PRIORITY FILING - INITIAL DRAFT
A fully revised and formatted version will follow.
Gregory Caldwell Beier
Susarb LLC, Cambridge, Massachusetts
March 19, 2026
From the Prolegomena to the Unification of Economics, Computer Science, Mathematics, and Physics by Gregory Caldwell Beier (forthcoming 2026).
© 2026 Gregory Caldwell Beier. Licensed under CC BY-NC-ND 4.0.
Every economic transaction is preceded by judgment. The buyer assessed value, weighed alternatives, considered risks, and formed an expectation about future states of the world. The seller did the same from a different vantage point. These judgments - often representing years of accumulated domain expertise applied to specific constraint environments - are the cognitive inputs to the transaction. The price retains none of them. The scalar that clears the market carries the intersection of willingness to pay and willingness to accept, compressing away the analytical reasoning that produced both. This paper argues that human judgment is the most abundant and most wasted resource in the global economy. It is present in every transaction, captured by no existing mechanism, and recoverable through the Disclosure Incentive (Beier, 2026b). The analogy to Kary Mullis's polymerase chain reaction is structural: before PCR, DNA was present in every biological sample but undetectable at useful concentrations. PCR did not create new DNA - it amplified what was already there. The Disclosure Incentive does not create new judgment - it preserves, structures, scores, and amplifies the judgment that economic agents already produce in the course of their ordinary activities but that the price system discards. The economics of judgment are the economics of a resource whose effective supply is zero not because the resource does not exist but because no mechanism has captured it. Tetlock's (2005) research program confirmed that judgment quality varies enormously and is measurable. Kahneman and Tversky's research program documented the systematic biases that degrade it. The missing piece was a mechanism that captures judgment at scale, scores it against outcomes, and feeds the scores back to improve the judgment itself. The Disclosure Incentive provides that mechanism. The economic value of the captured judgment is the difference between allocation decisions made on compressed price signals and allocation decisions made on the richer information environment that preserved judgment creates.
Keywords: judgment, bounded rationality, expertise, information architecture, price compression, amplification, calibration, disclosure, track record capital
JEL Codes: D82, D83, D91, J24, G41
Consider the information produced and destroyed in a single capital allocation decision. A portfolio manager deciding whether to invest in a water-intensive agricultural company in central California has consulted aquifer depletion data, assessed the regulatory trajectory for groundwater extraction, modeled the interaction between the company's water usage and the extraction patterns of neighboring operations, evaluated the company's disclosed climate adaptation strategy against her own assessment of precipitation trends, and formed a probability distribution over the company's earnings trajectory conditional on constraint outcomes she has spent fifteen years learning to evaluate.
The investment decision compresses all of this into a position size. If the manager buys, the market observes a purchase at the prevailing price. If she declines, the market observes nothing. In neither case does the manager's judgment - the aquifer assessment, the regulatory trajectory, the interaction model, the precipitation analysis, the conditional earnings distribution - enter the information environment available to other participants. The judgment existed. It informed a decision. And the price system discarded it.
This is not a special case. It is the universal condition of economic coordination. Every transaction is preceded by judgment that the transaction's scalar representation does not preserve. The cumulative volume of judgment produced and discarded across the global economy in any given day - by fund managers, engineers, farmers, doctors, teachers, analysts, builders, and every other agent whose decisions depend on assessments of uncertain future states - dwarfs the information content of the price system that records their decisions.
The present paper develops the economics of this wasted resource. The argument is not that the price system should be replaced - Hayek (1945) was right that prices aggregate dispersed knowledge with extraordinary efficiency - but that the judgment the price system discards has economic value that no existing mechanism captures. The Disclosure Incentive (Beier, 2026b) is the mechanism: it preserves judgment at the point of production, structures it in a format that can be evaluated, scores it against outcomes, and feeds the scores back to improve subsequent judgment. The economics of judgment become visible only when a mechanism for capturing judgment exists, just as the economics of genetic information became visible only when PCR provided a mechanism for amplifying it.
Section 2 characterizes judgment as an economic resource. Section 3 explains why the price system discards it. Section 4 develops the PCR analogy. Section 5 describes what changes when judgment is captured. Section 6 addresses the quality problem. Section 7 states limitations. Section 8 concludes.
Human judgment - the cognitive process through which agents assess uncertain situations, weigh evidence, form probability estimates, and reach decisions - has been studied extensively but has not been treated as a resource in the economic sense. Simon's (1955) bounded rationality framework recognizes that agents operate with limited information, processing limited information with limited computational capacity. The systematic biases catalogued by Kahneman and Tversky (1979) that distort probability assessment. The ecological rationality program (Gigerenzer, Todd, and the ABC Research Group, 1999) argues that simple heuristics can be ecologically rational - well-adapted to specific decision environments even when they violate the axioms of expected utility theory.
This research program has treated judgment primarily as a decision-making process to be understood and corrected. The present paper treats it as a resource to be captured and deployed.
The distinction matters. A process that is understood but not captured has no economic footprint - its value cannot be measured, traded, accumulated, or improved through market mechanisms. A resource that is captured, structured, and scored enters the economic system: it can be evaluated by principals (Beier, 2026r, Section 5), priced by markets, accumulated into track records, and improved through the feedback loop that scoring provides.
Tetlock (2005) provided the critical empirical foundation. His twenty-year study of expert political judgment demonstrated two facts that are essential to the economics of judgment. The first is that judgment quality varies enormously across individuals: the best forecasters dramatically outperform the median, and the median barely outperforms simple base-rate algorithms. The second is that judgment quality is measurable: calibration - the correspondence between stated confidence levels and observed outcome frequencies - provides a scalar summary of judgment quality that can be tracked over time and compared across individuals. Tetlock and Gardner (2015) extended these findings through the Good Judgment Project, showing that the best forecasters (superforecasters) maintain their calibration advantage over years, that the advantage persists across domains, and that training and feedback improve calibration measurably.
These findings establish that judgment is a heterogeneous resource with measurable quality variation. What they do not establish - and what the Disclosure Incentive adds - is a mechanism for capturing judgment at the scale of economic activity rather than at the scale of forecasting tournaments. The Good Judgment Project captured judgment by asking volunteers to forecast geopolitical events. The Disclosure Incentive captures judgment by asking economic agents to disclose the reasoning behind their ordinary professional decisions - capital allocation, risk assessment, infrastructure planning, resource management - in structured, scored, attributed form.
The price system discards judgment not through malice or design flaw but through the structural properties of market clearing. Three features of price formation ensure that the judgment behind transactions is lost.
The first is aggregation across agents. A market price aggregates the judgments of all participants into a single scalar. When two agents with opposing views transact, the price reflects the point of agreement but not the reasoning on either side. The bull's thesis about future water availability and the bear's thesis about regulatory risk both informed the transaction, but the price carries only the scalar intersection. The judgments cancel in the aggregation, even though both contain information that other agents could use.
The second is temporal compression. The price records the outcome of a judgment process at a single moment. The weeks of analysis that preceded the portfolio manager's investment decision - the data gathering, the model building, the scenario analysis, the constraint assessment - are compressed into the timestamp of the trade. No mechanism exists for the price to carry the temporal structure of the judgment that produced it: which data came first, how the assessment evolved as new information arrived, what alternative hypotheses were entertained and rejected, and what conditions would cause the manager to reverse the decision.
The third is format incompatibility. Judgment is high-dimensional and conditional: "I believe outcome X with probability p, conditional on assumptions A, B, and C, with the main risk being that assumption B may fail under scenario S." A price is unconditional and one-dimensional: the asset trades at $47.50. There is no field in the price format for conditional structure, for confidence levels, for the identification of key assumptions, or for the specification of scenarios under which the assessment changes. The format mismatch is total, and the information lost in the format conversion is irrecoverable from the price alone.
The companion paper on lossy compression (Beier, 2026c) formalizes this in rate-distortion terms. The present paper's contribution is to shift attention from the information lost (the compression artifact) to the resource wasted (the judgment that produced the information). The price system destroys information. But behind every piece of destroyed information is a human judgment that cost cognitive effort to produce, that reflected accumulated expertise, and that had economic value which no mechanism captured.
Before Kary Mullis developed the polymerase chain reaction in 1983, DNA was present in every biological sample but existed at concentrations too low for most analytical purposes. A single cell contains its full genome, but one copy of a gene in a sea of cellular material is invisible to detection methods that require millions of copies. PCR did not create new genetic information. It amplified existing information by providing a mechanism - thermal cycling with a heat-stable polymerase - that replicated specific DNA sequences exponentially until they reached detectable and analyzable concentrations.
The structural parallel to judgment in the economy is precise. Judgment is present in every economic transaction - the cognitive assessment that preceded the decision is as ubiquitous as DNA in biological tissue. Like pre-PCR DNA, it exists at concentrations too low for analytical use: one manager's aquifer assessment, unrecorded and unscored, is invisible in the noise of the market. The Disclosure Incentive amplifies judgment by providing a mechanism - structured disclosure with proper scoring and track record accumulation - that preserves individual judgments, verifies them against outcomes, and aggregates them into a public information resource whose concentration and quality are sufficient for economic analysis and coordination.
The analogy extends to the transformation that amplification enables. PCR did not merely make existing DNA detectable - it created the field of molecular biology, enabled forensic genetics, made gene therapy possible, and transformed medicine. The amplification was the enabling technology for an entire domain of science and application that was latent in the biology but inaccessible without the mechanism. Similarly, capturing judgment at economic scale does not merely make existing assessments visible - it creates a new information resource (the judgment data stream) that enables coordination improvements, AI training on structured human reasoning, and real-time constraint pricing that are latent in the economy's cognitive capacity but inaccessible without the capture mechanism.
The analogy has a limit. DNA replication is mechanical and exact - each PCR cycle doubles the target sequence with high fidelity. Judgment capture through disclosure is neither mechanical nor exact: agents must choose to disclose, the disclosure must be structured and honest, and the scoring must be accurate. The companion paper on incentive compatibility (Beier, 2026r) establishes the conditions under which honest disclosure is the dominant strategy. The companion paper on adversarial economics (Beier, 2026an) addresses the conditions under which the mechanism fails. The PCR analogy captures the structural insight (an abundant resource, invisible without amplification, transformative once captured) without implying that the capture mechanism is as reliable as enzymatic replication.
The effective supply of judgment in the current economy is near zero - not because agents lack judgment but because no mechanism captures it. When the Disclosure Incentive provides the capture mechanism, three things change.
The first is that allocation decisions improve. Capital allocators currently make decisions on the basis of prices, financial statements, and reconstructed analysis (sell-side research, ESG ratings, consultant reports). Each of these information sources is a compression of someone else's judgment, filtered through institutional incentives and format constraints. When allocators can additionally observe the disclosed constraint reasoning of agents with demonstrated calibration quality - the fund manager's aquifer assessment, the engineer's infrastructure resilience analysis, the agronomist's soil depletion forecast - the information environment for allocation decisions is qualitatively richer. The improvement is not additive (more data points) but structural (a different type of information that the existing sources cannot carry).
The second is that judgment itself improves. The improvement mechanism developed in the companion paper (Beier, 2026b, Section 3.4) is the distinctive property of judgment as a resource: capturing it makes it better. When a manager discloses structured constraint reasoning and receives calibration feedback - learning which assessments corresponded to outcomes and which did not - the feedback loop trains better judgment over successive cycles. The mechanism has experimental support: Lichtenstein and Fischhoff (1980) found that subjects who received structured calibration feedback became measurably better calibrated. Ericsson, Krampe, and Tesch-Romer (1993) identified the conditions (structured task, unambiguous feedback, sufficient iteration) under which practice produces genuine skill improvement. The Disclosure Incentive satisfies all three conditions for the domain of probabilistic judgment about constraint environments. No other economic resource has this property - capturing it increases its quality.
The third is that AI systems gain access to a training resource that does not currently exist. Current AI training corpora consist of text produced for purposes other than structured judgment disclosure - news articles, social media posts, financial reports, academic papers. Structured human judgments that are calibrated, attributed, and scored against outcomes constitute a qualitatively different training input, one that preserves the reasoning path rather than just the conclusion. The cooperative channel of the Disclosure Incentive (Beier, 2026b, Section 7.2) produces this training resource as a byproduct of its primary function. The competitive channel's incentive is secrecy, which is incompatible with training data production. Only the cooperative channel generates the disclosed, scored, structured reasoning that AI systems could train on.
If judgment is heterogeneous in quality - and Tetlock's evidence is unambiguous that it is - then capturing more judgment is only valuable if the capture mechanism differentiates quality. A flood of poorly calibrated disclosures would degrade rather than enrich the information environment.
The Disclosure Incentive addresses the quality problem through proper scoring. Each disclosed judgment is scored against outcomes using a strictly proper scoring rule, and the cumulative score constitutes the track record. Agents with high calibration quality accumulate track records that principals value; agents with poor calibration quality accumulate track records that principals discount. The market for judgment quality differentiates, and the information environment is enriched primarily by the disclosures of well-calibrated agents.
Kahneman and Klein (2009) identified the boundary conditions for this differentiation to work. Feedback improves judgment only in "high-validity environments" - domains where the relationship between cues and outcomes is stable enough for the practitioner to learn. The Disclosure Incentive's structured format, proper scoring, and explicit confidence levels are designed to engineer high-validity feedback in domains where the natural environment is low-validity (long-horizon constraint reasoning, where outcomes are delayed and noisy). Whether the engineering is sufficient is an empirical question. The companion paper (Beier, 2026b, Section 3.4) argues it is, grounded in the experimental evidence from Lichtenstein and Fischhoff (1980) and the deliberate practice framework from Ericsson et al. (1993). The present paper flags the boundary condition without resolving it: the economics of judgment depend on the improvement mechanism working, and the improvement mechanism depends on the feedback environment being sufficiently valid.
The Scaling field provides the engineering solution to this validity problem. A fifty-year aquifer depletion forecast is a low-validity prediction - the outcome is too distant for meaningful feedback. But the Scaling field decomposes this into intermediate milestones: three-year extraction trends, five-year recharge measurements, decade-scale water table trajectories. Each milestone is scoreable on a timeline short enough for calibration feedback to function. The disclosure format transforms a single low-validity long-horizon forecast into a sequence of high-validity intermediate assessments, enabling the deliberate practice mechanism that Ericsson et al. (1993) identified as the engine of genuine skill improvement.
The biases documented by Kahneman and Tversky (1979) - overconfidence, anchoring, availability, representativeness - are not eliminated by the Disclosure Incentive. They are measured by it. An overconfident agent who systematically states higher confidence than outcomes warrant accumulates a track record that reveals the overconfidence. The bias is not corrected automatically by the format; rather, it is made visible, and this visibility is what triggers the calibration feedback loop enabling self-correction (through calibration feedback) and market correction (through principals discounting the track records of agents whose biases the scores reveal).
The PCR analogy, while structurally informative, risks overstating the reliability of judgment capture. DNA amplification operates on a physical substrate with known chemistry. Judgment capture operates on a cognitive substrate with imperfectly understood dynamics. The fidelity of the capture (whether disclosed judgment accurately represents the agent's actual assessment), the reliability of the scoring (whether outcomes are observable and attributable within the relevant time horizon), and the effectiveness of the feedback loop (whether scored calibration actually improves subsequent judgment) are all empirical questions that the Disclosure Incentive's design addresses but does not guarantee.
The claim that judgment is "the most abundant and most wasted resource" is rhetorical and not precisely measurable. The total volume of judgment produced across the global economy cannot be estimated with existing methods. What can be estimated is the volume of judgment that is structured, disclosed, and scored - which is currently near zero. The gap between the total judgment produced and the total judgment captured is the economic opportunity, but the opportunity's magnitude depends on what fraction of uncaptured judgment has decision-relevant content, which varies by domain and agent.
The paper treats judgment as a single resource type with a single quality measure (calibration). In practice, judgment is heterogeneous across domains, time horizons, and decision contexts. An agent well-calibrated on short-horizon earnings forecasts may be poorly calibrated on long-horizon constraint assessments. The track record mechanism accommodates this heterogeneity (by scoring within domains), but the economics of judgment are more complex than a single-resource model suggests.
The paper does not address the labor economics of judgment capture. If the Disclosure Incentive requires agents to produce structured disclosures alongside their existing duties, the time cost is a real economic input. Adding an observable task to a multitask setting distorts effort allocation (Dewatripont, Jewitt, and Tirole, 1999). The companion paper (Beier, 2026b, Section 3.3) addresses this concern but does not resolve it fully. The economics of judgment must account for the production cost of disclosure alongside the information value of the captured judgment.
The resource was always there. Every buyer who assessed value, every seller who assessed cost, every analyst who modeled risk, every engineer who evaluated capacity, every farmer who judged the season - each produced judgment that the price system discarded. The economics of judgment begin with the observation that a resource present in every transaction and captured by none represents an opportunity whose scale is proportional to the volume of economic activity itself.
What was missing was not the resource but the capture mechanism. The Disclosure Incentive provides it: structured format, proper scoring, track record accumulation, and the improvement feedback loop that makes the captured resource grow in quality as it grows in volume. The analogy to PCR is not decorative. Before amplification, the resource is invisible. After amplification, it transforms the system. The judgment that economic agents already produce, once captured and scored and fed back, creates an information environment that the price system operating alone cannot approach - because the price system was designed to clear markets, not to preserve the reasoning of the people whose decisions the markets coordinate.
This paper presents results from the Prolegomena to the Unification of Economics, Computer Science, Mathematics, and Physics (Beier, forthcoming 2026). The economics of judgment developed here provides the resource-theoretic foundation for the Disclosure Incentive (Beier, 2026b) and connects the improvement mechanism of that paper to the broader literatures on bounded rationality, calibration, and deliberate practice. The author invites engagement from the behavioral economics, decision science, and human capital communities.