Methodology & Data Sources — US Federal Data
How we collect, process, and present federal economic data. Every claim we make about the data is documented here.
Core Principles
GOVSPENDING.ORG aggregates data from official U.S. government sources. We do not generate estimates, produce forecasts, or apply proprietary models to the underlying data. When we present a number, it comes directly from a federal agency's published dataset.
Our role is to normalize formatting, unify presentation across agencies, and make government data more accessible through search, charts, and export tools. We add context through titles, units, and frequency labels — but the data itself is unmodified.
Data Sources
Federal Reserve Economic Data (FRED)
Federal Reserve Bank of St. Louis
Over 800,000 U.S. and international time series from dozens of sources. The most comprehensive publicly available macroeconomic database.
FRED aggregates data from over 100 sources including the Bureau of Economic Analysis, Bureau of Labor Statistics, Census Bureau, and many others. We query the FRED API for both series metadata and observations. Date ranges begin at 2014-01-01 by default but can be extended. Seasonal adjustment status is preserved from the original source.
Bureau of Labor Statistics
U.S. Department of Labor
Primary source for employment, unemployment, prices, compensation, and working conditions data. Produces the Consumer Price Index and jobs reports.
BLS data is retrieved via the Public Data API v2. Series data includes monthly observations with period codes. We convert BLS period codes (M01-M12) to standard date formats. Footnotes from the BLS response are preserved when available. Rate-limited to 25 queries per 24-hour period without a registration key.
U.S. Treasury Fiscal Data
U.S. Department of the Treasury
Federal debt, receipts, outlays, interest rates on government securities, and Treasury auction data. The official record of the federal government's financial position.
Treasury Fiscal Data API provides daily debt figures, monthly treasury statements, and average interest rate data. No API key is required. We query the "debt to the penny" dataset for daily total public debt, debt held by the public, and intragovernmental holdings. Data is sorted by record date with the most recent values first.
Congress.gov API
Library of Congress
Legislative data including bills, amendments, committee actions, voting records, and member information for current and historical Congresses.
The Congress.gov API, maintained by the Library of Congress, provides structured data about bills, amendments, and legislative actions. We retrieve recent bills with their latest action dates and statuses. Bill text and full amendment details are available on Congress.gov directly.
Data Processing
Date Normalization
All dates are normalized to ISO 8601 format (YYYY-MM-DD). Monthly series use the first of the month. Quarterly series use the first day of the quarter. This ensures consistent sorting and charting across sources with different native date formats.
Missing Values
FRED observations with value "." (indicating missing or unavailable data) are excluded from our datasets. When gaps exist in a time series, they are not interpolated — we show only what the source provides. Charts skip missing data points rather than showing zeros.
Derived Calculations
Year-over-year percentage changes shown in KPI cards and dashboards are calculated from the raw index values: ((Current - YearAgo) / YearAgo × 100). Period-over-period changes use the same formula with consecutive observations. These calculations are clearly labeled as computed values.
Caching
API responses are cached server-side for 1 hour to reduce load on federal data services and improve page load performance. Timestamps shown on dashboards reflect the last successful data fetch. Stale cache is returned as a fallback if a live fetch fails.
Exports
CSV and JSON exports contain the raw data as received from the source (after date normalization and missing value exclusion). Export filenames follow the convention: source_indicator_YYYYMMDD.format. Metadata headers are not included in CSV exports — they contain only the column headers and data rows.
Caveats & Limitations
Revision risk: Government economic data is routinely revised. Monthly employment figures, GDP estimates, and CPI calculations are all subject to revision — sometimes months after initial release. We display the most recently published value, which may differ from the "advance" or "preliminary" estimate.
Lag time: Federal data is released on official schedules. Monthly indicators (like CPI or employment) typically reflect conditions from the prior month. Quarterly data (like GDP) reflects the prior quarter. This site shows data as fast as the sources publish it, but there is always an inherent reporting delay.
Seasonal adjustment: Many series are available in both seasonally adjusted and not seasonally adjusted forms. We default to seasonally adjusted versions where available, as they better reflect underlying trends. The adjustment methodology is determined by the publishing agency, not by us.
API availability: Federal data APIs may experience downtime, rate limiting, or service interruptions outside our control. If a data source is temporarily unavailable, cached data is served with a note about the cache age.
Not financial advice: This site provides publicly available government data for informational purposes. Nothing here constitutes financial, investment, legal, or policy advice. Consult qualified professionals for decisions that depend on economic data.
Questions about our methodology?
If you find a discrepancy between our data and the original source, or have questions about how we process specific datasets, please open an issue on the project repository. We take data accuracy seriously.