pith. sign in

arxiv: 2604.13885 · v2 · submitted 2026-04-15 · ❄️ cond-mat.stat-mech · physics.soc-ph

Role of volatility mixing in wealth condensation transition

Jaeseok Hur , Meesoon Ha , Hawoong Jeong This is my paper

Pith reviewed 2026-05-10 12:20 UTC · model grok-4.3

classification ❄️ cond-mat.stat-mech physics.soc-ph
keywords wealth condensationvolatility mixingstochastic block modelpower-law tail exponentBouchaud-Mézard modelnetworked wealth dynamicsnoise heterogeneitynonequilibrium condensation
0
0 comments X

The pith

Local mixing between high- and low-volatility nodes in a wealth network neutralizes their separate tail exponents and can push the overall distribution into condensation.

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

The paper extends the Bouchaud-Mézard wealth model by assigning each node one of two volatility levels and connecting the groups with a stochastic block model whose mixing parameter is varied. It finds that cross-group links cause the effective power-law tail of the combined wealth distribution to become shallower than either group would produce alone. This neutralization effect depends on both the global coupling strength and the degree of volatility mixing. When the resulting aggregate exponent falls below 2, wealth condenses into a small number of nodes. The work therefore treats volatility heterogeneity on networks as an independent control knob for the condensation transition.

Core claim

In the extended Bouchaud-Mézard model on a stochastic-block network with two volatility classes, local interactions between unlike-volatility nodes neutralize the group-specific exponents, yielding a lower effective tail exponent for the aggregate wealth distribution that can cross the condensation threshold γ_c = 2 even when the global parameter Λ would not produce condensation in a homogeneous case.

What carries the argument

Neutralization of group-wise exponents through local interactions between binary-volatility nodes in the stochastic block model.

If this is right

  • The effective tail exponent is set by both the global parameter Λ = 2J/β² and the volatility mixing configuration.
  • Greater mixing between volatility classes lowers the aggregate tail exponent relative to the segregated case.
  • Condensation can occur for parameter values that would remain non-condensed in a single-volatility network.
  • Noise heterogeneity on networks supplies an additional mechanism that can control the onset of wealth condensation.

Where Pith is reading between the lines

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

  • Real-world social or financial networks that mix investors of very different risk tolerances may therefore exhibit stronger inequality than their average volatility would suggest.
  • The neutralization effect could be even larger if volatility values were drawn from a continuous distribution rather than two discrete levels.
  • Changing the underlying network topology while keeping the same volatility groups would likely alter the critical mixing threshold for condensation.

Load-bearing premise

Binary volatility groups linked by a stochastic block model are enough to capture how noise heterogeneity shapes the wealth tail without continuous volatility distributions or extra economic rules.

What would settle it

A simulation in which the measured aggregate tail exponent stays above 2 when volatility groups are fully segregated but drops below 2 once the same nodes are rewired to increase cross-group links, while keeping Λ fixed.

Figures

Figures reproduced from arXiv: 2604.13885 by Hawoong Jeong, Jaeseok Hur, Meesoon Ha.

Figure 1
Figure 1. Figure 1: FIG. 1. (a) Visualization of ER network with average degree [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2. (a) Schematic illustrations of SBM for modified BM [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3. (a) Aggregate effective tail exponent ˆγ [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗
read the original abstract

We study the role of heterogeneous volatility in a networked wealth dynamics model and its impact on the wealth condensation transition. Extending the Bouchaud--M{\'e}zard framework, we introduce binary volatility in networks and investigate how its configuration affects the effective power-law tail exponent of the wealth distribution. Using a stochastic block model, we control the mixing between volatility groups and show that the effective exponent is governed not only by the global parameter $\Lambda=2J/\beta^2$ but also by the volatility configuration in the network. We find that local interactions between nodes with different volatility induce a neutralization of group-wise exponents, which lowers the aggregate tail exponent and can drive a condensation transition across $\gamma_{\rm c}=2$. Our results identify volatility mixing as another control mechanism for wealth condensation and highlight the importance of noise heterogeneity in nonequilibrium systems on networks.

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.

Referee Report

3 major / 2 minor

Summary. The paper extends the Bouchaud-Mézard wealth dynamics model on networks by introducing binary volatility heterogeneity whose mixing is controlled by a stochastic block model. Numerical simulations show that local interactions between nodes of differing volatility produce a neutralization of the group-wise power-law exponents, yielding a lower effective aggregate tail exponent that can drive the wealth distribution across the condensation threshold γ_c=2. The effective exponent is reported to depend on both the global parameter Λ=2J/β² and the volatility configuration.

Significance. If the neutralization mechanism is robust, the work identifies volatility mixing as an additional control parameter for the condensation transition in networked nonequilibrium systems, complementing the existing Λ dependence and highlighting the role of noise heterogeneity in shaping stationary distributions.

major comments (3)
  1. [Numerical results / Figures 3-5] The procedure used to extract the effective tail exponent from the simulated stationary wealth distributions is not described: no fitting method (maximum-likelihood, least-squares, etc.), no fitting range, and no error bars or bootstrap estimates are provided. This is load-bearing because the central claim that mixing drives the aggregate exponent below γ_c=2 rests on the numerical values of these exponents.
  2. [Model and simulation details] No systematic checks are reported on the dependence of the neutralization effect on network size N or on the volatility contrast ratio. Finite-size scaling or extrapolation to N→∞ is required to confirm that the reported lowering of the aggregate exponent is not a finite-N artifact.
  3. [Discussion / Conclusions] The neutralization result is demonstrated only for binary volatility groups coupled by an SBM. The manuscript does not test whether the same lowering of the aggregate exponent occurs for continuous volatility distributions drawn from, e.g., a log-normal or gamma distribution; this limits the generality of the claim that volatility mixing is a robust mechanism.
minor comments (2)
  1. [Model definition] Notation for the volatility parameters (β_i) and the SBM mixing parameter should be introduced explicitly in the model section before being used in the results.
  2. [Abstract] The abstract states that mixing 'can drive a condensation transition across γ_c=2'; the manuscript should clarify whether this threshold remains exactly 2 or is shifted by the heterogeneity.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the constructive and detailed report. The comments highlight important issues in the presentation of numerical results and the scope of the study. We have revised the manuscript to incorporate additional methodological details, finite-size checks, and an explicit discussion of limitations. Below we respond to each major comment.

read point-by-point responses

  1. Referee: [Numerical results / Figures 3-5] The procedure used to extract the effective tail exponent from the simulated stationary wealth distributions is not described: no fitting method (maximum-likelihood, least-squares, etc.), no fitting range, and no error bars or bootstrap estimates are provided. This is load-bearing because the central claim that mixing drives the aggregate exponent below γ_c=2 rests on the numerical values of these exponents.

    Authors: We agree that the exponent extraction procedure must be fully specified. In the revised manuscript we have added a new subsection (III.C) that describes the method: we employ maximum-likelihood estimation for power-law tails following the procedure of Clauset et al. (2009), with the lower cutoff x_min chosen by minimizing the Kolmogorov-Smirnov distance between the empirical distribution and the fitted power law. The fitting range is restricted to the upper 5–10 % of the wealth values, consistent with the visual tails in Figs. 3–5. Statistical uncertainties are obtained from 200 bootstrap resamples of each stationary distribution; these error bars are now shown in the revised figures. The updated text makes clear that the reported crossing of γ_c = 2 under strong mixing is robust to these choices. revision: yes

  2. Referee: [Model and simulation details] No systematic checks are reported on the dependence of the neutralization effect on network size N or on the volatility contrast ratio. Finite-size scaling or extrapolation to N→∞ is required to confirm that the reported lowering of the aggregate exponent is not a finite-N artifact.

    Authors: We acknowledge the absence of finite-size analysis in the original submission. We have performed additional simulations for N ranging from 500 to 10 000 and included a new finite-size scaling study (Appendix B and revised Fig. 6). The effective aggregate exponent converges with increasing N for fixed volatility contrast; the neutralization effect and the crossing below γ_c = 2 remain intact in the large-N limit. We have also varied the volatility contrast ratio between 1:2 and 1:10 and added a brief discussion showing that the qualitative lowering of the aggregate exponent is insensitive to the precise contrast once it exceeds a modest threshold. These results are now reported in the revised manuscript. revision: yes

  3. Referee: [Discussion / Conclusions] The neutralization result is demonstrated only for binary volatility groups coupled by an SBM. The manuscript does not test whether the same lowering of the aggregate exponent occurs for continuous volatility distributions drawn from, e.g., a log-normal or gamma distribution; this limits the generality of the claim that volatility mixing is a robust mechanism.

    Authors: We agree that the binary volatility setting, while sufficient to isolate the mixing mechanism, does not exhaust the space of possible heterogeneity. In the revised Conclusions we have added an explicit paragraph acknowledging this limitation and stating that the neutralization effect is demonstrated for the binary case as a proof of principle. We note that the underlying local-interaction mechanism (nodes with different volatilities exchanging wealth) is not restricted to two values, but we have not performed the corresponding continuous-distribution simulations in the present work. We therefore present the binary result as the core finding and flag continuous distributions as a natural direction for follow-up study. revision: partial

Circularity Check

0 steps flagged

No significant circularity; central claim rests on independent SBM simulations

full rationale

The derivation adopts the global parameter Λ from the established Bouchaud-Mézard model (external prior work) but obtains the neutralization effect and lowered aggregate tail exponent from fresh stochastic-block-model simulations of binary volatility groups. No equation reduces the new prediction to a redefinition or fit of the input data; the aggregate tail is measured numerically rather than forced by construction. The paper is self-contained against external benchmarks and does not rely on self-citation chains for its load-bearing step.

Axiom & Free-Parameter Ledger

2 free parameters · 2 axioms · 0 invented entities

The model inherits the multiplicative-noise dynamics of the original Bouchaud-Mézard framework and adds binary volatility groups whose specific numerical values are not stated; the condensation threshold γ_c=2 is taken as given from prior literature.

free parameters (2)
  • binary volatility contrast
    High- and low-volatility values are introduced without numerical specification or fitting procedure in the abstract.
  • SBM mixing parameter
    Controls inter-group connection probability and is treated as a free control variable.
axioms (2)
  • domain assumption Wealth updates follow the extended Bouchaud-Mézard multiplicative process with additive network interactions.
    Invoked when the authors state they are extending the Bouchaud-Mézard framework.
  • standard math Power-law tail exponent γ_c=2 marks the condensation transition.
    Standard result in wealth-distribution literature; used to interpret when the lowered effective exponent crosses the threshold.

pith-pipeline@v0.9.0 · 5446 in / 1443 out tokens · 49779 ms · 2026-05-10T12:20:09.863736+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

18 extracted references · 18 canonical work pages · 2 internal anchors

  1. [1]

    Bouchaud and M´ ezard [1] reported that a ran- dom network with a mean degree⟨k⟩= 4 exhibits a power-law tail

    Power-law test and size dependence In this appendix, we examine the statistical test for power-law tails and their robustness across parameter regimes. Bouchaud and M´ ezard [1] reported that a ran- dom network with a mean degree⟨k⟩= 4 exhibits a power-law tail. Garlaschelli [13] discussed the relevance of log-normal behavior in sparsely connected cluster...

    work page

  2. [2]

    Wealth distribution and asymmetric wealth shares in HBM model Figure A3 shows the normalized wealth distributions and the group-wise wealth shares for the two volatility groups. A comparison of the group-wise wealth distri- butions indicates that the effective tail exponent of the overall distribution and wealth shares are primarily gov- erned by the low-...

    work page

  3. [3]

    We consider the following three cases

    Modified numerical simulations In this subsection, we briefly present additional nu- merical results for variants of the model that go beyond the binary-volatility setting and the symmetric two-block SBM considered in the main text. We consider the following three cases. (i)Continuous volatility within each block. Instead of assigning a single volatility ...

    work page

  4. [4]

    Bouchaud and M

    J.-P. Bouchaud and M. M´ ezard, Physica A: Statistical Mechanics and its Applications282, 536 (2000)

    work page 2000

  5. [5]

    Erd¨ os and A

    P. Erd¨ os and A. R´ enyi, Publicationes Mathematicae De- brecen6, 290 (1959)

    work page 1959

  6. [6]

    Bernard, J.-P

    M. Bernard, J.-P. Bouchaud, and P. Le Doussal, Physical Review E113, L032101 (2026)

    work page 2026

  7. [7]

    J. Hur, M. Ha, and H. Jeong, arXiv preprint arXiv:2506.08175 (2025)

    work page internal anchor Pith review arXiv 2025

  8. [8]

    Bouchaud and M

    J.-P. Bouchaud and M. M´ ezard, Journal of Physics A: Mathematical and General30, 7997 (1997)

    work page 1997

  9. [9]

    For the original SDE of BM model:dC i =αC idt+ βCidWt,i + P j(̸=i) (JijCj −J jiCi)dt, the normalized wealth is defined asci =C i/ ¯Cwhich always gives⟨c⟩= 1

    work page

  10. [10]

    Cinelli, L

    M. Cinelli, L. Peel, A. Iovanella, and J.-C. Delvenne, Physical Review E102, 062310 (2020)

    work page 2020

  11. [11]

    Vicsek, A

    T. Vicsek, A. Czir´ ok, E. Ben-Jacob, I. Cohen, and O. Shochet, Physical review letters75, 1226 (1995)

    work page 1995

  12. [12]

    L. L. Bonilla, Physical Review E100, 022601 (2019)

    work page 2019

  13. [13]

    Imry and S.-k

    Y. Imry and S.-k. Ma, Physical Review Letters35, 1399 (1975)

    work page 1975

  14. [14]

    Ahrens and A

    B. Ahrens and A. K. Hartmann, Physical Review B—Condensed Matter and Materials Physics84, 144202 (2011)

    work page 2011

  15. [15]

    Clauset, C

    A. Clauset, C. R. Shalizi, and M. E. Newman, SIAM review51, 661 (2009)

    work page 2009

  16. [16]

    Garlaschelli and M

    D. Garlaschelli and M. I. Loffredo, Physica A: Statistical Mechanics and its Applications338, 113 (2004)

    work page 2004

  17. [17]

    Small-World Effects in Wealth Distribution

    W. Souma, Y. Fujiwara, and H. Aoyama, arXiv preprint cond-mat/0108482 (2001)

    work page internal anchor Pith review arXiv 2001

  18. [18]

    Q. H. Vuong, Econometrica: journal of the Econometric Society , 307 (1989)

    work page 1989