pith. sign in

arxiv: 2601.15537 · v2 · pith:E7AXBMKWnew · submitted 2026-01-21 · ⚛️ physics.soc-ph · econ.GN· q-fin.EC

Can Rising Consumption Deepen Inequality?

Jhordan Silveira de Borba , Celia Anteneodo , Sebastian Gon\c{c}alves This is my paper

Pith reviewed 2026-05-22 12:03 UTC · model grok-4.3

classification ⚛️ physics.soc-ph econ.GNq-fin.EC
keywords wealth inequalityagent-based modelconsumptionGini indexwealth distributioneconomic inequalitycapitalism modeltransaction dynamics
0
0 comments X

The pith

Higher wealth-to-salary ratios increase inequality in an extended agent-based model of capitalism, and this pattern holds after relaxing transaction assumptions.

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

This paper extends the Social Architecture of Capitalism agent-based model to check whether rising consumption deepens wealth inequality. The authors vary transaction frequencies for purchases, salaries, and revenues while adding caps on the fraction of wealth moved in each step. They report that inequality, tracked by the Gini index, continues to rise with the key ratio R of average wealth per capita to mean salary, though the detailed shapes of wealth distributions shift. A further version lets wages adjust from within the model dynamics, showing that such feedback can either dampen or boost inequality depending on overall economic conditions.

Core claim

The macroscopic behavior of the model is predominantly governed by the single dimensionless parameter R, with inequality increasing as R increases. This dependence of inequality on R remains qualitatively robust when transactions occur at different frequencies and when maximum fractions of wealth per transaction are imposed. In a variant with adaptive wages emerging endogenously, self-organized labor-market feedback can stabilize or amplify inequality depending on macroeconomic conditions.

What carries the argument

The dimensionless ratio R of average wealth per capita to mean salary, which controls the shape of the wealth distribution, the emergence of a two-class structure, and the Gini index.

If this is right

  • Inequality, measured by the Gini index, rises with increasing R even after allowing heterogeneous transaction frequencies.
  • Limits on transaction sizes change fine details of wealth distributions but leave the overall increase of inequality with R intact.
  • Different relative frequencies for purchases, salary payments, and revenue collections affect distribution patterns without overturning the R-dependence.
  • Endogenous wage adaptation can reduce inequality under some macroeconomic conditions and increase it under others.

Where Pith is reading between the lines

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

  • If real economies track the model, policies that raise wages relative to accumulated wealth could lower inequality without changing other rules.
  • The robustness finding implies that consumption growth outpacing wage growth tends to widen wealth gaps across varied transaction settings.
  • Testing the model against historical data on transaction timing and size limits could reveal where the qualitative R-dependence breaks down.
  • Incorporating external shocks or policy interventions into the same framework might show whether they can override the R-driven rise in inequality.

Load-bearing premise

The extended agent-based model captures the essential mechanisms of real economies sufficiently to draw conclusions about what drives inequality.

What would settle it

Empirical data showing that Gini coefficients do not rise, or even fall, in economies where average wealth per capita grows faster relative to mean salaries would contradict the claimed dependence on R.

Figures

Figures reproduced from arXiv: 2601.15537 by Celia Anteneodo, Jhordan Silveira de Borba, Sebastian Gon\c{c}alves.

Figure 1
Figure 1. Figure 1: Complementary cumulative distributions of (a) wealth and (b) income for [PITH_FULL_IMAGE:figures/full_fig_p008_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Effects of expenditure frequency: Complementary cumulative (a) wealth and (b) [PITH_FULL_IMAGE:figures/full_fig_p009_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Impact of changes in (a) ΩE and (b) ΦE over labor share (LS), Gini of income, normalized HHI, and rescaled GDP. The maximum measured value of GDP is ≈ 6 × 104 for ΩE = 5 and ≈ 1.2 × 104 for ΦE = 1. 9 [PITH_FULL_IMAGE:figures/full_fig_p009_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Effects of expenditure fraction ΦE: Complementary cumulative distributions of (a) wealth and (b) income for different values of ΦE indicated in the legend. The inset shows the average annual Gini coefficient. We can gain further insight into the underlying dynamics by examining and comparing different macroeconomic indices represented in [PITH_FULL_IMAGE:figures/full_fig_p010_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Effects of market revenue frequency ΩM: Complementary cumulative (a) wealth and (b) income distributions for different values of ΩM. The inset shows the average annual Gini. When we analyze the effects of changing ΦM, that is, the fraction withdrawn from the market in the Market revenue rule, we observe in [PITH_FULL_IMAGE:figures/full_fig_p011_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Effects of market revenue fraction ΦM: Complementary cumulative (a) wealth and (b) income distributions for different values of ΦM. The inset shows the average annual Gini. That is, modifications in how money is withdrawn from the market value do not affect consumption, therefore keeping GDP constant. Since, in this model, GDP is given by the aggregate annual income of capitalists, this also implies that t… view at source ↗
Figure 7
Figure 7. Figure 7: Complementary cumulative (a) wealth and (b) income distributions for different [PITH_FULL_IMAGE:figures/full_fig_p015_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Relationship of Gini with R and the unemployment rate, calculated for different values of average monthly wages obtained from different values of α. (approximately 18%, from the color scale in the figure), displaying increasing inequality as R increases. This behavior is consistent with results originally obtained under constant wages and is also observed in empirical data. This behavior can be explained a… view at source ↗
read the original abstract

The impact of rising consumption on wealth inequality remains an open question. Here we revisit and extend the Social Architecture of Capitalism agent-based model proposed by Ian Wright, which reproduces stylized facts of wealth and income distributions. In a previous study, we demonstrated that the macroscopic behavior of the model is predominantly governed by a single dimensionless parameter, the ratio between average wealth per capita and mean salary, denoted by R. The shape of the wealth distribution, the emergence of a two-class structure, and the level of inequality - summarized by the Gini index - were found to depend mainly on R, with inequality increasing as R increases. In the present work, we examine the robustness of this result by relaxing some simplifying assumptions of the model. We first allow transactions such as purchases, salary payments, and revenue collections to occur with different frequencies, reflecting the heterogeneous temporal dynamics of real economies. We then impose limits on the maximum fractions of wealth that agents can spend or collect at each step, constraining the amplitude of individual transactions. We find that the dependence of the inequality on R remains qualitatively robust, although the detailed distribution patterns are affected by relative frequencies and transaction limits. Finally, we analyze a further variant of the model with adaptive wages emerging endogenously from the dynamics, showing that self-organized labor-market feedback can either stabilize or amplify inequality depending on macroeconomic conditions.

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

2 major / 2 minor

Summary. The manuscript extends the Social Architecture of Capitalism agent-based model to examine whether the previously reported dependence of wealth inequality (Gini index) on the dimensionless parameter R (average wealth per capita divided by mean salary) remains robust when heterogeneous transaction frequencies are allowed and when limits are imposed on the fractions of wealth that agents can spend or collect. The authors conclude that the qualitative increase of inequality with R persists across these variants, although detailed distribution patterns are affected; they also analyze an endogenous adaptive-wage variant in which labor-market feedback can stabilize or amplify inequality depending on macroeconomic conditions.

Significance. If the reported qualitative robustness holds under further scrutiny, the work would strengthen the argument that a single control parameter R largely governs macroscopic outcomes in this family of models, thereby linking rising consumption levels to deeper wealth inequality in a manner that is at least partially insensitive to certain temporal and amplitude assumptions. The adaptive-wage extension adds a falsifiable prediction about self-organized labor-market dynamics that could be tested against empirical data.

major comments (2)
  1. [Abstract] Abstract and robustness section: the assertion that the R-inequality relation is 'qualitatively robust' supplies no quantitative measures (e.g., Gini-R slopes, confidence intervals, or Kolmogorov-Smirnov distances between curves) across the frequency and limit variants, so the strength of the invariance cannot be verified from the data presented.
  2. [Results] Robustness tests: only relative frequencies of purchases/salaries/revenues and caps on spendable/collectible fractions are relaxed; fixed network topology, uniform saving propensities, and absence of firm-level or external-demand dynamics remain unchanged. If any of these introduce an additional dimensionless group whose magnitude varies with R, the reported qualitative invariance may be an artifact of the still-narrow model class rather than a general property.
minor comments (2)
  1. Figure captions should explicitly state which curves correspond to the original model versus each frequency/limit variant so that visual comparison is immediate.
  2. [Adaptive-wage variant] Clarify the precise definition of R when adaptive wages are introduced, since mean salary is no longer an exogenous input.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive and detailed comments, which help clarify the scope and presentation of our robustness analysis. We address each major point below and indicate the revisions made to the manuscript.

read point-by-point responses

  1. Referee: [Abstract] Abstract and robustness section: the assertion that the R-inequality relation is 'qualitatively robust' supplies no quantitative measures (e.g., Gini-R slopes, confidence intervals, or Kolmogorov-Smirnov distances between curves) across the frequency and limit variants, so the strength of the invariance cannot be verified from the data presented.

    Authors: We agree that quantitative support would strengthen the claim. In the revised manuscript we have added explicit slopes of the Gini-R relation for the baseline and each variant, together with Kolmogorov-Smirnov distances between the resulting wealth distributions. These metrics are reported in a new subsection of the robustness analysis and referenced in the abstract. revision: yes

  2. Referee: [Results] Robustness tests: only relative frequencies of purchases/salaries/revenues and caps on spendable/collectible fractions are relaxed; fixed network topology, uniform saving propensities, and absence of firm-level or external-demand dynamics remain unchanged. If any of these introduce an additional dimensionless group whose magnitude varies with R, the reported qualitative invariance may be an artifact of the still-narrow model class rather than a general property.

    Authors: We acknowledge that the tested variants leave other structural assumptions (network topology, uniform saving rates, absence of firm-level dynamics) unchanged and that these could in principle introduce R-dependent dimensionless groups. Our analysis is deliberately scoped to the temporal and amplitude relaxations most directly tied to consumption. We have added an explicit limitations paragraph in the conclusions that states the result holds within the explored model class and outlines how future work could incorporate heterogeneous networks or endogenous firm demand. revision: partial

Circularity Check

0 steps flagged

Minor self-citation to prior R-governance result; current robustness tests are independent simulations

full rationale

The paper defines R explicitly as average wealth per capita divided by mean salary and treats it as an input control parameter in the agent-based model. New simulations vary transaction frequencies and spending/collecting limits independently of R, then measure emergent Gini and distribution shapes. These extensions supply non-circular empirical checks within the model. The only self-citation is to the authors' prior demonstration that the base model is governed by R; that prior result is not invoked as a uniqueness theorem or to forbid alternatives, and the present work's central claim (qualitative invariance under the listed relaxations) rests on the new runs rather than reducing to the citation by construction.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The central claim rests on the base model reproducing stylized facts, the definition of R as the dominant parameter, and the assumption that the listed extensions are sufficient to test robustness.

free parameters (1)
  • R
    Dimensionless ratio of average wealth per capita to mean salary; varied to control inequality level.
axioms (1)
  • domain assumption The Social Architecture of Capitalism agent-based model reproduces stylized facts of wealth and income distributions.
    Invoked as the foundation being extended and tested for robustness.

pith-pipeline@v0.9.0 · 5778 in / 1182 out tokens · 50952 ms · 2026-05-22T12:03:51.573626+00:00 · methodology

discussion (0)

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

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

22 extracted references · 22 canonical work pages

  1. [1]

    A potential employerjis selected from the setHof all agents except employees (ej ̸= 0), with probabilityP(j) =w j/P n∈H wn

    work page

  2. [2]

    A wage offerwis drawn uniformly from the interval [η i,(1 +α)η i], whereη i rep- resents the agent’s wage expectation and the parameterαtherefore controls the intensity of the wage increase sought by the agent. This choice is inspired by Isaac’s analysis of the CSA model, in which he proposes replacing Wright’s origi- nal interval [ηi,2η i] with a fixed i...

    work page

  3. [3]

    If agentiis not hired and was originally unemployed, a new expectation is drawn uniformly from the interval [0, ηi]

    If the final wage offer exceeds the agent’s expectation (w > η i), agentiis hired by employerj(e i =j), and the expectation is updated toη i =w. If agentiis not hired and was originally unemployed, a new expectation is drawn uniformly from the interval [0, ηi]. •Wage payment and firing– If agentiis an employer (e i <0), we define the setE of employees ofi...

    work page

  4. [4]

    If employerihas sufficient funds, an amount equal to the employee’s expectation ηj is transferred from employer (w i →w i −η j) to employee (w j →w j +η j), and the employee’s last received wage is updated toµ j =η j

    work page

  5. [5]

    That is, if the agent is fired before receiving the negotiated wage, the expectation is not realized

    If employeridoes not have sufficient funds, employeejis fired, and their wage expectation is reset to the last received wage,η j =µ j. That is, if the agent is fired before receiving the negotiated wage, the expectation is not realized

    work page

  6. [6]

    If, after processing the entire setE, all employees have been fired, employeri becomes unemployed (e i = 0). 7 III. RESULTS FROM AGENT-BASED SIMULATIONS Observables were collected annually, over 1000 years, after a short transient. Wealth was calculated as the amountw i held by agentiat the end of the year, and incomey i as the aggregate wealth received b...

    work page 2025

  7. [7]

    Inequality in a model of capitalist economy.Physica A: Statistical Mechanics and its Applications, 664:130457, 2025

    Jhordan Silveira Borba, Sebastian Gon¸ calves, and Celia Anteneodo. Inequality in a model of capitalist economy.Physica A: Statistical Mechanics and its Applications, 664:130457, 2025

    work page 2025

  8. [8]

    The social architecture of capitalism.Physica A: Statistical Mechanics and its Applications, 346(3):589–620, 2005

    Ian Wright. The social architecture of capitalism.Physica A: Statistical Mechanics and its Applications, 346(3):589–620, 2005

    work page 2005

  9. [9]

    Yakovenko

    Adrian Dr˘ agulescu and Victor M. Yakovenko. Exponential and power-law probability dis- tributions of wealth and income in the United Kingdom and the United States.Physica A: Statistical Mechanics and its Applications, 299(1):213–221, 2001

    work page 2001

  10. [10]

    Statistical physics perspective on economic inequality

    Victor M Yakovenko. Statistical physics perspective on economic inequality. InRoutledge International Handbook of Complexity Economics, pages 279–291. Routledge, 2024

    work page 2024

  11. [11]

    Plata-P´ erez, J

    L. Plata-P´ erez, J. S´ anchez-P´ erez, and F. S´ anchez-S´ anchez. An elementary characterization of the Gini index.Mathematical Social Sciences, 74:79–83, 2015

    work page 2015

  12. [12]

    Data - WID - World Inequality Database.https://wid.world/data/, 2025

    WID. Data - WID - World Inequality Database.https://wid.world/data/, 2025. Accessed: 2025-06-04

    work page 2025

  13. [13]

    Implicit microfoundations for macroeconomics.Economics, 3(1):20090019, 2009

    Ian Wright. Implicit microfoundations for macroeconomics.Economics, 3(1):20090019, 2009

    work page 2009

  14. [14]

    Alan G. Isaac. Exploring the social-architecture model.Eastern Economic Journal, 45:565–589, 2019

    work page 2019

  15. [15]

    Medidas de concentra¸ c˜ ao industrial: uma resenha.An´ alise Econˆ omica, 12(21 & 22):24–33, 2009

    Marcelo Resende. Medidas de concentra¸ c˜ ao industrial: uma resenha.An´ alise Econˆ omica, 12(21 & 22):24–33, 2009

    work page 2009

  16. [16]

    Lin Lin.Agent-Based Models, Macroeconomic Scaling Laws and Sentiment Dynamics, Chap

    work page

  17. [17]

    PhD thesis, Universit¨ at zu Kiel, 2012

    work page 2012

  18. [18]

    Econophysics review : II

    Anirban Chakraborti, Ioane Toke, Marco Patriarca, and Fr´ ed´ eric Abergel. Econophysics review : II. Agent-based models.Quantitative Finance, 11:1013–1041, 2011

    work page 2011

  19. [19]

    Fokker–Planck description of wealth dynamics and the origin of Pareto’s law.International Journal of Modern Physics C, 25(12):1441008, 2014

    Bruce Boghosian. Fokker–Planck description of wealth dynamics and the origin of Pareto’s law.International Journal of Modern Physics C, 25(12):1441008, 2014

    work page 2014

  20. [20]

    Chakrabarti, Anirban Chakraborti, Satya R

    Bikas K. Chakrabarti, Anirban Chakraborti, Satya R. Chakravarty, and Arnab Chatterjee. Econophysics of Income and Wealth Distributions. Cambridge University Press, 2013

    work page 2013

  21. [21]

    Boghosian

    Bruce M. Boghosian. Kinetics of wealth and the Pareto law.Physical Review E, 89:042804, 2014. 18

    work page 2014

  22. [22]

    Piketty.Le capital au XXIe si` ecle

    T. Piketty.Le capital au XXIe si` ecle. Livres du nouveau monde. ´Editions du Seuil, 2013. 19

    work page 2013