About this Project

The Economic Mobility Explorer is an interactive companion to Lands of Opportunity: Differences in the Geography of Wealth and Income Mobility in the United States (Binder, Risch & Voorheis 2026, NBER Working Paper No. 35219). The paper makes a careful case that the geography of wealth mobility looks markedly different from that of income mobility. This explorer lets anyone see and compare that finding county by county across the United States.

Read the working paper (NBER WP 35219) ↗

Why this explorer

In 2014, Chetty, Hendren, Kline, and Saez found that Charlotte ranked 50th out of the 50 largest U.S. commuting zones for upward mobility — children born in the bottom income quintile had the lowest chance of any major metro of reaching the top. The finding generated significant concern and policy conversation in the Charlotte region and beyond, and it shaped my own research focus on economic mobility.

When the Binder–Risch–Voorheis paper came out, I began analyzing its data for the Charlotte region (working paper to be linked soon) and decided to build a public resource for the area. The scope quickly expanded into a national one: I wanted states and metros to navigate these new measures easily and generate analytics quickly, so any reader can understand and study how different regions of the U.S. are doing.

The explorer opens at a national view. From there you can pick a metric (income or wealth, plus three ways of measuring mobility), search for a state, metro area, or county, and pin places into a side-by-side comparison. Its aim is to make empirical research about American opportunity tangible and explorable, not hidden behind a table of regression coefficients.

Methodology & measures

The source paper describes each county by three parameters (α, β, δ) for each of five wealth and income concepts. The map renders one of those parameters at a time, for one concept at a time, with the color showing the county's position on the national distribution.

The three measures

• α — Absolute mobility: The average rank of children whose parents are at the bottom of the distribution. A higher value means children from low-resource families end up higher on average in the adult distribution — higher is better.
• β — Persistence (rank-rank slope): How strongly a parent's rank carries forward to their child's rank. A county with high β has a steep parent-to-child gradient — rank is more heritable. Higher means less mobility.
• δ — Extensive-margin gap: The additional penalty for children whose parents had none of a resource (e.g., renters vs. homeowners, or those with zero wealth). Captures the discrete cliff at the bottom of the distribution. Higher means a larger penalty.

The five economic concepts

Each of the three measures above is computed separately for five different definitions of what is being passed across generations:

Total Income

All income a person receives, including wages, salaries, business income, capital income, and government transfers. The broadest income concept and the most directly comparable to prior intergenerational income-mobility work.

Labor Income

Income earned from work alone (wages and salaries), excluding capital income and transfers. Captures the role of earnings on the job, rather than asset returns or safety-net programs.

Homeownership (Extensive housing wealth)

The probability of owning a home at all; the "extensive margin" of housing wealth. A binary distinction that often matters more than the dollar value, because homeownership is a major pathway to long-run wealth accumulation.

Housing Wealth (Gross housing wealth)

The value of housing assets a person holds, conditional on owning. Captures variation in the magnitude of housing wealth, not just whether someone has any. Where housing markets have appreciated sharply, this concept can diverge from homeownership.

Total Wealth (Gross total wealth)

The total value of all assets a person holds: housing, financial assets, business equity, and more. The most encompassing wealth concept, and the one that diverges most clearly from income-based mobility measures.

Reading the map

Each county's fill color reflects its national percentile rank on the selected measure and concept. The color ramp runs from deep plum (low, or less mobile) through warm sand (middle) to teal-green (high, or more mobile). For persistence (β) the direction inverts: a lower slope is better, so teal reflects lower β.

The estimates are small-area (Fay–Herriot) estimates. This is a model-assisted approach that borrows strength from neighboring counties and covariate information to stabilize estimates for counties with sparse survey coverage. Counties with few observations are therefore less precise; treat point estimates in sparsely-populated counties with caution.

Where the data come from

The microdata come from linked administrative tax records and the U.S. Census Bureau's American Community Survey. The public-release county-level estimates are distributed as binder_risch_voorheis_county_mobility_stats.xlsx.

Metro-area estimates in the Compare counties panel and metro-navigation strip are population-weighted means of constituent county estimates, using OMB Core-Based Statistical Area (CBSA) delineations.

→ Read the user guide for a walkthrough of every panel and interaction.