Economic Inequality and Mental Health: A Null Effect?

Critical Analysis of the First Meta-Analysis Published in Nature (Sommet et al., 2025)

Abstract

The relationship between economic inequality and mental health has long been considered established fact in the psychological literature and in public policy debate. A landmark meta-analysis published in Nature (Sommet et al., 2025) challenges this assumption, finding that the average effect of economic inequality on subjective well-being and mental health is statistically null across 168 studies and over 11 million participants. Moderator analyses indicate, however, that the negative effect does emerge under specific contextual conditions — particularly in the presence of absolute poverty and high inflation — suggesting that poverty, not relative inequality, is the primary determinant of psychological distress. This article examines the methodology, findings, and clinical and policy implications of this study, situating it within the theoretical debate on the distinction between absolute poverty and relative inequality.

Keywords: economic inequality · mental health · subjective well-being · poverty · meta-analysis · social determinants of health

1. Introduction

The question seems straightforward: does living in an economically unequal society make people feel worse? For decades, the answer from the scientific literature — and adopted by public discourse — has been yes. Influential works such as The Spirit Level (Wilkinson & Pickett, 2009) systematised this hypothesis, arguing that income inequality erodes social capital, intensifies social comparison processes, and produces widespread negative effects on the physical and mental health of populations.

This narrative became so entrenched that, in many clinical and academic settings, it functioned almost as an unquestioned postulate. Its practical implications have been significant: it has shaped redistribution policies, informed explanatory models of psychological distress, and influenced the training of mental health professionals.

In November 2025, the publication in Nature of a meta-analysis by Sommet et al. (2025) challenged this consensus with evidence of unprecedented scale. It is the first social science meta-analysis ever published in Nature, based on 168 studies covering over 11 million participants across 38,335 geographical units worldwide (Sommet et al., 2025). Its findings are as clear as they are counterintuitive: the average effect of economic inequality on well-being and mental health is statistically null.

This article offers a critical analysis of this study: it examines the methodological structure, discusses the main findings and moderators, situates the implications within the theoretical debate on absolute poverty versus relative inequality, and reflects on the consequences for clinical practice and public health policy.

2. Theoretical Background: The Inequality Hypothesis and Its Limits

2.1 The Income Inequality Hypothesis

The Income Inequality Hypothesis (IIH) holds that it is not an individual's absolute income that determines health outcomes, but rather their relative position within the income distribution of a given society (Wilkinson & Pickett, 2009; Ribeiro et al., 2017). According to this view, more unequal societies generate steeper status hierarchies, intensifying upward social comparison processes that produce chronic psychosocial stress.

The proposed mechanisms include reduced interpersonal trust and social capital, erosion of community cohesion, and increased competition for resources perceived as scarce, with effects that vary across specific vulnerable groups such as young people and minoritised communities (Rakesh et al., 2025). A systematic review based on subnational data (Thomson et al., 2022) found partial support for this hypothesis, while also noting considerable heterogeneity across findings.

2.2 The Methodological Problem

The literature predating Sommet et al. (2025) suffered from well-documented systematic limitations. First, publication bias — the tendency to preferentially publish studies with significant results — likely inflated estimates of inequality's negative effect. Previous correlations between inequality and mental health problems may have been affected by this very phenomenon (Sommet et al., 2025; Wilkinson & Pickett, 2017).

Second, the wide variety of measures used to operationalise both inequality (Gini coefficient, decile ratio, top 1% share) and well-being (life satisfaction, absence of depressive symptoms, eudaimonia) made synthesis and cross-study comparison difficult. According to the IIH, it is not socioeconomic position per se that affects health, but one's position relative to others in a given context; to capture this gap, several indices have been developed that correlate with one another but are not interchangeable (Blesch et al., 2022).

Finally, the conceptual distinction between absolute poverty — the inability to meet basic material needs — and relative inequality — the distance between groups within a distribution — was not systematically maintained in prior literature (Brady et al., 2023). This conceptual conflation has hindered the cumulative interpretation of findings.

3. Sommet et al. (2025): Methodological Structure

3.1 Search Strategy and Study Selection

The research team conducted a systematic search across ten international databases, reviewing over 10,000 abstracts from psychology, sociology, economics, epidemiology, and public health (Sommet et al., 2025). After a rigorous selection process, 168 studies were retained, representing over 11 million participants across 38,335 geographical units — countries, regions, and municipalities — worldwide. The researchers extracted over 100 variables per article and cross-referenced them with more than 500 World Bank indicators, enabling a contextual analysis of unprecedented depth (Sommet et al., 2025).

3.2 Methodological Rigour: Specification Curve Analysis and Independent Replication

The study's most important methodological strength is the systematic verification of result robustness. The researchers tested 1,536 alternative statistical models through a specification curve analysis, and subsequently replicated their findings using Gallup World Poll data covering approximately 2 million respondents across more than 150 countries between 2005 and 2021 (Sommet et al., 2025). All data and analysis code were made openly available.

This level of transparency and verification remains rare in the psychological literature. The convergence between the meta-analysis and the primary data replication substantially raises confidence in the findings.

3.3 Moderator Identification via Machine Learning

To identify the conditions under which an effect might emerge, the researchers applied machine learning techniques to hundreds of candidate variables drawn from World Bank indicators. This data-driven approach — rather than a theory-driven one — allowed the identification of key moderators without distortions introduced by a priori researcher hypotheses (Sommet et al., 2025).

4. Main Findings

4.1 Null Effect of Inequality

The main finding is unequivocal: the average effect of economic inequality on well-being and mental health is statistically null (Sommet et al., 2025). This is not a small or marginal effect: it is equivalent to zero. The Gallup World Pollreplication confirmed this pattern at global scale, ruling out the possibility that it is an artefact of the studies included in the meta-analysis.

Consistent with the publication bias hypothesis, the authors found that previous studies systematically overestimated the negative effect of inequality, likely due to editorial preference for significant results (Sommet et al., 2025; Wilkinson & Pickett, 2017). The specification curve analysis confirmed that the null effect is robust across more than 1,500 analytical variants.

4.2 Moderators: When and Where the Effect Emerges

A null average effect does not imply an absence of variation. The moderation analyses identified two main conditions under which inequality is negatively associated with well-being and mental health (Sommet et al., 2025). The first is low income: the negative effect is confined to economically disadvantaged samples — it is not proximity to wealth that causes harm, but the insufficiency of one's own resources. The second is high inflation: in contexts of high inflation, the negative association between inequality and well-being emerges reliably, suggesting that acute economic pressure amplifies sensitivity to the distributive context (Sommet et al., 2025).

As Sommet et al. (2025) argued, inequality acts as a catalyst that amplifies other determinants of well-being and mental health — such as inflation and poverty — but is not in itself a direct cause of negative effects; this is a far more nuanced view that should inform public policy.

4.3 The Fundamental Distinction: Poverty vs. Inequality

The finding with the most significant practical implications concerns the distinction between absolute poverty and relative inequality. Policies focused solely on reducing economic inequality are unlikely to have a significant impact on the well-being and mental health of the general population; the researchers recommend prioritising the fight against poverty, whose detrimental effects on well-being and mental health are well documented (Sommet et al., 2025; Thomson et al., 2022; Ribeiro et al., 2017).

This conclusion aligns with a body of research on the psychology of poverty that distinguishes between the cognitive, emotional, and behavioural consequences of material scarcity in an absolute sense — resource-scarcity stress, cognitive tunnelling, reduced decision-making capacity (Galvan & Payne, 2024) — and those arising from the perception of relative social position, which shows a far weaker and less stable effect.

5. Critical Discussion

5.1 A Paradigm Shift or a Necessary Clarification?

Two opposite interpretive errors should be avoided. The first would be to conclude that economic inequality is irrelevant to mental health: the moderators identified by Sommet et al. (2025) clearly show otherwise. The second would be to minimise the significance of the finding, treating it as mere confirmation of what was already known.

What this study demands is a necessary clarification: the relationship between inequality and mental health is neither universal nor linear. It is contingent on the material context — particularly on the presence or absence of absolute poverty and inflationary pressure. This has non-trivial implications for how we build explanatory models of psychological distress, how we train practitioners, and what policies we support as a scientific community.

5.2 Limitations of the Study

Like all meta-analyses, this study has structural limitations that must be acknowledged. On causality: the study aggregates observational research, so a null average effect does not rule out causal effects mediated by unmeasured variables or operating over longer time horizons. On measurement heterogeneity: the 168 studies use different operationalisations of both "inequality" and "well-being," introducing variance that the moderators can only partially capture. On level of analysis: the study works with geographically aggregated data, and the subjective experience of inequality — perception, comparison, sense of injustice — is not directly captured. Finally, despite its global scale, some regions of the world remain underrepresented, limiting generalisability in certain low-income contexts.

5.3 Implications for Future Research

The findings of Sommet et al. (2025) open a new research agenda. The relevant questions are no longer "does inequality cause harm?" but rather: under what conditions and for which individuals? Through what specific mechanisms? At what level of analysis does the effect become detectable? How do inequality, inflation, and poverty interact over time? Future research should adopt longitudinal designs, systematically distinguish between absolute and relative poverty, and integrate subjective and objective measures of the socioeconomic context.

6. Implications for Clinical Practice and Policy

6.1 In the Consulting Room: Reformulating the Questions

For the mental health professional, the most practical contribution of this study is the invitation to reformulate the questions asked when assessing a patient's socioeconomic context. The relevant question is not "do you feel worse because others have more?" — which presupposes an upward social comparison mechanism — but rather "are you able to meet your basic needs?" and "are you experiencing acute economic pressure?".

This shift in focus has concrete implications for case conceptualisation and for building the therapeutic alliance: it helps distinguish between stress arising from material poverty — which may require interventions that include practical resources and psychoeducation about rights — and distress arising from relative social position, which responds differently to treatment (Galvan & Payne, 2024).

6.2 For Public Health Policy

At the level of public policy, Sommet et al. (2025) indicate that interventions focused exclusively on reducing income inequality cannot be supported by empirical evidence as sufficient tools for improving collective mental health. Resources dedicated to population mental health should prioritise the reduction of absolute poverty and economic stabilisation in high-inflation contexts.

This conclusion does not amount to a defence of the redistributive status quo: inequality may have harmful effects on dimensions not measured in this study — social cohesion, institutional trust, intergenerational mobility. But the field of mental health must be precise in attributing psychological effects to the right variables.

7. Conclusions

The meta-analysis by Sommet et al. (2025) represents a historically significant contribution to psychology and the social sciences. Its main finding — the null effect of economic inequality on well-being and mental health at the aggregate level — is not a defeat for research on inequality and psychological distress: it is an invitation to conceptual and empirical precision.

The true determinant of psychological distress, in this body of evidence, is absolute poverty — not the gap between rich and poor. This distinction, apparently technical, has profound consequences for how we think about distress, how we treat it, and what policies we advocate for. Science, when it works, does not give us simpler answers: it gives us more precise questions.

References

Blesch, K., Hauser, O. P., & Jachimowicz, J. M. (2022). Measuring inequality beyond the Gini coefficient may clarify conflicting findings. Nature Human Behaviour, 6, 1525–1536. https://doi.org/10.1038/s41562-022-01421-4

Brady, D., Curran, M., & Carpiano, R. M. (2023). A test of the predictive validity of relative versus absolute income for self-reported health and well-being in the United States. Demographic Research, 48(26). https://doi.org/10.4054/demres.2023.48.26

Galvan, M. J., & Payne, B. K. (2024). The inequality cycle: How psychology helps keep economic inequality in place. Current Directions in Psychological Science, 33(2). https://doi.org/10.1177/09637214241246553

Rakesh, D., Shiba, K., Lamont, M., Lund, C., Pickett, K. E., & VanderWeele, T. J. (2025). Economic inequality and mental health: Causality, mechanisms, and interventions. Annual Review of Clinical Psychology. https://doi.org/10.1146/annurev-clinpsy-081423-025710

Ribeiro, W. S., et al. (2017). Income inequality and mental illness-related morbidity and resilience: A systematic review and meta-analysis. Lancet Psychiatry, 4(7), 554–562. https://doi.org/10.1016/S2215-0366(17)30159-1

Sommet, N., Fillon, A. A., Rudmann, O., Rossi Saldanha Cunha, A., & Ehsan, A. (2025). No meta-analytical effect of economic inequality on well-being or mental health. Nature. https://doi.org/10.1038/s41586-025-09797-z

Thomson, R. M., et al. (2022). How do income changes impact on mental health and wellbeing for working-age adults? A systematic review and meta-analysis. Lancet Public Health, 7(6), e515–e528. https://doi.org/10.1016/S2468-2667(22)00058-5

Wilkinson, R. G., & Pickett, K. E. (2009). The spirit level: Why more equal societies almost always do better. Allen Lane.

Wilkinson, R. G., & Pickett, K. E. (2017). The enemy between us: The psychological and social costs of inequality. European Journal of Social Psychology, 47(1), 11–24. https://doi.org/10.1002/ejsp.2275