{"paper_id":"2506.17720","title":"Wealth Thermalization Hypothesis and Social Networks","authors":["Klaus M. Frahm","Dima L. Shepelyansky"],"paper_date":"2025-06-21","paper_type":"methodological","paper_url":"https://arxiv.org/pdf/2506.17720v3","page_url":"https://thequant.space/flowcharts/wealth-thermalization-hypothesis-and-social-networks/","scores":{"empirical_rigor":6,"math_complexity":8.5,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes the Wealth Thermalization Hypothesis, linking societal wealth to energy in a Rayleigh‑Jeans distribution and introducing nonlinear random‑matrix‑based models (RJS and RJE) that exhibit dynamical thermalization above a chaos border.","quote":"In this work we study more generic models based on Random Matrix Theory and social networks with a nonlinear perturbation leading to dynamical thermalization above a certain chaos border. We introduce the wealth thermalization hypothesis according to which the society wealth is associated with energy in the Rayleigh-Jeans distribution.","summary":"In 1955 Fermi, Pasta, Ulam and Tsingou performed first numerical studies with the aim to obtain the thermalization in a chain of nonlinear oscillators from dynamical equations of motion. This model happend to have several specific features and the dynamical thermalization was established only later…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"The systems have two conserved quantities: total energy and norm, so the Rayleigh‑Jeans thermal distribution depends on temperature and chemical potential.","type":"model","stated":"explicit","quote":"These systems have two integrals of motion being total energy and norm so that the theoretical Rayleigh-Jeans thermal distribution depends on temperature and chemical potential.","support":"full"},{"assumption":"In the limit a → 0 the RJE model reduces to the RJS model; increasing a introduces exponential growth of Em.","type":"model","stated":"explicit","quote":"In the limit a →0, we recover simply the RJS model while with increasing values of a the exponential growth of Em becomes more dominant.","support":"full"},{"assumption":"The parameter a is determined by fitting a reconstructed spectrum to the Lorenz curve of empirical data.","type":"statistical","stated":"explicit","quote":"To determine optimal values for the parameter a, we compute a reconstructed spectrum from a given Lorenz curve of some given data set (see Appendix Section 5 for a description and more detailed discussion of this spectral reconstruction with Figure A.13) and fit the reconstructed spectrum to the function Em ≈C(ea (m/N) − 1)/a with two parameters C and a.","support":"full"},{"assumption":"The Lorenz curves are compared using a fixed number of modes N = 10000 (or the actual data size) which does not affect the shape at graphical precision.","type":"data","stated":"explicit","quote":"Here, we choose for simplicity the value of N = 10000 for the curves of both RJE and RJS models (the RJS curves are also shown for comparison). Other values such as N = 1000 or the given size of the data set, give the same Lorenz curves at graphical precision.","support":"full"},{"assumption":"The rescaled energy ε is obtained by matching the model Gini coefficient to the empirical Gini coefficient.","type":"statistical","stated":"explicit","quote":"The rescaled energy values ε = 0.01233 (RJE) and ε = 0.07420 (RJS) are obtained by matching the Gini coefficient G = 0.8515. The rescaled energy values ε = 0.1332 (RJE) and ε = 0.1996 (RJS) are obtained by matching the Gini coefficient G = 0.6255.","support":"full"}]},"datasets":{"status":"available","items":[{"name":"Lorenz curves of US wealth (2019)","provider":null,"kind":"other","access":"unknown","period":"2019","universe":"US households","quote":"We show that this thermalization gives a good description of real data of Lorenz curves of US, UK, the whole world and capitalization of companies at Stock Exchange of New York SE (NYSE), London and Hong Kong. N = 10000) and the RJE model with a = 4.42 (blue curve; N = 10000).","support":"partial"},{"name":"Lorenz curves of UK wealth (2012‑2014)","provider":null,"kind":"other","access":"unknown","period":"2012‑2014","universe":"UK households","quote":"w h RJS RJE, a = 2.18 UK 2012-2014 0.4 0.6 0.8 1 0.8 0.85 0.9 0.95 1 w h RJS RJE, a = 2.18 UK 2012-2014 Fig. I.8 Comparison of the Lorenz curve for the data of UK 2012-2014 (black) with the correspond-ing curves for the RJS model (red curve; N = 10000) and the RJE model with a = 2.18 (blue curve; N = 10000). The rescaled energy values ε = 0.1332 (RJE) and ε = 0.1996 (RJS) are obtained by matching the Gini","support":"full"},{"name":"Lorenz curves of World wealth (2021)","provider":null,"kind":"other","access":"unknown","period":"2021","universe":"World households","quote":"0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 w h RJS RJE, a = 5.34 World 2021 0.2 0.4 0.6 0.8 1 0.8 0.85 0.9 0.95 1 w h RJS RJE, a = 5.34 World 2021 Fig. I.9 Comparison of the Lorenz curve for the data of World 2021 from [33] (black) with the corresponding curves for the RJS model (red curve; N = 10000) and the RJE model (I.2) with a = 5.34 (blue curve; N = 10000). The rescaled energy values ε = 0.008553 (RJE) and ε =","support":"full"},{"name":"Lorenz curve of NYSE company capitalization (June 16, 2025)","provider":"S&P500 data (Ref. [57])","kind":"fundamentals","access":"public","period":"June 16, 2025","universe":"504 NYSE companies","quote":"A.4 Data for companies of stock exchange at New York, London, Hong Kong We present here the Lorenz curves for the capitalization of companies at stock exchanges of New York, London, Hong Kong. First, we present in Fig. A.9 the Lorenz curve for the data of 504 S&P500 companies of the New York Stock Exchange (NYSE) of June 16, 2025 (see Ref. [57]).","support":"full"},{"name":"Lorenz curve of London Stock Exchange FTSE (31 Dec 2024)","provider":null,"kind":"other","access":"unknown","period":"2024‑12‑31","universe":"FTSE constituents","quote":"0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 w h RJS RJE, a = 3.13 FTSE 2024 0.2 0.4 0.6 0.8 1 0.8 0.85 0.9 0.95 1 w h RJS RJE, a = 3.13 FTSE 2024 Fig. I.10 Comparison of the Lorenz curve for the data of the London stock exchange FTSE at 31 December 2024 (black; data from Ref. [58]) with the corresponding curves for the RJS model (red curve; N = 10000) and the RJE model with a = 3.13 (blue curve; N = 10000). The rescaled","support":"full"},{"name":"Lorenz curve of Hong Kong Stock Exchange (19 Jun 2025)","provider":null,"kind":"other","access":"unknown","period":"2025‑06‑19","universe":"HK listed companies","quote":"This is because the construction procedure of Lorenz curve 13 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 w h RJS RJE, a = 6.83 HK 2025 0 0.2 0.4 0.6 0.8 1 0.8 0.85 0.9 0.95 1 w h RJS RJE, a = 6.83 HK 2025 Fig. I.11 Comparison of the Lorenz curve for the data of the Hong Kong stock exchange at 19 June 2025 (black; data from Ref. [59]) with the corresponding curves for the RJS model (red curve; N = 10000) and the RJE","support":"full"}],"known":[],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"unclear"},"transaction_costs":{"status":"not_applicable","modeled":"n/a","headline_net_of_costs":null,"detail":null,"quote":null,"source":"front_matter","rule":"R3","candidates":{"front_matter":"n/a","evidence_card":"n/a","evidence_card_quoted":false},"meaning":"the paper reports no trading or execution result"},"results":{"status":"available","evaluation_type":"unclear","findings":[{"finding":"The Rayleigh‑Jeans thermalization model provides a good description of empirical Lorenz curves for US, UK, world wealth and stock‑exchange capitalizations.","kind":"theoretical","quote":"We show that this thermalization gives a good description of real data of Lorenz curves of US, UK, the whole world and capitalization of companies at Stock Exchange of New York SE (NYSE), London and Hong Kong.","support":"full"},{"finding":"Across all cases the RJE model fits the data significantly better than the RJS model.","kind":"statistical","quote":"In all cases, the agreement of the RJE model with the data is significantly better than the RJS model.","support":"full"},{"finding":"For the Hong Kong 2025 data the RJE model matches the empirical Lorenz curve almost perfectly, even in the zoomed high‑wealth region.","kind":"performance","quote":"In particular for HK 2025, the agreement is close to perfect and even in the zoomed panel it is difficult to distinguish the theoretical RJE curve (blue) from the data (black).","support":"full"}],"reported_performance":null,"out_of_sample":{"method":"none","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":33,"citations":{"cited_by_in_corpus":0,"by_independent_authors":0,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":0,"unclassified":0},"independent_papers":[],"scope":"Citations between papers in this corpus only; citations from outside it are not counted."}},"implementation":{"status":"available","public":false,"code_url":null,"note":"No repository presented by the authors as their own was found in the paper. Absence here is not proof that no code exists."},"replication":{"status":"none_found","card":null,"note":"No replication of this paper has been run by thequant.space yet."},"extraction":{"text_source":"fulltext","model":"openai/gpt-oss-120b","version":2,"extracted_at":"2026-11-14T19:49:41Z"},"provenance":{"claims":"Everything here is what the authors state in the paper. Nothing was re-run or reproduced by the platform.","quotes":"Assumptions, datasets, cost detail and findings were extracted by a language model. Every quote is verbatim paper text. Where an item has `support`, \"full\" means every figure in it appears in the quoted sentences and \"partial\" means a figure is stated elsewhere in the paper; items with a figure found nowhere in the paper were discarded.","datasets_known":"datasets.known lists well-known vendors and public datasets by a stable id, found by a dictionary over the full text where the sentence says the data is used; it is independent of the model and covers only names in the dictionary.","flags":"transaction_costs.modeled, results.evaluation_type and results.out_of_sample.method are platform classifications (structured-decision models); see https://thequant.space/score-guide/#penalty-flags","comparisons":"supports / contradicts / qualifies are a language model's reading of two papers' quoted findings, restricted to closely related papers that share no author. They are pointers to read, not verified replications, and coverage is partial.","implementation":"code_url is a repository the paper itself presents as its own (rule-scored, ambiguous cases judged by a model), checked for liveness when it was recorded."},"docs":"https://thequant.space/agents/"}