Economy and Work: evidence

Economy and Work evidence in the United States, as of fall 2026, shows an unemployment rate of 4.2%, prices rising faster than average hourly earnings, a record real median household income for 2025, and a large share of households still worried about prices and exposed to small shocks. Research on labor policy, especially the minimum wage, contains credible studies pointing in different directions. The honest summary is that the data support precise statements about particular measures and dates, not a single verdict on whether the economy is “good.”

In plain words: some numbers look strong and some look weak, and both are true at the same time. Unemployment is moderate. Prices rose faster than paychecks over the past year. Median household income in 2025 was the highest ever recorded after adjusting for inflation. Many families still could not easily cover a surprise bill. Economists also disagree about some big questions, like what a higher minimum wage does to jobs. This page lays out the numbers and the disputes. Why people rank the economy as they do is on Economy and Work; who holds each power is on Economy and Work: who decides.

Defined terms

Economy and Work evidence relies on these measures.

Employment Situation report. The Bureau of Labor Statistics (BLS) monthly report combining a household survey (for the unemployment rate) and an establishment survey (for payroll jobs).

Labor force participation rate. The share of the adult civilian population that is working or looking for work.

CPI-U. The Consumer Price Index for All Urban Consumers. “Core” CPI excludes food and energy.

Real average hourly earnings. Average hourly pay adjusted for CPI inflation.

Real median household income. The income of the household in the middle of the distribution, adjusted for inflation.

Official poverty measure and Supplemental Poverty Measure (SPM). Two Census Bureau poverty measures; the SPM accounts for several government programs and expenses that the official measure does not.

Natural experiment. A study that uses a policy change, such as a state minimum wage increase, to compare groups affected and unaffected by it.

Evidence

Economy and Work data come from official statistical agencies, the Federal Reserve and peer-reviewed or working-paper research.

Jobs

The BLS reported on October 2, 2026 that in September 2026 the unemployment rate was 4.2%, with 7.1 million people unemployed, and that nonfarm payrolls rose by 29,000.⁠¹ The labor force participation rate was 61.8% and the employment-population ratio 59.2%.⁠¹ A monthly payroll gain of that size is small, and monthly estimates are revised, so a single month should not be read as a trend.

Prices and wages

The BLS reported on October 14, 2026 that the CPI-U rose 3.7% over the twelve months ending September 2026, before seasonal adjustment.⁠² Core CPI rose 2.6%, food 2.5%, food at home 1.8%, shelter 3.2% and energy 20.4%.⁠² Average hourly earnings for private nonfarm employees were $37.81, up 3.0% over the year.⁠¹ Real average hourly earnings fell 0.7% from September 2025 to September 2026.⁠³

These figures show where the pressure came from: energy prices rose much faster than other categories, while core inflation was lower than overall inflation. They also show why many households felt worse off in 2026: pay rose more slowly than prices over that twelve-month period.

Income and poverty

The Census Bureau reported on September 15, 2026 that real median household income in 2025 was $87,460, up 2.6% from $85,210 in 2024, and “the highest on record since 1967”.⁠⁴ Median income after taxes and credits was $76,060, up 3.1%, but not a record, because pandemic-era payments had raised after-tax income in 2020 and 2021.⁠⁴

The official poverty rate fell 0.5 percentage points to 10.2% in 2025, with 34.5 million people in poverty, and the official child poverty rate reached a historic low of 13.4%.⁠⁵ The Supplemental Poverty Measure was 13.1%, not statistically different from 2024.⁠⁵ The two measures differ because they count resources and needs differently; which one is “right” depends on the question being asked.

Note the timing: the income and poverty data describe calendar year 2025, while the price and wage data above describe the year ending September 2026. They cover different periods and can point in different directions without contradicting each other.

Household security

The Federal Reserve’s Survey of Household Economics and Decisionmaking, fielded in October 2025, found that 73% of adults were doing at least okay financially, consistent with 2024 but below the 2021 high of 78%.⁠⁶ Sixty-three percent said they would cover a $400 emergency expense with cash or its equivalent, unchanged from 2024.⁠⁶ Ninety-one percent said price increases were a major or minor concern, and 53% called them a major concern, down from 56% in 2024.⁠⁶

Perception

Gallup’s July 1 to 19, 2026 survey found 22% of adults rating economic conditions excellent or good and 44% poor.⁠⁷ Asked whether conditions were getting better or worse, 67% of Republicans said better and 26% worse, while 4% of Democrats said better and 93% worse.⁠⁷ A difference of that size between partisans facing the same data is evidence that economic perception is shaped by political identity as well as by personal experience, though a single survey cannot say how much of each.

Unions and licensing

The BLS reported that 10.0% of wage and salary workers were union members in 2025, or 14.7 million people, compared with 20.1% and 17.7 million in 1983.⁠⁸ The rate was 32.9% in the public sector and 5.9% in the private sector.⁠⁸ In 2025, 21.6% of employed people held a license and 24.0% held a certification or license.⁠⁹

The minimum wage: the central research dispute

Minimum wage research is the most cited and most contested labor policy evidence in the United States.

Card and Krueger (1993, 1994). After New Jersey raised its minimum wage from $4.25 to $5.05 on April 1, 1992, David Card and Alan Krueger surveyed 410 fast food restaurants in New Jersey and neighboring Pennsylvania. They found that “Relative to stores in Pennsylvania, fast food restaurants in New Jersey increased employment by 13 percent,” a finding that, in their words, challenged “the prediction that a rise in the minimum reduces employment”.⁠¹⁰

Cengiz, Dube, Lindner and Zipperer (2019). Studying “138 prominent state-level minimum wage changes between 1979 and 2016,” they found that “the overall number of low-wage jobs remained essentially unchanged” over five years after an increase.⁠¹¹

Jardim and coauthors on Seattle (2017, revised 2018). Studying Seattle’s increase to $13, they estimated that it “reduced hours worked in low-wage jobs by 6-7 percent,” and that the ordinance “lowered the amount paid to workers in low-wage jobs by an average of $74 per month per job in 2016”.⁠¹²

Neumark and Shirley (2021, revised 2022). Reviewing the research literature, they concluded that “there is a clear preponderance of negative estimates in the literature,” with the evidence “stronger for teens and young adults as well as the less-educated”.⁠¹³

These studies differ in place, period, size of the increase relative to local wages, and method. Supporters of higher wage floors emphasize studies finding small or no job losses and higher pay; critics emphasize studies finding reduced hours or employment, especially for the least experienced workers. Both camps include respected economists.

Where the evidence is uncertain or missing

Economy and Work data have gaps that matter for public debate.

Missing data. The BLS notes that its 2025 annual averages for licensing are based on eleven months, because October 2025 data were not collected.⁠⁹ Gaps like this weaken year-over-year comparisons.

Revisions. Monthly payroll estimates are revised as more data arrive, so the September 2026 figure of 29,000,⁠¹ is provisional.

Periods do not line up. Annual income data for 2025⁠⁴ and twelve-month price data through September 2026⁠² cannot be combined into one judgment without choosing a time frame, and the choice changes the conclusion.

Medians hide distribution. A record median household income⁠⁴ describes the household in the middle; it is compatible with gains or losses elsewhere in the distribution, which need separate measures to assess.

Untested questions. This page found no source it could verify on several questions people raise often, including how much of 2026 inflation came from tariffs versus energy markets, and how new technologies are changing hiring. These are stated as open questions, not findings.

Competing interpretations

Economy and Work evidence supports different readings.

One reading emphasizes record real median household income in 2025 and falling official poverty⁠^4,5 and concludes that the economy delivered broad gains. A second emphasizes falling real hourly earnings through September 2026 and high price concern⁠^3,6 and concludes that many households are being squeezed. A third points to the partisan perception gap⁠⁷ and argues that public assessments partly reflect who holds power. On labor policy, one reading treats Cengiz and colleagues and Card and Krueger as showing that moderate wage floors raise pay without measurable job loss; another treats Seattle and the Neumark and Shirley review as showing real costs for the least experienced workers. Each reading uses accurate numbers; they differ in which numbers they weigh and which time frame they choose.

Application to this framework

Economy and Work evidence matters to My Vote Forward because priority, perception and evidence are separate things. A participant can rank the economy first because of prices, jobs or wages, and the framework records the ranking, not the reason.

When a proposal is shown, such as a local minimum wage increase, the framework should show the strongest research on each side, with dates and settings, and label research as research. Hypothetical example: a city proposal to raise its minimum wage to a stated level would be presented alongside the New Jersey, multi-state and Seattle studies above, with a note that effects may differ with the size of the increase and local wage levels. See Proposals and Tradeoffs.

The framework should also show official data next to group priority results, so participants can compare what people feel with what agencies measure, without treating either as the whole truth. See Measurement.

Unresolved questions

Economy and Work evidence leaves important questions open:

• Which single time frame, if any, should a civic tool use when showing economic data next to priorities?
• How large are minimum wage effects at the higher nominal levels now being adopted, such as the $17.13 state rate in Washington as of July 1, 2026⁠¹⁴ compared with the increases in older studies?
• How much of the partisan perception gap reflects different personal experiences, and how much reflects political identity?
• How should the framework handle data gaps, such as uncollected months, when showing trends?
• What share of economic concern is really about housing, health or transport costs, which sit in Housing and Communities, Health and Care and Infrastructure and Transport?