Local changes in income inequality do not predict teen depression, massive study finds

A recent study published in Psychological Science suggests that local shifts in the gap between the rich and the poor do not meaningfully predict changes in depression among teenagers. While some social theories propose that economic disparities inherently harm public mental health, the findings provide evidence that fluctuating income inequality within a specific community has little practical impact on adolescent depressive symptoms.

The idea that how a society’s wealth is distributed affects the well-being of its citizens has gained immense traction over the past few decades. Popularized by social scientists, the income-inequality hypothesis proposes that wide economic gaps foster status anxiety, erode social trust, and degrade public services. This perspective suggests that inequality operates like an environmental pollutant, affecting everyone in the social climate regardless of their personal bank account.

Adolescence represents a particularly sensitive period for testing this idea. During these teenage years, individuals become highly attuned to peer comparison and social hierarchies. At the same time, most mental health disorders first emerge during this developmental window. Rates of teenage depression have noticeably increased in countries like Norway, leading some public health experts to point to rising national income inequality as a possible contributor to this worrying trend.

Sondre Aasen Nilsen, an associate professor in the Department of Special Needs Education at the University of Oslo and a senior researcher at the Regional Centre for Child and Youth Mental Health and Child Welfare at the NORCE Norwegian Research Centre, wanted to examine this presumed link. “Income inequality is often discussed as one of the major social determinants of health, and many studies seem to proceed from the assumption that it is an established causal factor for mental-health-related outcomes,” Nilsen said.

He noted that past evaluations often relied on comparing unequal countries to equal countries at a single point in time. “That is a major limitation,” Nilsen explained. “Such studies can show that more unequal places also tend to have worse mental health, but they cannot tell us whether changes in inequality actually lead to changes in mental health, which is the question we arguably care most about.”

Free daily newsletter

Comparing entirely different nations makes it difficult to rule out cultural or political factors that might explain varying depression rates. “We therefore wanted to test the question more directly: when income inequality changes within the same municipality over time, do adolescent depressive symptoms change as well?” Nilsen said.

To build a more accurate picture, Nilsen and his colleagues utilized a massive set of health and economic data from Norwegian municipalities. They analyzed a vast collection of responses from a national youth survey program in Norway, which routinely polls middle school and high school students across the country about their lifestyle, health, and well-being. The current study focused on survey responses collected from 2010 through 2019.

The final analytical sample included approximately 550,000 adolescent respondents nested within 863 municipality-years and 340 specific municipalities. To measure teenage depression, the researchers used a widely accepted six-question self-report tool. This questionnaire asked students to rate how often they felt affected by depressive moods over the previous week.

The authors also tracked students’ self-reported family financial situations. Students rated whether their family had experienced mostly good, neutral, or poor financial circumstances over the past two years. For the economic side of the equation, the authors relied on official government registry data to calculate income inequality for each municipality during each specific year.

They primarily used the Gini index, a standard statistical tool that gauges how wealth is distributed across a population. A score of zero represents total equality where everyone earns the exact same amount, while higher scores indicate greater disparity. To verify their work, the researchers also used a second economic measure that directly compares the income of the top 10 percent of earners to the bottom 10 percent.

The researchers controlled for various demographic and economic factors at both the local and individual levels. They accounted for the median income of the municipality, local poverty rates, age, gender, and the specific year the survey was taken. Factoring in the local poverty rate helps to separate the stress of absolute material hardship from the psychological effects of simply living near wealthier people.

Using advanced statistical models that isolated the specific changes happening within a single town over time, the authors found that changes in a municipality’s income inequality did not statistically predict changes in overall adolescent depressive symptoms. When the gap between the rich and the poor widened within a local area, the average mental health of the teenagers living there remained essentially unchanged. “Given the mixed prior literature and the results from recent high-quality studies, I thought a small or null effect was quite plausible,” Nilsen told PsyPost.

“What was perhaps most informative was how close to zero the overall within-municipality effect was, even in a large study designed to test this question directly,” he added. To verify this near-zero effect, the scientists employed a statistical technique known as equivalence testing. Traditional statistics usually focus on finding whether an effect simply exists or not.

“Because the analyses were based on repeated observations of many municipalities over time, the study had good power to detect small within-municipality effects,” Nilsen said. Equivalence testing helped determine if an observed effect fell within a predefined range of outcomes that are considered too tiny to be of any real-world consequence. “We therefore used equivalence testing to ask whether the effects were large enough to matter in practice,” Nilsen explained.

The equivalence tests confirmed that the overall effect of income inequality on depression was practically negligible. The researchers also tested whether this economic shift disproportionately harmed teenagers already living in poor financial situations. They found no meaningful interaction, meaning adolescents from struggling families were not uniquely impacted by the widening local wealth gap compared to their wealthier peers.

When looking at gender, the authors did observe a very slight divergence in the data. As local inequality rose, depressive symptoms tended to increase marginally for teenage girls and decrease marginally for teenage boys. “The gender interaction was statistically significant, but the underlying effects for both girls and boys were small,” Nilsen noted. “I therefore see it more as a nuance than as a surprising finding.”

Translating these statistical shifts to real-world terms helps put the findings into perspective. “For girls, a two-point increase in Gini corresponded to less than a one percentage-point increase in probable depression,” Nilsen said. “To explain the observed rise in depressive symptoms among girls, Gini would have needed to increase by roughly 20 points, which is far beyond the municipal changes observed in our data.”

The researchers even explored whether the effects of inequality might take a few years to manifest. They tested the data using time delays of up to seven years, looking to see if an economic shift in one year predicted depression rates several years later. Across almost all of these delayed timeframes, the results remained practically negligible.

While these findings are informative, readers should be careful not to misinterpret the results as proof that income inequality does not matter at all. “I would also preempt one possible misinterpretation: our findings should not be read as saying that inequality is desirable or irrelevant,” Nilsen said. “There may be many good reasons to reduce inequality, including reducing poverty and improving living conditions.”

The authors note that economic disparities might still influence other psychological or behavioral outcomes, such as risk-taking or extreme competitiveness. “Our study suggests that local changes in income inequality did not meaningfully predict adolescent depressive symptoms in this specific context,” Nilsen clarified.

The study also faces certain limitations related to its specific geographic setting. Norway is a strong welfare state with universal healthcare, free education, and robust social safety nets. This protective environment might blunt the harshest psychological impacts of economic stratification.

“The main caveat is that this study was conducted in Norway, which is a relatively equal country with limited within-municipality changes in income inequality,” Nilsen said. “Effects could be different in countries with much higher inequality or larger changes over time.”

Future research could benefit from applying this same rigorous approach to different nations. “For income inequality specifically, I think future studies should continue to use repeated local-area data, preregistered analyses, and clear thresholds for what would count as a meaningful effect,” Nilsen suggested. “It would also be useful to examine whether similar findings appear in countries with higher inequality and larger changes over time.”

“More broadly, I am interested in how social and contextual conditions shape adolescent mental health and behavior, and in using designs that better separate between-area differences from within-area changes over time,” Nilsen said. Expanding the scope to include other markers of psychological well-being could also provide a more comprehensive understanding of how wealth distribution shapes the developing mind.

“In my own work, I also plan to examine other adolescent outcomes, such as bullying and antisocial behavior,” he added. “It is possible that income inequality matters more for these kinds of social and behavioral outcomes than for depressive symptoms.”

“One thing I would emphasize is that null or very small effects can be informative when the study is well powered and the design directly addresses the research question,” Nilsen concluded. “In this case, the study does not show that inequality never matters for mental health. But it does suggest that, in Norway from 2010 to 2019, changes in local income inequality were unlikely to be a major explanation for changes in adolescent depressive symptoms.”

The study, “Does Income Inequality Predict Adolescent Depressive Symptoms?“, was authored by Sondre Aasen Nilsen, Kyrre Breivik, Kjell Morten Stormark, and Tormod Bøe.