A. Data Sources Guide

This appendix provides an overview of the primary data sources used throughout this book, along with guidance on how to access and interpret them.

Bureau of Economic Analysis (BEA)

The BEA, housed within the Department of Commerce, produces the foundational measures of the U.S. economy.

Key Products

Gross Domestic Product (GDP)

Industry Accounts

Regional Economic Accounts

International Transactions

How to Access

  • Interactive Tables: BEA's website offers customizable data retrieval

  • API: Free API access for programmatic retrieval (registration required)

  • FRED: Many BEA series available through Federal Reserve Economic Data


Bureau of Labor Statistics (BLS)

The BLS, within the Department of Labor, produces employment and price statistics.

Key Products

Current Employment Statistics (CES)

  • Monthly employment by industry (nonfarm payrolls)

  • Average hourly earnings

  • Hours worked

  • The "jobs report" released first Friday of each month

Quarterly Census of Employment and Wages (QCEW)

  • Employment and wages by industry and geography

  • Derived from unemployment insurance records

  • Covers 95%+ of jobs

Occupational Employment and Wage Statistics (OEWS)

Consumer Price Index (CPI)

Producer Price Index (PPI)

Productivity and Costs

How to Access

  • Data Retrieval Tools: BLS offers customizable series selection

  • API: Free API access for bulk retrieval

  • FRED: Most major BLS series available through FRED


Census Bureau

The Census Bureau, within the Commerce Department, conducts surveys and the decennial census.

Key Products

Economic Census (every 5 years)

  • Comprehensive establishment-level data by industry

  • Employment, payroll, revenue by NAICS code

  • Geographic detail to county level

Annual Business Survey

  • Annual updates between Economic Censuses

  • Business characteristics and innovation

County Business Patterns

American Community Survey (ACS)

Trade Data


Federal Reserve

The Federal Reserve System produces financial and monetary data.

Key Products

Financial Accounts of the United States (Z.1)

H.4.1 Release

  • Federal Reserve balance sheet

  • Weekly

H.8 Release (Assets and Liabilities of Commercial Banks)

Industrial Production and Capacity Utilization

Survey of Consumer Finances

FRED (Federal Reserve Economic Data)


Agency-Specific Sources

Energy

Energy Information Administration (EIA)

  • Comprehensive energy production, consumption, prices

  • State Energy Data System (SEDS)

Healthcare

Centers for Medicare & Medicaid Services (CMS)

Agriculture

USDA Economic Research Service

USDA National Agricultural Statistics Service (NASS)

Finance

FDIC

SEC EDGAR

Transportation

Bureau of Transportation Statistics (BTS)


International and Comparative Data

World Bank Open Data

OECD Statistics

UN Comtrade

IMF Data


Publication Calendar: Key Economic Data Releases

Knowing when data arrives---and how much it is revised---is essential for interpreting economic statistics in real time.

Release
Frequency
Typical Timing
Lag
Revision Magnitude

Employment Situation (BLS)

Monthly

First Friday of month

~5 days

±30,000-100,000 jobs in annual benchmark

GDP (BEA advance estimate)

Quarterly

~30 days post-quarter

30 days

±1.0-1.5 pp from advance to third estimate

GDP (BEA second estimate)

Quarterly

~60 days post-quarter

60 days

±0.5 pp from second to third

GDP (BEA third estimate)

Quarterly

~90 days post-quarter

90 days

Further revised in annual/comprehensive revisions

CPI (BLS)

Monthly

~15th of following month

15 days

Rarely revised; seasonal factors updated annually

PPI (BLS)

Monthly

~15th of following month

15 days

Minor revisions

Industrial Production (Fed)

Monthly

~15th of following month

15 days

Frequently revised

Personal Income and Spending (BEA)

Monthly

~30 days post-month

30 days

±0.2-0.5 pp on spending

JOLTS (BLS)

Monthly

~55 days post-month

55 days

Modest revisions

State GDP (BEA)

Quarterly

~6 months lag

180 days

Substantial revisions possible

Economic Census (Census)

Every 5 years

2-3 years after reference year

2-3 years

Benchmark data; not revised

Census of Agriculture (USDA)

Every 5 years

~2 years after reference year

2 years

Benchmark data

Revision Caveats

GDP estimates illustrate the revision problem. The advance estimate of Q1 2023 GDP growth was 1.1% (annualized); the third estimate revised this to 2.0%---a near-doubling. Annual comprehensive revisions can change GDP levels by tens of billions of dollars and alter the picture of entire business cycles. The lesson: treat early estimates as useful signals, not settled facts.

Seasonal vs. Non-Seasonally Adjusted Data

Most monthly and quarterly series are published in both seasonally adjusted (SA) and not seasonally adjusted (NSA) forms. Use seasonally adjusted data for tracking trends and comparing across months. Use non-seasonally adjusted data when analyzing a specific month's level or when seasonal patterns themselves are the object of study. Be aware that seasonal adjustment factors are re-estimated annually and can change historical data retroactively.

Tips for Using Government Data

  1. Check Vintage: Economic data is revised frequently. The first release differs from final estimates. Always note the release date of the data you cite.

  2. Understand Seasonal Adjustment: Most monthly/quarterly data is seasonally adjusted. Use adjusted figures for trend analysis, unadjusted for specific period comparisons.

  3. Note Geographic Coverage: Some series cover all establishments; others sample. The CES survey covers ~145,000 businesses; the QCEW covers 95%+ of all jobs. Sampling error increases at detailed geographic and industry levels.

  4. Mind NAICS Changes: The NAICS classification system is revised every 5 years. Historical comparisons may require concordances. The 2022 revision significantly reclassified parts of information and retail.

  5. Use APIs for Reproducibility: Programmatic access via APIs ensures your analysis can be replicated and updated.

  6. Beware of Coverage Gaps: The gig economy, informal work, and some financial activities are poorly captured by establishment surveys. The Quarterly Census of Employment and Wages misses self-employment entirely. The Current Population Survey captures it but with measurement error.


Further Resources

  • Data.gov: Central portal for federal open data

  • USAFacts: Non-profit aggregating government data

  • IPUMS: Harmonized Census and survey microdata

  • ICPSR: Academic data archive with many government series

Last updated