Useful Labor Data Links

Here are some useful links to publicly-available labor market data sources, many of which I use (and some I have even helped create!)

Survey of Consumer Expectations (Federal Reserve Bank of New York)

The Main SCE Survey

• FRBNY Survey of Consumer Expectations (SCE):  a monthly survey of ~1,200 heads of household per month that started in 2013. Individuals are in the survey longitudinally for 12 months and respond via an online, nationally representative survey. The survey asks a broad range of questions on expectations about prices, the labor market, the housing market, and other aspects of the economy. It also includes basic demographic and economic information about the individual and their household. 

  • Details on the monthly SCE, including its representativeness and survey design, are in "An Overview of the Survey of Consumer Expectations," NY Fed Economic Policy Review, 2017, by Olivier Armentier, Giorgio Topa, Wilbert van der Klaauw, and Basit Zafar.

Our Supplement to the SCE

• SCE Job Search Supplement (available via the SCE Data Bank): an annual supplement to the monthly SCE developed by myself, Andreas Mueller, Ayşegül Şahin, and Giorgio Topa. We have fielded the survey each October since 2013. The survey asks respondents a broad range of questions on their labor market experiences. The questions ask whether respondents looked for work, how much effort they put into it, the types of jobs they are looking for, and the characteristics of any job offer received. They also ask about respondents' reservation wage, desired work hours, and views on job disamenities, as well as their work histories and the usual labor market and demographic information included in the CPS.

  • Details on SCE Job Search Supplement survey design and questions are in our article, "Job Search Behavior Among the Employed and Non-Employed" Econometrica, 2022, with additional detail in its Supplemental Appendix.

  • The SCE Job Search Supplement Codebook is here (and also in the FRBNY SCE Data Bank).

Other Labor-Related SCE Supplements  (also available via the SCE Data Bank)

• SCE Labor Market Supplement: This is a regular quarterly supplement to the monthly SCE that asks respondents a range of more detailed labor market questions, including several from the annual Job Search Supplement.

• SCE Informal Work Participation Supplement: This is another annual supplement to the SCE fielded between 2013 and 2016 in conjunction with economists from the Federal Reserve Bank of Boston. The survey asks a range of questions on informal and contingent work arrangements, along with additional questions on respondents' labor market and household income situation.

U.S. Bureau of Labor Statistics

The Basics

• Current Population Survey (CPS) (a.k.a., the "Household Survey"): monthly survey of ~60,000 households per month. Individuals are in the survey longitudinally for two 4-month periods, with an 8-month gap between each period. The National Bureau of Economic Research also maintains a repository of the monthly micro data (and CPS supplements) here.

• Current Employment Statistics (CES) (a.k.a., the "Payroll Survey"): monthly survey of ~400,000 establishments per month. The survey has data on aggregate employment, average weekly and hourly earnings, and average weekly hours by detailed industry and region. 

Additional Establishment Surveys

• Job Openings and Labor Turnover Survey (JOLTS): monthly survey of ~16,000 establishments per month. Reports job vacancies, hires, and separations (by quits, layoffs, and other separations)  by broad industry, region, and establishment size class categories. The sample is nationally representative and drawn from the CES respondents, so the JOLTS and CES are directly comparable in terms of sample representativeness.

• Occupational Employment and Wage Statistics (OEWS) (formerly OES): semi-annual survey of ~400,000 establishments each period, with a full survey frame of ~1.1 million establishments over an 18-month period. The survey has detailed data on employment and wages (including distributional wage statistics) within occupations by detailed industry and geography.

Administrative Data

• Quarterly Census of Employment and Wages (QCEW): quarterly administrative data on the employment and payrolls of all establishments with at least one W-2 employee. The data cover roughly 98% of all U.S. workers and are the sampling frame for the CES, JOLTS, and nearly all other BLS establishment-based surveys. Published statistics include data on employment and payrolls by detailed industry, county, and establishment size class.

• Business Employment Dynamics Data (BED): a version of the QCEW data that are longitudinally linked over time, allowing the BLS (and researchers) to estimate statistics on gross job creation and gross job destruction, along with statistics on establishment openings, closings, expansions, and contractions, for nearly all U.S. establishments. Its coverage, frequency, and scope are identical to the QCEW data, though the BLS only publishes statistics for the private sector.

U.S. Census Bureau 

Administrative Data

• Business Dynamics Statistics (BDS): annual administrative data on employment, payrolls, and selected output statistics for nearly all establishments in the U.S. These data are the basis for the annual Census County Business Patterns estimates. The Census Bureau also uses these data to estimate annual statistics on gross job creation and job destruction and entry and exit at both the firm and establishment level. There is a wide variety of publicly available estimates by business age, size, industry, and geography, with most estimates going back to 1978. 

• Longitudinal Employer-Household Dynamics (LEHD): This is the most comprehensive matched employer-employee data set for the U.S. It uses quarterly administrative records on incorporated businesses and W-2 wage and salary employees to track the employment histories of workers across businesses over time. The program is the result of cooperative agreements between the Census Bureau and most U.S. states, so its coverage includes data for all participating states  back to the 1990s (in most cases). The program produces several sets of economic statistics on worker and business dynamics, the most notable of which is its Quarterly Workforce Indicators (QWI). 

Individual and Household Data

• IPUMS Census and American Community Survey (ACS) Data: a website with an extensive collection of microdata and aggregate statistics for the decennial Censuses from 1790 to 2010 and American Community Survey data from 2000 to the present.

Accessing Confidential U.S. Microdata

Both the BLS and Census Bureau have procedures in place for researchers to access the underlying confidential microdata for most of the above programs.  Both agencies require a detailed proposal for a specific project that will use one or more confidential data sources. Some details are below, though the U.S. Federal Government recently streamlined its confidential data access application process to work through a single portal at ResearchDataGov.org. At this site, you can browse all the data sets one can apply to access across multiple agencies (including BLS and Census) through the website, and use the Standard Application Process to request access to the data (though I would advise contacting the agency/data program directly to make sure you include all relevant information in your research proposal).  

• BLS Confidential Microdata Access: Researchers can submit a proposal to access a wide range of confidential BLS micro data for a specific research project. Projects must be statistical in nature and benefit or further the mission of the BLS in some way. Most projects require researchers to access the microdata onsite at the BLS' Washington, DC office, though selected data are available through the Federal Research Data Center (RDC) network. More information on the specific datasets available, and how one can access them, are here.

• Census Bureau Restricted-Use Microdata Access: Researchers can submit proposals to use confidential Census microdata in an approved research project. The project must be research-oriented in nature and provide some benefit to the Census Bureau. In addition, researchers working directly with the microdata will have to receive Special Sworn Status prior to starting the project. Researchers can usually access the micro data at one of the Census Bureau's Research Data Centers (RDCs).