The college pay gap grew in dollars—but the percent premium held steady

Twenty years of Education Pays: dollars up, percent premium steady

Biennial Education Pays 2026 (U.S. Census Bureau CPS) editions document median earnings by education level using full-time, year-round workers age 25+. From 2004 to 2026, the dollar gap between bachelor’s and high school medians widened. The percent premium stayed near 62% in the 2026 edition (and in a relatively narrow band across editions).

Census household and worker ratios (2025 story)

The Census Bureau (September 2025) reported 2024 inflation-adjusted figures: median household income for householders 25+ was $85,580 (up 18% since 2004). Household income was $132,700 for bachelor’s or higher vs $58,920 for high school only—a ratio of about 2.3, up from 2.0 in 2004. Among full-time workers 25+, medians were $49,050 (high school) and $85,810 (bachelor’s or higher)—ratio about 1.8, up from 1.7.

Household income includes all earners and non-wage income; individual earnings gaps can differ.

Do not mix lifetime synthetic dollars with annual CPS

Georgetown CEW lifetime earnings totals (2011/2021 College Payoff) sum decades of medians in nominal dollars for a synthetic career. They are not comparable to a single-year $81,800 CPS median without inflation and population notes—see historical section and attainment trends over time.

Recessions temporarily compress premiums when layoffs hit college graduates in cyclical industries; expansions widen dollar gaps when demand for credentialed workers rises. Readers should not infer from one recession that the long-run premium has permanently shrunk.

Inflation and real earnings

Education Pays 2026 (U.S. Census Bureau CPS) and Census Bureau stories adjust for inflation when comparing 2004 to 2024. Nominal dollars rise over time; real purchasing power matters for interpreting “growing gaps.” The percent premium staying near 62% means bachelor’s medians and high school medians rose at roughly similar percentage rates over many editions—not that individuals kept the same real wage.

Generational change

More adults hold bachelor’s degrees than in 2004. That supply shift can affect returns to additional schooling in some fields even while economy-wide medians by education level remain favorable to college completers in CPS data.

Read the full historical table in Education Pays 2004–2026 and the Census Bureau’s September 2025 education-and-income release.

Education Pays editions over time

Each Education Pays (U.S. Census Bureau CPS) edition updates medians with new CPS years. The dollar gap between bachelor’s and high school medians widened from about $19,100 in 2004 to $31,200 in 2026 while the percent premium stayed near 62%. That pattern means both groups gained in nominal dollars, with bachelor’s medians rising faster in dollar terms.

The Census Bureau household income story complements Education Pays worker earnings: households include multiple earners and non-wage income, so household ratios can move differently than individual earnings ratios.

Technology and globalization shocks

The college earnings premium widened in the 1980s as demand for skilled labor rose and manufacturing employment fell. CEW historical analyses connect automation, trade, and union decline to polarization between workers with and without postsecondary training. Recent AI adoption may shift which skills command premiums, but baseline literacy, numeracy, and adaptability from college still matter in service and health sectors.

Will the premium shrink in the 2030s?

Some economists argue that if more workers earn degrees, the marginal premium could fall; others point to continued demand in health and education. Watch cohort studies, not single-year spikes during labor shortages. This article’s time-series chart should be updated when new CPS and Scorecard releases land.

Premium in dollars and percent

Education Pays reports bachelor’s medians at $81,800 vs high school at $50,600 in 2024. The dollar gap grew as both levels gained, but the percentage premium stayed near the low-60s for years—useful for journalists who ask whether college “stopped working” because the percent gap narrowed slightly in some years.

Associate’s and master’s medians also rose in recent editions; stacking credentials changes lifetime paths. Premium articles should name which credential pair is compared.

Chart.js figure in this article visualizes the dollar gap series—update the data array when Education Pays 2028 publishes new CPS medians.

Historical chart discipline

Keep edition years explicit when citing Education Pays 2026 (U.S. Census Bureau CPS) tables.

Plot both dollar gaps and percent premiums for readers.

Update charts when the 2028 edition publishes new CPS medians.

How this article fits The College Payoff, 2026

This piece is one of twenty companion articles to The College Payoff, 2026. The parent report synthesizes College Board Education Pays, Georgetown CEW, Census, BLS, College Scorecard, Federal Reserve SCF wealth tables, and EDsmart institution-level findings. Each breakout answers one question journalists and families ask when those sources are quoted side by side.

When federal sources update on different calendars, start from the parent report’s definitions section, then return here for the topic-specific FAQ. Interactive tables and state pages on EDsmart Data—tuition trends, debt-to-earnings by major, attainment by state, STEM outcomes—let you filter national benchmarks beyond what a single government table provides.

If a number here differs from a chart in the parent report, use the parent report and the underlying federal publication as the authority. Breakouts explain how to read those sources; they are not a separate dataset.