative maneuvering. Rawls contributes the logic of pre-distribution, embedding dispersion mechanisms at the point of wealth generation rather than remedying ex post. Co-determination supplies the idea of embedding in the governance structure: once legislated, the anti-concentration mechanism is embedded in the institution itself and is not easily repealed.
Risks avoided. The most fatal weakness of the Meidner Plan was that converting profits to shares directly transferred ownership, giving capital the mobilizing slogan of "defend property rights." Universalization of benefit rights transfers only the return flow, not ownership and control rights. Shareholders still hold shares, companies still operate according to market logic, but a certain proportion of profits is not distributed to shareholders as dividends but is injected into a public fund held by all the people. Capital's counteroffensive will still occur, but the available ideological weapons are fewer. This is a hypothesis about political feasibility, not a hypothesis about the degree of harm to interests.
Another lesson from the Meidner Plan is: the 20% extraction rate was too aggressive, and the one-step design triggered the maximum counteroffensive. Universalization of benefit rights can operate at a lower initial rate (e.g., 5%), gradually rising after the public has built trust in the mechanism, giving the system and the political process time to adapt.
The lesson from co-determination is: mechanisms can be circumvented by cross-border restructuring. If universalization of benefit rights is implemented in only one country, capital can transfer profits to unregulated jurisdictions through cross-border restructuring. This means the scheme ultimately requires international cooperation or some constraint on cross-border capital flows; otherwise, the institutional design of a single country will be hollowed out by globalization.
Legal basis and precedent. The legal basis of universalization of benefit rights is: the sources of corporate returns are multiple, specifically including shareholder capital, the infrastructure provided by the whole society, public education, the legal framework, and (increasingly important in the digital age) the data contributions of all the people. Attributing part of the benefit rights to all the people is an institutional recognition of the multiplicity of sources of returns.
Here a direct objection must be addressed: companies have already paid for these public goods through taxation, so does demanding a share of profits constitute double taxation? This objection conflates two relations of different natures. Taxation is the payment of citizens/legal entities for public goods; once paid, the relation ends. Benefit rights are the economic realization of ownership; they reflect the institutional arrangement of residual claim rights. The essence of profit distribution is answering "to whom the residual belongs" rather than "paying for inputs." If "having paid through taxation" can exclude the benefit rights of other subjects, then the same logic applies to shareholders: shareholders have "paid" through investment, and by this logic, the relation should also end after payment. But shareholders receive profits not because they "paid," but because the institution allocates the residual claim right to them. This allocation should not be exclusively attributed to shareholders, because the generation of returns depends on multiple inputs.
The Alaska Permanent Fund Dividend, since 1982, has injected 25% of oil revenue into a fund and distributed dividends equally to all state residents each year. It proves that universal dividends are operationally feasible, but it also has clear limitations: the Alaska fund depends on natural resource rents rather than structural profit-sharing, the fund's scale is affected by oil price fluctuations, and it does not touch corporate governance. Universalization of benefit rights extends this logic from natural resource rents to corporate profits.
Control rights constraints. But benefit rights without the guarantee of control rights are a tree without roots. If the proportion of universal benefit rights can be unilaterally adjusted by the company or the legislature, it will be slowly hollowed out. Therefore, universalization of benefit rights must be paired with three control rights constraints:
First, the setting and modification of the universal benefit proportion cannot be unilaterally determined by the company, nor can it be arbitrarily adjusted by the regular legislative process; it requires a higher threshold for modification than ordinary law (such as a constitutional amendment or a referendum), so that the cost of capture is far higher than ordinary legislation.
Second, the company's political expenditures (lobbying, political contributions, media acquisitions) must be subject to restrictions symmetric with the individual wealth ceiling. If the wealth of natural persons is limited below the threshold to prevent political capture, the political expenditures of legal entities must also be limited; otherwise, companies become a channel of capture that bypasses the individual ceiling.
Third, the recipient of universal benefit rights (whether a public fund, a sovereign wealth fund, or a similar entity) must have a supervision mechanism independent of corporate governance and the legislature. The fund's managers may themselves form a new bureaucratic stratum; therefore, the fund internally also requires multiple supervision: independent audits, public inquiry rights, and regular disclosure of operational data.
Relation to ownership dispersion. Universalization of benefit rights is its prerequisite rather than a substitute for ownership dispersion. When the political conditions for ownership dispersion are not ripe, first accumulate economic rights for all the people through institutional embedding at the benefit-rights level, paving the way for deeper ownership transformation in the future. At the same time, it itself constitutes an anti-concentration mechanism: by continuously extracting part of profits for universal sharing, it structurally limits the speed of unlimited accumulation of corporate wealth, slowing the occurrence of legal entity phase transition.
But this progressive relationship is not automatic. The precedent of Norway's sovereign wealth fund shows: universal dividends can operate for decades without leading to any ownership dispersion; the public may form a vested interest in dividends and instead resist deeper transformation. The reason the progression may occur is that universalization of benefit rights creates three things for all the people: economic rights (the material basis of dividends), organizational experience (the practice of operating and supervising public funds), and economic consciousness (the self-perception as owners rather than passive welfare recipients). These three things constitute the material basis and political conditions for driving deeper transformation. But "possible" does not equal "inevitable"; the occurrence of progression requires the sustained impetus of solidarity forces. No institution can automatically lead to justice, and universalization of benefit rights is no exception.
An honest admission. The core mechanism of universalization of benefit rights derives from the Meidner Plan and is not an original conception. The modifications it makes are: removing the element of ownership transfer that most triggers counteroffensive, lowering the initial rate, adding an anti-capture protection layer, and expanding the recipient subject from trade unions to all the people. These modifications are incremental, not paradigmatic. Its fate may be the same as the Meidner Plan: weakened, diluted, and ultimately hollowed out by capital's counteroffensive. Any transformation of the existing system will inevitably be obstructed by entrenched interests, and the struggle will not be easily won. Universalization of benefit rights is not a guaranteed-winning scheme, nor even necessarily a winnable one. It is only an exit we can find under existing institutional constraints and theoretical resources, a direction that is correct and may be feasible under certain conditions. Finding an exit does not equal exiting the maze. But without finding an exit, there is not even the possibility of exiting the maze.
V. Design Principles
From the above theory and experience, principles of institutional design for preventing wealth's re-phase-transition can be extracted. This is not a complete blueprint, which would be neither realistic nor honest. These are several directional constraints.
5.1 Dispersion: Dual-Track Dispersion of Ownership and Benefit Rights
The first principle: the dispersion of economic power must proceed simultaneously on both the ownership and benefit-rights tracks.
Ownership dispersion is the fundamental solution. The ownership of productive assets must be broadly distributed so that no one can dominate others' political choices solely by the scale of assets. Specific instruments may include: the progressive structure of the estate tax (preventing intergenerational concentration), employee stock ownership plans (enabling workers to share in capital returns), public funds holding corporate shares (enabling all the people to share in capital returns), and strict enforcement of antitrust law (preventing market concentration).
Universalization of benefit rights is the prerequisite scheme. When the political conditions for ownership dispersion are not ripe, a mandatory institutional arrangement makes part of corporate profits shared by all the people. It does not alter the ownership structure but embeds a universal distribution mechanism at the level of return flows. The relationship between the two is progressive: universalization of benefit rights accumulates economic rights for all the people, creating conditions for deeper transformation at the ownership level in the future; at the same time, it itself constitutes an anti-concentration mechanism, slowing the accumulation speed of legal-entity wealth through the continuous extraction of part of profits.
The key requirement is: the dispersion mechanisms of both tracks must be structural and automatic, not dependent on annual legislative maneuvering. The lesson of the Meidner Plan is: any mechanism requiring sustained legislative support will be captured. The best mechanism is one that "is set once and then runs itself"; once the proportion of universalization of benefit rights is established through a high-threshold procedure, it cannot be arbitrarily adjusted by ordinary legislation. Of course, what extraction rate counts as "automatic" and effective still needs to be determined empirically in light of the specific economic structure and cannot be set a priori.
5.2 Redundancy: Multiple Anti-Capture Mechanisms
The second principle: no single mechanism can permanently resist capture. Institutions need multiple redundancy, resting on multiple independent lines of defense rather than a single one.
If antitrust law is captured, ownership dispersion still serves as backup, and universalization of benefit rights still acts as a buffer; if those in turn are eroded, co-determination and public funds remain as further checks. Each mechanism can be individually breached, but the difficulty of breaching all mechanisms simultaneously increases significantly.
A condition needs to be qualified: only when the mechanisms are mutually independent does the difficulty of simultaneous breaching increase significantly. If all mechanisms depend on the same infrastructure (such as judicial independence), the failure of one mechanism may have a cascading effect on other mechanisms. This is precisely the reason the redundancy principle requires that the mechanisms be as mutually independent as possible in design: antitrust law depends on the judiciary, co-determination depends on labor law, universalization of benefit rights depends on a constitutional-level high-threshold procedure; they depend on different institutional bases, so capture in one domain does not automatically lead to failure in others.
What particularly needs emphasis is the symmetric constraint on natural persons and legal entities. If the wealth of natural persons is limited below the threshold to prevent political capture, the political expenditures of legal entities must also be symmetrically limited; otherwise, companies become a channel of capture that bypasses the individual ceiling. This symmetry is a concrete application of the redundancy principle: one cannot defend only at one end while leaving the other end unchecked.
This is consistent with the logic of checks and balances in federalism; Madison's insight in The Federalist Papers, No. 51, remains valid: "Ambition must be made to counteract ambition." But in the context of this essay, checks and balances are not only horizontal checks among branches of government; they must also include vertical checks between capital and labor, between ownership and decision-making rights, between natural persons and legal entities, and between market and democracy. What needs to be guarded against is that multiple mechanisms do not mean the more the better; the coordination costs and mutual constraints among mechanisms are also real constraints.
5.3 Transparency: Phase Transition Monitoring
The third principle: institutions need to be able to monitor the degree of wealth concentration and issue warnings before the phase transition threshold is reached.
The first essay argued for the existence of the phase transition threshold; no precise value is needed, only the knowledge that it exists somewhere. Institutional design can utilize this argument: establish systematic monitoring of wealth concentration (such as the wealth distribution tracking done by Piketty's team), set warning lines, and when concentration approaches the danger zone, automatically trigger anti-concentration measures (such as raising the estate tax, launching antitrust investigations, expanding public fund shareholding, or raising the proportion of universal benefit rights).
Monitoring must simultaneously cover both the natural-person level and the legal-entity level. Monitoring at the natural-person level tracks the concentration of individual wealth and the scale of political expenditures; monitoring at the legal-entity level tracks the concentration of corporate market capitalization, changes in market share, political lobbying expenditures, and cross-corporate equity affiliation networks. As diagnosed in the first essay, legal entity phase transition and natural person phase transition are different carriers of the same mechanism; if the monitoring system covers only one side, wealth will flow from the monitored side to the unmonitored side.
The key phrase is "automatic triggering": a reflex built into the institution rather than depending on the legislature's annual decisions. But an operational difficulty needs to be acknowledged: the first essay argued for the existence of the phase transition threshold but did not give a precise value, which means "automatic triggering" cannot switch on at a precise degree like a thermostat. A more accurate analogy is the immune system: the immune system does not activate at a precise threshold but gradually activates when the threat level exceeds a certain fuzzy zone: first warning signals, then investigative responses, and finally full-scale intervention. Anti-concentration mechanisms likewise require gradient design: when concentration enters the "zone of concern," information warnings and investigation launches are triggered; when concentration enters the "danger zone," the gradual escalation of anti-concentration measures is triggered. The fuzziness of the threshold does not cancel the necessity of automatic triggering, but it requires that the triggering mechanism be gradient rather than binary. In practice, the independence and credibility of the monitoring data itself are equally critical; otherwise, the monitoring system itself will be captured or hollowed out.
5.4 Correction: Structural Exit Mechanisms
The fourth principle: institutions must preserve channels of sustained intervention for the deprived, as a structural self-correction mechanism rather than charity.
The second essay argued that solidarity and collective action are necessary conditions for changing the structure. They are equally necessary conditions for maintaining the structure. The new institution cannot assume that a one-time institutional design is sufficient to prevent degradation; it must build in the institutional participation rights of the deprived, so that when concentration trends emerge, those affected have the capacity to signal within the structure and push for correction.
This means: union rights, community self-governance, direct democracy mechanisms, and legal protection of civil disobedience constitute the institution's immune system. Without them, the institution will slowly degrade under the sustained pressure of capital; with them, the institution at least retains the possibility of self-correction.
Under the framework of universalization of benefit rights, this principle has an additional concrete meaning: the people, as holders of benefit rights, must enjoy supervision rights over public funds: the right to know (access to fund operational data), the right to inquire (consultation of fund investment details at any time), and the right to accountability (interpellation and impeachment of fund managers). The managers of public funds may themselves form a new bureaucratic stratum; this principle requires: the people must be active supervisors rather than merely passive dividend recipients. Passively received benefit rights can be hollowed out at any time; actively supervised benefit rights alone have sustained vitality. It needs to be acknowledged that the institutionalization of participation rights does not automatically translate into the actual occurrence of participation; there remains a distance between the existence of rights and the exercise of rights.
VI. The Core Paradox: Who Guards the Guardians?
6.1 The Recursive Problem
The above four principles appear reasonable. But they face a fundamental challenge, one symmetrical to the "self-protection of structure" in the second essay.
Any anti-capture mechanism may itself be captured.
The enforcement of antitrust law depends on the independence of judges and regulators, but judges can be ideologically appointed, and regulators can be bought through revolving doors. The automatic mechanism of ownership dispersion requires legislation to set it, but legislation can be modified. The legal framework of co-determination requires political will to maintain, but political will can be eroded. Public funds require managers, and managers can form a new bureaucratic stratum. The proportion of universalization of benefit rights requires a high-threshold procedure to protect it, but the threshold itself can be gradually eroded; each "small adjustment" appears harmless, but accumulated, it suffices to hollow out the entire mechanism. Companies can use accounting means to transfer profits into internal pricing, shrinking the dividend base; they can circumvent the jurisdiction of co-determination and benefit-rights distribution through cross-border restructuring; they can render actual control rights opaque through networks of shell companies.
This is the recursive problem of institutional design: you have designed a mechanism to prevent capture, but who prevents this mechanism from being captured? You can design another meta-mechanism to supervise it, but who supervises the meta-mechanism? This recursion has no endpoint.
This recursion is not a design oversight that can be patched by a more ingenious scheme. It is an intrinsic property of the logic of power. Any mechanism that constrains power is itself a kind of power — the power to set rules, the power to interpret rules, the power to enforce rules — and therefore itself falls into the category that can be captured. You can disperse power, but the manner of dispersion requires someone to set and maintain, and the power to set and maintain can be captured. You can establish independent agencies, but the determination of "independence" requires some standard to measure, and the power to interpret the standard is itself a kind of power. This is not a problem of a certain type of institutional design but a structural limitation faced by all schemes that constrain power from within the institution. Acknowledging this is to redirect institutional design from "seeking the perfect scheme" to "designing schemes that can be sustained and maintained," rather than to abandon it.
6.2 No Permanent Solution
The honest answer is: there is no once-for-all institutional solution.
Any institution, no matter how ingeniously designed, will face the pressure of degradation over time. The pressure of degradation comes from two directions. One is endogenous: interest groups continuously search for the crevices of institutions, the loopholes of rules, and the blind spots of regulation. Institutional capture is a continuous process, not a one-time event, like water continuously seeking cracks in a dam. The other is exogenous: the world changes, and the conditions for which the institution was designed no longer exist. The welfare-state institutions designed after the war appeared inadequate in the face of financialization and globalization, because they could not foresee the magnitude of capital mobility forty years later. Antitrust law was effective in the industrial age; in the era of the platform economy, the meaning of "monopoly" needs to be re-understood. The degradation of institutions is not only being breached but also being bypassed by the times.
History is not short of precedents of institutional degradation. The checks and balances of the Roman Republic — consuls, senate, tribunes — operated for over four hundred years and ultimately collapsed into imperial rule under the pressure of wealth concentration and military expansion. The postwar Western welfare-state consensus lasted about thirty years and was systematically dismantled after the 1970s. The commercial oligarchy of the Republic of Venice endured for nearly a thousand years and ultimately declined because entrenched interests refused to open up to new capital. The common pattern of these cases is: institutions are effective when established, accumulate degradation pressure during operation, and are breached at a certain critical point.
This means the goal of institutional design should be "building a dam that can be continuously repaired, and ensuring that someone is always repairing it," rather than "building an unbreachable dam," which does not exist.
"Someone is always repairing it": this "someone" is the solidarity argued in the second essay. Institutions buy time; solidarity provides energy. Solidarity without institutional guarantees is fragile: individual actors, without legal protection, organizational tools, or information channels, can be crushed one by one. Institutions without solidarity to maintain them degrade: anti-capture mechanisms can be flawless on paper, but if no one continuously uses them, supervises them, and raises the alarm when they are eroded, they will be hollowed out in silence. The two are not merely complementary but mutually constitutive: institutions provide tools and channels for solidarity, and solidarity provides the energy for institutions to operate continuously.
6.3 Structural Sleeplessness
Acemoglu and Robinson in Why Nations Fail distinguished between "extractive institutions" and "inclusive institutions". Inclusive institutions disperse power, protect property rights, and provide public goods; extractive institutions concentrate power, plunder resources, and exclude the participation of the majority. Their core finding is: institutions are not immutable; inclusive institutions can degenerate into extractive institutions, and vice versa. The key lies in "critical junctures": major events (wars, crises, technological changes) open windows for institutional change, and at this point the direction of collective action determines whether institutions move toward inclusiveness or extractiveness.
This is consistent with the argument of this essay. Institutional design is not done in a vacuum; it occurs in the window period when the old structure has been broken and the new structure has not yet taken shape. The choices made in this window period will long affect the trajectory of the structure. But the choice in the window period is not the endpoint; it will face sustained erosion. Each generation faces the same set of questions: Is wealth reconcentrating? Are institutions being captured? Do the deprived still retain the capacity for intervention?
An important qualification needs to be added: Acemoglu and Robinson's framework also contains a dimension: the degree of political centralization. They explicitly argue that without a certain degree of political centralization, inclusive institutions cannot function, because dispersed political power cannot enforce property-rights protection and anti-concentration measures. The "dispersion" in this essay refers to the dispersion of economic power, not the fragmentation of political power. On the contrary, the enforcement of anti-capture mechanisms requires sufficient state capacity, including judicial independence, regulatory enforcement capacity, and public audit capacity. The dispersion of economic power and the effective centralization of political power are not contradictory but complementary: the dispersion of economic power prevents a few from capturing the state, and the centralization of state capacity ensures that anti-capture mechanisms can be enforced.
Whether the current global situation constitutes such a critical juncture is not a question this essay can answer. But certain symptoms are visible: wealth concentration in multiple developed countries has reached its highest level since World War II; signs of institutional capture are everywhere from tax law to media ownership; technological change is reshaping the relationship between labor and capital. If this is a critical juncture, then the direction of institutional design at this moment will long affect the trajectory of the structure. If not, the pressure will continue to accumulate.
Any transformation of the existing system will inevitably be obstructed by entrenched interests. The Meidner Plan was diluted and ultimately abolished. Co-determination has been circumvented through cross-border restructuring. Rawls's property-owning democracy, for its part, never left the thought experiment. These are repeated replays of capital's counteroffensive rather than isolated failures. Every anti-concentration measure triggers a round of counteroffensive, and the strategies of counteroffensive include media mobilization, legal challenges, capital flight, and political bribery. The struggle will not be easily won. But the significance of institutional design lies in providing tools for the struggle and a position for sustained contestation rather than in guaranteeing victory.
This is why the conclusion of the second essay, that solidarity is structurally the most promising response, applies not only to changing the old structure but also to maintaining the new structure: there are no eternal guardians, only eternal vigilance. The justice of structure is not a state but a process.
VII. Conclusion: The Closure of Three Essays
7.1 From Diagnosis to Design
The three essays constitute a complete logical chain:
The first essay (diagnosis): The functional phase transition of wealth strips excess holdings of moral protection. When wealth shifts from a resource for living into a control variable, institutional capture skews the rules, and deprivation of survival escalates injustice into a violation of basic rights. Phase transition spans both the natural-person and legal-entity levels; any constraint targeting only one level cannot block the phase transition mechanism. The structure is unjust.
Diagnosis necessarily points to action: having confirmed the injustice of the structure, the next step is how to change it.
The second essay (action): Individual reclaiming is a dead end; it does not touch the structure. Liberal action theory loses efficacy under deep capture. The coordination dilemma explains why the deprived do not automatically unite. The self-protection mechanisms of structure (hegemony, exhaustion, coercion) sustain the unjust structure. Solidarity is structurally the most promising response. Class struggle is the most fundamental historical mechanism by which the oppressed change the structure. The legitimacy of rebellion derives from the oppressiveness of the structure itself.
After action breaks the old structure, the truly difficult question emerges: how to avoid the new structure repeating the same mistakes.
The third essay (design): The lesson of the Soviet Union is that revolution replaced the persons while the structure remained as it was. Since the cause of the disease lies in the phase transition mechanism of wealth, the prescription must also be structural. The first essay has already diagnosed that phase transition spans both the natural-person and legal-entity levels; on this basis, constraints must symmetrically cover both. Hence the dual-track design is proposed: ownership dispersion as the fundamental, universalization of benefit rights as the prerequisite; the latter does not touch the property-rights framework but only embeds universal distribution in the return flows, supplemented by three control rights constraints. Four principles then unfold: dual-track dispersion, multiple redundancy, phase transition monitoring, and structural correction. Yet the guardians themselves may be captured, and a permanent solution does not exist. Institutions procure only time; the energy that sustains justice always comes from solidarity. The justice of structure is a process, not a state.
This chain is not circular. Circularity is returning to the origin; this chain is a spiral ascent. Each iteration operates at a higher level: the diagnosis of the first essay was completed within the normative framework of institutions, using Rawls's diagnostic tools to point out that institutions deviate from norms. The action of the second essay turned to the logic of collective action after the institutional framework failed. The design of the third essay inquires into the maintenance of the new structure after collective action has broken the old structure. If the spiral continues, the next round of diagnosis will face a more complex question: where does the new institution deviate from the new norms, given that the new norms themselves need to be established in practice and cannot be given a priori.
7.2 Back to Action
The endpoint of this logical chain returns to the starting point: action.
Diagnosis tells us the structure is unjust. Action tells us that change requires solidarity. Design tells us the new structure requires sustained maintenance. Maintenance requires solidarity.
This is the dialectic of structure, not circular reasoning. No institutional arrangement settles the question once and for all. The wealth phase transition mechanism diagnosed in the first essay will not disappear because of a one-time institutional design; it will reappear in new forms within the new structure. The logic of solidarity argued in the second essay will not become obsolete because of the establishment of the new structure; it still needs to operate within the new structure, only shifting from "breaking the old structure" to "maintaining the new structure."
Solidarity under the new structure differs substantively from solidarity under the old structure. Solidarity under the old structure was bare-handed, without institutional tools, without organizational experience, without monitoring capacity. Solidarity under the new structure has gained three things: institutional tools (legal supervision channels, statutory participation rights, the accountability mechanism of public funds), organizational experience (the organizational capacity accumulated through the practice of operating and supervising public funds), and monitoring capacity (the early warning provided by the phase transition monitoring system). These differences make the solidarity that maintains the new structure an upgrade rather than a simple repetition — from "the force that breaks the structure" upgraded to "the immune system that maintains the structure."
The process is continuous: diagnosis, action, design, then re-diagnosis, re-action, and re-design. Justice is continually approximated rather than finally completed.
This is honesty, not pessimism. The three essays have traversed a path: from diagnosis to action, from action to design, from design back to action. This path has no endpoint. Universalization of benefit rights may be hollowed out; anti-capture mechanisms may be captured; solidarity may be dismantled. But universalization of benefit rights may also not be hollowed out; anti-capture mechanisms may also continue to operate; solidarity may also re-coalesce after each dismantling. Which side it specifically goes to is not decided in theory but in struggle. What theory can do is point out for the struggle an exit in the correct direction. The path out of the exit must be walked by the actors themselves.
The justice of structure cannot be declared "complete" on some day; it is a responsibility that each generation must take up anew. The previous two essays argued why action is necessary; the remaining question is what is needed after action. But the three essays together ultimately point in one direction: to resist wealth's re-phase-transition and to maintain the justice of structure continuously, collectively, and in an organized manner.
This is a condition, not a choice; as long as the phase transition mechanism of wealth exists, this condition will not change.