Kenneth Arrow

Record year1950–1963

summaryFormalizes individual preference orderings, social welfare functions, and a set of normative conditions, proving that with at least three options these conditions generally cannot be simultaneously satisfied by a non-dictatorial social ordering rule.

M rationaleThis result provides a formal boundary that is continuously reused across economics, politics, and group decision-making, and can change how institutional questions are posed, earning M6.6. Condition dependence and the non-prescriptive nature regarding actual institutions cap the upper bound; no adjustment is made for reputation or awards.

attribution rationaleArrow 0.46; prehistory of voting and welfare theory 0.26; discussion review community 0.09; Cowles institution 0.07; unresolved 0.12.

overlap boundaryIn the 1950 paper, the 1951 "Social Choice and Individual Values," and the 1963 second edition, with individual weak orderings as input, under conditions of non-dictatorship, unrestricted domain, Pareto/citizen sovereignty, and independence of irrelevant alternatives, the impossibility of aggregating social orderings is proved. Only the conditional formal framework and theorems are attributed; not all voting systems, democratic practices, or subsequent mechanism designs are attributed to Arrow.

Evidence gapsIt is necessary to check chapter by chapter the axiom statements and proof revisions in the 1951 and 1963 editions. It is also necessary to further separate contemporaneous independent results by Black and others from direct precursor inputs.

Record year1954

summaryArrow and Gérard Debreu, under explicit assumptions on preferences, technology, and convexity, proved the existence of equilibrium for a competitive economy including production, exchange, and consumption.

M rationaleThis result provides a long-reused unified existence framework for compatibility in complex markets, rated M6.7. Abstract assumptions and non-empirical prediction boundaries enter the scope judgment, avoiding attributing the entire consequences of the market economy back to it.

attribution rationaleArrow and Debreu each 0.24; equilibrium, fixed point, and prehistory of game theory 0.25; Cowles community 0.09; institutions 0.07; unresolved 0.11.

overlap boundaryIn 1954, Arrow and Debreu proved the existence of equilibrium for a competitive economy with production, exchange, and consumption, under explicit assumptions such as preferences, technology, and convexity. Only existence and the abstract economy proof framework are counted; equilibrium stability, computation, uniqueness, empirical realism, and all macroeconomic policy are excluded.

Evidence gapsAuthor correspondence or archival calibration is needed for the item-by-item division of proof labor between Arrow and Debreu. A separate review is needed to clarify the boundaries with McKenzie's parallel proof and Debreu's early social equilibrium theorem.

Record year1962

summaryIt analyzes how information is difficult to price before production, cheap to reproduce afterward, and difficult to fully exclude, and explains why these characteristics lead markets to underinvest in invention.

M rationaleThis work identifies a resource allocation mechanism that is pervasive in knowledge production and reusable across industries, rated M5.7. It is normative analysis, not evidence of the effectiveness of any patent or subsidy scheme.

attribution rationaleArrow 0.42; prehistory of public goods, patents, and welfare theory 0.25; NBER conference community 0.12; institutions 0.09; unresolved 0.12.

overlap boundaryIn the 1962 'Economic Welfare and the Allocation of Resources for Invention', welfare analysis is conducted on the uncertainty of invention, difficulties in information dissemination/appropriation, and insufficient market incentives. Only the general information—innovation allocation framework is counted; specific patent policies, R&D institutions, internet information, or subsequent results in information economics are all excluded.

Evidence gapsIts conditional judgments on patents, monopoly, government, and basic research require page-by-page verification. Direct transmission to contemporaneous innovation-economics work such as Nelson's must also be distinguished from mere historical co-occurrence.

Record year1963

summaryAn open economic theoretical analysis of the medical service system, explaining how the uncertainty of illness and treatment and the differences in professional knowledge relate to insurance, trust, professional norms, and market boundaries; does not include clinical advice.

M rationaleThis work connects general economic theory with the specific uncertainties of the medical system into an independent, long-term reusable analytical framework, scoring M5.8. The scope is strictly limited to public theoretical/institutional knowledge, consistent with the current non-clinical boundary; the evidence does not support any treatment or policy effectiveness.

attribution rationaleArrow 0.35; prehistory of welfare, insurance, and medical markets 0.20; knowledge community of medical system 0.15; named commentators on the original text 0.10; institutions 0.08; unresolved 0.12.

overlap boundaryThe 1963 public paper conducts normative economic analysis of the medical service industry, not individual health or treatment plans, explaining how the uncertainty of disease occurrence, diagnosis/treatment effectiveness, and professional knowledge correspond to insurance, information asymmetry, trust, licensing, and non-market norms. Only public theoretical and institutional knowledge is counted; no case data, clinical materials, or treatment recommendations are involved.

Evidence gapsA full section-by-section verification of the conditional conclusions on insurance, professional norms, and licensing systems is required; a medical economics review is needed to distinguish its conceptual influence from the independent increments of subsequent empirical and policy research.

Record year1952–1965

summaryIt uses state securities to show how risk is allocated under specific assumptions and develops a theoretical language for analyzing risk aversion as wealth changes; it does not imply that real financial markets are necessarily efficient.

M rationaleThis result provides a general, reusable theoretical structure for cross-state risk allocation, rated M5.6. It is separated from the 1954 certainty/integrated-equilibrium existence result by its increment on uncertainty and risk allocation, and the shared equilibrium baseline cannot be accounted for twice.

attribution rationaleArrow 0.40; expected utility, insurance, and pre-equilibrium history 0.30; conference community 0.10; institutions 0.07; unresolved 0.13.

overlap boundaryThe optimal risk allocation via state securities in the 1952 report, 1953 French publication, and 1964 English version, together with the risk-aversion framework in 1965 'Aspects of the Theory of Risk-Bearing', form one family. Only the general theory of securities/preferences under uncertainty is counted; Pratt's independent results, actual financial market performance, derivatives, and all insurance products are excluded.

Evidence gapsNeed to verify the 1965 text and strictly separate it from Pratt's contemporaneous risk-aversion measures; also need to distinguish the original proof of state securities from the later overlap with the Arrow–Debreu uncertainty model.

Record year1962

summaryIt incorporates knowledge formed through production and investment experience into a growth model and analyzes how learning spillovers cause private investment and social returns to learning to diverge.

M rationaleThis work elevates the empirical learning curve into an analyzable growth mechanism, with reusable value across productivity and innovation policy, and is rated M5.4. The model's simplified empirical proxy and the independent contributions of subsequent theory cap the rating.

attribution rationaleArrow 0.43; prehistory of learning curves and growth accounting 0.28; empirical community 0.09; institutions 0.07; unresolved 0.13.

overlap boundaryIn 1962, it treated knowledge growth as a byproduct of production and investment experience and analyzed the implications for long-run growth and resource allocation. Only this endogenous learning mechanism is counted; later endogenous growth, firm learning curves, all productivity growth, and policy effects are excluded.

Evidence gapsThe model's precise restrictions on capital stock, experience, and knowledge spillovers require section-by-section confirmation. The independent shares of subsequent endogenous-growth work and of the empirical prehistory associated with Alchian and others require a dedicated review.

Record year1951

summaryArrow, Theodore Harris, and Jacob Marschak combined intertemporal inventory with stochastic demand, analyzing parameters and discounted costs within a prespecified (s,S) policy class; they do not claim Scarf's later broader optimality proof.

M rationaleM5.5 reflects a widely reused inventory-decision method under stochastic demand and fixed adjustment costs. The original authors explicitly optimized only within a prespecified policy family, so this entry does not claim a general optimality proof; its independent increment is the inventory stochastic process and policy parameters, rather than a second claim on general recursive decision-making.

attribution rationaleSL-0138 0.24: jointly organized the stochastic intertemporal inventory problem with Marschak; EXT-NAME:JACOB-MARSCHAK 0.24: joint problem construction and inventory model; EXT-NAME:THEODORE-HARRIS 0.27: analysis of the inventory stochastic-process distribution and discounted costs; ARROW::INVENTORY-AND-CONSUMED-SEQUENTIAL-METHODS 0.13: prior inventory-theory and recursive-method inputs already accounted for; ARROW::RAND-COWLES-RESEARCH-CONTEXT 0.05: RAND and Cowles research collaboration; ARROW::LOGISTICS-BIBLIOGRAPHY-INPUT 0.03: inventory-research literature and logistics-problem support; ARROW::UNRESOLVED-1951-ROLES 0.04: original research records not fully preserved

overlap boundaryIn 1951 Arrow, Harris, and Marschak combined intertemporal inventory with stochastic demand, deriving parameters and discounted costs within a prespecified (s,S) policy class. The deterministic two-threshold form existed earlier; this entry excludes Scarf's 1960 proof of optimality over a broader policy class and does not claim all modern supply-chain optimization.

Evidence gapsAll technical assumptions and the scope of modern applications of the 1951 paper still require item-by-item verification. Arrow's account is a recollection from years later, explicitly lacking original records, so fine-grained attribution remains uncertain.

Record year1949

summaryArrow, David Blackwell, and Meyer Girshick analyzed the choice of continuing sampling or stopping action when sampling has costs, characterizing optimal sequential strategies for bounded problems via recursive structures.

M rationaleM5.3 reflects reusable sequential-decision forms and their early input into the expression of dynamic programming. It concerns statistical decision problems and does not claim all of dynamic programming on the grounds of influence on Bellman.

attribution rationaleSL-0138 0.25: joint modeling and recursive decision expression; EXT-NAME:DAVID-BLACKWELL 0.25: joint statistical decision derivation; EXT-NAME:MEYER-GIRSHICK 0.25: joint statistical decision derivation; ARROW::WALD-SEQUENTIAL-ANALYSIS-DECISION-PREHISTORY 0.15: prior sequential analysis and statistical decisions; ARROW::RAND-RESEARCH-CONTEXT 0.05: research seminar and collaborative environment; ARROW::UNRESOLVED-1949-ROLES 0.05: section-by-section manuscripts and division of labor pending analysis

overlap boundaryIn 1949, Arrow, Blackwell, and Girshick studied costly choices between continued sampling and stopping, characterizing optimal sequential decisions and recursive structures. This entry does not claim Wald's prior sequential analysis or Bellman's later full dynamic-programming system; specific inventory policies are accounted for in a separate item.

Evidence gapsAll technical assumptions in the 1949 original text and the three authors' section-by-section contributions still require verification. The direct-accounting boundary with later dynamic programming and the 1951 inventory method requires further topic-specific calibration.

Record year1958–1959

summaryArrow, with Leonid Hurwicz and H. D. Block, studied when price adjustment tends toward competitive equilibrium under conditions such as gross substitutability, supplementing the convergence question left unanswered by the existence proof.

M rationaleThis is an independent, condition-specific family of methods in general-equilibrium dynamics, addressing convergence questions not covered by the existence result, and is rated M5.2. Because its assumptions are narrow and it is not a general law of actual markets, it is rated below the existence framework.

attribution rationaleArrow and Hurwicz each 0.22, Block 0.12; prehistory of dynamic stability 0.20; community 0.08; institutions 0.06; unresolved 0.10.

overlap boundaryThe 1958 Arrow–Hurwicz and 1959 Arrow–Block–Hurwicz studies on competitive-equilibrium stability analyze when price adjustment converges locally or globally to equilibrium under conditions such as gross substitutability. They are separate from the 1954 existence proof; this entry does not claim real-market adjustment speeds or all computational equilibrium algorithms.

Evidence gapsThe two proofs and version differences of the gross substitutability condition have not yet been read page by page; subsequent counterexamples and stability theory need to be reviewed to calibrate the scope of applicability.