Human Opportunity Index: Measuring Access to a Decent Life

What the Human Opportunity Index measures

The Human Opportunity Index, as the World Bank originally constructed it for Latin America in the late 2000s, was an answer to a single question: what share of children in this country have access to a set of basic services that determine the chance of a decent life, conditional on factors outside their control like the income or location of their family? The original World Bank work, Measuring Inequality of Opportunities in Latin America and the Caribbean (2009), set the template that has since been adapted to dozens of contexts.

This article presents an adaptation of that template to the United States at the state level. The US Human Opportunity Index combines four sub-indices: education quality and attainment, health access and outcomes, economic mobility, and digital and physical connectivity. Each sub-index is built from public micro-data, and each can be re-weighted independently. The composite is bounded between 0 and 100, and the policy reading is straightforward: the higher the score, the higher the probability that a person born in that state will have meaningful access to the foundations of a decent life by age 30.

The point of the Index is not to rank states for tourism. It is to give state-level policymakers, philanthropic funders, and Workforce Innovation and Opportunity Act (WIOA) intermediaries a single integrated picture of where opportunity is structurally most and least available, so they can compare against peer states and prioritize the binding constraint.

The education sub-index

The education component combines four primary inputs. The first is the share of fourth-grade and eighth-grade students reading at or above proficient on the National Assessment of Educational Progress (NAEP). The second is the four-year on-time high-school graduation rate from the National Center for Education Statistics. The third is the share of 25-to-34-year-olds with a post-secondary credential, drawn from the Census Bureau's American Community Survey. The fourth is the published per-pupil instructional expenditure inflation-adjusted to a common base year.

The first three are outcomes; the last is an input that conditions outcomes. Together they capture both how the system performs today and what resources it has to perform with. Massachusetts, New Jersey, New Hampshire, Connecticut, and Maryland consistently score in the top tier on this sub-index, primarily because their NAEP proficiency rates are at or near the top of the distribution and their attainment rates among young adults are well above 50 percent. New Mexico, Mississippi, West Virginia, Louisiana, and Alabama consistently score lowest, driven primarily by NAEP outcomes and lower attainment rates.

The education sub-index has the strongest correlation with the eventual composite Human Opportunity Index, because education access shapes the other three sub-indices over the medium term. This is consistent with the long-running finding in the Raj Chetty research program on intergenerational mobility (Chetty et al., Nature 2022) that childhood-environment characteristics, including school quality, are first-order determinants of adult outcomes.

The health sub-index

The health sub-index combines four inputs. The first is life expectancy at birth from the CDC National Vital Statistics System. The second is the rate of uninsured adults under 65 from the Census Bureau Small Area Health Insurance Estimates. The third is age-adjusted preventable mortality, also from CDC. The fourth is access to primary care, measured as the share of the state population residing in a federally designated Health Professional Shortage Area.

The pattern is striking. The interquartile range in life expectancy across US states is roughly four years. The states at the bottom on life expectancy, in the Mississippi Delta and parts of Appalachia, also tend to score at the bottom on uninsured rates, preventable mortality, and primary-care access. The states at the top, Hawaii, California, Connecticut, Minnesota, and Massachusetts, combine longer life expectancy with broader coverage and better access. The single most consequential structural variable in this sub-index is whether the state expanded Medicaid under the Affordable Care Act, which the Kaiser Family Foundation tracks at the KFF Medicaid expansion dashboard.

The health sub-index is the most reactive to policy in the medium term. States that expanded Medicaid saw measurable improvement in uninsured rates within two years of expansion, and longer-run reductions in preventable mortality have been documented in the peer-reviewed literature. The Index thus tends to register the effect of policy choices in this dimension faster than in the education dimension.

The economic mobility sub-index

The economic mobility sub-index uses the Opportunity Atlas methodology developed by Chetty, Hendren, Jones, and Porter (2018) and updated through the Opportunity Insights data portal. Three inputs are combined: the mean adult earnings rank for children raised in low-income families, the probability of moving from the bottom quintile of household income to the top quintile, and the standard intergenerational income elasticity for the state.

This sub-index is the most powerful at separating states with similar education and health scores. Several Midwestern states, Iowa, Minnesota, and the Dakotas, score higher than expected on mobility given their education and health scores, because their adult labor markets historically absorbed workers from low-income families into stable middle-class trajectories. Several Southern states score lower than expected on mobility given their education and health scores, because the gradient between childhood circumstance and adult earnings is unusually steep there. The geographic pattern is one of the most reliable findings in the Opportunity Insights program and has been replicated in independent work by the Federal Reserve Banks of Atlanta and Minneapolis.

The connectivity sub-index

The connectivity sub-index combines digital access and physical access. The digital input is the share of households with fixed broadband at the FCC 100/20 Mbps definition from the FCC Broadband Deployment Report. The physical input is a transit-and-roads composite from the US Department of Transportation Bureau of Transportation Statistics state profile data, supplemented by the AARP Livability Index for state-level walkability and proximity-to-services measures.

Connectivity is the most under-discussed driver of opportunity. The Chetty social-capital work (Nature 2022) showed that cross-class network connections, which depend partly on physical and digital infrastructure that mixes people across income strata, are stronger predictors of upward mobility than many traditional school and neighborhood measures. States that score high on connectivity, including parts of the upper Midwest, the Pacific Northwest, and the Northeast, are also disproportionately states that score high on mobility.

Reading the composite

When the four sub-indices are combined with equal weights, three patterns dominate.

A cluster of high-opportunity states scores above 70: Massachusetts, Connecticut, New Hampshire, Minnesota, New Jersey, Maryland, Vermont, and Washington. The common features are high education attainment, high life expectancy, strong intergenerational mobility, and high broadband coverage.

A middle cluster scores between 50 and 70 and contains most US states, including California, New York, Illinois, Pennsylvania, Wisconsin, Iowa, Colorado, and Virginia. Each member of this cluster has at least one sub-index in the top tier and at least one in the bottom tier, which is why the composite lands near the median.

A low-opportunity cluster scores below 50: Mississippi, West Virginia, Louisiana, New Mexico, Alabama, Arkansas, Oklahoma, and Tennessee. The common features are low education outcomes, lower life expectancy, steep intergenerational mobility gradients, and lower broadband coverage. These are not coincidences. The four sub-indices are positively correlated, and structural patterns at the state level tend to cumulate.

What the Index is not

The Human Opportunity Index is a state-level summary statistic. It is not a verdict on any individual life, and it is not a sufficient guide to within-state inequality. A state can have a high HOI score and still contain high-inequality cities or communities; California is the most obvious example. Users who care about within-state inequality should pair the HOI with the Opportunity Atlas county-level data, which preserves the high-resolution geography that the state aggregation washes out.

The Index is also not a forecast. It captures the structural conditions a person born today is most likely to encounter. It does not predict the trajectory of any individual cohort, and it does not project forward how the four sub-indices will evolve under particular policy choices. For projection work, the OECD Better Life Index and the UNDP Human Development Index provide international benchmarking that is useful but not directly comparable on methodology.

What the Index implies for policy

Three operational implications follow from a careful reading.

First, the binding constraint differs by state. A state in the middle cluster that scores low on connectivity but high on education and health should prioritize broadband and transit, not school reform. A middle-cluster state that scores low on health but high on education should prioritize health-coverage expansion. Generic ranking moves nothing; identifying the binding constraint by state is the actionable step.

Second, mobility is partly a function of cross-class network density, which is partly a function of how communities are built. The Chetty social-capital findings imply that policies that mix income strata, in schools, housing, and public space, have larger effects on mobility than equally expensive policies that do not. This is the same finding that surfaces in the design of opportunity-rich places, covered in our Geography of Opportunity Rankings.

Third, low-opportunity states are not all alike. Mississippi, West Virginia, and New Mexico share a low composite score but differ sharply on which sub-index drives the score. The cluster-level finding is correct; the prescription must be state-specific.

For context on adjacent measurement frames, see our companion pieces on the American Dream Affordability Index, America's Loneliness Map, and AI job displacement by county. All four series, including this one, are different lenses on the same underlying question of structural access to a decent life in twenty-first-century America.

Sources

• World Bank, *Measuring Inequality of Opportunities in Latin America and the Caribbean*, 2009.
• Chetty, Hendren, Jones, Porter, *The Opportunity Atlas*, 2018; and Chetty et al., *Social capital and economic mobility*, Nature 2022.
• US Census Bureau, American Community Survey and Small Area Health Insurance Estimates.
• CDC, National Vital Statistics System.
• National Center for Education Statistics, Nation's Report Card and Condition of Education.
• Federal Communications Commission, Broadband Deployment Reports.
• AARP, Livability Index.
• Kaiser Family Foundation, Status of State Medicaid Expansion Decisions.
• OECD, Better Life Index, and UNDP, Human Development Index.