arxiv: 2604.13794 · v1 · submitted 2026-04-15 · 💰 econ.TH · math.CO

Recognition: unknown

Balanced Contributions in Networks and Games with Externalities

Frank Huettner

Authors on Pith no claims yet

Pith reviewed 2026-05-10 12:18 UTC · model grok-4.3

classification 💰 econ.TH math.CO

keywords allocation rulesnetwork gamesexternalitiesbalanced contributionscomponent efficiencyMyerson value

0 comments

The pith

A unique component-efficient allocation rule satisfies balanced contributions for every edge in networks with externalities.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

In networks where the worth of any group can depend on connections outside it, the paper establishes that exactly one way exists to divide the total value while meeting two conditions: each connected component gets exactly its own worth, and any single link removal changes the shares of its two endpoints by the same amount. This matters because the usual fairness principle and the efficiency requirement no longer pick out the same rule once externalities are allowed, unlike in standard games. The proof constructs the rule from a spanning tree and then invokes a cycle-sum identity to confirm the equal-impact condition holds for all remaining edges. The resulting BCE rule recovers the Myerson value when externalities vanish and the externality-free value on the complete network, yet it cannot be obtained simply by ignoring the network structure.

Core claim

The BCE rule is the unique component-efficient allocation rule that satisfies balanced contributions on every possible edge. Its existence is shown by first defining shares on the edges of any spanning tree so that the balanced-contributions condition holds there, then using the cycle-sum identity to verify that the same condition automatically holds for every non-tree edge; this identity reduces the required equality on a cycle to a sum of balanced-contributions relations already satisfied in strictly smaller subnetworks.

What carries the argument

The cycle-sum identity, which equates a linear combination of balanced-contributions differences around any cycle to a relation among allocations in the proper subnetworks obtained by deleting edges of that cycle.

If this is right

The BCE rule coincides with the Myerson value on transferable-utility games.
It coincides with the Jackson-Wolinsky value on network games that have no externalities.
On the complete network it returns the externality-free value.
Unlike the fairness-based FCE rule, the BCE rule cannot be obtained by applying a graph-free formula to the graph-restricted game.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

The same construction technique may extend to other fairness axioms once an analogous cycle-reduction identity is identified.
In applications such as supply chains or social networks, the BCE rule supplies a computable benchmark that treats every possible link symmetrically rather than privileging a particular spanning tree.
The distinction between BCE and FCE suggests that empirical tests could separate whether agents care more about equal marginal impact of links or about equal treatment of symmetric positions.

Load-bearing premise

The cycle-sum identity holds for any cycle, so that balanced contributions on non-tree edges reduce to already-established relations in smaller networks.

What would settle it

Exhibit a concrete network with externalities in which the candidate BCE shares violate the balanced-contributions equality on at least one non-tree edge, or show that two distinct component-efficient rules both satisfy balanced contributions on all edges.

Figures

Figures reproduced from arXiv: 2604.13794 by Frank Huettner.

**Figure 2.** Figure 2: A fundamental cycle: tree edges (solid blue) satisfy balanced contributions ( [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗

**Figure 3.** Figure 3: Failure of pair-wise balanced contributions ( [PITH_FULL_IMAGE:figures/full_fig_p010_3.png] view at source ↗

**Figure 4.** Figure 4: Graph-projection can change w at networks h that are not subnetworks of g: here w({1, 2, 3}, h) = 0 but w g ({1, 2, 3}, h) = 1. Yet the BCE(w, g) only evaluates w on subnetworks of g, where w g = w, hence BCE(w g , g) = BCE(w, g). changes w at h: the link {1, 2} is absent in h, so w({1, 2, 3}, h) = 0, but the g-projection reads off g instead and returns w g ({1, 2, 3}, h) = w({3}, g) + w({1, 2}, g) = 1. Ye… view at source ↗

**Figure 5.** Figure 5: Two networks with v g = v g ′ for v = u ≼ ({3}, {{1,2},{3}}) . The FCE value is the same for both; the BCE value differs. To the FCE value, it does not matter whether player 1 can communicate the threat of deleting {1, 2} to player 3; the BCE value accounts for this through balanced contributions. two networks in [PITH_FULL_IMAGE:figures/full_fig_p016_5.png] view at source ↗

read the original abstract

For networks with externalities, where each component's worth may depend on the full network structure, balanced contributions and fairness lead to distinct component-efficient allocation rules. We characterize the unique component-efficient allocation rule satisfying balanced contributions -- the BCE rule. Existence is the main challenge: balanced contributions must hold on every edge, but the construction uses only spanning-tree edges. A cycle-sum identity bridges this gap by reducing balanced contributions on non-tree edges to relations in proper subnetworks. The BCE rule coincides with the Myerson value for TU games and with its generalization by Jackson--Wolinsky for network games without externalities, it recovers the externality-free value on the complete network, and -- unlike the fairness-based FCE rule -- it does not reduce to a graph-free formula applied to the graph-restricted game.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

Huettner pins down a unique BCE rule for networks with externalities via balanced contributions and a cycle-sum identity that recovers the standard values in the no-externality cases.

read the letter

The paper separates balanced contributions from fairness when component values depend on the full network. It defines the BCE rule as the unique component-efficient allocation that satisfies balanced contributions on every edge and shows it matches the Myerson value for ordinary TU games, the Jackson-Wolinsky value for network games without externalities, and the externality-free value on the complete graph. The construction proceeds inductively along a spanning tree and then uses the cycle-sum identity to extend the condition to the remaining edges without solving the entire system simultaneously. That move keeps the rule well-defined and single-valued on arbitrary networks, which is the main technical achievement. The distinction from the FCE rule is also drawn cleanly: BCE does not collapse to a graph-free formula applied to the restricted game. The special cases and recovery results are stated precisely enough to see the difference in approach. The soft spot is the cycle-sum identity itself. If that identity only holds under additional restrictions on how v(S, g) depends on edges outside S, or if overlapping cycles produce inconsistent requirements, the uniqueness claim would not go through for general externality structures. The abstract presents the identity as holding identically, but the proof details would need checking to confirm it imposes no hidden linearity or independence assumptions. This work is for people already working on axiomatic cooperative game theory on networks. A reader who cares about how balanced contributions behave once externalities are allowed will find the separation from fairness useful and the construction instructive. It is technically non-trivial and internally coherent on the terms it sets, so it deserves a serious referee who can verify the cycle-sum step and the examples.

Referee Report

1 major / 1 minor

Summary. The manuscript characterizes the unique component-efficient allocation rule satisfying the balanced contributions axiom in network games with externalities, termed the BCE rule. It addresses the challenge of ensuring balanced contributions on all edges by constructing the rule using spanning trees and invoking a cycle-sum identity to handle non-tree edges. The rule is shown to coincide with the Myerson value for TU games and the Jackson-Wolinsky value for network games without externalities, while differing from the fairness-based FCE rule by not reducing to a graph-free formula.

Significance. If the cycle-sum identity holds as claimed, the BCE rule offers a principled way to allocate value in networks where component worth depends on the entire structure, providing an alternative to fairness axioms. This could have implications for economic models involving externalities in networks, such as in collaboration or trade networks. The paper credits the axiomatic approach and shows consistency with prior special cases.

major comments (1)

[Existence proof (cycle-sum identity)] The central existence argument relies on a cycle-sum identity to extend balanced contributions from spanning-tree edges to all edges. However, it is unclear whether this identity holds for arbitrary externality dependencies in v(S, g) without additional assumptions, or if overlapping cycles could lead to inconsistent allocations. This is load-bearing for the uniqueness claim, as failure would mean the rule either violates BC or is not well-defined.

minor comments (1)

[Abstract] The abstract could more explicitly reference the theorem number for the main characterization result.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their careful reading of the manuscript and for identifying the cycle-sum identity as central to the existence and uniqueness argument. We respond to the major comment below.

read point-by-point responses

Referee: [Existence proof (cycle-sum identity)] The central existence argument relies on a cycle-sum identity to extend balanced contributions from spanning-tree edges to all edges. However, it is unclear whether this identity holds for arbitrary externality dependencies in v(S, g) without additional assumptions, or if overlapping cycles could lead to inconsistent allocations. This is load-bearing for the uniqueness claim, as failure would mean the rule either violates BC or is not well-defined.

Authors: The cycle-sum identity is proved by induction on network size. The allocation rule is first defined on spanning trees to satisfy balanced contributions and component efficiency on those edges. For any non-tree edge, the identity equates the balanced-contributions difference on that edge to a signed sum of the same differences over a collection of proper subnetworks obtained by deleting edges from the cycle. Because each subnetwork is strictly smaller, the inductive hypothesis applies and the identity holds for arbitrary v(S,g). Overlapping cycles do not produce inconsistency: the relations are linear and the cycle space is spanned by the fundamental cycles of any spanning tree; once the tree edges satisfy the axiom, every linear combination (including overlapping cycles) is automatically satisfied. The construction therefore yields a unique well-defined rule that meets balanced contributions on every edge for any externality structure. No additional assumptions on v are required. revision: no

Circularity Check

0 steps flagged

No circularity: axiomatic characterization remains independent of its own outputs

full rationale

The paper defines the BCE rule via the standard axioms of component efficiency and balanced contributions, then proves uniqueness and existence by an inductive construction on spanning trees together with a cycle-sum identity that reduces non-tree edges to subnetwork relations. No quoted step shows the rule being fitted to data, renamed from a known result, or derived from a self-citation whose content is itself unverified; the cycle-sum identity is presented as a derived property of the value function rather than an assumption smuggled in to force the conclusion. The derivation is therefore self-contained against the stated axioms and does not reduce to its inputs by construction.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The characterization relies on standard graph-theoretic notions (spanning trees, cycles) and the existence of a cycle-sum identity that reduces balanced-contribution conditions on non-tree edges. No free parameters or invented entities are mentioned in the abstract.

axioms (2)

domain assumption Balanced contributions must hold on every edge, yet the allocation is constructed using only spanning-tree edges.
Stated in the abstract as the main existence challenge that the cycle-sum identity resolves.
domain assumption Component efficiency requires each connected component to divide exactly the value it creates.
Core axiom used to define the class of rules under consideration.

pith-pipeline@v0.9.0 · 5417 in / 1361 out tokens · 35139 ms · 2026-05-10T12:18:33.967500+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

12 extracted references

[1]

Decomposition of solutions and the Shapley value

Casajus, A., Huettner, F., 2018. Decomposition of solutions and the Shapley value. Games and Economic Behavior 108, 37–48. De Clippel, G., Serrano, R., 2008. Marginal contributions and externalities in the value. Econometrica 76, 1413–1436

2018
[2]

Bargaining, coalition formation and value

Feldman, B., 1996. Bargaining, coalition formation and value. Ph.D. thesis. State University of New York at Stony Brook

1996
[3]

Potential, value, and consistency

Hart, S., Mas-Colell, A., 1989. Potential, value, and consistency. Econometrica 57, 589–614

1989
[4]

A strategic model of social and economic networks

Jackson, M.O., Wolinsky, A., 1996. A strategic model of social and economic networks. Journal of Economic Theory 71, 44–74

1996
[5]

Implementing cooperative solution concepts: a generalized bidding approach

Ju, Y., Wettstein, D., 2025. Implementing cooperative solution concepts: a generalized bidding approach. Economic Theory 79, 221–258

2025
[6]

Game Theory

Maschler, M., Solan, E., Zamir, S., 2013. Game Theory. Cambridge University Press, Cambridge

2013
[7]

Conference structures and fair allocation rules

Myerson, R.B., 1980. Conference structures and fair allocation rules. International Journal of Game Theory 9, 169–182

1980
[8]

Fair allocation in networks with externalities

Navarro, N., 2007. Fair allocation in networks with externalities. Games and Economic Behavior 58, 354–364

2007
[9]

A Course in Game Theory

Osborne, M.J., Rubinstein, A., 1994. A Course in Game Theory. MIT Press. P´ erez-Castrillo, D., Wettstein, D., 2001. Bidding for the surplus: A non-cooperative ap- proach to the Shapley value. Journal of Economic Theory 100, 274–294. Pham Do, K.H., Norde, H., 2007. The Shapley value for partition function form games. International Game Theory Review 9, 353–360

1994
[10]

A value forn-person games, in: Kuhn, H.W., Tucker, A.W

Shapley, L.S., 1953. A value forn-person games, in: Kuhn, H.W., Tucker, A.W. (Eds.), Contributions to the Theory of Games II. Princeton University Press. number 28 in Annals of Mathematics Studies, pp. 307–317

1953
[11]

Bidding for surplus in network allocation problems

Slikker, M., 2007. Bidding for surplus in network allocation problems. Journal of Economic Theory 137, 493–511

2007
[12]

Naval Research Logistics Quarterly 10, 281–298

Thrall, R.M., Lucas, W.F., 1963.n-person games in partition function form. Naval Research Logistics Quarterly 10, 281–298. 20

1963