Every day, millions of Americans make decisions based on federal statistics, often without even realizing it. When businesses decide where to open new stores, when local governments plan roads, schools, and emergency services, when families compare cost of living across cities—they’re all relying on behind-the-scenes data from a network of federal statistical agencies that costs remarkably little to maintain.
In fiscal year 2023, the entire federal statistical system—including the Census Bureau, Bureau of Labor Statistics, Bureau of Economic Analysis, and others—operated on a budget of about $6.8 billion. That might sound like a lot, but it represented just 0.03% of the nation’s $27 trillion economy, and less than one-tenth of what the government spent on education, transportation, health, income security, or veterans’ affairs (see figure).
Figure: Federal Statistics Cost Little Compared With Other Discretionary Spending
Federal discretionary spending by major function, fiscal year 2023 (in billions)
Note: *Other includes general government administrative costs of Medicare, Social Security, agriculture, energy, and commerce and housing credit programs.
Source: Congressional Budget Office, Discretionary Spending in Fiscal Year 2023, March 5, 2024; Office of Management and Budget, Statistical Programs of the United States Government: Fiscal Year 2023, Executive Office of the President, January 2025.
The Return Far Exceeds the Investment
Federal statistics function as cross-cutting infrastructure, informing how funds are spent across the federal government—and often saving taxpayers far more than they cost.
Each year, accurate and timely federal statistics guide the distribution of trillions of dollars to the people and places they’re meant to serve. When data are timely and precise, resources flow efficiently to the people who need them. Working with bad data increases the risk of poor decision-making (out-of-touch program design, ineffective or even harmful policies), as well as long-term costs that outweigh any near-term savings. Here are a few examples of bad data in action:
- Due to a major undercount of children in the 2020 Census, Texas’s federal funding will be slashed by $25 billion over the decade, researchers estimate.
- The Federal Highway Administration maintains a bridge inventory that guides the allocation of billions of dollars in transportation funds, but data errors can delay critical repairs, increase long-term infrastructure costs, and elevate safety risks.
- Missing about 750,000 Floridians in the 2020 Census could cost the Sunshine State at least $11.4 billion in federal funding from 2020 through 2029, according to an estimate from the Florida Chamber Foundation.
- When inflation is mismeasured, interest rate decisions may be poorly timed, affecting borrowing, investment, and economic stability.
- The recent elimination of the Current Population Survey Food Security Supplement prevents policymakers and researchers from tracking changes to household food insecurity—just as the largest-ever cuts to food assistance (through SNAP) hit families and as food prices continue to rise.
The U.S. decennial census draws much attention from data advocates because of its far-reaching implications for federal funding. In FY21, census data guided the distribution of $2.8 trillion in funds to states, local and tribal governments, organizations, households and individuals. That’s crucial money for health care (Medicaid, Medicare, CHIP), nutrition and social services (SNAP, TANF, child welfare), education (including special education, Head Start), infrastructure and community development (highways, housing assistance), workforce development (job training, American Job Centers), and more.
A System Under Strain, Despite Its Low Cost
While the stakes have never been higher for accurate data, statistical agency budgets have remained essentially flat for years. Meanwhile, the obstacles these agencies face in producing reliable data have mounted:
Attempts to save money by further cutting statistical budgets are shortsighted. Even small budget cuts can force agencies to conduct surveys less frequently, reduce sample sizes, delay data releases, or provide less geographic detail, according to the National Academy of Sciences. These changes save relatively little money while degrading the quality of critical data.
What Protecting This Investment Means
High-quality federal statistics serve as critical infrastructure for democracy. They provide the shared, nonpartisan facts used to allocate funds, evaluate programs, guide economic policy, and hold America’s institutions accountable.
Policymakers can help by:
- Ensuring adequate funding that keeps pace with rising costs and new demands for data.
- Protecting agencies from budget cuts that save pennies while costing dollars in lost data quality.
- Investing in modernization to improve data collection methods and reduce long-term costs.
- Supporting transparency and methodological rigor that maintains public trust.
The modest investment in federal statistics generates big returns. We’d be foolish not to preserve this bargain for future generations.