Bor Cohen and Galea 2017 — Population Health in an Era of Rising Income Inequality USA 1980–2015

Summary

A systematic review of US evidence on trends in income- and education-related survival inequalities 1980–2015, published in The Lancet as the fifth paper in its "America: Equity and Equality in Health" series. The review synthesizes 15+ primary studies and contributes an original decomposition of Chetty et al. (2016) data showing that the steepening income-survival gradient since 2001 is driven more by a change in the slope of the gradient than by falling incomes at the bottom. The paper concludes with a framework centered on a potential 21st-century health-poverty trap.

Key Claims

• Gaps in mortality and life expectancy (LE) by income and education have widened during 1980–2014, robust across all socioeconomic status (SES) measures, outcomes, and study designs.
• Since 2001, the poorest $5\%$ of Americans experienced near-zero LE gains while middle- and upper-income Americans gained over $2$ years; this divergence occurred specifically below the $\$60{,}000$ income level.
• The two-channel decomposition of Chetty et al. data (2001–14): falling real incomes at the bottom explain $\approx$ one-third (men) to one-sixth (women) of the growing top–bottom LE gap; a steepening income–survival gradient explains the remainder and is the dominant factor.
• The income–survival gradient (measured as LE per log-point of income) grew from $1.7$ to $2.4$ yrs/log-point (women) and from $3.2$ to $3.7$ yrs/log-point (men) 2001–14.
• The gradient steepened only for households earning below $\approx \$60{,}000$/year; the rich–upper-middle-class gap did not widen.
• Cohort gaps dwarf period gaps: National Academy of Sciences (NAS) (2015) projects the cohort LE gap between top and bottom income quintile will widen from $\approx 5$ years (1930 birth cohort) to $\approx 12$–$14$ years (1960 birth cohort).
• LE declined for white women with less than high-school (HS) education; survival gaps did not increase among Hispanic Americans.
• Proximate mechanisms (partially supported): smoking burden shifted downward disproportionately; substance abuse and self-harm concentrated at low SES; underuse of essential medical care; obesity differential (emerging).
• Distal mechanisms (suggestive): geographic income segregation; reduced intergenerational mobility; mass incarceration; differential access to health-improving technologies; erosion of subsidies shielding the poor from health input prices.
• Health can reproduce income inequality through a feedback loop ("health-poverty trap"), especially if health deficits in childhood compound into earnings deficits in adulthood.

Concepts Introduced or Extended

• Health-Poverty Trap
• Income-Mortality Gradient
• Education-Mortality Gradient
• SES Health Behavior Gradient

Entities Mentioned

• Jacob Bor
• Gregory H. Cohen
• Sandro Galea
• Raj Chetty
• Hilary Waldron

Quotes

"Having missed out on decades of income growth and longevity gains, low-income Americans are increasingly left behind."

"Without interventions to decouple income and health, or to reduce inequalities in income, we might see the emergence of a 21st century health-poverty trap and the further widening and hardening of socioeconomic inequalities in health."

"Since 2001, annual inflation-adjusted household earnings have fallen by 17% for men and women in households at the 25th income percentile. Americans at the top of the distribution experienced much smaller (or non-existent) relative losses."

My Take

The paper's main original contribution is the two-channel decomposition reanalyzing Chetty et al. supplementary data — the finding that the gradient steepening matters more than income distribution changes is substantively important and not obvious. The review function is also valuable: it synthesizes a fragmented literature in a structured table with comparable statistics. The distal-mechanism section is speculative and the paper is honest about this. Limitation: as a review rather than an original analysis, Bor et al. can synthesize but not causally identify. The health-poverty trap framing is conceptually useful but remains largely theoretical — no causal evidence links income inequality per se (as opposed to low income) to worse health outcomes at the individual level.