Three's Company: There are good reasons that the Labor Market Information Office publishes three different sets of employment data.

By John Berglund
May 2008

PDF of article (4 pages)

The Labor Market Information (LMI) Office of the Minnesota Department of Employment and Economic Development (DEED) publishes three different employment series — the Local Area Unemployment Statistics series (LAUS), the Current Employment Statistics (CES) series, and the Quarterly Census of Employment and Wages (QCEW) series. They don’t match. What’s going on?

The short answer is: They measure different things. The LMI Office is funded by the Department of Labor to produce the three series because the federal government wants to see the fullest picture possible of the economy. Our job at the Minnesota LMI Office is to produce the best possible data and then send it to Washington where it is published as part of the national picture. Once it is approved for release nationally, Minnesota and other states can publish their state data.

Local Area Unemployment Statistics (LAUS)

The LAUS employment figures are the most inclusive of the three series we produce.  The program estimates how many people living in Minnesota worked or were actively looking for work, no matter what sector or type of work, during the week that includes the 12th of the month. These data include the following workers who are not included in the other series: independent contractors, railroad employees, elected officials, unpaid family workers and self-employed individuals including farmers. Military personnel and those in institutions are not included, nor is anyone younger than 16. It also includes the unemployed who are either separated from a job, coming into the labor force for the first time or coming back into the labor force after a period away. This is the one series that reports by worker residence rather than job site.

The data we collect are published as part of the national Current Population Survey (CPS), which comes out monthly. The raw survey data are very good on the national level, but at the state level there are often significant variations because of the smaller sample size. To avoid the problems associated with small sample size, econometric methods using local data, including claims for Unemployment Insurance and job growth, are used to improve the CPS estimates. This is especially important since the “local area” part of the LAUS name refers to county and city-level estimates which are used extensively in Minnesota by a wide variety of customers.

The LAUS employment figures are published monthly, usually on the third Tuesday of the month in Minnesota, and can be found on the Internet at www.deed.state.mn.us/lmi/tools/laus/default.aspx.

Current Employment Statistics (CES)

The Current Employment Statistics (CES) program is the product of a monthly survey of businesses in nonagricultural industries.  The Bureau of Labor Statistics began collecting employment and payroll information in 1915, but only for a few manufacturing industries. This grew into the CES program, which collects employment data from a sample of all employers in the state who produce goods or services. The data are collected and published by the location of the business. Full-time, part-time and temporary employees are all counted as employed if they were on the payroll, in other words on the job or on paid leave from the job. The military is not included in this survey, and the series does not include the self-employed, unpaid family workers or the agricultural sector. As with the LAUS program, the employment, hours, and wages reported here are for the week that includes the 12th of the month.

The CES data are collected on a monthly questionnaire sent to the employers in the sample or by phone. Nationally the sample includes 160,000 employers. In Minnesota estimates are based on data from about 4,800 employers each month.  One of the most valuable aspects of this series is that the data can be published so promptly.  This has made the series a high profile economic indicator which is reported widely in the media both in Minnesota and nationwide.

The Bureau of Labor Statistics is implementing some changes to this series.  One is to collect and publish hours and wages for all employees, not just production workers as has been the practice until now. Another change will be to collect gross wage data for the entire month rather than the pay period that includes the 12th. There are other minor changes, but these will make the biggest difference in the data available to users.  Data from these new standards will be published beginning in February of 2008.  To read more about these changes go to "What's New in the 2008 Current Employment Statistics Program?".

The CES data are usually available on the second Tuesday of each month at  www.deed.state.mn.us/lmi/tools/ces/default.aspx. 

Quarterly Census of Employment and Wages (QCEW)

The Quarterly Census of Employment and Wages (QCEW) is the series that provides the most detailed employment data by industry and geography. This data series is compiled from administrative reports for every employer covered by Unemployment Insurance in the state of Minnesota and therefore reports about 97 percent of all the employment in Minnesota. Because data are gathered from every employer represented in the data set, these data can be sliced and diced in much more detailed ways -- by specific industry, firm size or geographic location, for example -- than is possible with smaller survey-based samples. Moreover, one of this series’ many functions is to serve as the universe from which survey samples are drawn and to which survey-based estimates are benchmarked (i.e. adjusted). 

There are some restrictions and drawbacks to the use of these data, however. First, these data are collected quarterly and must be edited and summarized before they can be published. This means that the data are not published until approximately six months after the period they represent. Second, being based on individual employers’ tax reports, they fall under statutory data confidentiality rules. 

Data for QCEW come from the reports employers submit with their quarterly unemployment tax payments. These reports include summarized employment in the pay period that includes the 12th of each month, total wages paid for the quarter, and the amount of those wages covered by the state’s unemployment tax law. These reports are not due in an unemployment agency’s office until the last day of the month following the quarter to which the report refers.

Minnesota is in the vanguard of states trying to get the unemployment tax report in electronically. Every employer is now required to submit total wage data online for each employee every quarter. The agency calculates the taxable wage figure and the contributions due and bills employers. The data are then summarized, edited, analyzed and published for the QCEW report.

As with CES , the data published for the QCEW are based on employer location. The data are quite detailed in that we collect by individual location for employers who have multiple plant locations in Minnesota. This is important for county or community data since, for example, if we published the post office as one location, St. Paul would look like it has far more jobs than it actually has only because the post office is headquartered in St. Paul. In fact, each post office is reported to the community in which it operates as is every Target, every Walgreens, and so on.

Coverage has also increased since the source of these data, the unemployment tax files, was first developed and the industrial coding system that is used to code and aggregate the data has changed. This means that data are not comparable across all years for which they are available. Below is a summary of the changes to the law that affected the totals reported.

The data also were affected by the change from Standard Industry Classification (SIC) codes to the North American Industry Classification System (NAICS) in 2000. Minor revisions were made in 2005 and 2007 as well.  These changes are an example of what are called “non-economic code changes” because they cause the employment level in the industry to change but are not the result of a change in the economy, only the coding structure. If you need to compare data from before 2000 to data after 2000, check carefully to be sure the data are comparable or if they can be adjusted.

The data are posted about five months after the quarter to which they refer at  www.deed.state.mn.us/lmi/tools/qcew/default.aspx.

Conclusion

The three employment series regularly produced by the LMI Office are complementary.  LAUS data are collected by place of residence and include some types of employment not included in the others. The series is available monthly for state, county and large cities. CES data, also available monthly, are summarized by industry and plant location.  The data are available at the total state and metropolitan statistical area (MSA) level. The employment data include railroad employment and students working for the college they are attending, but not self-employment. The QCEW covers 97 percent of all the employment in Minnesota. It comes out five to six months after the quarter to which the data refer but includes very detailed information, including data at the six-digit NAICS level and at a very fine geographic level. What is published is somewhat limited by the state’s data privacy laws.

Top