Framework note: This essay applies the full arity triad to employment structure. Signal legibility is persistence made explicit — can the system state be tracked by all parties? Revenue coupling is distinction made explicit — is the gap between contribution and compensation visible? Bounded engagement windows are selection made explicit — does the system produce criterion-governed verdicts at defined boundaries? The irreducibility claim is the role-arity diagnostic applied to labor: all three levels of explicitness are required, and removing any one produces a characteristic failure. The hourly wage is the static income fallacy — treating a dynamic, process-level quantity as a fixed invariant. See Running Comes First for the ontological claim and Role-Arity and the Structure of Emergence for the diagnostic.
The Static Income Fallacy
Three Primitives for Structurally Sound Employment
Tom Passarelli February 2026
Working Paper v2
The Static Income Fallacy is the treatment of a worker’s income as a predetermined quantity — a number negotiated once and paid repeatedly regardless of system performance — rather than a value dynamically generated at each distribution from the transactional revenue of the system in which the worker operates. This severance of compensation from revenue is an engineering failure that produces systematic misallocation of value and necessitates every major compensatory mechanism in modern employment (annual reviews, performance bonuses, managerial authority, retention packages) that would be unnecessary under a structurally sound arrangement.
Abstract
The standard employment contract makes an invalid assertion: that a worker’s income is static—a fixed quantity detached from the performance of the system in which the worker operates. This paper argues that the hourly wage model is not merely imperfect but structurally unsound, constituting an engineering failure that produces predictable dysfunction across all employment relationships. The paper identifies three primitives—signal legibility, revenue coupling, and bounded engagement windows—and demonstrates that each is irreducible: removing any one produces a known and documented failure mode already present in the labor market. The employer-employee hierarchy, typically treated as a cultural or organizational question, is shown to be a structural consequence of information asymmetry—itself downstream of signal illegibility. The paper situates this framework against the principal-agent, profit-sharing, and open-book management literatures, identifying what it adds that each omits: the irreducibility claim. The paper addresses the major structural objections—risk transfer in downturns, roles without direct transactional endpoints, regulatory compatibility, and measurement brittleness—and shows that each either rests on a false premise or identifies a modeling failure the framework corrects. The paper then names the honest case for hourly pay’s persistence: not theoretical superiority but enforcement simplicity. Hours times rate is trivially verifiable; revenue coupling historically required forensic access to business internals. This enforcement advantage, however, is technological rather than structural, and the paper proposes AI-mediated auditing as the first technology capable of producing low-bias, reproducible signal legibility at near-zero cost — making the enforcement simplicity advantage of hourly pay, for the first time, surmountable. A worked contract template, open falsification criteria, and actor-specific implications for adoption are provided.
Core Definitions
The Static Income Fallacy: The erroneous treatment of a worker’s income as a fixed quantity independent of the system in which the worker operates—an abstraction that severs the relationship between compensation and the enterprise’s actual revenue generation, producing systematic misallocation of value and a compensatory architecture of workarounds (annual reviews, performance bonuses, retention packages, managerial authority) that would be unnecessary under a structurally sound arrangement.
The Triadic Claim: Every structurally sound employment arrangement—defined as one in which incentives are aligned, information is symmetric, and commitment is mutual—requires exactly three irreducible primitives: signal legibility (shared observability of system state), revenue coupling (structural linkage between income and a modelable transactional endpoint), and bounded engagement windows (defined commitment periods with predefined terms for early termination). No subset of two is sufficient; each omission produces a known failure mode. Roles that appear to lack transactional endpoints are modelable through cost-of-failure valuation, inherited upstream coupling, or opportunity-cost pricing.
Outcome Evaluation Function: When this paper claims the triadic framework produces “better outcomes” than the hourly-at-will-opaque alternative, “outcomes” is defined as: (1) allocative accuracy of compensation relative to contribution as modeled by the coupling rule, evaluated over engagement-window time rather than instantaneously, (2) incentive alignment toward enterprise value creation, (3) reduction of renegotiation friction via structural self-correction at window boundaries, and (4) reduction of catastrophic downside via legible risk and bounded exit costs. The evaluation function explicitly excludes minimizing income variance at all costs — income variance under revenue coupling is a feature (honest signal), not a defect (instability). An arrangement that produces lower variance by concealing risk is not outperforming; it is obscuring.
Formal Propositions
Proposition 1 (The Invalid Assertion). Any employment contract in which compensation is fixed per unit of time asserts, implicitly, that the worker’s marginal contribution to enterprise revenue is constant. This assertion is false for all workers, because the marginal product of labor is a function of the worker and the system, not the worker alone. Roles that appear to lack direct transactional endpoints (support, prevention, infrastructure) have modelable endpoints through cost-of-failure valuation, inherited upstream coupling, or opportunity-cost pricing; the claim that such roles are exempt reflects a modeling failure, not a structural limitation.
Proposition 2 (Irreducibility). The three primitives—signal legibility, revenue coupling, and bounded engagement windows—are mutually independent and jointly necessary. For each pair of primitives, there exists a documented employment arrangement exhibiting that pair without the third, and each such arrangement produces a characteristic dysfunction: (a) legibility + coupling without windows yields indefinite mutual hostage-taking; (b) coupling + windows without legibility yields blind risk transfer; (c) legibility + windows without coupling yields perverse incentives against efficiency.
Proposition 3 (Hierarchy as Artifact). The employer-employee hierarchy is not an organizational design choice but a compensatory mechanism for information asymmetry between parties. In the limit of full signal legibility, hierarchical management converges to peer coordination, because the informational advantage that makes directive authority functional ceases to exist.
Proposition 4 (Self-Correction). The triadic framework produces a self-correcting arrangement in which engagement window boundaries generate data that updates signal legibility, which informs renegotiation of revenue coupling terms. Traditional employment lacks a structural self-correction mechanism; error correction requires one party to initiate renegotiation, imposing social costs that suppress legitimate adjustment.
Proposition 5 (The Rivalry Dissolution). Revenue coupling reverses the incentive structure that makes competitive replication rational. A worker whose income is tethered to enterprise transactional revenue has a stronger incentive to optimize the existing system than to exit and replicate it. The rivalry objection to signal legibility is therefore self-defeating under the triadic framework: it assumes adversarial incentives that the framework itself eliminates.
Proposition 6 (Reproducible Legibility). When signal legibility is produced by a deterministic or near-deterministic auditing process applied to shared artifacts, the resulting assessment is independently reproducible by any party, eliminating the requirement for trust in the auditor and reducing the cost of verification to near zero. Reproducibility here means: given a fixed set of artifacts, a specified model version, logged prompts, and defined audit parameters, any party running the same protocol produces the same assessment — and deviations between runs are themselves auditable artifacts rather than undetectable noise. This property — reproducible legibility — is uniquely available through AI-mediated auditing and has no equivalent in human-mediated signal production (consulting, management reporting, peer review), all of which require trust in the intermediary. Reproducible legibility extends beyond system-state observation to compensation assessment: an AI pointed at revenue data, contribution artifacts, and before-and-after performance deltas can produce an evidence-derived compensation estimate that neither party authored, both parties can verify, and neither party can dismiss without engaging the underlying evidence.
Proposition 7 (Enforcement Simplicity as the Persistence Mechanism). The persistence of hourly compensation is not explained by any theoretical advantage but by enforcement simplicity: hours times rate is trivially verifiable by either party, any labor board, or any court, with no access to business internals required. Revenue coupling historically required forensic access to accounting systems that employers resisted providing and regulatory infrastructure not optimized for variable compensation. This enforcement advantage is technological, not structural. AI-mediated auditing makes it feasible for the first time to verify revenue-coupled compensation as simply and reproducibly as counting hours — at which point hourly pay retains no advantage, operational or theoretical, over the triadic alternative.
I. The Invalid Assertion
Every hourly wage contract contains an implicit assertion: that the value of the worker’s contribution to the enterprise is a constant. The employer offers a rate—say, $40 per hour—and in doing so claims that each hour of the worker’s labor produces, on average, value equivalent to or greater than $40, and that this relationship holds regardless of what the enterprise earns, how the market shifts, or whether the worker’s specific contribution becomes the binding constraint on the company’s growth or its most redundant input.
No one believes this is literally true. The hourly rate is understood, by everyone involved, as an administratively convenient averaging rule — a fiction introduced to make labor transactions computationally tractable at the cost of representing reality. The question is whether the distortions produced by this convenience are small enough to tolerate. This paper argues they are not. The hourly wage treats income as though it exists in a vacuum, unconnected to the living system in which the work occurs. It asserts that income is not growing, not declining, not tethered to anything real. That is the fiction. That is the point at which the entire architecture fails — not because anyone is confused about reality, but because the approximation produces predictable, structural dysfunction that compounds over the life of every employment relationship that uses it.
To be clear: the claim is not that hourly employment is impossible. People work hourly jobs and survive. The claim is that hourly employment is always suboptimal. It is never the best available arrangement for either party. It is a Rube Goldberg machine—a needlessly complex set of compensatory mechanisms (annual reviews, performance bonuses, equity vesting schedules, retention packages, cost-of-living adjustments) bolted onto a fundamentally broken foundation. Each of these mechanisms exists solely to patch the distortions introduced by decoupling income from revenue. None of them would be necessary if the foundation were sound.
The economics literature has long recognized that the marginal product of labor is context-dependent (see Clark, 1899; Hicks, 1932; more recently, Lazear, 2000 on sorting effects and Bartel et al., 2007 on IT-enhanced productivity variation). The same developer might generate $50 per hour equivalent in one system and $500 per hour in another, not because their skill changed but because their skill happens to be the binding constraint in one system and redundant capacity in the other. The hourly rate erases this context entirely. It prices the worker as though they were a commodity—interchangeable, system-independent, valued in isolation. This is a convenience for the market, not a description of reality. It is a coordination shortcut that sacrifices accuracy for liquidity.
The question, then, is not whether the current model can be improved. It is why the current model persists when a structurally superior alternative is available. The answer requires identifying what the superior alternative actually consists of.
II. Relation to Prior Work
The triadic framework does not emerge from a vacuum. Several substantial literatures have addressed parts of the problem, and locating this paper’s contribution relative to them is essential.
Principal-Agent Theory
The canonical formulation (Jensen & Meckling, 1976; Holmström, 1979; Grossman & Hart, 1983) diagnoses the core information asymmetry between employer and worker and prescribes monitoring, incentive alignment through variable compensation, and contractual design to mitigate moral hazard. The principal-agent framework correctly identifies that the employer (principal) and worker (agent) have divergent interests under conditions of asymmetric information. However, its prescribed solution is to manage the asymmetry through monitoring and incentive design—not to dissolve it through shared signal legibility. The principal-agent tradition accepts the hierarchy as given and optimizes within it. This paper argues that the hierarchy itself is the artifact to be eliminated.
Profit-Sharing and Gainsharing
Weitzman’s The Share Economy (1984) argued that profit-sharing arrangements produce macroeconomic stability by making labor costs flexible, reducing the incentive to lay off workers during downturns. Empirical studies (Kruse, 1993; Blasi, Freeman & Kruse, 2013) have broadly supported the claim that profit-sharing correlates with higher productivity and lower turnover. The Scanlon Plan, Rucker Plan, and Improshare represent specific implementations of gainsharing tied to productivity metrics rather than raw profit.
This paper’s framework differs from Weitzman in two critical respects. First, Weitzman advocates coupling to profit (revenue minus costs), which is manipulable through cost allocation; this paper advocates coupling to transactional revenue at a specific endpoint, which is legible and auditable. Second, Weitzman treats profit-sharing as a standalone policy prescription. This paper demonstrates that revenue coupling without signal legibility produces blind risk transfer (the commission-based sales dysfunction), and revenue coupling without bounded engagement windows produces indefinite mutual hostage-taking (the startup equity dysfunction). Revenue coupling is necessary but not sufficient. The irreducibility claim is the novel contribution.
Open-Book Management
Jack Stack’s The Great Game of Business (1992) and the open-book management movement advocate sharing financial information with all workers. Case studies (e.g., Springfield Remanufacturing Corporation) demonstrate that transparency can produce dramatic improvements in worker engagement and operational performance. The movement correctly identifies signal legibility as a lever for organizational improvement.
However, open-book management does not structurally couple transparency to compensation. Workers in an open-book company may see the financials without having their income tied to them. This produces an informed but structurally powerless workforce—workers who can see that the enterprise is underperforming but whose hourly income remains unchanged regardless. Open-book management achieves signal legibility without revenue coupling, which this paper identifies as a specific failure mode: the informed-but-misaligned arrangement in which visibility produces frustration rather than agency.
Worker Cooperatives
Cooperative structures (Mondragon, John Lewis Partnership, the Emilia-Romagna cooperative ecosystem) achieve all three primitives simultaneously: worker-owners have signal legibility (access to financials), revenue coupling (profit distribution), and bounded governance cycles (annual elections, defined terms). The cooperative model is proof of concept that the triad is functional.
However, cooperatives require ownership transfer, which introduces capital requirements, governance complexity, and legal restructuring that make the model inaccessible for most employment relationships. This paper’s framework explicitly does not require ownership. It achieves the functional benefits of cooperative structure—aligned incentives, symmetric information, mutual commitment—through contractual design rather than ownership transfer. The framework is cooperativism without the cooperative.
What This Paper Adds
The novel contribution is the irreducibility claim: that signal legibility, revenue coupling, and bounded engagement windows are three independent primitives, not a menu of interchangeable options, and that the absence of any one produces a characteristic and predictable dysfunction. No prior framework makes this claim. Principal-agent theory optimizes within asymmetry rather than dissolving it. Profit-sharing isolates revenue coupling from its necessary preconditions. Open-book management isolates signal legibility from its necessary consequences. Cooperatives achieve the triad but require ownership transfer. This paper identifies the minimal structural requirements for sound employment and demonstrates that they are jointly necessary and individually insufficient.
III. The Three Primitives
Complex systems, when properly understood, reduce to a small number of irreducible components. The argument of this paper is that structurally sound employment reduces to exactly three: signal legibility, revenue coupling, and bounded engagement windows. These are primitives in the formal sense—they cannot be derived from each other, and no subset of two is sufficient to produce a functional arrangement.
A. Signal Legibility
Signal legibility is the capacity of all parties in an employment relationship to observe and interpret the state of the system in which they operate. This includes revenue flows, cost structures, customer acquisition metrics, operational bottlenecks, and the causal relationship between individual contributions and enterprise-level outcomes.
In the standard employment arrangement, signal legibility is radically asymmetric. The employer sees the full picture—or at least a substantially larger portion of it—while the worker sees only their own tasks, their own compensation, and whatever narrative the employer chooses to provide. This asymmetry is not incidental. It is the structural foundation of the employer-employee hierarchy. The employer “manages” the worker precisely because the employer possesses information the worker does not. Management, in its most common form, is not coordination. It is the exercise of an informational advantage.
Signal legibility is the precondition for everything else. A worker cannot evaluate whether their compensation is fair without visibility into what the enterprise earns. They cannot assess whether their role is strategically valuable or operationally redundant without understanding the system’s constraints. They cannot negotiate from a position of knowledge without signal legibility, which means every negotiation conducted under signal illegibility is structurally biased toward the party with more information—invariably, the employer.
The immediate objection to signal legibility is rivalry: if the worker can see the system, they can leave and replicate it. This objection is empirically weak and legally addressed. Non-disclosure agreements, non-compete clauses (where enforceable), and trade secret protections already exist as instruments for managing the informational risks of transparency. A company that cannot afford to show its workers how the business operates is not protecting a legitimate competitive advantage. It is protecting an information asymmetry that subsidizes its ability to underpay. The rivalry concern is a Band-Aid solution to a problem that signal legibility itself would dissolve. If the worker can see the system clearly, and the worker’s compensation is coupled to the system’s performance, the worker’s incentive is to improve the system—not to replicate it. Rivalry logic assumes adversarial actors. Revenue coupling creates aligned actors. The concern evaporates once the structural incentives are corrected.
B. Revenue Coupling
Revenue coupling is the structural linkage between a worker’s income and the revenue generated at a specific transactional endpoint within the enterprise. The emphasis on transactional endpoint is deliberate. The claim is not that workers should share in “the company’s revenue” as an undifferentiated aggregate. It is that each worker’s compensation should be tethered to an identifiable point in the system where revenue is generated—a product sold, a service delivered, a contract executed. This specificity is essential because it makes the coupling legible (the worker can see the transaction their income depends on) and fair (the worker is compensated for what they actually influence, not for macroeconomic fluctuations or the performance of unrelated divisions).
Revenue coupling eliminates the fundamental invalid assertion of the hourly model. Under revenue coupling, income is not static. It is not arbitrary. It is not detached from reality. It rises when the enterprise succeeds and declines when the enterprise struggles. This is not a punishment for the worker during downturns—it is an honest representation of the situation. The hourly worker whose company is failing still receives $40 per hour until the day they are terminated. The revenue-coupled worker sees the decline in real time, has the signal legibility to diagnose why, and has the structural incentive to intervene. The hourly worker is a passenger. The revenue-coupled worker is a participant.
This model does not require ownership. It does not require equity. It does not require that every worker have a stake in the entire enterprise. A company operating two entirely distinct business lines under a single brand can couple workers in Division A to Division A’s transactional revenue and workers in Division B to Division B’s transactional revenue. The compartmentalization is modular. What matters is that the coupling exists and that it points at something real—a measurable, auditable transaction rather than a negotiated abstraction.
C. Bounded Engagement Windows
A bounded engagement window is a defined period of commitment during which neither party may unilaterally terminate the arrangement without a predefined cost. The minimum viable window is approximately two weeks, though the specific duration is negotiable between parties. What is not negotiable is the existence of the window itself.
The at-will employment framework—in which either party can terminate the relationship at any moment for any reason—is structurally incompatible with productive collaboration. Work is an experiment. Every meaningful contribution requires a hypothesis, an implementation period, and an observation window. At-will employment allows the experiment to be terminated before results are observed, which means the worker can never commit fully to a long-term strategy because the strategy might be cancelled before it bears fruit. The result is systematic bias toward short-term, visible contributions over long-term, structural ones. Workers under at-will arrangements rationally prioritize looking busy over being effective, because effectiveness requires time horizons that the arrangement does not guarantee.
Bounded engagement windows function identically to experimental protocols in science. An experiment has a start date and an end date. Data is collected during the observation period. Conclusions are drawn at the boundary. Early termination is permitted only when the acceptance criteria defined at the outset are overwhelmingly met—and critically, those criteria must be defined before the experiment begins, not speculated about during its execution. Terminating an experiment because intermediate results are disappointing is not science. It is anxiety. The same logic applies to employment: terminating a worker before their engagement window closes because early results are ambiguous is not management. It is a failure to commit to the observational structure that produces knowledge.
The engagement window also provides the scaffolding for the graduated commitment model. Successive windows can carry increasing structural commitments: the first window is a mutual trial with defined terms and minimal entanglement; subsequent windows extend in duration and deepen the revenue coupling; eventually, the arrangement stabilizes into a long-term partnership with clear terms, transparent metrics, and mutual investment. This progression mirrors the natural evolution of any trust-based relationship—from first contact to casual engagement to formal commitment—but makes it structurally explicit rather than leaving it to accumulate informally and asymmetrically.
IV. Irreducibility: The Two-Primitive Test
The claim that all three primitives are necessary is testable by elimination. For each possible pair, there exists a documented employment arrangement exhibiting that pair without the third, and each such arrangement produces a characteristic dysfunction.
Signal Legibility + Revenue Coupling, Without Engagement Windows
Real-world analog: Early-stage equity-compensated startup employment.
The worker can see the metrics (investor dashboards, burn rate, MRR). The worker shares in the upside (equity, SAFEs, options). But there is no structural commitment to a defined term. Either party can exit at any time—the founder can fire the engineer, the engineer can leave for a larger company.
Characteristic dysfunction: Perpetual mutual hostage-taking. The worker stays because equity might vest; the employer retains them because replacement is expensive. Neither has committed to anything. The absence of a defined window means every day is implicitly a renegotiation, creating chronic low-grade anxiety that undermines long-term planning even when the relationship is going well. Empirically, early-stage startups exhibit the highest voluntary turnover rates in the technology sector (Wasserman, 2012), consistent with the prediction that legibility and coupling without windows fail to produce stable arrangements.
Revenue Coupling + Engagement Windows, Without Signal Legibility
Real-world analog: Commission-based outside sales with quarterly quotas.
The worker has a defined term (quarterly targets, annual reviews) and revenue-coupled income (commissions on closed deals). But the worker cannot see the system clearly enough to diagnose whether their territory is viable, whether leads are being routed equitably, whether the product is deteriorating, or whether management is cannibalizing their pipeline for house accounts.
Characteristic dysfunction: Blind risk transfer. The worker is coupled to an outcome they cannot diagnose, which forces them into either blind trust or premature departure. The revenue coupling, without signal legibility, becomes a mechanism for transferring risk to the party with the least information—the exact opposite of fair allocation. The commonly observed phenomenon of “commission breath” — the desperation of salespeople working bad territories they cannot evaluate — is a direct expression of this failure mode.
Signal Legibility + Engagement Windows, Without Revenue Coupling
Real-world analog: Fixed-fee management consulting engagement (e.g., a six-month McKinsey project billed at a flat rate).
The consultant can see the client’s system (full access to financials, operations, strategy during the engagement). The consultant operates within a defined scope and timeline. But the consultant is paid for time and deliverables rather than in proportion to the value they create.
Characteristic dysfunction: Perverse incentives against efficiency. The consultant who identifies a $500,000 efficiency gain in week two is structurally incentivized to take four months implementing it rather than two weeks, because their income tracks with project duration rather than value delivered. Signal legibility and bounded terms without revenue coupling produce the paradoxical outcome of punishing efficiency. The consulting industry’s well-known tendency to extend engagements and generate follow-on work is a rational response to this incentive structure, not a moral failure.
Each elimination produces a recognizable dysfunction. No pair is sufficient. The triad is irreducible.
V. The Hierarchy as Downstream Artifact
The employer-employee hierarchy is conventionally treated as a design choice—an organizational structure selected for its efficiency in coordinating labor (Coase, 1937; Williamson, 1975). This paper contends that the hierarchy is not a design choice. It is an emergent consequence of information asymmetry, which is itself a consequence of signal illegibility.
The logic is straightforward. When the employer possesses system-level visibility that the worker lacks, the employer must translate that visibility into directives: do this task, in this order, by this deadline. The worker, unable to see the system’s constraints and priorities directly, must either follow the directive or negotiate from ignorance. This dynamic is called management. But it is not an inherent feature of collaboration. It is a compensatory mechanism for the absence of shared signal legibility.
If both parties can see the system, both can independently identify what needs to be done, why it matters, and how their contribution fits into the whole. Coordination under these conditions looks nothing like traditional management. It looks like two informed agents negotiating a shared strategy—which is to say, it looks like partnership. The hierarchy dissolves not because anyone decided to flatten the org chart, but because the information asymmetry that made the hierarchy functional no longer exists. To be precise: shared legibility does not remove the need for coordination — it removes the structural justification for unilateral directive authority rooted in informational advantage. Two partners coordinating a shared strategy is not hierarchy. A manager issuing directives because only they can see the dashboard is.
This reframing has a critical implication: efforts to “flatten hierarchies” or “empower workers” without first establishing signal legibility are solving the wrong problem. They attempt to remove the symptom (authority) without addressing the cause (information asymmetry). Predictably, such efforts fail—as documented extensively in the literature on failed holacracy implementations (Bernstein et al., 2016; Robertson, 2015). The hierarchy reasserts itself because the underlying condition that produced it—asymmetric access to system-level information—remains intact. The solution is not to fight the hierarchy. It is to dissolve the information asymmetry that makes the hierarchy necessary.
VI. The Rivalry Objection
The strongest intuitive objection to signal legibility is that transparency invites competition. If the worker can see how the business operates, they can leave and do it themselves. This objection, while psychologically compelling, is empirically weak and structurally addressed.
First, the empirical weakness: most workers who leave an employer do not start competing businesses. They go to other employers. Bureau of Labor Statistics data consistently shows that the overwhelming majority of job transitions are lateral moves within the same industry, not entrepreneurial departures. The fear of replication is disproportionate to its actual incidence, which suggests that the fear is serving a function other than risk mitigation—namely, justifying the information asymmetry that subsidizes the employer’s bargaining position.
Second, the structural address: non-disclosure agreements, non-compete clauses (in jurisdictions where enforceable), trade secret protections under the Defend Trade Secrets Act (2016), and confidentiality provisions in employment contracts already provide a comprehensive legal toolkit for managing the informational risks of transparency. A company that refuses to provide signal legibility while simultaneously failing to use any of these instruments is not protecting itself. It is exploiting an asymmetry. A company that provides signal legibility while deploying appropriate legal protections is doing what any rational enterprise should do: aligning its workers’ incentives with its own success while protecting against the specific, bounded risk of competitive replication.
Third, and most fundamentally: revenue coupling reverses the incentive structure that makes rivalry rational. A worker whose income is tethered to the enterprise’s transactional revenue has no incentive to leave and compete. They have an incentive to stay and optimize. The rivalry concern assumes adversarial actors operating under misaligned incentives. Revenue coupling creates aligned actors for whom rivalry is irrational. The objection refutes itself once the primitives are in place.
VII. Structural Objections and Responses
Objection 1: Risk Transfer to Workers in Downturns
“Revenue coupling transfers business risk to workers. In a recession, the hourly worker still gets paid; the revenue-coupled worker takes a pay cut.”
This objection presupposes that the hourly model protects workers during downturns. It does not. There is no credible argument that it ever has. Workers under the hourly model are the first to be eliminated in a downturn—not the last. They are not “protected” by their fixed rate; they are anesthetized by it. The hourly rate creates the illusion of stability right up until the moment of termination, at which point income drops from its fixed level to zero in a single event, with no warning, no diagnostic information, and no opportunity to intervene.
The hourly model does not reduce the worker’s exposure to business risk. It makes that exposure invisible until it becomes catastrophic. It converts a continuous signal (revenue declining over weeks or months) into a binary event (employed one day, fired the next) and delivers that event with minimal lead time and zero explanatory context. The worker cannot see the downturn coming because they lack signal legibility. They cannot prepare because their engagement has no structural boundary at which renegotiation occurs. They cannot intervene because their compensation is decoupled from the revenue they might otherwise fight to preserve. Every feature of the hourly model that appears to “protect” the worker is in fact a mechanism for concealing risk until the moment of maximum damage.
Revenue coupling does not transfer risk to the worker. The risk was always there. Revenue coupling makes the risk legible. The revenue-coupled worker watching income decline over weeks has three options the hourly worker does not: (1) diagnose the cause using signal legibility and intervene to reverse the decline; (2) prepare for exit while still earning income rather than after a sudden termination; (3) renegotiate terms at the next engagement window boundary with full information. The hourly worker gets none of these options. They get a paycheck followed by a termination letter.
Furthermore, the framework is compatible with a minimum income floor. Revenue coupling does not mean income can reach zero. A contract can specify a base rate (equivalent to or above minimum wage) plus a revenue-coupled variable component. This structure already exists in commissioned sales; the framework generalizes it and removes the signal illegibility that makes commissioned sales dysfunctional.
Worked example: A developer earning $40/hr under the standard model generates $160,000/year. Under revenue coupling at a 5% share of a $4M transactional endpoint, the same developer earns $200,000 when the business performs well and $150,000 during a 25% revenue decline—but retains signal legibility to diagnose the decline, structural incentive to reverse it, and advance warning to prepare alternatives. Under the hourly model, the same 25% revenue decline results in a layoff with two weeks’ severance and no diagnostic information. The question is not “who bears the risk.” Both workers bear identical risk. The question is which one can see it coming.
Objection 2: Roles Without Direct Transactional Endpoints
“Not every role has a measurable transactional endpoint. What about HR, IT support, internal operations?”
The premise of this objection is false. These roles do have transactional endpoints—they are simply modeled incorrectly as overhead rather than as functions with measurable economic outputs. The fiction that support roles lack transactional endpoints is itself a symptom of signal illegibility: because the enterprise does not model the economic value of these functions, it treats them as cost centers rather than value contributors, which produces the familiar pattern of support roles being chronically underfunded, undervalued, and first to be cut.
There are three approaches to modeling transactional endpoints for support roles, each appropriate to different contexts:
Approach 1: Cost-of-failure modeling. HR’s transactional endpoint is dispute resolution and retention. Every employee who quits over a preventable dispute represents a retraining cost—typically estimated at 50–200% of annual salary for knowledge workers (SHRM, 2022). HR’s measurable output is retraining cost avoided. Every dispute resolved, every retention event, every avoided termination-and-rehire cycle has a calculable dollar value. IT operations can be modeled identically: the transactional endpoint is system availability, and the measurable output is the cost of downtime, data breach, or system failure avoided. If the database leaks, the cost is quantifiable. If the system goes down for four hours during peak transaction volume, the lost revenue is calculable. These are not abstract values. They are auditable events with dollar figures attached.
Approach 2: Inherited transactional endpoints. Support roles exist to serve functions that generate transactional revenue. An IT department that maintains the infrastructure for a sales division is not separate from that division’s revenue generation—it is a scaling factor on it. The natural model is for support roles to inherit the transactional endpoint of the system they support, with their revenue coupling scaled by their contribution to upstream efficiency. This is literally how the relationship works: the IT team’s output is the sales team’s ability to operate. The developer who builds the CRM that the sales team uses to close deals is coupled to those deals. The coupling is indirect but the causal chain is real and measurable. Model the scaling factor on upstream efficiency and the “no transactional endpoint” problem dissolves.
Approach 3: Opportunity-cost pricing for pure prevention roles. Some roles are genuinely defensive—security, compliance, disaster prevention—where the value created is the absence of a catastrophic event rather than the presence of a positive transaction. These roles are the hardest to couple because their success condition is that nothing happens. A reasonable heuristic: the person in a prevention role gets paid at the rate they could earn doing scale-based work in the same system. This prices their contribution at the opportunity cost of their talent rather than at the (unmeasurable) expected value of prevented disasters. The effect is that prevention work is never structurally punished relative to growth work—the enterprise pays the same rate for protecting value as for creating it, which eliminates the chronic underinvestment in infrastructure, security, and maintenance that plagues organizations that treat these functions as cost centers.
The broader point: the claim that support roles lack transactional endpoints is not an observation about reality. It is a modeling failure. The enterprise that cannot identify the economic output of its HR department is the same enterprise that cuts HR first in a downturn and then spends three times the savings on emergency recruiting and wrongful termination settlements. The inability to model the endpoint is the problem. The framework requires that the modeling be done—and the act of doing it produces the signal legibility that makes the entire arrangement functional.
Objection 3: Regulatory Compatibility
“Labor law requires minimum wage, overtime pay, and specific classifications. Revenue coupling may violate wage and hour regulations.”
Revenue coupling is structurally compatible with existing wage and hour regulations by construction. The framework specifies a base rate at or above the applicable minimum wage plus a revenue-coupled variable component. This is legally identical to the base-plus-commission structure already used in millions of employment relationships and explicitly contemplated by the Fair Labor Standards Act (29 U.S.C. § 207(i) exemption for commission-based retail employees; various state-level commission pay statutes). Overtime requirements apply to the base rate and can be calculated accordingly. Jurisdictional specifics vary; the point is structural compatibility, not legal advice.
The more substantive regulatory question is classification: does a revenue-coupled worker with signal legibility and bounded engagement windows look more like an employee or an independent contractor? Under the IRS common-law test and the Department of Labor’s economic reality test, the key factors are behavioral control, financial control, and the nature of the relationship. Revenue coupling shifts the financial control factor toward contractor-like treatment, but signal legibility (integration into the enterprise’s information systems) and bounded engagement windows (structural commitment to defined terms) shift back toward employee treatment. The framework is classification-agnostic; it can be implemented under either classification with appropriate legal structuring.
Objection 4: Measurement Brittleness and Gaming
“If compensation is tied to a dashboard, workers will game the dashboard. Goodhart’s Law applies: when a measure becomes a target, it ceases to be a good measure.”
This objection confuses unilateral measurement with bilateral measurement. Goodhart’s Law applies when the measured party controls their behavior in response to a metric set by the measuring party. In the standard employment model, the employer defines the metrics and the worker games them—because the worker has no visibility into the system-level consequences of their gaming. They optimize locally (hitting the metric) at the expense of global performance (actual enterprise value), and they cannot see the global consequences because signal legibility is absent.
Under the triadic framework, both parties observe the same system. Gaming the dashboard requires falsifying shared information—which is fraud, not optimization. A worker who inflates a revenue metric under signal legibility is committing the same act as an accountant who falsifies financial statements: they are lying about observable reality to a party who can check. The deterrent is not a monitoring regime imposed from above. The deterrent is that the other party can see the same data and will notice the discrepancy.
More fundamentally, revenue coupling at the transactional endpoint is resistant to Goodhart effects because the endpoint itself is a natural measurement: did the customer pay or not? Revenue at the point of transaction is the least gameable metric in business because it represents an actual exchange of value with an external party. Unlike productivity metrics, engagement scores, or lines of code committed, transactional revenue cannot be inflated without the cooperation of the customer. This is why the framework specifies coupling to transactional revenue rather than to profit, productivity, or any internally generated metric.
VIII. The Steel Man for Hourly Pay
The preceding sections establish that hourly compensation is allocatively inaccurate, incentive-misaligned, and structurally inferior to the triadic alternative. If these claims are correct — and no falsifying counterexample has been produced — then the persistence of hourly pay requires explanation. Why does a demonstrably inferior model remain the default?
The answer is not intellectual. The theoretical defenses of time-based compensation that exist in the literature — transaction-cost arguments (Coase, 1937), measurement difficulty under multi-causal production, insurance and risk-sharing models, and principal-agent variants that justify fixed pay in specific regimes (Holmström, 1979) — all reduce, upon examination, to the same underlying claim: that measuring actual contribution is too expensive relative to the gains from accuracy. These are not defenses of hourly pay as optimal. They are defenses of hourly pay as cheaper to administer than the alternative. The distinction matters, because if the administration cost falls, the defense collapses.
The real defense of hourly pay is operational, and it is stronger than any theoretical defense could be: hourly pay is trivially enforceable.
Hours times rate equals pay. A child can verify it. A labor board can verify it. A judge can verify it with no discovery, no forensic accounting, and no access to business internals. The worker counts their hours. The employer counts the same hours. If the numbers on the paycheck don’t match, someone broke the law. The enforcement infrastructure — timesheets, pay stubs, overtime regulations, minimum wage statutes — is simple, universal, and battle-tested over a century of labor law. The entire regulatory apparatus of employment law is built on the assumption that compensation is a function of time, and the enforcement mechanisms are optimized for exactly that assumption.
Revenue coupling disrupts this enforcement model at every level.
Verification requires access. Under hourly pay, the worker needs access to exactly two numbers: their hours and their rate. Both are in their possession. Under revenue coupling, the worker needs access to the enterprise’s revenue data — or at minimum, the revenue data for their coupled transactional endpoint. This requires the employer to expose financial internals that most employers experience as proprietary. Not because they’re hiding fraud. Because they’ve been trained to treat financial information as competitive leverage, and the idea of sharing it with employees feels like surrendering power even when the math would favor the employer.
Disputes require forensic capacity. When an hourly worker is shorted, the dispute is arithmetic: did you work 40 hours at $20/hr? Then you’re owed $800. If the check says $750, the employer owes $50. A labor board can adjudicate this in minutes. When a revenue-coupled worker suspects they’ve been shorted, the dispute is epistemic: is the revenue number real? Are all relevant transactions included? Were any reclassified, deferred, or routed through a different account? Answering these questions requires access to accounting systems, understanding of revenue recognition practices, and potentially forensic auditing. The cost of adjudicating a $50 discrepancy under revenue coupling can exceed the cost of adjudicating a $50,000 discrepancy under hourly pay.
Regulation assumes time-based compensation. Minimum wage is defined per hour. Overtime is defined per hour. The Fair Labor Standards Act, state wage-and-hour statutes, and the entire compliance infrastructure of employment law operate on the assumption that pay is a function of time. Revenue coupling is legal — commission structures prove that — but the regulatory infrastructure isn’t optimized for it. Filing a wage claim for unpaid commissions is categorically harder than filing one for unpaid hours, not because the law doesn’t protect commission workers, but because proving the violation requires proving what the revenue was, which requires the very transparency that most employers resist providing.
Administrative complexity compounds. Under hourly pay, the employer’s payroll calculation is: count hours, multiply by rate, withhold taxes at a known bracket, cut check. Under revenue coupling, the calculation is: pull revenue report for the relevant accounts, filter for correct transaction types, apply the coupling percentage, add the base rate, calculate tax withholding on a variable amount that changes every pay period, cut check. This is not technically difficult. But it is more complex than the alternative, and the person who bears the additional complexity — the employer — is the same person who must voluntarily adopt the system.
These four layers — transparency resistance, forensic verification cost, regulatory mismatch, and administrative overhead — compound into a single practical reality: hourly pay persists not because anyone believes it is fair, accurate, or incentive-aligned, but because it is the local minimum of operational friction. It is the easiest compensation structure to administer, verify, and enforce. The system optimizes for enforcement simplicity over allocative accuracy, and it has done so for over a century.
This is the honest steel man, and it is important to name it explicitly because it changes the nature of the argument. The case against hourly pay is not primarily intellectual — the theoretical defenses all reduce to enforcement and measurement convenience. The case is operational: can revenue-coupled compensation be made as easy to verify as hourly pay? If verification remains expensive and complex, the triadic framework will remain theoretically superior and practically unadopted. If verification can be made trivially simple, the enforcement advantage of hourly pay — its only remaining advantage — disappears.
This is precisely what AI-mediated auditing achieves, as described in Section X. A reproducible AI audit pointed at the accounting system’s API produces a verification result that either party can independently regenerate — making revenue-coupled compensation as easy to verify as counting hours on a timesheet. The enforcement simplicity advantage of hourly pay is not permanent. It is a technological limitation, and for the first time, the technology to overcome it exists and is deployable. The only remaining argument for hourly pay is inertia.
IX. The Dependency Structure
While all three primitives are equally necessary, they are not symmetrically ordered. They form a dependency graph in which signal legibility is the precondition for revenue coupling, and both are preconditions for meaningful engagement windows.
You cannot negotiate a fair revenue share without signal legibility, because you do not know what “fair” means without being able to read the system. And you cannot evaluate whether a revenue share is working without a bounded engagement window, because you need a stable observation period to generate data. The dependency runs: signal legibility enables revenue coupling, which is tested within engagement windows.
But the structure is cyclical, not linear. Engagement windows produce the data that updates signal legibility, which informs renegotiation of the revenue coupling at the next window boundary. Each primitive feeds the others. The engagement window is both the last thing established (because it requires the other two to be meaningful) and the first thing that produces value (because without a bounded observation period, neither signal legibility nor revenue coupling generates actionable feedback).
This cyclical dependency is the mechanism by which the arrangement self-corrects. Traditional employment has no self-correcting mechanism. If the hourly rate is wrong on day one, it remains wrong indefinitely unless one party initiates a renegotiation—which always carries the social cost of being the person who “brought it up.” Under the triadic framework, correction is structural. The window closes, the data is reviewed, the terms are recalibrated. Neither party is the aggressor. The system is.
X. AI as Signal Legibility Infrastructure
The enforcement and implementation challenges described in the preceding sections share a common structural bottleneck: signal legibility requires that both parties can observe and interpret system state, but in most employment relationships, the parties have asymmetric domain expertise. The employer may understand the business but cannot read a codebase. The worker may understand the technical system but cannot interpret the financial pipeline. Each party possesses legibility in their own domain and opacity in the other’s. Traditional solutions to this asymmetry — managers who translate between domains, consultants who audit and report, weekly status meetings where one party verbally represents system state to the other — all introduce the same problem: the translator has stakes, and stakes distort translation.
A manager who translates technical progress into business terms has a career incentive to present their team’s work favorably. A consultant who audits project state has a financial incentive to find problems that generate follow-on engagements. A worker who verbally reports “we’re 30% done” has a reputational incentive to project confidence. Every human intermediary in the signal chain has an interest that is not perfectly aligned with accurate reporting, which means every human intermediary degrades signal legibility rather than producing it.
Artificial intelligence has no direct stakes in the outcome of the assessment. It cannot be hired for follow-on work. It cannot be fired for delivering an unfavorable assessment. It does not benefit from finding problems and does not benefit from rubber-stamping progress. It has no career, no reputation to manage, no family dynamics, no anxiety about the relationship’s future. This is not a claim that AI is perfectly neutral — the model provider has interests, the party selecting the model and constructing the prompt makes choices, and reproducibility across different models and configurations is imperfect. But the incentive contamination is categorically lower than in any human intermediary, because the AI’s assessment does not feed back into its own compensation, employment, or professional advancement. It is, structurally, the first available technology that can serve as a low-bias translator between parties with asymmetric domain expertise and conflicting incentive structures.
The Audit Model
The practical implementation is straightforward. On a defined cadence (weekly, biweekly, or at engagement window boundaries), the technical party provides the AI with raw artifacts: the project specification, the git log for the period, completed and remaining tickets, and access to a staging environment or codebase. The AI produces a progress report that includes:
- Percentage complete against specification, verified against code artifacts rather than self-reported
- Current velocity measured in tickets closed, features demonstrable, or other objective units
- Projected completion date if current velocity holds
- Risks or blockers identified from the ticket backlog or codebase analysis
- A plain-language summary readable by a non-technical party in under two minutes
Critically, the technical party does not edit or filter the report before it reaches the non-technical party. The report goes directly from AI to both parties simultaneously. This is the mechanism that produces signal legibility: the non-technical party receives a translation of system state that was not produced by anyone with an incentive to distort it.
What This Solves
The AI audit model addresses three problems simultaneously.
First, the technical literacy gap. In any employment relationship where one party’s contribution is technical and the other party’s evaluation capacity is non-technical, signal legibility is impossible without translation. The employer who cannot read a git diff cannot independently verify that meaningful work occurred. The AI bridges this gap by translating technical artifacts into business-legible assessments — not by simplifying the technical reality, but by mapping it onto metrics the non-technical party already understands (timeline, percentage, risk).
Second, the trust displacement problem. When a worker says “we’re on track,” the non-technical employer must decide whether to trust that assertion. Trust is a scarce and depletable resource, particularly in relationships with existing strain (family businesses, early-stage engagements, cross-cultural collaborations). The AI removes trust from the equation entirely. The employer does not need to trust the worker’s self-report because the employer is not relying on the worker’s self-report. They are reading an independent assessment produced by an agent with no relationship to either party. The question “are you sure about the deadline?” — which is really a question about trust — becomes “what does the audit say?” — which is a question about data.
Third, the confrontation cost of inquiry. In many employment relationships, the act of asking “how is the project going?” carries implicit social weight — it signals doubt, it creates pressure, it can feel like surveillance. This is particularly acute in family businesses and other arrangements where the professional and personal relationships are entangled. The AI audit eliminates this cost because the inquiry is structural rather than interpersonal. The report is generated on a cadence. Nobody “asked.” The system produced it. This is the same dynamic as the engagement window boundary in the triadic framework: error correction happens because the structure requires it, not because one party initiated it.
AI as Dispute Resolution Substrate
The AI audit model also provides a foundation for the lightweight dispute resolution described in Section XI. If the parties disagree about whether a milestone was met, whether a deadline was realistic, or whether the quality of delivered work matches the specification, the AI can be pointed at the artifacts and asked to adjudicate. This is not legally binding arbitration. But it produces a credible, evidence-based assessment that shifts the burden of proof: whichever party disagrees with the AI’s assessment must explain why, using specific evidence, rather than simply asserting a different interpretation. The social dynamics of the dispute are inverted — the default is the AI’s reading of the artifacts, and deviation from that default requires justification.
This positions AI not as a replacement for formal arbitration but as a pre-arbitration filter that resolves the majority of disputes before they escalate. Most employment disputes are not genuinely ambiguous. They are disagreements that persist because neither party has access to a neutral, evidence-based assessment. The AI provides that assessment at near-zero marginal cost, on demand, without lawyers, without consultants, and without the social overhead of “bringing it up.”
AI as Compensation Assessor
The AI audit model extends beyond progress verification to the question that produces the most friction in any employment relationship: what should compensation be?
Under the current model, compensation is determined by negotiation — a rhetorical contest in which the outcome tracks with leverage, personality, and willingness to tolerate discomfort rather than with actual contribution. The employer names a number or the worker requests one, and the resulting figure reflects the power dynamics of the conversation rather than the economics of the arrangement. Even under signal legibility and revenue coupling, the specific percentage — the coupling rate — must be proposed by someone, and the act of proposing a number is where social friction concentrates.
AI dissolves this friction by deriving the number from the data rather than from the parties. The process is straightforward: point the AI at the system state before the worker arrived, the system state now, the revenue delta over the period, the specific artifacts the worker produced (features shipped, infrastructure built, processes automated), and the causal chain from those artifacts to revenue outcomes. Ask: given this contribution and this delta, what revenue-coupling percentage represents fair compensation? The AI produces a range. Both parties can verify the inputs, reproduce the assessment independently, and negotiate within an evidence-derived range rather than starting from competing anchors rooted in leverage rather than value.
This eliminates the single most destructive dynamic in employment negotiations: the requirement that one party must “ask.” In the traditional model, the worker who requests a raise is the aggressor — they disrupted the default, they “brought it up,” they bear the social cost of the conversation regardless of outcome. Under AI-mediated compensation assessment, neither party proposes a number. The data proposes the number. The engagement window boundary arrives, the AI reads the artifacts and the revenue, and a compensation assessment is produced that neither party authored. The conversation shifts from “I think I deserve more” (a claim about self-worth, easily dismissed) to “the evidence suggests the coupling rate should be X” (a claim about data, requiring counter-evidence to dismiss).
The reproducibility property is critical here. If the worker runs the assessment and gets one number, and the employer runs the same assessment with the same inputs and gets the same number, the negotiation is effectively over. Disagreement requires one party to argue that the AI’s reading of the evidence is wrong, which means identifying specific inputs that were missing, specific artifacts that were misattributed, or specific causal claims that don’t hold. That’s a productive conversation about evidence. It is a fundamentally different kind of conversation than two people with asymmetric leverage asserting competing numbers at each other.
Broader Implications
The application of AI as signal legibility infrastructure extends beyond individual employment relationships. The same model applies to any arrangement where value is created by one party and evaluated by another across a domain expertise gap: freelance contracts, vendor relationships, open-source contributions, research collaborations, and any context where “how much progress was made?” is a contested question answered by the party with the most to gain from a favorable answer.
The conventional framing of AI in the workplace focuses on AI as a tool for productivity — writing code faster, generating documents, automating tasks. The framing proposed here is orthogonal: AI as a tool for legibility. Not making the work happen faster, but making the work visible to parties who cannot directly observe it. This is a fundamentally different value proposition and one that maps directly onto the first primitive of the triadic framework. Signal legibility has historically been expensive to produce (requiring auditors, consultants, or managers) and easy to distort (because every producer of legibility had stakes in the outcome). AI collapses the cost and dramatically reduces the distortion. It is, for the first time, feasible to make system state legible to all parties continuously, cheaply, and with categorically less incentive contamination than any human-mediated alternative.
XI. The Enforcement Question
The most practical challenge facing this framework is enforcement. Lightweight contracts are only as strong as the mechanisms available to adjudicate their breach. The legal system is too expensive and too slow for the routine disputes that arise in employment relationships. Social media—the de facto court of public opinion—is too blunt: it offers only silence or thermonuclear reputational destruction, with no proportional intermediate.
The solution space exists in nascent form. Online arbitration platforms—including the American Arbitration Association’s small business services, standalone platforms like Rapid Ruling and Arbitration Resolution Services, and decentralized protocols like Kleros—offer binding dispute resolution without lawyers, at low cost, with resolution timelines measured in days rather than months. Freelancing platforms like Upwork and Freelancer.com have demonstrated that escrow-backed, platform-arbitrated employment relationships can function at scale for project-based work.
What does not yet exist is the assembly of these components into a coherent product for ongoing employment relationships rather than discrete transactions. The infrastructure for low-friction escrow, binding online arbitration, and revenue-transparent employment arrangements exists in separate silos. The product gap is integration: a platform that combines rolling escrow deposits, real-time revenue dashboards, defined engagement windows with automatic renewal or renegotiation, and automated arbitration escalation paths. This is not a technical challenge. It is a coordination challenge—the same type of coordination failure that produced the current dysfunctional equilibrium.
XII. Falsification Criteria
The triadic claim is universal and falsifiable. This section states the exact conditions under which the claim would be refuted, so that critics are not required to guess what would count as a counterexample and proponents are not permitted to move the goalposts after the fact.
Falsification Criterion 1: Superior Allocative Accuracy Under Hourly-At-Will-Opaque. Produce a concrete employment arrangement using hourly compensation, at-will termination, and asymmetric signal access that achieves higher allocative accuracy of compensation relative to contribution — measured over a bounded observation period of at least one engagement window — than the triadic equivalent applied to the same role, same enterprise, and same time period. The comparison must hold under the evaluation function defined in the Core Definitions (allocative accuracy, incentive alignment, structural self-correction, and catastrophic downside reduction). Demonstrating lower income variance alone does not constitute falsification, as variance reduction through risk concealment is not an outcome improvement.
Falsification Criterion 2: A Genuinely Unmodelable Role. Produce a role for which no transactional endpoint can be modeled under any of the three approaches specified in this paper: cost-of-failure valuation (the economic cost of the role’s absence or failure), inherited upstream coupling (tethering to the transactional endpoint of the system the role supports), or opportunity-cost pricing (compensating at the rate the worker could earn doing scale-based work in the same system). The role must be one that exists within an enterprise that generates revenue — purely volunteer or non-economic arrangements are outside the framework’s domain by construction, not by failure.
Falsification Criterion 3: Signal Legibility Producing Worse Incentive Alignment. Demonstrate that providing a revenue-coupled worker with full signal legibility — real-time visibility into system state, revenue flows, and the causal relationship between their contribution and enterprise outcomes — produces worse incentive alignment than withholding that information. The demonstration must show that the worker, upon gaining visibility, behaves in ways that reduce enterprise value relative to a counterfactual in which the same worker, with the same revenue coupling and the same engagement window, operates without signal legibility. The mechanism by which opacity outperforms transparency must be specified — it is not sufficient to assert that “some workers might game the system” without showing that gaming under legibility exceeds the misallocation produced by opacity.
If none of these criteria can be met, the triadic claim stands — not because it has been proven in the mathematical sense, but because no arrangement has been shown to outperform it under the conditions it specifies. The claim is dominant until a concrete counterexample displaces it. This is how engineering standards work: the superior design holds until someone builds a better one, not until someone imagines a hypothetical objection.
XIII. The Convergence Thesis
The triadic framework is an engineering solution. It is not the structural solution. This section names the difference.
The structural solution to every problem identified in this paper is co-ownership. When workers are owners, signal legibility is not a negotiated concession — it is a legal right. Owners see the books. Revenue coupling is not a contractual add-on — it is the default. Owners share in surplus. Bounded engagement windows are not artificial constraints — they are governance cycles. Owners vote on terms at regular intervals. The three primitives that this paper labors to establish as contractual requirements are automatic features of any arrangement in which the worker and the owner are the same person.
The Mondragón Cooperative Corporation has demonstrated this for seventy years. Founded in 1956 in the Basque region of Spain, Mondragón is a federation of over 100 worker-owned cooperatives employing more than 70,000 people across industrial manufacturing, retail, finance, and education, with annual revenues exceeding $14 billion. Every worker-member buys in at a fixed price, receives one vote in the General Assembly regardless of role or seniority, and shares in the cooperative’s surplus through capital accounts that track enterprise performance. Executive compensation is capped at a ratio of 5:1 to 9:1 relative to the lowest-paid member — compared to ratios exceeding 300:1 in conventional US corporations. When one of Mondragón’s founding cooperatives went bankrupt in 2013, the federation absorbed displaced workers across other cooperatives rather than executing mass layoffs, using solidarity reserves funded by the revenue-coupling mechanism that had been accumulating for decades.
Mondragón does not solve the individual attribution problem. It dissolves it. Because every worker is an owner, the question “how much of this collective output belongs to you specifically?” is replaced by the question “what is your ownership share?” The first question is often unanswerable — most value creation is systemic, not individual, and attempting to decompose collective output into individual contributions is a category error applied to emergent phenomena. The second question is a one-time negotiation, revisable at governance boundaries, and enforceable by the same mechanisms that govern any ownership agreement. The coarse-grained revenue split — a democratically agreed ratio applied to the whole enterprise — is not a simplification of the fine-grained model. It is the structurally correct model, because it matches the granularity of the actual causal structure (systemic production) rather than imposing a granularity (individual attribution) that the system does not support.
This means the triadic framework, as presented in this paper, is duct tape. Sophisticated, well-engineered, deployable-tomorrow duct tape — but duct tape nonetheless. It exists because the current legal and cultural paradigm maintains a distinction between owners and employees, and within that paradigm, the three primitives must be individually negotiated into existence through contractual mechanisms rather than emerging automatically from the ownership structure. Every section of this paper — signal legibility as a contractual right, revenue coupling as a compensation formula, engagement windows as commitment periods, AI auditing as a verification layer, enforcement mechanisms as dispute resolution — is a workaround for the fact that the worker is not an owner.
The honest conclusion is therefore two-layered:
Within the current paradigm, the triadic framework is the best available engineering. It produces strictly better outcomes than the hourly-at-will-opaque alternative across every dimension of the evaluation function. It is deployable tomorrow, within existing legal structures, using existing technology. It requires no change in corporate law, no restructuring of ownership, and no cultural revolution. It is the correct move for any worker or employer operating within the constraints of the current system.
Beyond the current paradigm, the triadic framework converges toward co-ownership if followed to its logical completion. Once both parties have full signal legibility, revenue coupled to shared transactional endpoints, and bounded engagement windows with structural self-correction — the functional distinction between “employee” and “owner” has already collapsed. The worker who can see all the numbers, whose income tracks with enterprise revenue, who renegotiates terms at regular intervals based on shared data, and whose engagement is bounded by mutual commitment rather than at-will disposability — that worker is an owner in every functional sense except the legal one. The triadic framework rebuilds co-ownership from first principles without using the word, and without requiring the ownership transfer that makes cooperative formation politically and legally difficult within the current system.
This is not a weakness of the framework. It is its deployment strategy. The cooperative movement has spent a century arguing that workers should be owners. The argument is correct and the evidence — Mondragón chief among it — is overwhelming. But the adoption rate remains marginal because co-ownership requires a structural leap: someone must relinquish the ownership distinction entirely, which means restructuring corporate governance, legal liability, capital access, and decision-making authority all at once. The triadic framework offers an incremental path to the same destination. Signal legibility first. Revenue coupling once the numbers are visible. Engagement windows once both parties have something worth committing to. Each step is individually rational and individually negotiable. The endpoint is functional co-ownership, arrived at through a sequence of engineering improvements rather than a single structural revolution.
The framework is the bridge. Co-ownership is the other side.
XIV. Implications for Action
If the assertions in this paper are correct — and the falsification criteria in Section XII remain unmet — then specific actors in the current system are making specific mistakes that produce specific, avoidable costs. This section names the mistakes, the actors, and the corrections.
For Workers
Stop negotiating hourly rates. The hourly rate is the endpoint of a negotiation conducted in signal darkness — you are guessing what your time is worth because you cannot see what the enterprise earns from it. The first negotiation should not be about compensation at all. It should be about visibility.
Ask for signal legibility. Dashboard access. Read-only visibility into the metrics your work affects. Revenue reports for the division you support. This is the least expensive ask available — it requires no change to compensation structure, no legal restructuring, and no capital outlay, though it does carry real costs in competitive sensitivity and internal politics that the employer will feel even if they don’t appear on a balance sheet. Frame it as diligence rather than what it actually is: the installation of the first primitive.
Once you can see the numbers, revenue coupling becomes arithmetic rather than rhetoric. You are no longer saying “I think I deserve more” — a claim about self-worth that is easily dismissed. You are saying “I can see that my work correlates with a measurable revenue delta, and I would like my compensation to reflect that.” The contract template in Appendix A provides the structure. The AI auditing model described in Section X provides the verification layer. Use both.
If the employer refuses signal legibility — if the response to “can I see the numbers my work affects?” is no — that refusal is itself diagnostic. It tells you that the arrangement depends on your inability to evaluate it. Proceed accordingly.
For Employers
The resistance to signal legibility and revenue coupling is understandable but economically irrational. The paper’s own analysis shows that the triadic framework produces higher retention (workers coupled to revenue have structural reasons to stay), lower renegotiation friction (engagement window boundaries replace adversarial salary negotiations with data-driven recalibration), self-correcting compensation (the system adjusts without either party initiating an uncomfortable conversation), and stronger incentive alignment (the worker who can see the revenue and is paid from it has more reason to grow it than the worker earning the same rate regardless of enterprise performance).
The enforcement simplicity objection — the real reason hourly pay persists — is now addressable through AI-mediated auditing. Revenue-coupled compensation can be verified as cheaply and reproducibly as counting hours on a timesheet. The transparency fear is the remaining barrier, and it is worth examining honestly: what exactly is the employer protecting by keeping revenue data from the worker? If the answer is “negotiating leverage” — the ability to claim the business cannot afford a raise while revenue grows — then the employer is acknowledging that the current arrangement depends on information asymmetry that benefits one party at the other’s expense. That is the definition of the problem this paper identifies.
The low-cost experiment: try one triadic contract with one worker. Use the template in Appendix A. Run it for two engagement windows — four weeks. Compare the outcomes against a counterfactual hourly arrangement on the four evaluation criteria defined in the Core Definitions. If the triadic arrangement produces worse results, the paper’s falsification criteria have been met and the framework is refuted. If it produces better results, the path forward is self-evident.
For Platform Builders and Technologists
The product gap identified in Section XI is the binding constraint on adoption. The components exist — accounting systems track revenue, payroll systems distribute compensation, AI can audit and verify — but no product integrates them into a single workflow that treats compensation as a computed output from revenue rather than a stored input from the employer.
The specific product that does not exist and should: a layer that connects to the employer’s accounting system (QuickBooks, Xero, FreshBooks), listens for revenue events on specified accounts (invoice paid, transaction cleared), applies a contractually defined coupling percentage, computes the variable compensation for the engagement window, generates an AI-auditable verification report that both parties receive simultaneously, and triggers payroll distribution. The core integration logic is straightforward — a webhook listener, a percentage calculation, and a payroll API call. The non-trivial parts are the edges: refund handling, revenue recognition timing, chargebacks, access controls, audit trails, tax withholding on variable amounts, and jurisdictional variance. These are real engineering problems, but they are solved problems — every commission-based payroll system already handles them. The bottleneck is not technical complexity but the absence of a product that assembles existing solutions under a mental model where payroll is a function of revenue rather than a stored input.
The AI audit layer is the second product: point an AI at the accounting system’s API and the employment contract’s terms, and produce a reproducible verification of whether the compensation paid matches the compensation owed. Either party can run this independently. The result is the revenue-coupling equivalent of a timesheet — a simple, verifiable artifact that makes enforcement trivial.
Build these two products and the enforcement simplicity advantage of hourly pay — its only remaining advantage — ceases to exist at the infrastructure level, not just in theory.
For Policymakers and Regulators
The regulatory infrastructure of employment law is optimized for time-based compensation. Minimum wage is per hour. Overtime is per hour. The enforcement mechanisms — wage claims, labor board adjudication, statutory penalties — assume that the disputed quantity is hours times rate. Revenue-coupled compensation is legal under existing law (commission structures prove this), but the enforcement infrastructure is not optimized for it, which means that workers under revenue coupling have weaker practical protections than workers under hourly pay even when their legal protections are formally equivalent.
Two interventions would correct this without new legislation. First, recognize AI-generated audit reports as admissible evidence in wage claims involving variable compensation. This gives revenue-coupled workers the same enforcement simplicity that hourly workers currently enjoy: a simple, verifiable artifact that proves the amount owed. Second, develop standardized reporting templates for revenue-coupled employment — the equivalent of the W-2 for triadic contracts — so that tax withholding, benefits eligibility, and labor protections can be administered on variable compensation without requiring bespoke legal structuring for each arrangement.
For the Cooperative Movement
The triadic framework is your incremental deployment strategy. The cooperative movement has spent over a century making the correct argument — that workers should be owners — and adoption remains marginal because co-ownership requires a structural leap. The legal restructuring, the capital access challenges, the governance redesign, the cultural shift — all of it must happen simultaneously for a cooperative to form. This is why Mondragón succeeded in a specific cultural and historical context and has proven difficult to replicate at scale elsewhere.
The triadic framework offers the bridge. Signal legibility first — this is a single contractual clause, not a governance restructuring. Revenue coupling second — this is a compensation formula, not an equity transfer. Engagement windows third — this is a commitment mechanism, not a corporate charter. Each step is individually rational, individually negotiable, and individually reversible. No step requires the employer to relinquish the ownership distinction. But the cumulative effect of all three steps is functional co-ownership: the worker who can see all the numbers, whose income tracks with enterprise revenue, who renegotiates terms at regular intervals based on shared data, and whose engagement is bounded by mutual commitment rather than unilateral disposability — that worker is an owner in every sense except the legal one.
Deploy the triad as the on-ramp to cooperative formation. Let workers experience functional co-ownership through contractual mechanisms before asking them to commit to the structural leap of actual co-ownership. Let employers experience the benefits of aligned, transparent, self-correcting employment relationships before asking them to restructure their corporate governance. The destination is the same. The path is less steep.
XV. Conclusion
The standard employment contract is a bundle of workarounds for the absence of three primitives. The hourly rate compensates for the absence of revenue coupling. Annual reviews compensate for the absence of bounded engagement windows. Managerial authority compensates for the absence of signal legibility. Performance bonuses, retention packages, equity vesting schedules, and cost-of-living adjustments are all downstream patches on a foundation that does not support the weight placed on it.
The framework proposed here does not add complexity to employment. It removes the complexity that was created by the absence of these fundamentals. Signal legibility dissolves the information asymmetry that necessitates hierarchical management. Revenue coupling eliminates the fiction of static income. Bounded engagement windows replace the anxious indeterminacy of at-will arrangements with the structured clarity of experimental protocols.
The honest case for hourly pay was never theoretical — it was operational. Hourly compensation is trivially enforceable: hours times rate equals pay, and any deviation is immediately verifiable by either party, any labor board, or any court. Revenue coupling, by contrast, historically required forensic access to business internals that employers resisted providing and workers lacked the capacity to audit. This enforcement simplicity advantage was real, and it explains a century of persistence better than any economic argument ever could. But the advantage was technological, not structural — and for the first time, AI-mediated auditing makes it possible to verify revenue-coupled compensation with the same ease and reproducibility as checking a timesheet. The last operational argument for hourly pay is no longer insurmountable. What remains is adoption.
The core claim is simple and falsifiable: there is no employment relationship in which the triadic framework produces worse outcomes than the hourly-wage, at-will, signal-opaque alternative. Roles that appear to resist revenue coupling—support, prevention, infrastructure—are not exceptions to the framework but modeling failures within the current paradigm, correctable through cost-of-failure valuation, inherited transactional endpoints, or opportunity-cost pricing. The current model is not wrong because it fails catastrophically. It is wrong because it is unnecessary. It is bad engineering—a set of compensatory mechanisms layered on top of a flawed premise when the correct premise was available all along. Your income is not arbitrary. It is not static. It is not detached from the system in which you work. Any contract that asserts otherwise is, at best, a convenient fiction—and at worst, a structural mechanism for transferring value from those who create it to those who control access to information about it.
References
Bartel, A., Ichniowski, C., & Shaw, K. (2007). How does information technology affect productivity? Plant-level comparisons of product innovation, process improvement, and worker skills. Quarterly Journal of Economics, 122(4), 1721–1758.
Bernstein, E., Bunch, J., Canner, N., & Lee, M. (2016). Beyond the holacracy hype. Harvard Business Review, 94(7), 38–49.
Blasi, J. R., Freeman, R. B., & Kruse, D. L. (2013). The Citizen’s Share: Putting Ownership Back into Democracy. Yale University Press.
Clark, J. B. (1899). The Distribution of Wealth: A Theory of Wages, Interest, and Profits. Macmillan.
Coase, R. H. (1937). The nature of the firm. Economica, 4(16), 386–405.
Errasti, A. M., Heras, I., Bakaikoa, B., & Elgoibar, P. (2003). The internationalisation of cooperatives: The case of the Mondragon Cooperative Corporation. Annals of Public and Cooperative Economics, 74(4), 553–584.
Grossman, S. J., & Hart, O. D. (1983). An analysis of the principal-agent problem. Econometrica, 51(1), 7–45.
Hicks, J. R. (1932). The Theory of Wages. Macmillan.
Holmström, B. (1979). Moral hazard and observability. Bell Journal of Economics, 10(1), 74–91.
Jensen, M. C., & Meckling, W. H. (1976). Theory of the firm: Managerial behavior, agency costs and ownership structure. Journal of Financial Economics, 3(4), 305–360.
Kruse, D. L. (1993). Profit Sharing: Does It Make a Difference? W.E. Upjohn Institute.
Lazear, E. P. (2000). Performance pay and productivity. American Economic Review, 90(5), 1346–1361.
Robertson, B. J. (2015). Holacracy: The New Management System for a Rapidly Changing World. Henry Holt.
Society for Human Resource Management (SHRM). (2022). Retaining Talent: A Guide to Analyzing and Managing Employee Turnover. SHRM Foundation.
Stack, J. (1992). The Great Game of Business. Currency Doubleday.
Wasserman, N. (2012). The Founder’s Dilemmas: Anticipating and Avoiding the Pitfalls That Can Sink a Startup. Princeton University Press.
Weitzman, M. L. (1984). The Share Economy: Conquering Stagflation. Harvard University Press.
Whyte, W. F., & Whyte, K. K. (1991). Making Mondragon: The Growth and Dynamics of the Worker Cooperative Complex (2nd ed.). ILR Press.
Williamson, O. E. (1975). Markets and Hierarchies: Analysis and Antitrust Implications. Free Press.
Appendix A: Contract Skeleton
The following is a minimal worked example demonstrating the triadic framework applied to a non-executive role at a technology hardware sales company. The contract is structured in three sections corresponding to the three primitives. Each section is necessary; omitting any one produces a documented failure mode (see Section IV).
TRIADIC EMPLOYMENT AGREEMENT
Between: TechCo Inc. (“Company”) and Jane Doe (“Contributor”)
Role: Sales Operations & CRM Developer
Effective Date: March 1, 2026
SECTION 1: SIGNAL LEGIBILITY
1.1. Within five (5) business days of the Effective Date, Company shall provision Contributor with read access to the following systems and data:
- Real-time revenue dashboard showing unit sales, transaction volume, and gross revenue at the Transactional Endpoint defined in Section 2.1
- Cost structure for the division in which Contributor operates, including headcount costs, infrastructure costs, and marketing spend
- Customer acquisition metrics including lead volume, conversion rate, and average deal size
- Historical revenue data for the Transactional Endpoint for the twelve (12) months preceding the Effective Date
1.2. Dashboard access shall be continuous and real-time. Company shall not delay, obscure, aggregate, or editorialize revenue data provided to Contributor. If the reporting system experiences downtime exceeding twenty-four (24) hours, Company shall provide equivalent data in written form within forty-eight (48) hours.
1.3. Contributor agrees to maintain confidentiality of all data accessed under this Section in accordance with the Non-Disclosure Agreement executed concurrently with this Agreement.
1.4. Rationale. Signal legibility is the precondition for fair negotiation, accurate self-assessment, and aligned incentive structures. Without it, Contributor cannot evaluate whether their compensation reflects their contribution, and Company cannot expect Contributor to optimize for outcomes Contributor cannot observe.
SECTION 2: REVENUE COUPLING
2.1. Transactional Endpoint. The Transactional Endpoint for this Agreement is defined as: completed laptop unit sales processed through TechCo’s point-of-sale system, net of returns and chargebacks, attributed to the sales division supported by Contributor’s role.
2.2. Compensation. Contributor shall receive:
- Base Rate: $2,400 per engagement window (equivalent to minimum viable floor; see Section 3.1 for window definition)
- Revenue-Coupled Component: 0.8% of gross revenue generated at the Transactional Endpoint during each engagement window
2.3. Payment. Revenue-coupled compensation shall be calculated and paid within five (5) business days of the close of each engagement window, using revenue figures from the dashboard specified in Section 1.1.
2.4. Illustrative Scenario. If the sales division generates $600,000 in laptop revenue during a two-week engagement window, Contributor’s compensation for that window is: $2,400 (base) + $4,800 (0.8% × $600,000) = $7,200. Annualized at consistent revenue: approximately $187,200. If revenue declines 30% to $420,000/window: $2,400 + $3,360 = $5,760/window (~$149,760 annualized) — but Contributor has real-time visibility into the decline, structural incentive to diagnose it, and a window boundary at which to renegotiate terms.
2.5. Attribution for Non-Direct Roles. Contributor’s role (Sales Operations & CRM Developer) does not generate laptop sales directly. Contributor’s revenue coupling is to the Transactional Endpoint of the system Contributor supports: the sales division’s completed transactions. Contributor inherits the transactional endpoint of the upstream function their work enables. The coupling percentage (0.8%) reflects the scaling factor of Contributor’s infrastructure work on sales efficiency, negotiated at the Effective Date and renegotiable at each window boundary per Section 3.3.
SECTION 3: BOUNDED ENGAGEMENT WINDOW
3.1. Window Duration. Each engagement window is fourteen (14) calendar days, beginning on the Effective Date and renewing automatically per Section 3.3.
3.2. Early Termination. Either party may terminate this Agreement prior to the close of an engagement window. The terminating party shall pay the terminated party an amount equal to the revenue-coupled compensation that would have been earned for the remainder of the window, calculated using the average daily revenue at the Transactional Endpoint over the preceding thirty (30) days. This early termination payment is due within five (5) business days of termination.
- Example. If Company terminates Contributor on Day 4 of a 14-day window, and the trailing 30-day average daily revenue at the Transactional Endpoint is $42,857 (~$600K/14 days), Company owes Contributor: 10 remaining days × (($42,857 × 0.8%) + ($2,400 / 14)) = 10 × ($342.86 + $171.43) = $5,142.86.
3.3. Automatic Renewal and Renegotiation. At the close of each engagement window, the Agreement renews automatically under the same terms unless either party provides written notice of intent to renegotiate at least three (3) business days before the window closes. Renegotiation may address: the revenue coupling percentage, the Transactional Endpoint definition, the window duration, the base rate, or any term of this Agreement. If renegotiation does not produce agreement within five (5) business days of the window close, either party may elect to terminate under the standard terms of Section 3.2, or both parties may jointly submit the dispute to binding arbitration per Section 4.
3.4. Rationale. The two-week window length corresponds to standard payroll cycles, agile sprint durations, and the minimum observation period required to generate meaningful revenue data at most transactional endpoints. Shorter windows produce insufficient signal for evaluation; longer initial windows require higher trust than early-stage arrangements warrant. Window duration is itself a renegotiable term and is expected to lengthen as the arrangement matures — a structural analog to the progression from casual engagement to committed partnership.
SECTION 4: DISPUTE RESOLUTION
4.1. Any dispute arising under this Agreement shall first be addressed through direct negotiation between the parties within five (5) business days of written notice of dispute.
4.2. If direct negotiation fails, the dispute shall be submitted to binding online arbitration through [AAA / Rapid Ruling / platform of mutual selection], with costs split equally between the parties. The arbitrator’s decision shall be final.
4.3. Neither party shall make public statements regarding the dispute or the terms of this Agreement prior to the conclusion of the arbitration process. Escalation to public forums (social media, review platforms, press) is available only after the arbitration process has concluded without compliance by the non-prevailing party.
SECTION 5: SIGNATURES
Company: Date:
Contributor: Date:
Reading the Contract Against the Framework
This contract contains no annual review mechanism because the engagement window boundary replaces it. It contains no performance bonus structure because revenue coupling replaces it. It contains no managerial authority clause because signal legibility makes directive management unnecessary — Contributor can see the system and identify priorities independently. It contains no retention package because the automatic renewal with renegotiation rights makes retention a structural outcome of fair terms rather than a bribe to tolerate unfair ones.
Every compensatory mechanism present in standard employment contracts is absent here — not because it was removed, but because the conditions that necessitated it do not exist. The contract is shorter than a standard employment agreement because it does less work. It does less work because the foundation is sound.