pith. sign in

arxiv: 2604.13890 · v1 · submitted 2026-04-15 · ⚛️ physics.soc-ph · cs.LG· econ.EM· econ.TH· stat.ML

Sandpile Economics: Theory, Identification, and Evidence

Diego Vallarino This is my paper

Pith reviewed 2026-05-10 12:09 UTC · model grok-4.3

classification ⚛️ physics.soc-ph cs.LGecon.EMecon.THstat.ML
keywords sandpile economicsinput-output networkseconomic cascadespower-law distributionsnetwork curvaturemacroeconomic instabilityproduction networkscrisis amplification
0
0 comments X

The pith

Production networks evolve toward a critical state where small shocks generate large cascades with power-law tails.

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

The paper claims that recurrent macroeconomic crises arise because capitalist economies, through ongoing specialization and competitive selection, push their inter-sectoral linkages into geometrically fragile configurations. The central state variable is the curvature of the input-output graph, which tracks how readily sectors can substitute for missing inputs when chains break. Once curvature drops below an endogenous threshold, the sizes of resulting cascades obey a power law with tail index between 1 and 2, so that average amplification diverges and arbitrarily small shocks can produce arbitrarily large outcomes. Global input-output data show economies persistently occupy negative-curvature regimes, and lower curvature forecasts weaker three-year cumulative growth while outperforming conventional network statistics in predicting resilience. Because the process is driven by path-dependent reconfiguration after each crisis, it lies outside the reach of representative-agent equilibrium models.

Core claim

When the curvature of the input-output graph falls below a threshold set by evolutionary pressures, the distribution of cascade sizes follows a power law with tail index α in (1,2). This regime produces unbounded amplification because specialization has already reduced local input substitutability, leaving the network unable to absorb shocks locally. Crises then trigger endogenous rewiring of linkages, yet the system remains non-ergodic and repeatedly returns toward the same critical geometry.

What carries the argument

The Forman-Ricci curvature of the input-output graph, which measures local substitution possibilities among sectors when supply chains are disrupted.

If this is right

  • Cascade sizes acquire a power-law tail with exponent between 1 and 2, so the expected size of the largest events diverges.
  • A one-standard-deviation rise in curvature is associated with higher cumulative output growth over three-year horizons.
  • Curvature outperforms standard network metrics in accounting for cross-country differences in resilience to shocks.
  • Each crisis induces endogenous network reconfiguration that creates lasting path dependence rather than restoring equilibrium.

Where Pith is reading between the lines

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

  • Policy interventions that raise input substitutability could lift average curvature and thereby dampen amplification of future shocks.
  • The same curvature-driven mechanism may govern instability in other adaptive networks such as financial exposures or supply chains within single firms.
  • Standard equilibrium models that assume ergodicity will systematically understate the structural contribution of network geometry to instability.

Load-bearing premise

That curvature on the input-output graph accurately reflects how easily sectors can substitute for disrupted inputs and that evolutionary specialization reliably drives networks toward the low-curvature threshold.

What would settle it

Historical or simulated data in which economies with more negative curvature show neither heavier-tailed crisis sizes nor systematically weaker resilience and growth forecasts.

Figures

Figures reproduced from arXiv: 2604.13890 by Diego Vallarino.

Figure 1
Figure 1. Figure 1: Theoretical architecture of Sandpile Economics. Ricci curvature [PITH_FULL_IMAGE:figures/full_fig_p014_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Log-log CCDF of absolute output contractions by curvature quartile. Dashed blue: most [PITH_FULL_IMAGE:figures/full_fig_p021_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Distribution of cascade episode sizes (log-binned frequency vs. log size). A cascade episode [PITH_FULL_IMAGE:figures/full_fig_p023_3.png] view at source ↗
read the original abstract

Why do capitalist economies recurrently generate crises whose severity is disproportionate to the size of the triggering shock? This paper proposes a structural answer grounded in the evolutionary geometry of production networks. As economies evolve through specialization, integration, and competitive selection, their inter-sectoral linkages drift toward configurations of increasing geometric fragility, eventually crossing a threshold beyond which small disturbances generate disproportionately large cascades. We introduce Sandpile Economics, a formal framework that interprets macroeconomic instability as an emergent property of disequilibrium production networks. The key state variable is the Forman--Ricci curvature of the input--output graph, capturing local substitution possibilities when supply chains are disrupted. We show that when curvature falls below an endogenous threshold, the distribution of cascade sizes follows a power law with tail index $\alpha \in (1,2)$, implying a regime of unbounded amplification. The underlying mechanism is evolutionary: specialization reduces input substitutability, pushing the economy toward criticality, while crisis episodes induce endogenous network reconfiguration and path dependence. These dynamics are inherently non-ergodic and cannot be captured by representative-agent frameworks. Empirically, using global input--output data, we document that production networks operate in persistently negative curvature regimes and that curvature robustly predicts medium-run output dynamics. A one-standard-deviation increase in curvature is associated with higher cumulative growth over three-year horizons, and curvature systematically outperforms standard network metrics in explaining cross-country differences in resilience.

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 paper introduces 'Sandpile Economics,' a framework that interprets macroeconomic instability as an emergent property of production networks' geometric fragility. The key variable is the Forman-Ricci curvature of the input-output graph, which is claimed to capture local substitution possibilities. The central theoretical result is that when curvature falls below an endogenous threshold, cascade sizes follow a power law with tail index α ∈ (1,2), implying unbounded amplification. This is attributed to evolutionary specialization reducing substitutability, with crises inducing network reconfiguration. Empirically, global IO data are used to show persistently negative curvature regimes and that curvature predicts medium-run output growth and cross-country resilience better than standard network metrics.

Significance. If the curvature-to-substitutability mapping and the power-law derivation are established rigorously, the framework could provide a structural, non-ergodic explanation for why small shocks produce large crises, offering falsifiable predictions that contrast with representative-agent models. The empirical claim that curvature outperforms standard metrics in growth regressions would be a notable contribution to network-based macroeconomics if robust. The evolutionary mechanism and emphasis on path dependence add conceptual value, though the overall significance hinges on resolving the geometric-to-economic translation.

major comments (2)
  1. [Theory / mechanism section] Theory section on the sandpile dynamics: The claim that Forman-Ricci curvature directly encodes local substitution possibilities (and thus controls the toppling threshold) lacks an explicit reduction. The combinatorial formula (e.g., Ric(uv) = 4 − deg(u) − deg(v) + 3 × triangles, or its weighted form) does not incorporate input coefficients, supplier shares, or elasticities; high-degree nodes can still exhibit low substitutability if one supplier dominates. This mapping is load-bearing for the power-law result with α ∈ (1,2) and must be derived or simulated explicitly rather than asserted.
  2. [Empirical identification / results section] Empirical section on growth predictions: The reported association (one-standard-deviation increase in curvature linked to higher three-year cumulative growth, outperforming standard metrics) is presented without the regression specification, controls, fixed effects, sample details, or robustness checks. Without these, it is impossible to assess whether the correlation survives alternative network measures or endogeneity concerns, undermining the claim that curvature is a superior predictor of resilience.
minor comments (2)
  1. [Abstract] Abstract and introduction: The phrase 'unbounded amplification' should be tied more precisely to the implication of α < 2 (infinite variance) to avoid ambiguity.
  2. [Model setup] Notation: Define the endogenous curvature threshold explicitly (e.g., how it is computed from the network or calibrated) rather than leaving it as a model-internal quantity.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their thoughtful comments on our manuscript. We believe the suggested revisions will strengthen the paper and address the concerns raised. Below we respond to each major comment.

read point-by-point responses

  1. Referee: [Theory / mechanism section] Theory section on the sandpile dynamics: The claim that Forman-Ricci curvature directly encodes local substitution possibilities (and thus controls the toppling threshold) lacks an explicit reduction. The combinatorial formula (e.g., Ric(uv) = 4 − deg(u) − deg(v) + 3 × triangles, or its weighted form) does not incorporate input coefficients, supplier shares, or elasticities; high-degree nodes can still exhibit low substitutability if one supplier dominates. This mapping is load-bearing for the power-law result with α ∈ (1,2) and must be derived or simulated explicitly rather than asserted.

    Authors: We appreciate the referee pointing out the need for a clearer derivation of the curvature-substitutability link. In the current manuscript, the weighted Forman-Ricci curvature is used, where edge weights are the input-output coefficients normalized by total inputs, which directly incorporates supplier shares. The curvature value decreases when weights are concentrated on few edges, reflecting low substitutability. However, we acknowledge that an explicit mapping from CES elasticities to the toppling threshold is not derived in detail. We will revise the theory section to include a derivation showing how the curvature threshold emerges from a linearized production function with limited substitution, and add simulation results on toy networks to illustrate the emergence of the power-law with α in (1,2). This addresses the load-bearing aspect. revision: yes

  2. Referee: [Empirical identification / results section] Empirical section on growth predictions: The reported association (one-standard-deviation increase in curvature linked to higher three-year cumulative growth, outperforming standard metrics) is presented without the regression specification, controls, fixed effects, sample details, or robustness checks. Without these, it is impossible to assess whether the correlation survives alternative network measures or endogeneity concerns, undermining the claim that curvature is a superior predictor of resilience.

    Authors: The referee correctly notes that the main text summarizes the empirical findings without full econometric details. The full specifications, including controls for standard network metrics (degree, betweenness, clustering), country and year fixed effects, and the sample of global IO tables from WIOD and OECD, are provided in the appendix. To improve accessibility, we will move the key regression table and specification to the main text, add robustness checks using alternative curvature definitions and lagged instruments to address endogeneity, and explicitly compare against other metrics in the reported results. This will allow readers to evaluate the claim directly. revision: yes

Circularity Check

0 steps flagged

No significant circularity detected in derivation chain

full rationale

The paper's central theoretical claim—that Forman-Ricci curvature below an endogenous threshold yields power-law cascade sizes with α ∈ (1,2)—is presented as a derived property of the sandpile dynamics on the input-output graph. The abstract and description frame this as an emergent result from evolutionary specialization and network reconfiguration, not as a direct renaming or redefinition of the inputs. No quoted equations reduce the power-law tail or threshold to a fitted parameter or self-citation by construction. The empirical section uses global IO data to test curvature's predictive power for growth and resilience, which is falsifiable against external benchmarks and does not rely on self-referential fitting. Self-citations are absent (single author, no load-bearing prior theorems invoked). The derivation remains self-contained against the stated assumptions, with the curvature-to-substitutability mapping treated as a modeling choice rather than a tautology.

Axiom & Free-Parameter Ledger

2 free parameters · 2 axioms · 1 invented entities

The framework rests on interpreting curvature as a direct proxy for substitutability and on an evolutionary drift toward criticality; both are introduced as modeling choices rather than derived from independent evidence. The power-law tail and endogenous threshold are stated without shown derivation or external calibration.

free parameters (2)
  • curvature threshold
    Endogenous value below which power-law cascades with α ∈ (1,2) occur; no external calibration shown.
  • tail index α = in (1,2)
    Stated to lie in (1,2) for the unbounded-amplification regime; likely obtained from model or data.
axioms (2)
  • domain assumption Forman-Ricci curvature of the input-output graph measures local substitution possibilities under supply-chain disruption
    Invoked to interpret the key state variable.
  • domain assumption Evolutionary specialization, integration, and selection drive production networks toward increasing geometric fragility
    Explains why economies recurrently reach the critical regime.
invented entities (1)
  • Sandpile Economics framework no independent evidence
    purpose: Interpret macroeconomic instability as emergent property of disequilibrium production networks
    New interpretive lens proposed to link network geometry to crisis cascades.

pith-pipeline@v0.9.0 · 5556 in / 1489 out tokens · 72929 ms · 2026-05-10T12:09:31.392900+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

4 extracted references · 4 canonical work pages

  1. [1]

    M., Ozdaglar, A., and Tahbaz-Salehi, A

    Acemoglu, D., Carvalho, V. M., Ozdaglar, A., and Tahbaz-Salehi, A. (2012). The network origins of aggregate fluctuations.Econometrica, 80(5):1977–2016. Allen, F. and Gale, D. (2000). Financial contagion.Journal of Political Economy, 108(1):1–33. Andrews, D. W. K. (1991). Heteroskedasticity and autocorrelation consistent covariance matrix estimation.Econom...

    work page 2012

  2. [2]

    and Saxena, S

    Cerra, V. and Saxena, S. C. (2008). Growth dynamics: The myth of economic recovery.American Economic Review, 98(1):439–457. Clauset, A., Shalizi, C. R., and Newman, M. E. J. (2009). Power-law distributions in empirical data. SIAM Review, 51(4):661–703. Davidson, J. (1994).Stochastic Limit Theory. Oxford University Press, Oxford. Di Giovanni, J., Levchenko...

    work page 2008

  3. [3]

    Dosi, G. (1988). Sources, procedures, and microeconomic effects of innovation.Journal of Economic Literature, 26(3):1120–1171. Dosi, G., Fagiolo, G., and Roventini, A. (2010). Schumpeter meeting Keynes: A policy-friendly model of endogenous growth and business cycles.Journal of Economic Dynamics and Control, 34(9):1748–1767. Driscoll, J. C. and Kraay, A. ...

    work page 1988

  4. [4]

    40 Nelson, R. R. and Winter, S. G. (1982).An Evolutionary Theory of Economic Change. Harvard University Press, Cambridge, MA. Newey, W. K. and West, K. D. (1987). A simple, positive semi-definite, heteroskedasticity and autocorrelation consistent covariance matrix.Econometrica, 55(3):703–708. Ni, C.-C., Lin, Y.-Y., Luo, F., and Gao, J. (2019). Community d...

    work page 1982