pith. sign in

arxiv: 2509.07793 · v4 · pith:EYJ32RMLnew · submitted 2025-09-09 · 💰 econ.GN · cs.AI· cs.CY· q-fin.EC

Individual utilities of life satisfaction reveal inequality aversion unrelated to political alignment

Crispin Cooper , Ana Fredrich , Tommaso Reggiani , Wouter Poortinga This is my paper

Pith reviewed 2026-05-21 23:09 UTC · model grok-4.3

classification 💰 econ.GN cs.AIcs.CYq-fin.EC
keywords life satisfactionutility functionsinequality aversionrisk aversionpolitical alignmentwell-being measurementexpected utilitystated preferences
0
0 comments X

The pith

Most people display concave utility curves for life satisfaction and stronger aversion to inequality across society than to personal risk, with no link to political alignment.

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

The paper runs a stated-preference experiment on a representative UK sample in which participants choose between uncertain life-satisfaction outcomes for themselves and for society. Individual utility functions are recovered under an expected-utility framework and compared with cumulative prospect theory. A majority of respondents show concave curves indicating risk aversion, yet their revealed aversion to unequal societal distributions exceeds their aversion to personal risk. These patterns appear independent of left-right political self-placement. The results suggest that average life satisfaction is an inadequate policy metric and that nonlinear, inequality-sensitive measures would better match observed collective preferences.

Core claim

Using choices over hypothetical life-satisfaction lotteries for self and others, the authors recover individual utility functions that are predominantly concave; the implied inequality aversion parameter for societal outcomes is larger than the risk-aversion parameter for personal outcomes, and neither parameter correlates with political orientation.

What carries the argument

Individual-level utility functions estimated from stated choices under uncertainty, separating personal risk aversion from societal inequality aversion.

If this is right

  • Policy evaluation should replace simple averages of life satisfaction with nonlinear utility aggregates that penalize inequality.
  • Well-being metrics can be designed to command support from across the political spectrum.
  • Value-aligned AI reward functions can incorporate the observed preference for reducing dispersion in societal well-being.
  • Redistribution or levelling policies focused on life-satisfaction outcomes may enjoy broader legitimacy than those framed purely in income terms.

Where Pith is reading between the lines

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

  • The same experimental design could be run in other countries to test whether the independence from political alignment is culturally specific.
  • If the inequality-aversion finding holds, cost-benefit analyses of health or education programs should weight improvements for the lowest-satisfaction groups more heavily.
  • Designers of collective decision algorithms could embed the revealed societal utility function as a default objective.

Load-bearing premise

Participants' answers to the hypothetical scenarios accurately reflect their stable underlying preferences rather than being shaped by the way the questions are worded or by social-desirability pressures.

What would settle it

A replication in which the same participants make incentivized choices with real money or real policy consequences and the inequality-aversion coefficient is no longer larger than the risk-aversion coefficient, or becomes correlated with political views.

read the original abstract

How should well-being be prioritised in society, and what trade-offs are people willing to make between fairness and personal well-being? We investigate these questions using a stated preference experiment with a nationally representative UK sample (n = 300), in which participants evaluated life satisfaction outcomes for both themselves and others under conditions of uncertainty. Individual-level utility functions were estimated using an Expected Utility Maximisation (EUM) framework and tested for sensitivity to the overweighting of small probabilities, as characterised by Cumulative Prospect Theory (CPT). A majority of participants displayed concave (risk-averse) utility curves and showed stronger aversion to inequality in societal life satisfaction outcomes than to personal risk. These preferences were unrelated to political alignment, suggesting a shared normative stance on fairness in well-being that cuts across ideological boundaries. The results challenge use of average life satisfaction as a policy metric, and support the development of nonlinear utility-based alternatives that more accurately reflect collective human values. Implications for public policy, well-being measurement, and the design of value-aligned AI systems are discussed.

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 / 1 minor

Summary. The manuscript reports results from a stated-preference experiment with a nationally representative UK sample (n=300). Participants evaluated hypothetical life-satisfaction outcomes for themselves and for society under uncertainty. Individual-level utility functions are recovered via an Expected Utility Maximization (EUM) framework and subjected to robustness checks under Cumulative Prospect Theory (CPT). The central claims are that a majority of participants exhibit concave (risk-averse) utility curves, display stronger aversion to inequality in societal outcomes than to personal risk, and that these preferences are uncorrelated with political alignment.

Significance. If the recovered parameters accurately capture stable preferences, the results would indicate a widely shared normative preference for inequality aversion in well-being distributions that cuts across ideological lines. This would provide empirical support for moving beyond linear or mean-based well-being metrics in policy and measurement, with potential implications for value-aligned AI design. The individual-level estimation and CPT sensitivity analysis are positive features that address heterogeneity.

major comments (2)
  1. [Methods] Methods section: the manuscript provides insufficient detail on the exact functional form adopted for the individual utility function u(·) in the EUM estimation, the rules used to exclude participants or choice observations, and the treatment of inconsistent or noisy responses. These choices are load-bearing for the headline result that a majority display concave utilities and for the inequality-aversion versus risk-aversion comparison.
  2. [Results] Results section: the claim of stronger societal inequality aversion than personal risk aversion requires explicit comparison of the estimated parameters across the two classes of scenarios. Without the relevant equations, tables, or derivation showing how the inequality metric is constructed from the societal lotteries (and how it differs from the personal-risk parameter), it is not possible to evaluate whether the reported difference is robust to framing or probability presentation.
minor comments (1)
  1. [Abstract] Abstract: the phrase 'tested for sensitivity to the overweighting of small probabilities' should be accompanied by the specific CPT weighting function and parameter values used in the robustness check.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their careful reading and constructive comments. We have revised the manuscript to provide the requested details on functional forms, exclusion rules, and parameter comparisons. Our responses to the major comments are as follows.

read point-by-point responses

  1. Referee: [Methods] Methods section: the manuscript provides insufficient detail on the exact functional form adopted for the individual utility function u(·) in the EUM estimation, the rules used to exclude participants or choice observations, and the treatment of inconsistent or noisy responses. These choices are load-bearing for the headline result that a majority display concave utilities and for the inequality-aversion versus risk-aversion comparison.

    Authors: We agree that additional detail improves transparency. In the revised Methods section we now specify that the individual utility function is the constant relative risk aversion form u(s) = s^(1-ρ)/(1-ρ) for ρ ≠ 1 (logarithmic for ρ = 1), estimated separately for each participant via maximum likelihood on the stated-preference choices. Exclusion criteria are now listed explicitly: participants were removed if they failed any of the three attention checks or violated first-order stochastic dominance in more than 25 % of trials (resulting in the exclusion of 14 respondents). Noisy or inconsistent responses were retained in the main analysis but subjected to two robustness checks: (i) a trimmed sample excluding the 5 % most inconsistent responders and (ii) a Bayesian hierarchical model that down-weights outliers. These clarifications directly underpin the reported share of concave utilities and the subsequent parameter comparisons. revision: yes

  2. Referee: [Results] Results section: the claim of stronger societal inequality aversion than personal risk aversion requires explicit comparison of the estimated parameters across the two classes of scenarios. Without the relevant equations, tables, or derivation showing how the inequality metric is constructed from the societal lotteries (and how it differs from the personal-risk parameter), it is not possible to evaluate whether the reported difference is robust to framing or probability presentation.

    Authors: We accept that an explicit derivation and side-by-side comparison are necessary. The revised Results section now contains two new equations: (1) the personal-risk parameter ρ_p estimated from individual lotteries over own life satisfaction, and (2) the societal inequality-aversion parameter ρ_s recovered from choices over mean-preserving spreads in the societal distribution of life satisfaction. Inequality aversion is isolated by comparing expected utility under the actual distribution versus a distribution with identical mean but lower variance. A new Table 3 reports means, standard deviations, and a paired t-test (ρ_s − ρ_p = 0.31, p < 0.01). We also add a CPT robustness panel that re-estimates both parameters under probability weighting; the inequality-aversion gap remains statistically significant. These additions demonstrate that the difference is not an artifact of framing or probability presentation. revision: yes

Circularity Check

0 steps flagged

No significant circularity in experimental utility estimation

full rationale

This is an empirical paper that collects primary data from a nationally representative sample using a stated preference experiment. Individual-level utility functions are estimated using the Expected Utility Maximisation framework applied to participants' evaluations of life satisfaction outcomes. The reported majority concave utilities and stronger inequality aversion are direct outputs of this estimation process on the collected choices, not equivalent to inputs by construction. No self-citation chains or fitted parameters renamed as predictions are evident in the derivation of the main results. The analysis is self-contained against external benchmarks of experimental data.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The central claim rests on the applicability of the Expected Utility Maximisation framework to stated life satisfaction choices and on the representativeness of the UK sample for broader normative conclusions.

free parameters (1)
  • individual utility function parameters
    Estimated per participant from choice data under the EUM model
axioms (1)
  • domain assumption Expected Utility Maximisation framework accurately describes choices over life satisfaction outcomes
    Invoked to estimate individual utilities from the stated preference experiment

pith-pipeline@v0.9.0 · 5724 in / 1369 out tokens · 35056 ms · 2026-05-21T23:09:13.471174+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

8 extracted references · 8 canonical work pages

  1. [1]

    A New Explanation for the Difference between Time Trade-off Utilities and Standard Gamble Utilities

    https://doi.org/10.1016/j.jhealeco.2013.08.006. Bleichrodt, Han. 2002. “A New Explanation for the Difference between Time Trade-off Utilities and Standard Gamble Utilities.” Health Economics 11 (5): 447–56. https://doi.org/10.1002/hec.688. Bonem, Emily M., Phoebe C. Ellsworth, and Richard Gonzalez. 2015. “Age Differences in Risk: Perceptions, Intentions a...

    work page doi:10.1016/j.jhealeco.2013.08.006 2013

  2. [2]

    Intertemporal Differences Among MTurk Workers: Time-Based Sample Variations and Implications for Online Data Collection

    “Intertemporal Differences Among MTurk Workers: Time-Based Sample Variations and Implications for Online Data Collection.” SAGE Open 7 (2): 2158244017712774. https://doi.org/10.1177/2158244017712774. Cato, Molly Scott. 2008. Green Economics: An Introduction to Theory, Policy and Practice. 1st edition. London, UK: Earthscan. Chmielewski, Michael, and Sarah...

    work page doi:10.1177/2158244017712774 2008

  3. [3]

    The Nature of Belief Systems in Mass Publics (1964)

    https://doi.org/10.1177/1948550619875149. Converse, Philip E. 2006. “The Nature of Belief Systems in Mass Publics (1964).” Critical Review 18 (1–3): 1–74. https://doi.org/10.1080/08913810608443650. Cooper, Crispin H. V. 2020. “Quantitative Models of Well-Being to Inform Policy: Problems and Opportunities.” Sustainability 12 (8): 3180. https://doi.org/10.3...

    work page doi:10.1177/1948550619875149 2006

  4. [4]

    Principled Artificial Intelligence: Mapping Consensus in Ethical and Rights-Based Approaches to Principles for AI

    https://doi.org/10.24963/ijcai.2018/768. Fjeld, Jessica, Nele Achten, Hannah Hilligoss, Adam Nagy, and Madhulika Srikumar. 2020. “Principled Artificial Intelligence: Mapping Consensus in Ethical and Rights-Based Approaches to Principles for AI.” SSRN Electronic Journal. https://doi.org/10.2139/ssrn.3518482. Fleurbaey, Marc. 2011. “Willingness-to-pay and t...

    work page doi:10.24963/ijcai.2018/768 2018

  5. [5]

    Imagine you are in a situation where you rate your life satisfaction as 8 out of 10

    Personal risk gamble, life satisfaction basis Now, consider the following scenario. Imagine you are in a situation where you rate your life satisfaction as 8 out of 10. Imagine that all your life, you have had a chronic health condition which you were born with, and which restricts your life somewhat. One day your doctor says you must choose between two t...

    work page

  6. [6]

    The context of this scenario is different - please read carefully! Imagine you are a policymaker who must make a choice affecting a large number of people

    Policymaker scenario – life satisfaction basis Now, consider the following new scenario. The context of this scenario is different - please read carefully! Imagine you are a policymaker who must make a choice affecting a large number of people. Currently, everyone in the affected group rates their life satisfaction as 2. You have the following options: • ...

    work page

  7. [7]

    The pay is low, and sometimes I struggle to pay the bills

    Personal risk scenario – vignette basis Now, consider this scenario, based on the following life situations: Situation Career Relationships Physical Fitness D I don't enjoy my job. The pay is low, and sometimes I struggle to pay the bills. I only occasionally visit my family and friends, and I often feel lonely. I am unhappy about my relationship status. ...

    work page

  8. [8]

    I am currently taking an online survey

    Political alignment The political questions are as follows. “I am currently taking an online survey” is an attention check complying with the Prolific guidelines

    work page