Understanding the Labor Department’s Jobs Day Reports
There’s a moment each month when financial markets hold their collective breath: 8:30 AM Eastern Time in the United States. That is when the U.S. Bureau of Labor Statistics (BLS) releases its latest estimates on jobs and unemployment.
This so-called first Friday “Numbers Day” can affect not only financial markets but also federal economic policy, as the White House and the Federal Reserve closely follows these data. Indeed, so important are the estimates to economic decisions that many financial markets worldwide take an unofficial pause just prior to 8:30 AM on Numbers Day.
BLS publishes summaries of two surveys on “Numbers Day”: the Current Employment Statistics (CES) and the Current Population Survey (CPS). The CES gives us the net number of payroll jobs created last month and a lot more industry details; the CPS gives us the unemployment rate, among many other labor force statistics.
Key Data Found in the Two Surveys Behind the Numbers Day Report
| Current Employment Statistics | Current Population Survey |
|---|---|
| Total Number of Monthly Job Changes | National Unemployment Rate |
| Average Weekly Hours | Labor Force Participations Rates |
| Average Hourly Earnings | Total Size of the Labor Force |
| Total Job Changes by Industrial Sector | Unemployment Rates by Age and Sex |
| Government Job Changes | Unemployment Rates by Race and Ethnicity |
| Changes in Number of Women Employees | Characteristics of People Not in the Labor Force |
Less appreciated, however, are the enormous differences between the two surveys, and it is worth spending a little time understanding what each one tells us.
The Current Employment Survey
The CES is a monthly survey of about 120,000 businesses and government organizations that contain about 630,000 work sites. These firms constitute a representative sample of the over 11 million firms that employ workers covered by state and federal unemployment insurance laws. Each firm is classified by a code that tells data users about the firm’s industry: for example, construction, manufacturing, or health care.
- Firms that participate in the CES survey are asked to complete their electronic questionnaire for the pay period containing the 12th day of the month. This is called the reference period. BLS asks questions about the total number of employees, women employees in that total, the pre-tax payroll for all employees and production employees, and their hours. BLS collects these data at two electronic data collection centers, which send their survey returns to the BLS headquarters in Washington, DC, where BLS analysts process the information for publication. BLS multiplies the data it processes by weights or factors to get to estimated national numbers. For example, the total number of jobs created in the previous month is a national estimate based on the weighted sample. These estimates are published each month in The Employment Situation report at 8:30 AM on the first Friday of the month.
- Careful readers of the monthly jobs report will notice that the CES estimates are revised for two subsequent months. For example, the January 2024 estimate will be revised in February and then again in March. BLS makes these revisions because not all the participating firms get their reports in on time. Somewhere around 70 percent of reports come in the first month, another 15 percent in the second month, and another 5 to 10 percent in the third month. Thus, by the third revision, about 95 percent of the respondents have reported. This percentage is called the response rate.
- All the monthly employment totals from the CES are also revised once a year, usually in January. Annual revisions are necessary primarily due to estimates made throughout the year about the number of new firms in the economy that did not exist when BLS created the most recent representative sample. Thus, BLS reconciles to more current data about all firms once a year to add or subtract created jobs from its monthly estimates. Usually, the revisions are well under 0.3 percent of total employment.
The Current Population Survey
The CPS surveys households, not firms, each month to determine the labor force characteristics of everyone in a household who is 16 years of age or older. The CPS asks questions (usually in person) about individuals in approximately 60,000 households. BLS and the Census Bureau select this sample from Census records to reflect the demographic, geographic, and labor force characteristics of the 134 million
U.S. households. Unlike the CES, which is collected directly by BLS, the CPS is collected by the Census Bureau under a contract with BLS and is fielded in the middle of each month.
- The CPS digs deeply into household labor activity. It asks if individuals are working or have looked for work in the past four weeks (and also are available for work). If the answer is yes to either of these questions, then that individual is considered in the labor force. If no, then broadly speaking that individual is not in the labor force. A person who answers “yes” to working in the past four weeks can be self-employed, on someone’s payroll, or involved in a business that pays partnership or income shares, etc.
- A person who is looking for work and available if a job comes up may or may not also be getting payments from a government unemployment insurance fund. That is why the total number of unemployed in the CPS rarely agrees with the total nationwide number of individuals receiving unemployment benefits. Also worth noting is that the CPS unemployment rate (technically the U3 unemployment rate) is the total number of people who are looking for work divided by the labor force (the sum of those working and looking for work). That means an unemployed person in CPS may not have applied for unemployment insurance.
- The CPS asks a host of questions about such individual characteristics as race, sex, ethnicity, educational level, age, disability status, and military service. These demographic data provide policy makers and researchers insights on how key segments of the labor force are faring. In addition, the CPS asks people working part time if that is because full time work is unavailable or because part-time work better suits their current needs. CPS asks those who are unemployed why they have left employment, which is where we get our monthly data on temporary and permanent layoffs and people who are in between employment. It is worth noting that CPS interviewers ask about a person’s country of birth, but they do not ask questions on immigrant status or citizenship.
- The CPS is revised only once a year when BLS checks the latest Census Bureau estimates of the U.S. population to see if BLS’s estimate of the civilian non- institutional population is correct. This population is aged 16 and above and excludes active duty members of the armed forces, those in prisons and jails, and individuals living in skilled nursing facilities. These population revisions do not affect rates, such as the unemployment rate, but they do affect such population concepts as the size of the labor force, the number of unemployed, the number of people employed, and ration of employment to population. See the January 2024 Employment Situation for details.
Policymakers and their advisors who know the difference between these two labor market surveys will more consistently choose the right data for the arguments they are making and the policy goals they want to achieve. They will, as well, discover extraordinary detail on American workers that the BLS provides each and every month.
