Interactions between resource dependent branching processes and equilibria

F.Thomas Bruss

REVIEW 3 major objections 5 minor 2 references

Interactions between resource dependent branching processes and equilibria

T0 review · 3 major / 5 minor · reviewed 2026-08-09 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A stable population ratio forces a unique resource threshold.

desk verdict The necessity half of the equilibrium theorem is honestly proved; the sufficiency half has a genuine gap that needs a serious referee, but the paper deserves one. read the letter →

arxiv 2502.03872 v1 pith:XCTV6GKJ submitted 2025-02-06 math.PR

F.Thomas Bruss This is my paper · ORCID

classification math.PR MSC 60J85

keywords equilibriumequationresource-dependentbranchingprocessBorel-Cantellilemmabisexualmulti-typeoptimaltransportBRS-inequalitycontrol

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper proves a precise condition under which two sub-populations competing for a shared, accumulated resource can settle into a lasting equilibrium. If the ratio of immigrant to home population sizes converges to a finite positive limit, then both populations must survive indefinitely, and the resource threshold each generation uses converges to a unique level. That level is determined by a linear balance equation equating expected resource production with expected consumption in both groups. The proof also yields a constraint: at the threshold, the expected reproduction rates of the two sub-populations must be equal and at least one, and if that common rate is strictly larger than one, such an equilibrium can actually occur.

What carries the argument

The central object is the resource-dependent branching process (RDBP) with two sub-populations drawing claims from the same accumulated resource space. The argument is carried by the limiting equilibrium equation, obtained by dividing the resource balance by the home population size, writing the immigrant contribution as $\alpha$ times a per-capita term, and applying the strong law of large numbers conditioned on survival. The companion constraint $r_h F_h(F) = r_i F_i(F)$ emerges from taking limits on both sides of the recurrence expressing the next generation's ratio, and the Borel-Cantelli arguments control extinction and supercritical growth.

What would settle it

Construct a two-population resource-dependent branching process satisfying all assumptions except that the two sub-populations draw from separate resource spaces; if a finite positive limiting ratio $\alpha$ still exists, the claimed equilibrium equation need not hold. Alternatively, simulate the model with claim densities such that $r_h F_h(F) = r_i F_i(F) = 1$ and check whether the ratio never converges to a finite positive constant, contradicting positive-probability existence.

Watch

Extended reading notes

Core claim

The central claim is that if the ratio $\Gamma^i_t/\Gamma^h_t$ of immigrant to home population sizes converges almost surely to some $\alpha \in (0,\infty)$, then both sub-processes tend to infinity almost surely and the resource threshold $F_t$ converges almost surely to a unique value $F := F(\alpha)$. The limiting pair satisfies the equilibrium equation $$ r_h \int_0^F x\,dF_h(x) + \$\alpha$\, r_i \int_0^F x\,dF_i(x) = R_h + \$\alpha$\, R_i, $$ together with the constraint qualification $r_h F_h(F) = r_i F_i(F) \ge 1$. If the common value is strictly greater than one, then such an equilibrium exists with strictly positive probability. This is the content of Theorem 3 of the companion paper, for which the present supplement supplies the full proof.

Load-bearing premise

The entire argument relies on the assumption that all resource claims of both sub-populations are submitted to the same accumulated resource space, so total consumption is the additive sum of the two groups' claims; if the groups drew on separate pools or one had priority access, the equilibrium equation would be different.

Editorial extensions

If this is right

If the paper is right, lasting coexistence of two sub-populations in this model requires both to be supercritical at the equilibrium threshold; a sub-population whose reproduction rate at the threshold is below one is driven to extinction.
The equilibrium threshold $F(\alpha)$ is unique whenever the claim densities do not both vanish near it, so the eventual composition of the population is determined by the balance equation.
The condition $r_h F_h(F) = r_i F_i(F) \ge 1$ is testable: it gives a concrete inequality that parameters and claim distributions must satisfy before any finite-positive-ratio equilibrium is possible.
With the strict inequality $>1$, the equilibrium is not just a limiting possibility but occurs with strictly positive probability, so the model predicts that such coexistence is a stable outcome.
The author indicates that the same approach, via the BRS-inequality, can be extended to an arbitrary number of sub-populations, though only the two-population case is fully proved here.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

The equilibrium equation suggests a direct calibration recipe: given an observed ratio $\alpha$, the implied threshold $F(\alpha)$ solves the balance equation, so the model yields a concrete, falsifiable prediction linking population ratios, resource production means, and claim distributions.
Because the proof requires all claims to be submitted to one shared resource space, the model would not apply to settings with segregated resource pools or priority access; testing those settings would need a modified equilibrium condition.
One could test the boundary case $r_h F_h(F) = r_i F_i(F) = 1$ numerically to see whether equilibria are indeed of probability zero, sharpening the paper's dichotomy.
The optional transport formulation suggests that policy interventions that alter claim distributions shift both $\alpha$ and $F$; this could be used to compute the minimal-cost redistribution of claims needed to reach a target equilibrium.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Dear colleague,

The paper delivers what it promises: complete proofs of the equilibrium theorem announced in the Festschrift article. Parts (I)-(III), the necessity direction, are mostly sound. The SLLN derivation of the equilibrium equation in Part (II) is clean, and the argument that an existing finite ratio forces both sub-processes to infinity is fine. The Borel-Cantelli step in Part (III) is compressed but plausible, and citing Bruss-Duerinckx (2015) for the survival criterion is legitimate when the cited theorem is proved there.

The soft spot is Part (IV), and it is not a minor compression. To prove sufficiency the paper needs a strictly positive probability that both sub-processes go to infinity. What (12) proves is only that the home process goes to infinity with positive probability, via the 2015 theorem. The next sentence claims that if one sub-process tends to infinity then both must, 'according to the definition of α'. That is circular: α is defined as the a.s. limiting ratio, so invoking it presupposes exactly the joint convergence at issue. On the event where the home process grows and the immigrant process dies out, the ratio tends to 0, and nothing in Part (I) rules that out, because Part (I) applies only when a finite positive limit already exists. The initial conditioning in Part (IV) on the ratio neither going to 0 nor ∞ is not used in (12), so it cannot supply the missing jointness. As written, the positive-probability existence of an equilibrium under the strict >1 condition is not established.

The reader's conditional verdict is fair, but I'd go a bit further: the gap is load-bearing, not stylistic. The additive resource-claims assumption is stated clearly and is genuinely necessary for the linear equation; that is a modeling constraint, not a flaw. Minor issues: the text says 'Part (VI)' where it means 'Part (IV)', and the theorem statement lives only in the Festschrift paper, which is awkward for a supplement. The optimal transport and control sections are programmatic, as the author acknowledges.

Who gets value from this? Anyone working on resource-dependent branching processes or demographic coexistence. The necessity result is a useful criterion, and the sufficiency gap is instructive — it marks exactly where a two-type survival argument must go. This deserves a serious referee: the core idea is valuable and the flaw is identifiable and likely fixable with an additional argument. Send it to review, but the referee should focus on Part (IV).

Referee Report

3 major / 5 minor

Summary. This supplement to the author's Festschrift paper (Bruss, 2024) provides the proof of Theorem 3 for a model of two interacting resource-dependent branching processes, one 'home' and one 'immigrant' population. The theorem characterizes equilibria via a limiting ratio alpha of population sizes, a common resource threshold F, and an equilibrium equation balancing expected resource consumption against expected resource production, together with a constraint qualification on the reproduction rates times the claim distributions at F. The proof has four parts: (I) necessity of both sub-processes tending to infinity if a finite positive limit alpha exists; (II) derivation of the equilibrium equation from the common-resource balance via the strong law of large numbers; (III) necessity of the constraint qualification r_h F_h(F)=r_i F_i(F) >= 1; and (IV) sufficiency of the strictly greater-than-1 condition for the existence of an alpha-equilibrium with strictly positive probability. The paper also contains brief discussion of optimal transport, control, and multi-population extensions.

Significance. If the theorem is correct, it gives a clean equilibrium condition for two interacting sub-populations sharing a common resource, with a transparent economic interpretation and potential applications to population dynamics and resource allocation. The proof strategy is largely elementary, combining strong-law-of-large-numbers arguments with Borel-Cantelli techniques, and Part (II) in particular is a clear and convincing derivation of the equilibrium equation from the assumed resource balance. The necessity direction (Parts I-III) appears sound modulo minor presentation issues. However, the sufficiency direction (Part IV) contains a serious logical gap: it attempts to prove positive-probability existence of an equilibrium by first proving marginal survival of the home process and then inferring joint survival from the 'definition' of alpha, which presupposes the very convergence that must be established. This gap is load-bearing for the sufficiency claim, and the manuscript as it stands does not prove the existence part of the theorem.

major comments (3)

[Part (IV), Eq. (11)-(12)] The sufficiency proof is circular. Equation (12) establishes only that the home process survives with positive probability, P(Gamma^h_t -> infinity | r_h F_h(F) > 1) > 0, by invoking Theorem 4.4 of Bruss-Duerinckx (2015). The next sentence claims that if one sub-process tends to infinity then both must do so 'according to the definition of alpha', but the definition of alpha as the a.s. limiting ratio Gamma^i_t/Gamma^h_t presupposes exactly the joint convergence that needs to be shown. Nothing in Part (I) rules out the positive-probability event Gamma^h_t -> infinity while Gamma^i_t -> 0, since Part (I) applies only when a finite limit alpha already exists. The conditioning event {Gamma^i_t/Gamma^h_t not-> 0} ∩ {Gamma^i_t/Gamma^h_t not-> infinity} mentioned at the start of Part (IV) does not appear in (12), so it cannot supply the missing jointness. Consequently, the claimed positive-probability existence of an alpha-equilibrium under the strict inequality is unproved.
[Part (IV), line 'the process (Gamma^h_t) is a RDBP by definition'] The marginal home process is not a resource-dependent branching process in the sense of Bruss-Duerinckx (2015), because in the two-population coupled model the resource threshold F_t depends on the total claims and total resources of both sub-populations (see equation (4) and the sentence before it). Therefore Theorem 4.4 of Bruss-Duerinckx (2015) cannot be applied directly to the marginal process (Gamma^h_t) without an additional argument that the coupling does not substantially change its survival behavior. Such an argument is not provided. This is a second, independent reason why the sufficiency proof fails.
[Part (III), necessity of >= 1] The proof that r_h F_h(F) >= 1 and r_i F_i(F) >= 1 are necessary for an equilibrium is sketched via a Borel-Cantelli type argument citing Bruss (1978), but the argument is not fully written out. In particular, the step from bounded conditional expectation to almost-sure extinction is stated without derivation. This is a weaker concern than the Part (IV) gap, but if the authors intend this supplement to be a complete proof, this step should be expanded.

minor comments (5)

[Part (III), paragraph after Eq. (8)] The text reads 'the conditions r_h F_h(F) >= 1 and r_h F_h(F) >= 1 are necessary'; the second condition should be r_i F_i(F) >= 1.
[Throughout] Equation numbering is confusing: the paper refers to equation (14) of the main paper but also introduces a numbered equation (14) in Section 0.6 for the optimal transport problem. Please renumber or distinguish the supplement's equations to avoid ambiguity.
[Notation] Notation for resource production means is inconsistent: the abstract uses R_h and R_i, while Part (II) uses F_h and F_i for both claim distributions and production means in the same formulas. Please introduce distinct, consistently used symbols.
[Sections 0.6-0.8] The optimal transport and control sections are speculative and not clearly connected to the theorem proved in the earlier parts. They would be better placed in a discussion section or in the main Festschrift paper, and their current inclusion distracts from the proof.
[Proof of Part (IV), Eq. (11)] The event in (11) is stated with conditioning only on r_h F_h(F)=r_i F_i(F)>1, but the introductory sentence of Part (IV) conditions on the ratio not tending to 0 or infinity; the relationship between these two statements should be clarified.

Circularity Check

1 steps flagged · score 6.0 of 10

Part (IV) proves only marginal survival; the step to joint survival invokes the limiting ratio α whose existence is the target.

self definitional [Section 0.4, Part (IV), immediately after equation (11) and before equation (12)]
"Now, if at least one of the two sub-processes tends to infinity with strictly positive probability, then both must do so according to the definition of α (0 < α < ∞) as being the a.s. limiting ratio of Γ^i_t/Γ^h_t as t → ∞. Hence, recalling part (I), it suffices to show that P(Γ^h_t → ∞ | r_hF_h(F) > 1) = 1 − P(Γ^h_t → 0 | r_hF_h(F) > 1) > 0."

The sufficiency proof must demonstrate that Γ^i_t/Γ^h_t converges to a finite positive α on a positive-probability event. Instead, it proves only P(Γ^h_t→∞ | r_hF_h(F)>1)>0 and then uses 'the definition of α as being the a.s. limiting ratio' to infer that Γ^i_t must also tend to ∞. That inference is valid only if α already exists, which is exactly the conclusion at stake. The conditioning {Γ^i_t/Γ^h_t ↛0} ∩ {↛∞} does not imply convergence of the ratio, and (12) is not conditioned on that event, so it does not exclude the scenario Γ^h_t→∞ while Γ^i_t→0 (ratio→0). The existence of α is thus presupposed, making the step circular.

full rationale

The necessity direction (Parts I–III) is a genuine model derivation: equilibrium equation (14) is obtained from the additive resource balance (4) via the strong law of large numbers, and the constraint qualification is derived from the fraction equation (7). The paper does lean on several self-citations (Bruss 1978, 1980, 2021; Bruss–Duerinckx 2015) for extinction criteria and the RDBP survival theorem, but these are prior peer-reviewed results with stated assumptions, not restatements of the present theorem; citing them is not itself circular. The one genuinely circular step is in Part (IV), where the sufficiency argument invokes the definition of α as the a.s. limiting ratio to pass from home-process survival to joint survival, thereby assuming the existence of the limit it is supposed to prove. That affects the positive-probability existence claim, so the score is 6 rather than a lower value.

Assumptions & free parameters 0 free parameters · 7 assumptions · 0 invented entities

The central theorem rests on the RDBP model from Bruss-Duerinckx (2015), the common-resource additivity assumption, absolute continuity of claim distributions, and several cited Borel-Cantelli and survival theorems. No parameters are fitted to data.

assumptions (7)

domain assumption Each sub-population evolves as a resource-dependent branching process with independent reproduction, fixed claim distribution F_h or F_i, reproduction mean r_h or r_i, and resource production mean R_h or R_i.
Model definition carried over from Bruss-Duerinckx (2015); used in all four parts of the proof.
domain assumption All resource claims are submitted to the same accumulated resource space, so total consumption is C^h_t(Gamma^h_t) + C^i_t(Gamma^i_t).
Invoked before equation (4) in Section 0.4; this additivity produces the equilibrium equation (14).
domain assumption Claim-size distributions F_h and F_i are absolutely continuous.
Stated in Section 0.3; used to define F and prove uniqueness.
domain assumption No new immigrants arrive after time 0, so each sub-population's zero state is absorbing.
Used in part (I) to rule out extinction of one sub-population in an equilibrium.
standard math Strong law of large numbers for i.i.d. offspring counts and resource production within each sub-population.
Used in part (II) to pass from Gamma-scaled sums to means r and R.
standard math Borel-Cantelli counterpart lemmas of Bruss (1978, 1980) imply almost sure extinction when conditional extinction probabilities diverge.
Cited in parts (I) and (III) to conclude extinction from boundedness and expectation bounds.
standard math Theorem 4.4 ii) b) of Bruss-Duerinckx (2015) states that an RDBP with r_h F_h(F) > 1 has positive probability of non-extinction.
Quoted in part (IV); it carries the sufficiency half of the theorem and is not re-proved.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Interactions between resource dependent branching processes and equilibria." pith.science (2026). https://pith.science/paper/XCTV6GKJ

@misc{pith2026250203872,
  author       = {Pith},
  title        = {Pith review of: Interactions between resource dependent branching processes and equilibria},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/XCTV6GKJ}},
  note         = {Machine review of arXiv:2502.03872}
}

abstract

This paper is a supplement to the paper "Interactions between Human Populations and Related Problems of Optimal Transport" written by the same author in honour of Marc Hallin, Universit\'e Libre de Bruxelles, at the occasion of Hallin's $75$th birthday. It was announced in the main paper (Bruss (2024)) published in the Springer Festschrift entitled {\it Recent Advances in Econometrics and Statistics}. It contains the proofs which, given the space constraints required for the Festschrift, could not appear in the main paper. Moreover, we complement in the present supplement the main paper by brief comments on related problems which are likely to turn up in practice for problems of guiding human populations, namely problems of control and problems of optimal stopping.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

2 extracted references · 2 canonical work pages

[1]

A general relationship between extinction risk and carrying capacity

Alsmeyer G. and R ¨osler U. (2005) Asexual Versus Promiscuous Bisexual Galton- Watson Processes: The Extinction Probability Ratio, Ann. Appl. Probab., Vol. 12, No. 1, 125-142. Ball T. S., Balmford B., Balmford A., Rinaldo D., Visconti P . and Green R. (2024), A general relationship between population size and extinct ion risk , arXiv:2411.13228v1. Bansaye...

work page Pith review arXiv 2005
[44]

(2016), The Bruss-Robertson Inequality: Elaborations, Extension s, and Applications, Math

Steele J.M. (2016), The Bruss-Robertson Inequality: Elaborations, Extension s, and Applications, Math. Applicanda, Vol. 44(1), Tom 22/60, 3-16. Villani C. (2009), Optimal Transport, Old and New, Vol. 334 of Grundlehren der Mathematischen Wissenschaften, Springer, New Y ork. Zubkov A.M. (1970), A degeneracy condition for a bounded branching process , Mat. ...

work page 2016

Pith tools

Reviewed August 9, 2026 · model on record in the stance chip above.

[1] [1]

A general relationship between extinction risk and carrying capacity

Alsmeyer G. and R ¨osler U. (2005) Asexual Versus Promiscuous Bisexual Galton- Watson Processes: The Extinction Probability Ratio, Ann. Appl. Probab., Vol. 12, No. 1, 125-142. Ball T. S., Balmford B., Balmford A., Rinaldo D., Visconti P . and Green R. (2024), A general relationship between population size and extinct ion risk , arXiv:2411.13228v1. Bansaye...

work page Pith review arXiv 2005

[2] [44]

(2016), The Bruss-Robertson Inequality: Elaborations, Extension s, and Applications, Math

Steele J.M. (2016), The Bruss-Robertson Inequality: Elaborations, Extension s, and Applications, Math. Applicanda, Vol. 44(1), Tom 22/60, 3-16. Villani C. (2009), Optimal Transport, Old and New, Vol. 334 of Grundlehren der Mathematischen Wissenschaften, Springer, New Y ork. Zubkov A.M. (1970), A degeneracy condition for a bounded branching process , Mat. ...

work page 2016