Measuring Information Burden: From Coalition-Based Reasoning to the Price System
Listen
Radio episode about this paper
Transcript
Introduction to the show: ident: Economics. The day in economics: theory, mechanisms and the evidence beneath them.
Ruth: I'm Ruth.
Warren: And I'm Warren, here with Iris.
Ruth: We're looking at the paper: "Measuring Information Burden".
Warren: Here's the gist of the paper. The price system is often said to economize on information <ref:1802.04595#pg2> This paper asks how much information it saves <ref:1802.04595#pg4>.
Ruth: Formal Framework for Measuring Burden The authors develop a formal framework for measuring the informational burden that agents would have to bear if they had to reason individually over feasible coalitional alternatives <ref:1802.04595#pg5>.
Warren: Taking the convergence theorem of Debreu and Scarf (1963) as the benchmark, they ask how much coalition-feasibility information must be held, and how it must be distributed, for individual acceptance decisions to recover the core <ref:1802.04595#pg5>.
Ruth: Agents’ information is represented as finite sets of meaning-bearing sentences in a formal language, and reasoning is modeled as proof in Gentzen’s sequent calculus <ref:1802.04595#pg8>.
Title and authors: Ruth: So we're talking about the paper "Measuring Information Burden: From Coalition-Based Reasoning to the Price System," and it’s about quantifying the informational savings of using prices instead of individual reasoning over coalitions.
Warren: It’s essentially building a formal way to measure that burden, looking at how much information agents need to hold if they have to reason individually over every possible combination of players >
Iris: The paper sets up this framework by using a formal language for agent information and modeling the decision-making process as a proof in Gentzen’s sequent calculus >
Ruth: And they use the convergence theorem of Debreu and Scarf from one thousand nine hundred sixty-three as their benchmark to figure out exactly how much coalition feasibility information is needed to recover the core > <ref:1802.04595#pg1,the convergence theorem of Debreu and Scarf>
Warren: The main finding there is that for unanimous acceptability to match the core across all transferable utility games, every coalition must be known by at least one member >
Iris: They also show that if you remove even a single coalition's information from everyone in it, you can break that equivalence for certain games >
Ruth: So they redefine the core as this exact boundary for what we can unanimously justify with a minimal but well-defined information structure >
Warren: Then they take this and apply it to the k-fold replica economy to count exactly which coalitions' feasibility information must be distributed across the entire population >
Iris: This leads them to show that the average per agent informational burden scales as Theta of four k over k cubed over two, which is exponential in terms of the economy's size >
The paper's summary: Ruth: To summarize what they’re doing, this paper takes the classical idea that institutions economize on knowledge and shows how much that saving actually is when we switch from individual reasoning to a price system >
Warren: They use a formal proof structure to determine the exact minimal informational structure required for unanimous acceptability to coincide with the core in these types of games >
Iris: They establish that this minimal structure requires every coalition to be known by at least one agent, and that removing just one piece of information from a coalition breaks the agreement >
Ruth: The paper then moves into a quantitative measure by applying this to replica economies to count the load-bearing coalitions and find their average per agent informational burden >
Warren: They find that this burden grows exponentially in terms of the number of players or agents, specifically as Theta of four k over k cubed over two >
Iris: This result is significant because it gives a precise, formal answer to the Hayekian question about how much information the price system actually saves >
Ruth: It moves beyond just saying prices save info; it shows exactly *how much* information, quantified by this specific scaling factor, is being saved in terms of agent cognitive load >
The paper's improvements: Warren: The authors suggest that the framework itself is a major improvement because it provides a rigorous method for measuring relevant information that isn't just some vague intuition >
Ruth: They introduce the formal language and proof-theoretic apparatus, which gives them exactly the kind of structure needed to make information measurable, not just an abstract concept >
Iris: The real methodological value is in providing this formal setting where relevant information can be measured, showing that institutions economize not only on transaction costs but also on this informational burden >
Ruth: They’ve also applied it to the pure exchange economy and found that the load-bearing coalitions are determined by the geometry of the core in replica economies >
Warren: In an exchange economy setting, they show that for a k-fold replica economy where k is greater than or equal to two, there’s a set Lk of load-bearing coalitions that includes individual coalitions and mixed pairs and one-asymmetric coalitions and the grand coalition >
Iris: And they confirm that the average per agent informational burden in this exchange economy setting grows as Theta of four k over k cubed over two, which diverges to infinity >
Conclusion: Ruth: So to wrap up, the paper "Measuring Information Burden: From Coalition-Based Reasoning to the Price System" gives us a formal answer on how much information is saved by the price system in terms of agent cognitive burden >
Warren: They show that this saving is quantified by an average per agent informational burden growing as Theta of four k over k cubed over two, which means this saving gets exponentially bigger as the economy gets larger >
Iris: The paper provides a rigorous way to measure information, using formal logic and proof theory to determine the minimal knowledge structure needed for equilibrium >
Ruth: It really frames institutional efficiency in terms of information cost, showing that mechanisms economize on cognitive simplicity alongside transaction costs >
Warren: So this is a precise answer to the Hayekian question about how much information the price system saves, quantified by this specific scaling factor and diverging as we scale up >
Department of Decision Sciences, Bocconi University
econ.TH
Submitted: 2018-02-13
Updated: 2026-11-11
License: http://creativecommons.org/licenses/by-nc-sa/4.0/
Importance score: 84/100
The gist: The price system is often said to economize on information <ref:1802.04595#pg2> This paper asks how much information it saves <ref:1802.04595#pg4>.
Key concepts
- Informational Burden
- This is the cognitive load agents face when they must reason individually over all possible coalition alternatives to make a decision. The paper formalizes this burden using logic, showing how much information must be held for individual acceptance decisions to recover the core.
- Core Boundary
- The core is reinterpreted as the exact boundary where unanimous acceptability matches the core. The authors find that this boundary is characterized by requiring every coalition to be known to at least one member, establishing a minimal informational structure needed for unanimous justification.
- Load-Bearing Coalitions
- These are specific coalitions whose feasibility information must be distributed across the population in a k-fold replica economy. Counting these coalitions provides a quantitative measure of the extra information that agents must bear due to market structures.
- Gentzen’s Sequent Calculus
- This is a formal system used to model agent reasoning as mathematical proofs. By using this calculus, the authors can rigorously verify whether an agent can derive a decision criterion from their specific set of held sentences, linking information structure directly to logical conclusions.
Terminology
Summary
The price system is often said to economize on information <ref:1802. p. 0 it
June 2, 2026
Abstract. The price system is often said to economize on information. it
June 2, 2026
Abstract. The price system is often said to economize on information. The price system is often said to economize on information. This paper asks how
much information it saves. The price system is often said to economize on information. This paper asks how
much information it saves.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Formal Framework for Measuring Burden
The authors develop a formal framework for measuring the informational burden that agents would have to bear if they had to reason individually over feasible coalitional alternatives <ref:1802.04595#pg5>. Taking the convergence theorem of Debreu and Scarf (1963) as the benchmark, they ask how much coalition-feasibility information must be held, and how it must be distributed, for individual acceptance decisions to recover the core <ref:1802. p. 0 I develop a formal framework for measuring the informational
burden that agents would have to bear if they had to reason individually over feasible
coalitional alternatives. Taking the convergence theorem of Debreu and Scarf (1963) as the
benchmark, I ask how much coalition-feasibility information must be held, and how it must
be distributed, for individual acceptance decisions to recover the core. Taking the convergence theorem of Debreu and Scarf (1963) as the
benchmark, I ask how much coalition-feasibility information must be held, and how it must
be distributed, for individual acceptance decisions to recover the core. Agents’ information
is represented as finite sets of meaning-bearing sentences in a formal language, and reasoning is modeled as proof in Gentzen’s sequent calculus.Where the paper says it
Where the paper says it
Characterizing the Core Boundary
The main result characterizes the minimal informational structure under which unanimous acceptability coincides with the core for all transferable-utility games on a fixed player set: every coalition must be known to at least one of its members <ref:1802.04595#pg9>. Removing information about even a single coalition from all of its members suffices to break the equivalence for some game <ref:1802. p. 0 The main result characterizes the
minimal informational structure under which unanimous acceptability coincides with the
core for all transferable-utility games on a fixed player set: every coalition must be known
to at least one of its members. Removing information about even a single coalition from all
of its members suffices to break the equivalence for some game. Theorem 1 gives the tight answer: unanimous acceptability coincides with core membership if and
only if every non-empty coalition is known to at least one of its members; in particular, removing
information of even a single coalition from all of its members suffices to break the equivalence for some
game, in which a non-core payoff vector goes unanimously unchallenged. The core is thereby reinterpreted not merely as a normative benchmark, but as the exact boundary of what can be unanimously
justified under a minimal but well-defined information structure, which might be hard to be satisfied
when the game (objectively) becomes large: as the number of participants grows, the informational
burden cannot be kept uniformly low across agents.Where the paper says it
Where the paper says it
Quantitative Measure in Replica Economies
Applying this result to the k-fold replica economy, they identify and count the load-bearing coalitions whose feasibility information must be distributed across the population. The associated average per-agent informational burden grows as Θ(4k/k 3/2)—exponentially in the size of the economy. p. 0 Applying this result to the
k-fold replica economy, I identify and count the load-bearing coalitions whose feasibility
information must be distributed across the population. The associated average per-agent
informational burden grows as Θ(4
k/k
3/2)—exponentially in the size of the economy. The associated average per-agent
informational burden grows as Θ(4
k/k
3/2)—exponentially in the size of the economy. This
is, precisely and formally, what the price system saves.Where the paper says it
Where the paper says it
Methodological Significance
The broader ambition of this paper is methodological, as it provides a framework in which relevant information is measurable. The formal language and proof-theoretic apparatus introduced here provide exactly such a framework, and the application is a demonstration for a general methodology. p. 4 Without a framework in which information is a measurable syntactic object and
reasoning is an explicit proof, the intuition cannot be made precise, cannot be applied systematically,
and cannot be generalized. The formal language and proof-theoretic apparatus introduced here provide
exactly such a framework, and the application is a demonstration for a general methodology. The answer
suggested by this paper is that such reasoning is itself costly. Institutions economize not only on
transaction costs, but also on the informational burden.Where the paper says it
Where the paper says it
Application to Exchange Economies
In the pure-exchange economy setting, the load-bearing coalitions are determined by the geometry of the core in the replica economy. For a k-fold replica Ek (k ≥ 2), the set Lk of load-bearing coalitions consists of individual coalitions, mixed pairs, one-asymmetric coalitions, and the grand coalition. p. 0 Removing information about even a single coalition from all
of its members suffices to break the equivalence for some game. Applying this result to the
k-fold replica economy, I identify and count the load-bearing coalitions whose feasibility
information must be distributed across the population. Applying this result to the
k-fold replica economy, I identify and count the load-bearing coalitions whose feasibility
information must be distributed across the population. The associated average per-agent
informational burden grows as Θ(4
k/k
3/2)—exponentially in the size of the economy.Where the paper says it
Where the paper says it
Conclusion on Information Cost
The total number of load-bearing coalitions grows as Lk ∼ 4k/√πk, while the number of agents grows only linearly. Hence the average informational burden per agent grows as Θ(4k/k 3/2) and diverges. p. 0 Applying this result to the
k-fold replica economy, I identify and count the load-bearing coalitions whose feasibility
information must be distributed across the population. The associated average per-agent
informational burden grows as Θ(4
k/k
3/2)—exponentially in the size of the economy. The collective question of this literature
is whether prices successfully incorporate dispersed information, how completely they do so, and at
what welfare cost. The present paper studies a complementary side of the same Hayekian idea: not
how information flows through prices, but how much information and reasoning the price system saves
relative to direct coalitional reasoning.Where the paper says it
Where the paper says it
References
Albagli, E., Hellwig, C. p. 4 A large literature following Hayek studies the informational role
of prices. Radner (1968, 1972, 1979) established the foundations of equilibrium under asymmetric
information; subsequent work, most influentially Grossman and Stiglitz (2025) and Hellwig (1980),
asked how well prices aggregate dispersed private information and with what welfare consequences. Measuring Information Burden Shuige Liu
consequences of dispersed private information (Chabakauri, Yuan, and Zachariadis, 2022; Galanis,
Ioannou, Kotronis, 2024; Albagli, Hellwig, and Tsyvinski, 2023). The collective question of this literature
is whether prices successfully incorporate dispersed information, how completely they do so, and at
what welfare cost. In this
respect, the argument is parallel in spirit to Coase’s (1937) theory of the firm. Coase asked why firms
exist if markets can coordinate economic activity, and his answer was that using markets is itself costly. An agent’s
information is then represented as a finite, consistent set of such meaning-bearing sentences. The point
is not merely to count sentences mechanically, but to make information decomposable into explicit
units of economic content.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Barelli, P. p. 5 The present paper studies a complementary side of the same Hayekian idea: not
how information flows through prices, but how much information and reasoning the price system saves
relative to direct coalitional reasoning. A related literature asks whether decentralized market mechanisms converge to competitive outcomes when traders have private information (Wilson, 1977; Milgrom, 1979, 1981; Palfrey, 1985; Vives,
1988; Reny and Perry, 1979; Barelli, Govindan, Wilson, 2024). The present paper studies a complementary side of the same Hayekian idea: not
how information flows through prices, but how much information and reasoning the price system saves
relative to direct coalitional reasoning. A related literature asks whether decentralized market mechanisms converge to competitive outcomes when traders have private information (Wilson, 1977; Milgrom, 1979, 1981; Palfrey, 1985; Vives,
1988; Reny and Perry, 1979; Barelli, Govindan, Wilson, 2024). The present paper studies a complementary side of the same Hayekian idea: not
how information flows through prices, but how much information and reasoning the price system saves
relative to direct coalitional reasoning. A related literature asks whether decentralized market mechanisms converge to competitive outcomes when traders have private information (Wilson, 1977; Milgrom, 1979, 1981; Palfrey, 1985; Vives,
1988; Reny and Perry, 1979; Barelli, Govindan, Wilson, 2024). Our paper poses the inverse question: taking the outcome concept—the
core—as given, we ask how much information agents must hold, and how that burden grows, for decentralized coalition-based reasoning to sustain stability. The strategic-foundations literature takes the
information structure as given and asks whether mechanisms achieve efficiency; we take the efficiency
concept as given and measure the informational cost of reaching it through explicit reasoning. Earlier proof-theoretic
work in game theory studies the structure of interpersonal reasoning itself. Here, the same constructive
perspective is brought to bear on a classical problem in economic theory: the informational conditions
under which decentralized reasoning recovers the core, and the informational burden hidden behind
the Debreu-Scarf convergence.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Battigalli, Pierpaolo and Giacomo Bonanno (1999), “Recent results on belief, knowledge and the epistemic foundations of game theory” Research in Economics, 53(2), 149–225. p. 0 JEL Codes: C71, D51, D83, D90
Early versions of this paper had been circulated under the title “Knowledge and Unanimous Acceptance of Core Payoffs:
An Epistemic Foundation for Cooperative Game Theory”. The author grateful to Pierpaolo Battigalli, Giacomo Bonanno,
Justus Preusser, and Niccolò Urbinati for their detailed comments.Where the paper says it
Bonanno, Giacomo (2008), “A syntactic approach to rationality in games with ordinal payoffs. p. 6 In particular, epistemic game theory has used formal languages to represent players’
beliefs, higher-order beliefs, and the informational structure of strategic situations (for surveys, see
Fagin, Halpern, Moses, and Verdi, 1995; Battigalli and Bonanno, 1999; Bonanno, 2015). In particular,
this paper is related in spirit to Bonanno’s (2015) syntactic treatment of common belief of rationality in
non-cooperative games. Keywords: Information burden, Price system, Cooperative game theory, Core, Debreu-Scarf
theorem, Formal language, Proof theory. JEL Codes: C71, D51, D83, D90
Early versions of this paper had been circulated under the title “Knowledge and Unanimous Acceptance of Core Payoffs:
An Epistemic Foundation for Cooperative Game Theory”. JEL Codes: C71, D51, D83, D90
Early versions of this paper had been circulated under the title “Knowledge and Unanimous Acceptance of Core Payoffs:
An Epistemic Foundation for Cooperative Game Theory”. The author grateful to Pierpaolo Battigalli, Giacomo Bonanno,
Justus Preusser, and Niccolò Urbinati for their detailed comments.Where the paper says it
Where the paper says it
Where the paper says it
Bonanno, Giacomo (2015), “Epistemic foundation of game theory. p. 0 JEL Codes: C71, D51, D83, D90
Early versions of this paper had been circulated under the title “Knowledge and Unanimous Acceptance of Core Payoffs:
An Epistemic Foundation for Cooperative Game Theory”. The author grateful to Pierpaolo Battigalli, Giacomo Bonanno,
Justus Preusser, and Niccolò Urbinati for their detailed comments. Keywords: Information burden, Price system, Cooperative game theory, Core, Debreu-Scarf
theorem, Formal language, Proof theory. JEL Codes: C71, D51, D83, D90
Early versions of this paper had been circulated under the title “Knowledge and Unanimous Acceptance of Core Payoffs:
An Epistemic Foundation for Cooperative Game Theory”. Formal languages and logical methods have long been useful in
economic theory. In particular, epistemic game theory has used formal languages to represent players’
beliefs, higher-order beliefs, and the informational structure of strategic situations (for surveys, see
Fagin, Halpern, Moses, and Verdi, 1995; Battigalli and Bonanno, 1999; Bonanno, 2015).Where the paper says it
Where the paper says it
Where the paper says it
Bondareva, Olga N. p. 7 6Not every TU game has a non-empty core. Bondareva (1963) and Shapley (1967) characterize the non-emptiness of the
core in terms of balancedness. Seen in this way, the comparison helps
isolate both a shared logical core and the additional structure needed for cooperative games. By contrast, proof-theoretic methods, introduced into game theoretical research by Kaneko and
Nagashima (1996, 1997), have been used much less frequently (see Kaneko (2002) for a survey).Where the paper says it
Where the paper says it
Boolos, George. (1984) “Don’t eliminate cut.” Journal of Philosophical Logic, 13: 373–378.
Chabakauri, G. p. 5 Measuring Information Burden Shuige Liu
consequences of dispersed private information (Chabakauri, Yuan, and Zachariadis, 2022; Galanis,
Ioannou, Kotronis, 2024; Albagli, Hellwig, and Tsyvinski, 2023). The collective question of this literature
is whether prices successfully incorporate dispersed information, how completely they do so, and at
what welfare cost. Measuring Information Burden Shuige Liu
consequences of dispersed private information (Chabakauri, Yuan, and Zachariadis, 2022; Galanis,
Ioannou, Kotronis, 2024; Albagli, Hellwig, and Tsyvinski, 2023). The collective question of this literature
is whether prices successfully incorporate dispersed information, how completely they do so, and at
what welfare cost. Radner (1968, 1972, 1979) established the foundations of equilibrium under asymmetric
information; subsequent work, most influentially Grossman and Stiglitz (2025) and Hellwig (1980),
asked how well prices aggregate dispersed private information and with what welfare consequences. This program continues in recent contributions that examine, among other things, multi-asset rationalexpectations models, the effects of ambiguity on information aggregation, and the real investment
5 Earlier proof-theoretic
work in game theory studies the structure of interpersonal reasoning itself. Here, the same constructive
perspective is brought to bear on a classical problem in economic theory: the informational conditions
under which decentralized reasoning recovers the core, and the informational burden hidden behind
the Debreu-Scarf convergence.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Chade, Hector (2006) “Matching with noise and the acceptance curse. p. 5 The coverage condition in Theorem 1
and the quantitative measure behind the Debreu–Scarf application follow from this shift. 2A closely related body of work studies incomplete information in stable matching, where similar tensions between
stability and efficiency arise; see Roth (1989); Chade (2006); Ehlers and Massó (2007); Hoppe, Moldovanu, and Sela (2009); Liu,
Mailath, Postlewaite, and Samuelson (2014); Pomatto (2022), and Chade, Lewis, and Smith (2014). In this
respect, the argument is parallel in spirit to Coase’s (1937) theory of the firm. Coase asked why firms
exist if markets can coordinate economic activity, and his answer was that using markets is itself costly.Where the paper says it
Where the paper says it
Chade, Hector, Gregory Lewis, and Lones Smith (2014) “Student portfolios and the college admissions problem. p. 5 The coverage condition in Theorem 1
and the quantitative measure behind the Debreu–Scarf application follow from this shift. 2A closely related body of work studies incomplete information in stable matching, where similar tensions between
stability and efficiency arise; see Roth (1989); Chade (2006); Ehlers and Massó (2007); Hoppe, Moldovanu, and Sela (2009); Liu,
Mailath, Postlewaite, and Samuelson (2014); Pomatto (2022), and Chade, Lewis, and Smith (2014).Where the paper says it
Coase, R. H. (1937). “The Nature of the Firm. p. 4 Institutions do not merely coordinate actions, allocate resources, or enforce rules; they also absorb
informational and cognitive burdens that agents would otherwise have to bear themselves. In this
respect, the argument is parallel in spirit to Coase’s (1937) theory of the firm.Where the paper says it
Debreu, Gerard and Scarf, Herbert (1963), “A limit theorem on the core of an economy. p. 0 I develop a formal framework for measuring the informational
burden that agents would have to bear if they had to reason individually over feasible
coalitional alternatives. Taking the convergence theorem of Debreu and Scarf (1963) as the
benchmark, I ask how much coalition-feasibility information must be held, and how it must
be distributed, for individual acceptance decisions to recover the core. An agent’s
information is then represented as a finite, consistent set of such meaning-bearing sentences. The point
is not merely to count sentences mechanically, but to make information decomposable into explicit
units of economic content.Where the paper says it
Where the paper says it
de Clippel, Geoffroy (2007), “The type-agent core for exchange economies with asymmetric information. p. 5 Incomplete information and the core. The literature on the core under incomplete information,
beginning with Wilson (1978) and developed by Vohra (1999); Dutta and Vohra (2005); Serrano and
Vohra (2007); de Clippel (2007), and Myerson (2007), among others, transforms the cooperative setting
into a decentralized problem of individual strategic reasoning and endows each player with a doxastic
or epistemic structure representing her beliefs. In this
respect, the argument is parallel in spirit to Coase’s (1937) theory of the firm. Coase asked why firms
exist if markets can coordinate economic activity, and his answer was that using markets is itself costly.Where the paper says it
Where the paper says it
Dekel, Eddie and Siniscalchi, Marciano (2015), “Epistemic game theory. p. 6 Formal languages and logical methods have long been useful in
economic theory. In particular, epistemic game theory has used formal languages to represent players’
beliefs, higher-order beliefs, and the informational structure of strategic situations (for surveys, see
Fagin, Halpern, Moses, and Verdi, 1995; Battigalli and Bonanno, 1999; Bonanno, 2015). Formal languages and logical methods have long been useful in
economic theory. In particular, epistemic game theory has used formal languages to represent players’
beliefs, higher-order beliefs, and the informational structure of strategic situations (for surveys, see
Fagin, Halpern, Moses, and Verdi, 1995; Battigalli and Bonanno, 1999; Bonanno, 2015). Formal languages and logical methods have long been useful in
economic theory. In particular, epistemic game theory has used formal languages to represent players’
beliefs, higher-order beliefs, and the informational structure of strategic situations (for surveys, see
Fagin, Halpern, Moses, and Verdi, 1995; Battigalli and Bonanno, 1999; Bonanno, 2015). Wáng, and participants of
13th LOFT Conference, IV Italian Junior Theory Workshop, seminars in Bocconi University, University of Cologne, and Otago
University for valuable discussions and encouragements. The paper was partially supported by Grant-in-Aids for Young
Scientists (B) of JSPS No.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Dutta, Bhaskar and Vohra, Rajiv (2005), “Incomplete information, credibility and the core. p. 5 Incomplete information and the core. The literature on the core under incomplete information,
beginning with Wilson (1978) and developed by Vohra (1999); Dutta and Vohra (2005); Serrano and
Vohra (2007); de Clippel (2007), and Myerson (2007), among others, transforms the cooperative setting
into a decentralized problem of individual strategic reasoning and endows each player with a doxastic
or epistemic structure representing her beliefs. The next step is to specify what it means for Ci(x) to be derivable from an agent’s information. In
much of the existing literature, this question is left implicit: decision-makers are assumed to have
whatever mathematical or probabilistic reasoning capacity needed to draw the relevant inferences,
so that the connection between information and conclusion is absorbed into the model’s background
assumptions rather than examined in its own right.Where the paper says it
Where the paper says it
Ebbinghaus, Heinz-Dieter, Flum, Jörg, and Thomas, Wolfgang (2021), Mathematical Logic. p. 3 The next step is to specify what it means for Ci(x) to be derivable from an agent’s information. In
much of the existing literature, this question is left implicit: decision-makers are assumed to have
whatever mathematical or probabilistic reasoning capacity needed to draw the relevant inferences,
so that the connection between information and conclusion is absorbed into the model’s background
assumptions rather than examined in its own right.Where the paper says it
Ehlers, Lars and Jordi Massó (2007) “Incomplete information and singleton cores in matching markets. p. 5 The coverage condition in Theorem 1
and the quantitative measure behind the Debreu–Scarf application follow from this shift. 2A closely related body of work studies incomplete information in stable matching, where similar tensions between
stability and efficiency arise; see Roth (1989); Chade (2006); Ehlers and Massó (2007); Hoppe, Moldovanu, and Sela (2009); Liu,
Mailath, Postlewaite, and Samuelson (2014); Pomatto (2022), and Chade, Lewis, and Smith (2014).Where the paper says it
Fagin, Ronald, Halpern, Joseph Y. p. 6 Formal languages and logical methods have long been useful in
economic theory. In particular, epistemic game theory has used formal languages to represent players’
beliefs, higher-order beliefs, and the informational structure of strategic situations (for surveys, see
Fagin, Halpern, Moses, and Verdi, 1995; Battigalli and Bonanno, 1999; Bonanno, 2015). Formal languages and logical methods have long been useful in
economic theory. In particular, epistemic game theory has used formal languages to represent players’
beliefs, higher-order beliefs, and the informational structure of strategic situations (for surveys, see
Fagin, Halpern, Moses, and Verdi, 1995; Battigalli and Bonanno, 1999; Bonanno, 2015). The literature on the core under incomplete information,
beginning with Wilson (1978) and developed by Vohra (1999); Dutta and Vohra (2005); Serrano and
Vohra (2007); de Clippel (2007), and Myerson (2007), among others, transforms the cooperative setting
into a decentralized problem of individual strategic reasoning and endows each player with a doxastic
or epistemic structure representing her beliefs. 2 A central finding of this literature is that stability
under incomplete information generally diverges from efficiency under complete information: solution
concepts based on agents’ limited knowledge may yield outcomes that are stable given what agents
know yet fail to coincide with the efficient core.Where the paper says it
Where the paper says it
Where the paper says it
Galanis, S. p. 5 Measuring Information Burden Shuige Liu
consequences of dispersed private information (Chabakauri, Yuan, and Zachariadis, 2022; Galanis,
Ioannou, Kotronis, 2024; Albagli, Hellwig, and Tsyvinski, 2023). The collective question of this literature
is whether prices successfully incorporate dispersed information, how completely they do so, and at
what welfare cost. Measuring Information Burden Shuige Liu
consequences of dispersed private information (Chabakauri, Yuan, and Zachariadis, 2022; Galanis,
Ioannou, Kotronis, 2024; Albagli, Hellwig, and Tsyvinski, 2023). The collective question of this literature
is whether prices successfully incorporate dispersed information, how completely they do so, and at
what welfare cost. Measuring Information Burden Shuige Liu
consequences of dispersed private information (Chabakauri, Yuan, and Zachariadis, 2022; Galanis,
Ioannou, Kotronis, 2024; Albagli, Hellwig, and Tsyvinski, 2023). The collective question of this literature
is whether prices successfully incorporate dispersed information, how completely they do so, and at
what welfare cost. Radner (1968, 1972, 1979) established the foundations of equilibrium under asymmetric
information; subsequent work, most influentially Grossman and Stiglitz (2025) and Hellwig (1980),
asked how well prices aggregate dispersed private information and with what welfare consequences. This program continues in recent contributions that examine, among other things, multi-asset rationalexpectations models, the effects of ambiguity on information aggregation, and the real investment
5 Earlier proof-theoretic
work in game theory studies the structure of interpersonal reasoning itself. Here, the same constructive
perspective is brought to bear on a classical problem in economic theory: the informational conditions
under which decentralized reasoning recovers the core, and the informational burden hidden behind
the Debreu-Scarf convergence.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Gentzen, Gerhard Karl Erich (1935a), “Untersuchungen über das logische Schließen. p. 3 If our goal is to identify which pieces of information do which inferential work—which sentences are
load-bearing for a given conclusion and which are redundant—then derivability itself must be made a
precise, transparent object. We formalize it as the existence of a proof in Gentzen’s (1935a; 1935b) sequent calculus: a finite,
tree-structured process that begins from the atomic expression and proceeds via explicit inference rules
which captures an agent’s reasoning structure/ability, with every step on record.Where the paper says it
Gentzen, Gerhard Karl Erich (1935b), “Untersuchungen über das logische Schließen. p. 3 If our goal is to identify which pieces of information do which inferential work—which sentences are
load-bearing for a given conclusion and which are redundant—then derivability itself must be made a
precise, transparent object. We formalize it as the existence of a proof in Gentzen’s (1935a; 1935b) sequent calculus: a finite,
tree-structured process that begins from the atomic expression and proceeds via explicit inference rules
which captures an agent’s reasoning structure/ability, with every step on record.Where the paper says it
Grossman, S. p. 4 A large literature following Hayek studies the informational role
of prices. Radner (1968, 1972, 1979) established the foundations of equilibrium under asymmetric
information; subsequent work, most influentially Grossman and Stiglitz (2025) and Hellwig (1980),
asked how well prices aggregate dispersed private information and with what welfare consequences. A large literature following Hayek studies the informational role
of prices. Radner (1968, 1972, 1979) established the foundations of equilibrium under asymmetric
information; subsequent work, most influentially Grossman and Stiglitz (2025) and Hellwig (1980),
asked how well prices aggregate dispersed private information and with what welfare consequences. The framework in hand, we can now characterize precisely how information must be distributed for
unanimous acceptance to coincide with the core—and the answer turns on a simple but consequential
asymmetry. Core payoff vectors are informationally robust: since no profitable coalitional objection
exists, they are accepted under any information structure. An agent’s
information is then represented as a finite, consistent set of such meaning-bearing sentences. The point
is not merely to count sentences mechanically, but to make information decomposable into explicit
units of economic content.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Hayek, Friedrich A. p. 2 It is clear that if the concept is
to have any empirical significance it cannot presuppose that everybody knows
everything. Hayek, “Economics and Knowledge” (1937)
1 Introduction
One of the classical insights of economics is that institutions do not merely allocate resources; they
also economize on what agents need to know. It is clear that if the concept is
to have any empirical significance it cannot presuppose that everybody knows
everything. Hayek, “Economics and Knowledge” (1937)
1 Introduction
One of the classical insights of economics is that institutions do not merely allocate resources; they
also economize on what agents need to know. It is clear that if the concept is
to have any empirical significance it cannot presuppose that everybody knows
everything. Hayek, “Economics and Knowledge” (1937)
1 Introduction
One of the classical insights of economics is that institutions do not merely allocate resources; they
also economize on what agents need to know. It is clear that if the concept is
to have any empirical significance it cannot presuppose that everybody knows
everything. Hayek, “Economics and Knowledge” (1937)
1 Introduction
One of the classical insights of economics is that institutions do not merely allocate resources; they
also economize on what agents need to know. It is clear that if the concept is
to have any empirical significance it cannot presuppose that everybody knows
everything. Hayek, “Economics and Knowledge” (1937)
1 Introduction
One of the classical insights of economics is that institutions do not merely allocate resources; they
also economize on what agents need to know. The point
is not merely to count sentences mechanically, but to make information decomposable into explicit
units of economic content. Once this is done, the framework can identify which units of information is
1This question has a classic origin in the history of economics. In the socialist calculation debates of the 1930s and 1940s,
Hayek emphasized two complementary informational roles of the price system. First, echoing earlier arguments by von
von Mises (1920), Hayek (1935a,b, 1940, 1942) argued that competitive prices integrate dispersed, local, and tacit knowledge
that no central planning committee could construct from scratch. First, echoing earlier arguments by von
von Mises (1920), Hayek (1935a,b, 1940, 1942) argued that competitive prices integrate dispersed, local, and tacit knowledge
that no central planning committee could construct from scratch. Second, following a tradition running from Adam Smith
through Walras, Hayek (1935a, 1945)argued that, once such a system exists, each participant can act on a dramatically reduced
informational basis: her own preferences, her endowments, and the prevailing price vector. Measuring Information Burden:
From Coalition-Based Reasoning to the Price System
Shuige Liu
Department of Decision Sciences, Bocconi University
Milan, Italy
shuige. liu@unibocconi.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Hayek, Friedrich A. (1937). “Economics and knowledge. p. 2 It is clear that if the concept is
to have any empirical significance it cannot presuppose that everybody knows
everything. Hayek, “Economics and Knowledge” (1937)
1 Introduction
One of the classical insights of economics is that institutions do not merely allocate resources; they
also economize on what agents need to know.Where the paper says it
Hayek, Friedrich A. (1940). “Socialist calculation: The competitive ‘solution’. p. 2 In the socialist calculation debates of the 1930s and 1940s,
Hayek emphasized two complementary informational roles of the price system. First, echoing earlier arguments by von
von Mises (1920), Hayek (1935a,b, 1940, 1942) argued that competitive prices integrate dispersed, local, and tacit knowledge
that no central planning committee could construct from scratch.Where the paper says it
Hayek, Friedrich A. (1943). “The facts of the social sciences. p. 0 Measuring Information Burden:
From Coalition-Based Reasoning to the Price System
Shuige Liu
Department of Decision Sciences, Bocconi University
Milan, Italy
shuige. liu@unibocconi.Where the paper says it
Hayek, Friedrich A. (1945) “The use of knowledge in society. p. 2 First, echoing earlier arguments by von
von Mises (1920), Hayek (1935a,b, 1940, 1942) argued that competitive prices integrate dispersed, local, and tacit knowledge
that no central planning committee could construct from scratch. Second, following a tradition running from Adam Smith
through Walras, Hayek (1935a, 1945)argued that, once such a system exists, each participant can act on a dramatically reduced
informational basis: her own preferences, her endowments, and the prevailing price vector. An agent’s
information is then represented as a finite, consistent set of such meaning-bearing sentences. The point
is not merely to count sentences mechanically, but to make information decomposable into explicit
units of economic content.Where the paper says it
Where the paper says it
Heifetz, Aviad, Martin Meier and Burkhard C. p. 0 Bach, Lorenzo Bastianello,
Chih Chang, Pietro Dall’Ara, Meng-Jhang Fong, Satoshi Fukuda, Yukihiko Funaki, Peter Gibbard, Pierfrancesco Guarino,
Mamoru Kaneko, Nenad Kos, Fabio Maccheroni, Suraj Malladi, Massimo Marinacci, Zsombor Z. Méder, Antonio Penta,
Andrés Perea, Ronald Peeters, Paolo Pin, Rasto Ratislav, Marten Ritterath, Pietro Salmaso, Enrico Mattia Salonia, Burkhard
Schipper, Christoph Schottmüller, Fernando Vega-Redondo, Murat Üngor, Yanjing Wang, Yì N. Bach, Lorenzo Bastianello,
Chih Chang, Pietro Dall’Ara, Meng-Jhang Fong, Satoshi Fukuda, Yukihiko Funaki, Peter Gibbard, Pierfrancesco Guarino,
Mamoru Kaneko, Nenad Kos, Fabio Maccheroni, Suraj Malladi, Massimo Marinacci, Zsombor Z. Méder, Antonio Penta,
Andrés Perea, Ronald Peeters, Paolo Pin, Rasto Ratislav, Marten Ritterath, Pietro Salmaso, Enrico Mattia Salonia, Burkhard
Schipper, Christoph Schottmüller, Fernando Vega-Redondo, Murat Üngor, Yanjing Wang, Yì N. An agent’s
information is then represented as a finite, consistent set of such meaning-bearing sentences. The point
is not merely to count sentences mechanically, but to make information decomposable into explicit
units of economic content.Where the paper says it
Where the paper says it
Where the paper says it
Hellwig, M. p. 4 A large literature following Hayek studies the informational role
of prices. Radner (1968, 1972, 1979) established the foundations of equilibrium under asymmetric
information; subsequent work, most influentially Grossman and Stiglitz (2025) and Hellwig (1980),
asked how well prices aggregate dispersed private information and with what welfare consequences. Radner (1968, 1972, 1979) established the foundations of equilibrium under asymmetric
information; subsequent work, most influentially Grossman and Stiglitz (2025) and Hellwig (1980),
asked how well prices aggregate dispersed private information and with what welfare consequences. This program continues in recent contributions that examine, among other things, multi-asset rationalexpectations models, the effects of ambiguity on information aggregation, and the real investment
5 In this
respect, the argument is parallel in spirit to Coase’s (1937) theory of the firm. Coase asked why firms
exist if markets can coordinate economic activity, and his answer was that using markets is itself costly.Where the paper says it
Where the paper says it
Where the paper says it
Hoppe, Heidrun C. p. 5 The coverage condition in Theorem 1
and the quantitative measure behind the Debreu–Scarf application follow from this shift. 2A closely related body of work studies incomplete information in stable matching, where similar tensions between
stability and efficiency arise; see Roth (1989); Chade (2006); Ehlers and Massó (2007); Hoppe, Moldovanu, and Sela (2009); Liu,
Mailath, Postlewaite, and Samuelson (2014); Pomatto (2022), and Chade, Lewis, and Smith (2014). The coverage condition in Theorem 1
and the quantitative measure behind the Debreu–Scarf application follow from this shift. 2A closely related body of work studies incomplete information in stable matching, where similar tensions between
stability and efficiency arise; see Roth (1989); Chade (2006); Ehlers and Massó (2007); Hoppe, Moldovanu, and Sela (2009); Liu,
Mailath, Postlewaite, and Samuelson (2014); Pomatto (2022), and Chade, Lewis, and Smith (2014).Where the paper says it
Where the paper says it
Hurwicz, L. p. 2 A natural starting point is syntactic measurement: to count symbols, expressions, or sentences. Yet Hurwicz (1969) already raised a pointed objection to this approach: without knowing what a symbol
means, counting symbols is meaningless. Our paper poses the inverse question: taking the outcome concept—the
core—as given, we ask how much information agents must hold, and how that burden grows, for decentralized coalition-based reasoning to sustain stability. The strategic-foundations literature takes the
information structure as given and asks whether mechanisms achieve efficiency; we take the efficiency
concept as given and measure the informational cost of reaching it through explicit reasoning. An agent’s
information is then represented as a finite, consistent set of such meaning-bearing sentences. The point
is not merely to count sentences mechanically, but to make information decomposable into explicit
units of economic content.Where the paper says it
Where the paper says it
Where the paper says it
Indrzejczak, Andrzej (2021) Sequents and Trees. Springer.
Kaneko, Mamoru (2002), “Epistemic logics and their game theoretic applications: Introduction. p. 6 Seen in this way, the comparison helps
isolate both a shared logical core and the additional structure needed for cooperative games. By contrast, proof-theoretic methods, introduced into game theoretical research by Kaneko and
Nagashima (1996, 1997), have been used much less frequently (see Kaneko (2002) for a survey). In this
respect, the argument is parallel in spirit to Coase’s (1937) theory of the firm. Coase asked why firms
exist if markets can coordinate economic activity, and his answer was that using markets is itself costly.Where the paper says it
Where the paper says it
Kaneko, Mamoru and Nagashima, Takashi (1996) “Game logic and its applications I” Studia Logica, 57 (2-3), 325–354. p. 6 Seen in this way, the comparison helps
isolate both a shared logical core and the additional structure needed for cooperative games. By contrast, proof-theoretic methods, introduced into game theoretical research by Kaneko and
Nagashima (1996, 1997), have been used much less frequently (see Kaneko (2002) for a survey).Where the paper says it
Kaneko, Mamoru and Nagashima, Takashi (1997) “Game logic and its applications II” Studia Logica, 58 (2), 273–303. p. 6 Seen in this way, the comparison helps
isolate both a shared logical core and the additional structure needed for cooperative games. By contrast, proof-theoretic methods, introduced into game theoretical research by Kaneko and
Nagashima (1996, 1997), have been used much less frequently (see Kaneko (2002) for a survey).Where the paper says it
Kaneko, Mamoru and Suzuki, Nobu-Yuki (2002), “Bounded interpersonal inferences and decision making. p. 6 By contrast, proof-theoretic methods, introduced into game theoretical research by Kaneko and
Nagashima (1996, 1997), have been used much less frequently (see Kaneko (2002) for a survey). 3
Existing work shows that they can illuminate the logical structure of strategic reasoning and bounded
interpersonal inference. In this
respect, the argument is parallel in spirit to Coase’s (1937) theory of the firm. Coase asked why firms
exist if markets can coordinate economic activity, and his answer was that using markets is itself costly.Where the paper says it
Where the paper says it
Kannai, Yakar (1992), “The core and balancedness. p. 6 2 TU Games and the Core: An Overview
This subsection provides an overview of cooperative games with transferable utility, as well as the
concept of the core. For a more detailed treatment, see Kannai (1992) and sections 2 and 3 of Peleg and
Sudhoelter (2007).Where the paper says it
Kaplow, L. (2011). “On the optimal burden of proof. p. 0 This
is, precisely and formally, what the price system saves. Keywords: Information burden, Price system, Cooperative game theory, Core, Debreu-Scarf
theorem, Formal language, Proof theory. Removing information about even a single coalition from all
of its members suffices to break the equivalence for some game. Applying this result to the
k-fold replica economy, I identify and count the load-bearing coalitions whose feasibility
information must be distributed across the population.Where the paper says it
Where the paper says it
Li, S. (2017). “Obviously strategy-proof mechanisms. p. 4 Coase asked why firms
exist if markets can coordinate economic activity, and his answer was that using markets is itself costly. The question here is analogous: why do institutional mechanisms matter if agents could, in principle,
each acting on her own initiative, determine what is collectively acceptable without them?Where the paper says it
Liu, Qingmin, Mailath, George J. p. 5 The coverage condition in Theorem 1
and the quantitative measure behind the Debreu–Scarf application follow from this shift. 2A closely related body of work studies incomplete information in stable matching, where similar tensions between
stability and efficiency arise; see Roth (1989); Chade (2006); Ehlers and Massó (2007); Hoppe, Moldovanu, and Sela (2009); Liu,
Mailath, Postlewaite, and Samuelson (2014); Pomatto (2022), and Chade, Lewis, and Smith (2014). The coverage condition in Theorem 1
and the quantitative measure behind the Debreu–Scarf application follow from this shift. 2A closely related body of work studies incomplete information in stable matching, where similar tensions between
stability and efficiency arise; see Roth (1989); Chade (2006); Ehlers and Massó (2007); Hoppe, Moldovanu, and Sela (2009); Liu,
Mailath, Postlewaite, and Samuelson (2014); Pomatto (2022), and Chade, Lewis, and Smith (2014).Where the paper says it
Where the paper says it
Mancosu, Paolo, Galvan, Sergio and Zach, Richard (2021) An Introduction to Proof Theory: Normalization, Cut-Elimination, and Consistency Proofs. p. 0 Méder, Antonio Penta,
Andrés Perea, Ronald Peeters, Paolo Pin, Rasto Ratislav, Marten Ritterath, Pietro Salmaso, Enrico Mattia Salonia, Burkhard
Schipper, Christoph Schottmüller, Fernando Vega-Redondo, Murat Üngor, Yanjing Wang, Yì N. Wáng, and participants of
13th LOFT Conference, IV Italian Junior Theory Workshop, seminars in Bocconi University, University of Cologne, and Otago
University for valuable discussions and encouragements. Measuring Information Burden:
From Coalition-Based Reasoning to the Price System
Shuige Liu
Department of Decision Sciences, Bocconi University
Milan, Italy
shuige. liu@unibocconi.Where the paper says it
Where the paper says it
Mendelson, Eliott (2015) Introduction to Mathematical Logic, 6th edition CRC Press.
Milgrom, P. p. 5 The present paper studies a complementary side of the same Hayekian idea: not
how information flows through prices, but how much information and reasoning the price system saves
relative to direct coalitional reasoning. A related literature asks whether decentralized market mechanisms converge to competitive outcomes when traders have private information (Wilson, 1977; Milgrom, 1979, 1981; Palfrey, 1985; Vives,
1988; Reny and Perry, 1979; Barelli, Govindan, Wilson, 2024). The present paper studies a complementary side of the same Hayekian idea: not
how information flows through prices, but how much information and reasoning the price system saves
relative to direct coalitional reasoning. A related literature asks whether decentralized market mechanisms converge to competitive outcomes when traders have private information (Wilson, 1977; Milgrom, 1979, 1981; Palfrey, 1985; Vives,
1988; Reny and Perry, 1979; Barelli, Govindan, Wilson, 2024). I develop a formal framework for measuring the informational
burden that agents would have to bear if they had to reason individually over feasible
coalitional alternatives. Taking the convergence theorem of Debreu and Scarf (1963) as the
benchmark, I ask how much coalition-feasibility information must be held, and how it must
be distributed, for individual acceptance decisions to recover the core.Where the paper says it
Where the paper says it
Where the paper says it
Milgrom, P. R. (1981). “Rational expectations, information acquisition, and competitive bidding. p. 5 The present paper studies a complementary side of the same Hayekian idea: not
how information flows through prices, but how much information and reasoning the price system saves
relative to direct coalitional reasoning. A related literature asks whether decentralized market mechanisms converge to competitive outcomes when traders have private information (Wilson, 1977; Milgrom, 1979, 1981; Palfrey, 1985; Vives,
1988; Reny and Perry, 1979; Barelli, Govindan, Wilson, 2024).Where the paper says it
Myerson, Roger B. p. 5 Incomplete information and the core. The literature on the core under incomplete information,
beginning with Wilson (1978) and developed by Vohra (1999); Dutta and Vohra (2005); Serrano and
Vohra (2007); de Clippel (2007), and Myerson (2007), among others, transforms the cooperative setting
into a decentralized problem of individual strategic reasoning and endows each player with a doxastic
or epistemic structure representing her beliefs. 2 TU Games and the Core: An Overview
This subsection provides an overview of cooperative games with transferable utility, as well as the
concept of the core. For a more detailed treatment, see Kannai (1992) and sections 2 and 3 of Peleg and
Sudhoelter (2007). In this
respect, the argument is parallel in spirit to Coase’s (1937) theory of the firm. Coase asked why firms
exist if markets can coordinate economic activity, and his answer was that using markets is itself costly.Where the paper says it
Where the paper says it
Where the paper says it
Negri, Sara and Jan Von Plato. p. 2 In the socialist calculation debates of the 1930s and 1940s,
Hayek emphasized two complementary informational roles of the price system. First, echoing earlier arguments by von
von Mises (1920), Hayek (1935a,b, 1940, 1942) argued that competitive prices integrate dispersed, local, and tacit knowledge
that no central planning committee could construct from scratch. This
is, precisely and formally, what the price system saves. Keywords: Information burden, Price system, Cooperative game theory, Core, Debreu-Scarf
theorem, Formal language, Proof theory. Measuring Information Burden:
From Coalition-Based Reasoning to the Price System
Shuige Liu
Department of Decision Sciences, Bocconi University
Milan, Italy
shuige. liu@unibocconi.Where the paper says it
Where the paper says it
Where the paper says it
Palfrey, T. p. 5 The present paper studies a complementary side of the same Hayekian idea: not
how information flows through prices, but how much information and reasoning the price system saves
relative to direct coalitional reasoning. A related literature asks whether decentralized market mechanisms converge to competitive outcomes when traders have private information (Wilson, 1977; Milgrom, 1979, 1981; Palfrey, 1985; Vives,
1988; Reny and Perry, 1979; Barelli, Govindan, Wilson, 2024). Radner (1968, 1972, 1979) established the foundations of equilibrium under asymmetric
information; subsequent work, most influentially Grossman and Stiglitz (2025) and Hellwig (1980),
asked how well prices aggregate dispersed private information and with what welfare consequences. This program continues in recent contributions that examine, among other things, multi-asset rationalexpectations models, the effects of ambiguity on information aggregation, and the real investment
5 Earlier proof-theoretic
work in game theory studies the structure of interpersonal reasoning itself. Here, the same constructive
perspective is brought to bear on a classical problem in economic theory: the informational conditions
under which decentralized reasoning recovers the core, and the informational burden hidden behind
the Debreu-Scarf convergence.Where the paper says it
Where the paper says it
Where the paper says it
Peleg, Bezalel and Sudhoelter, Peter (2007), Introduction to the Theory of Cooperative Games. p. 6 2 TU Games and the Core: An Overview
This subsection provides an overview of cooperative games with transferable utility, as well as the
concept of the core. For a more detailed treatment, see Kannai (1992) and sections 2 and 3 of Peleg and
Sudhoelter (2007). Institutions economize not only on
transaction costs, but also on the informational burden. This suggests a new dimension along which
institutions can be compared and evaluated: not only by what they achieve, but by what they spare
agents from having to know.Where the paper says it
Where the paper says it
Pomatto, Luciano (2022). p. 5 The coverage condition in Theorem 1
and the quantitative measure behind the Debreu–Scarf application follow from this shift. 2A closely related body of work studies incomplete information in stable matching, where similar tensions between
stability and efficiency arise; see Roth (1989); Chade (2006); Ehlers and Massó (2007); Hoppe, Moldovanu, and Sela (2009); Liu,
Mailath, Postlewaite, and Samuelson (2014); Pomatto (2022), and Chade, Lewis, and Smith (2014). The coverage condition in Theorem 1
and the quantitative measure behind the Debreu–Scarf application follow from this shift. 2A closely related body of work studies incomplete information in stable matching, where similar tensions between
stability and efficiency arise; see Roth (1989); Chade (2006); Ehlers and Massó (2007); Hoppe, Moldovanu, and Sela (2009); Liu,
Mailath, Postlewaite, and Samuelson (2014); Pomatto (2022), and Chade, Lewis, and Smith (2014). Seen in this way, the comparison helps
isolate both a shared logical core and the additional structure needed for cooperative games. By contrast, proof-theoretic methods, introduced into game theoretical research by Kaneko and
Nagashima (1996, 1997), have been used much less frequently (see Kaneko (2002) for a survey).Where the paper says it
Where the paper says it
Where the paper says it
Radner, R. (1968). “Competitive Equilibrium Under Uncertainty. p. 2 This paper therefore begins with the reasoning that would be required
without the price system. Among many possible formulations, the replica process in the theorem
of Debreu and Scarf (1963) is a natural benchmark: it provides an implicit process in which agents,
acting on their own initiative, progressively reject allocations for which they can individually identify
a feasible improvement, until competitive equilibrium—both the allocation and the price system—
emerges as the only outcome that survives.Where the paper says it
Radner, R. (1972). “Existence of Equilibrium of Plans, Prices, and Price Expectations in a Sequence of Markets. p. 4 A large literature following Hayek studies the informational role
of prices. Radner (1968, 1972, 1979) established the foundations of equilibrium under asymmetric
information; subsequent work, most influentially Grossman and Stiglitz (2025) and Hellwig (1980),
asked how well prices aggregate dispersed private information and with what welfare consequences.Where the paper says it
Radner, R. (1979). “Rational Expectations Equilibrium: Generic Existence and the Information Revealed by Prices. p. 4 A large literature following Hayek studies the informational role
of prices. Radner (1968, 1972, 1979) established the foundations of equilibrium under asymmetric
information; subsequent work, most influentially Grossman and Stiglitz (2025) and Hellwig (1980),
asked how well prices aggregate dispersed private information and with what welfare consequences.Where the paper says it
Reny, P. J., Perry, M. p. 5 The present paper studies a complementary side of the same Hayekian idea: not
how information flows through prices, but how much information and reasoning the price system saves
relative to direct coalitional reasoning. A related literature asks whether decentralized market mechanisms converge to competitive outcomes when traders have private information (Wilson, 1977; Milgrom, 1979, 1981; Palfrey, 1985; Vives,
1988; Reny and Perry, 1979; Barelli, Govindan, Wilson, 2024). Keywords: Information burden, Price system, Cooperative game theory, Core, Debreu-Scarf
theorem, Formal language, Proof theory. JEL Codes: C71, D51, D83, D90
Early versions of this paper had been circulated under the title “Knowledge and Unanimous Acceptance of Core Payoffs:
An Epistemic Foundation for Cooperative Game Theory”.Where the paper says it
Where the paper says it
Roth, Alvin E. p. 5 The coverage condition in Theorem 1
and the quantitative measure behind the Debreu–Scarf application follow from this shift. 2A closely related body of work studies incomplete information in stable matching, where similar tensions between
stability and efficiency arise; see Roth (1989); Chade (2006); Ehlers and Massó (2007); Hoppe, Moldovanu, and Sela (2009); Liu,
Mailath, Postlewaite, and Samuelson (2014); Pomatto (2022), and Chade, Lewis, and Smith (2014). The coverage condition in Theorem 1
and the quantitative measure behind the Debreu–Scarf application follow from this shift. 2A closely related body of work studies incomplete information in stable matching, where similar tensions between
stability and efficiency arise; see Roth (1989); Chade (2006); Ehlers and Massó (2007); Hoppe, Moldovanu, and Sela (2009); Liu,
Mailath, Postlewaite, and Samuelson (2014); Pomatto (2022), and Chade, Lewis, and Smith (2014). An agent’s
information is then represented as a finite, consistent set of such meaning-bearing sentences. The point
is not merely to count sentences mechanically, but to make information decomposable into explicit
units of economic content.Where the paper says it
Where the paper says it
Where the paper says it
Serrano, Roberto and Vohra, Rajiv (2007), “Information transmission in coalitional voting games. p. 5 Incomplete information and the core. The literature on the core under incomplete information,
beginning with Wilson (1978) and developed by Vohra (1999); Dutta and Vohra (2005); Serrano and
Vohra (2007); de Clippel (2007), and Myerson (2007), among others, transforms the cooperative setting
into a decentralized problem of individual strategic reasoning and endows each player with a doxastic
or epistemic structure representing her beliefs. In this
respect, the argument is parallel in spirit to Coase’s (1937) theory of the firm. Coase asked why firms
exist if markets can coordinate economic activity, and his answer was that using markets is itself costly.Where the paper says it
Where the paper says it
Shapley, Lloyd S. p. 4 But the reduction obtained
in Theorem 1 remains available: what matters is which coalitions’ feasibility sets must be known
somewhere among their members. Using results in Shapley (1975), Section 5 identifies the minimal
collection of such load-bearing coalitions in the k-fold replica economy, counts them exactly, and shows
that the associated average per-agent (coalition-level) informational burden grows as Θ(4
k/k
3/2)——
exponentially in the size of the economy. But the reduction obtained
in Theorem 1 remains available: what matters is which coalitions’ feasibility sets must be known
somewhere among their members. Using results in Shapley (1975), Section 5 identifies the minimal
collection of such load-bearing coalitions in the k-fold replica economy, counts them exactly, and shows
that the associated average per-agent (coalition-level) informational burden grows as Θ(4
k/k
3/2)——
exponentially in the size of the economy. 6Not every TU game has a non-empty core. Bondareva (1963) and Shapley (1967) characterize the non-emptiness of the
core in terms of balancedness. The paper was partially supported by Grant-in-Aids for Young
Scientists (B) of JSPS No. 17K13707 and Grant for Special Research Project No.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Shapley, Lloyd S. (1975), “An example of a slow-converging core. p. 4 But the reduction obtained
in Theorem 1 remains available: what matters is which coalitions’ feasibility sets must be known
somewhere among their members. Using results in Shapley (1975), Section 5 identifies the minimal
collection of such load-bearing coalitions in the k-fold replica economy, counts them exactly, and shows
that the associated average per-agent (coalition-level) informational burden grows as Θ(4
k/k
3/2)——
exponentially in the size of the economy. An agent’s
information is then represented as a finite, consistent set of such meaning-bearing sentences. The point
is not merely to count sentences mechanically, but to make information decomposable into explicit
units of economic content.Where the paper says it
Where the paper says it
Vohra, Rajiv (1999), “Incomplete information, incentive compatibility, and the core. p. 5 Incomplete information and the core. The literature on the core under incomplete information,
beginning with Wilson (1978) and developed by Vohra (1999); Dutta and Vohra (2005); Serrano and
Vohra (2007); de Clippel (2007), and Myerson (2007), among others, transforms the cooperative setting
into a decentralized problem of individual strategic reasoning and endows each player with a doxastic
or epistemic structure representing her beliefs. In this
respect, the argument is parallel in spirit to Coase’s (1937) theory of the firm. Coase asked why firms
exist if markets can coordinate economic activity, and his answer was that using markets is itself costly.Where the paper says it
Where the paper says it
von Mises, Ludwig (2910) “Die Wirtschaftsrechnung im sozialistischen Gemeinwesen”, Sozialwissenschaft und Sozialpolitik 47: 86-121. p. 2 In the socialist calculation debates of the 1930s and 1940s,
Hayek emphasized two complementary informational roles of the price system. First, echoing earlier arguments by von
von Mises (1920), Hayek (1935a,b, 1940, 1942) argued that competitive prices integrate dispersed, local, and tacit knowledge
that no central planning committee could construct from scratch.Where the paper says it
Wilson, R. (1977). “A bidding model of perfect competition. p. 3 The next step is to specify what it means for Ci(x) to be derivable from an agent’s information. In
much of the existing literature, this question is left implicit: decision-makers are assumed to have
whatever mathematical or probabilistic reasoning capacity needed to draw the relevant inferences,
so that the connection between information and conclusion is absorbed into the model’s background
assumptions rather than examined in its own right. Earlier proof-theoretic
work in game theory studies the structure of interpersonal reasoning itself. Here, the same constructive
perspective is brought to bear on a classical problem in economic theory: the informational conditions
under which decentralized reasoning recovers the core, and the informational burden hidden behind
the Debreu-Scarf convergence.Where the paper says it
Where the paper says it
Wilson, Robert (1978), “Information, efficiency, and the core of an economy. p. 5 The literature on the core under incomplete information,
beginning with Wilson (1978) and developed by Vohra (1999); Dutta and Vohra (2005); Serrano and
Vohra (2007); de Clippel (2007), and Myerson (2007), among others, transforms the cooperative setting
into a decentralized problem of individual strategic reasoning and endows each player with a doxastic
or epistemic structure representing her beliefs. 2 A central finding of this literature is that stability
under incomplete information generally diverges from efficiency under complete information: solution
concepts based on agents’ limited knowledge may yield outcomes that are stable given what agents
know yet fail to coincide with the efficient core.Where the paper says it
--- Page 1 ---
Measuring Information Burden: From Coalition-Based Reasoning to the Price System Shuige Liu Department of Decision Sciences, Bocconi University Milan, Italy June 2, 2026 Abstract. The price system is often said to economize on information <ref:1802.04595#pg2>. This paper asks how much information it saves <ref:1802.04595#pg4>.
Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.
Improvements for AI systems
-
Precise measurement of informational burden: The improved AI system can formally calculate how much information is saved by a price system by counting
load-bearing coalitions
whose feasibility information must be distributed across the population, following the formula for average per-agent informational burden growth as Theta(4k/k(3/2)). -
Formal reasoning under explicit belief structures: The system can use Gentzen's sequent calculus to verify whether an agent can derive a decision criterion Ci(x) from her specific set of held sentences, making the connection between information and conclusion transparent.
-
Identifying minimal information requirements for equilibrium: The AI can determine the exact
coverage condition
that ensures unanimous acceptability coincides with the core for all games on a fixed player set, which is the weakest sufficient condition. -
Evaluating mechanism design via informational cost: The system can assess mechanisms not only by efficiency and incentive compatibility but also by
what a mechanism demands that agents know and infer,
providing an explicit account of informational and cognitive simplicity to mechanism evaluation.
Abstract
The price system is often said to economize on what individuals need to know. This paper measures a particular informational burden associated with decentralized reasoning in its absence: the coalition-feasibility information required for individual rejection decisions to recover the core. We develop a proof-theoretic framework in which information is represented by economically meaningful sentences and reasoning by derivations in Gentzen's sequent calculus. An agent rejects a proposed payoff vector only when she can establish the existence of a profitable coalitional deviation; failure to reject may instead reflect ignorance. For each transferable-utility game, the main theorem characterizes the informational distributions under which the set of unopposed payoff vectors coincides with the core: every noncore payoff vector must have a blocking coalition known to at least one of its members. This yields a game-specific measure of minimum informational burden. We extend the characterization to pure-exchange economies through a first-order framework that preserves finite informational premises. Applying the result to Shapley's (1975) replica economy, within the core-convergence tradition of Debreu and Scarf (1963), we show that the minimum average coalition-level informational burden grows as Θ(4 k/k 3/2), exponentially with replication. The result quantifies an informational requirement of direct decentralized coalitional reasoning that price-mediated coordination can spare agents, without claiming to measure the total informational savings of prices.