Why are estimates revised and when are they final?
On what basis are the industries in the Current Employment Statistics survey classified?
How are the data in the CES survey collected?
How are CES estimates developed?
What is a seasonally adjusted estimate?
Do hours and earnings statistics include overtime?
How are the estimates organized?
How can I get employment data for all private and public hospitals or schools?
What is a benchmark?
What is the UI universe count?
Why are the payroll survey estimates benchmarked to UI universe counts?
How does the benchmark revision affect the employment data for months prior to the benchmark month?
How does the benchmark revision affect the employment data for months subsequent to the benchmark month?
What are the causes of benchmark revisions?
What is the birth/death adjustment? Why is it used?
How are the birth/death adjustment amounts calculated?
How do strikes affect CES estimates?
(1) Why are estimates revised and when are they final?
Estimates are presented as soon as sufficient data have been collected to meet standards of accuracy and reliability so that they can be used to guide policy decisions. Aggregate level estimates (all 3-digit NAICS industry groups and above) are published with the first release of preliminary data, usually three Fridays after the survey reference week. At this point, about 72 percent of the sample have been collected and used in the estimates. One month later, when over 92 percent of the sample has been collected, estimates are published for the first time for all of the detailed industries, and the second set of preliminary estimates are published for the aggregate levels. The "first final" estimates are published the following month, when over 94 percent of the sample reports have been collected. These estimates, published the third month after the month of reference, are the official estimates until the next benchmark revision which is published each February.
All estimates, including annual averages, are subject to two revisions in connection with benchmarking, and seasonally adjusted series may be revised slightly three additional times, in conjunction with reseasonal adjustment. See the question on benchmarking in this section for further discussion.
(2) On what basis are the industries in the Current Employment Statistics survey classified?
A sample establishment in the CES survey is an economic unit, such as a factory, which produces goods or services. It is generally at a single location and engaged predominantly in one type of economic activity. Establishments reporting on the schedule (form BLS 790) are classified into industries based on their principal product or activity. Ideally, the principal good or service should be determined by its relative share of current production costs and capital investment at the establishment. In practice, however, it is often necessary to use other variables such as revenue, shipments, or employment as proxies for measuring significance. Industry classification, based on the North American Industry Classification System (NAICS) 2012, is determined from a supplement to the quarterly unemployment insurance tax reports filed by each employer. NAICS was developed through a cooperative effort between the United States, Mexico, and Canada. NAICS is based on a production-oriented concept in which industries with similar production processes are classified together.
(3) How are the data in the CES survey collected?
BLS as well as BLS Data Collection Centers collect data on employment, hours, and earnings from a sample of about 145,000 businesses and government agencies, which cover approximately 557,000 individual worksites drawn from a sampling frame of approximately 9 million Unemployment Insurance tax accounts. The active CES sample includes approximately one-third of all nonfarm payroll employees. Sample respondents extract the requested data from their payroll records, which must be maintained for a variety of tax and accounting purposes. Data are collected by telephone, touch-tone self response, computer-assisted interviews, fax technology, internet, and mail. The use of electronic media results in more rapid response times and higher response rates.
(4) How are CES estimates developed?
Data submitted on the 790 schedule are used in developing National, Statewide, and major metropolitan area estimates. All States' samples are combined to form a collective sample for developing National industry estimates. Statewide samples range from nearly 30,000 sample units in California to about 1,000 units in smaller States. It should be noted that State estimation procedures are designed to produce accurate data for each individual State. BLS independently develops National and State and area employment, hours, and earnings series and does not force State estimates to sum to National totals nor vice versa. Because each State series is subject to larger sampling and nonsampling errors than the National series, summing them cumulates individual State level errors and can cause significant distortions at an aggregate level. Due to these statistical limitations, BLS does not compile a "sum of States" employment series and cautions users that such a series is subject to a relatively large and volatile error structure.
(5) What is a seasonally adjusted estimate?
Seasonal adjustment removes the change in employment that is due to normal seasonal hiring or layoffs, thus leaving an over-the-month change that reflects only employment changes due to trend and irregular movements. Seasonally adjusted estimates of employment and other series are generated using the X-12 ARIMA program developed by the United States Census Bureau. This program adjusts estimates for fluctuations that occur on a regular basis within a year. For example, employment in Retail trade rises prior to the Christmas holiday season and then falls following the holiday. Annual averages, however, are computed using data that are not seasonally adjusted.
(6) Do hours and earnings statistics include overtime?
Yes, employers report payroll and hours including overtime. Overtime hours are published for Manufacturing industries only.
(7) How are the estimates organized?
The data are first separated by ownership — private and public. The public ownership is further divided into Federal, State, and Local. Each of these is then organized by industry (NAICS codes). Thus, for example, employment in all hospitals would be the sum of the estimates for Private, Federal, State, and Local hospitals. Federal government estimates also are published for the Department of Defense, the U.S. Postal Service, Ship building, Hospitals, and Other federal government.
(8) How can I get employment data for all private and public hospitals or schools?
See above answer.
(9) What is a benchmark?
The benchmark adjustment, a standard part of the payroll survey estimation process, is a once-a-year re-anchoring of the sample-based employment estimates to full population counts available principally through Unemployment Insurance (UI) tax records filed by employers with State Employment Security Agencies. By late September of each year, BLS completes preliminary tabulations of these universe counts for the first quarter of the year and routinely shares that information with the public.
(10) What is the UI universe count?
The Bureau's UI universe count is a quarterly tabulation from administrative records of the number of employees covered by UI laws. UI universe counts, available on a lagged basis, contain individual employer records for approximately 9 million establishments and cover nearly 97 percent of Total nonfarm employment; they thus provide a benchmark for the sample-based estimates. For the small segment of the population not covered by UI, BLS develops employment benchmarks from several alternative sources.
(11) Why are the payroll survey estimates benchmarked to UI universe counts?
The CES survey, like many other surveys, establishes benchmarks on a periodic basis in order to adjust its sample-based estimates to complete population counts available from administrative records.
Because of their much smaller size, sample surveys offer an ability to produce very timely estimates along with a greater ability to control the data quality of individual reports. There is a need, however, to recalibrate sample estimates periodically against full population counts. The use of a population count, or benchmark, allows a sample survey to adjust the results of estimation processes for new birth units in the population frame and to adjust for sampling and other nonsampling errors.
(12) How does the benchmark revision affect the employment data for months prior to the benchmark month?
Following standard BLS methodology, the March UI-based benchmark employment level replaces the March sample-based employment estimate, and then the difference between the benchmark level and the sample-based estimate is wedged back to the previous benchmark level. For example, the benchmark revision that was released in February 2013 replaced the March 2012 estimate with the benchmark level, increasing the employment level for that month by 424,000. To wedge this adjustment over the prior year, one-twelfth of the difference was added to April 2011, two-twelfths to May, and so forth, through February 2012 which received eleven-twelfths of the difference.
(13) How does the benchmark revision affect the employment data for months subsequent to the benchmark month?
Estimates for the period after the benchmark month (the post-benchmark period) are calculated for each month based on the new benchmark level, new net birth/death figures, and the annual sample update, which is implemented in November following the benchmark month.
(14) What are the causes of benchmark revisions?
In general, differences between universe counts and sample-based estimates result from both sampling and nonsampling error. Although sampling error is present in the payroll survey, as it is in all surveys, the CES sample is so large that sampling error is not usually an important factor in explaining the differences.
Nonsampling error arises in the survey estimates and in the universe counts from both the UI and the alternative sources used to establish the noncovered population benchmarks. Nonsampling error is a more significant cause of benchmark revisions. Sources of nonsampling error include coverage, response, and processing errors in both data series. Additionally, the survey is potentially subject to sample design and estimator biases.
(15) What is the birth/death adjustment? Why is it used?
To derive a complete count of Total nonfarm employment, a two-part estimator is required. First, a sample-based estimate of the over-the-month employment change is made using the CES sample, which represents about 557,000 business establishments. The sample is drawn from the population of all employers who have filed Unemployment Insurance tax returns. The sample does not include employers who have recently formed new businesses but who have not yet been added to the Unemployment Insurance tax files. Business births occur every month, and failure to include an estimate for these units would result in a consistent underestimation of employment totals, that is, a downward bias. Therefore, BLS utilizes a model-based technique to estimate for this part of the population.
In a dynamic economy, firms are continually opening and closing. These two occurrences offset each other to some extent. That is, firms that are born replace firms that die. CES uses this fact to account for a large proportion of the employment associated with business births. This is accomplished by excluding such business death units from the matched sample definition. Effectively, business deaths are not included in the sample-based link portion of the estimate, and the implicit imputation of their previous month's employment is assumed to offset a portion of the employment associated with births.
There is an operational advantage associated with this approach as well. Most firms will not report that they have gone out of business; rather, they simply cease reporting and are excluded from the link, as are all other nonrespondents. As a result, extensive follow-up with monthly nonrespondents to determine whether a company is out of business or simply did not respond is not required.
Employment associated with business births will not exactly equal that associated with business deaths. The amount by which it differs varies by month and by industry. As a result, the residual component of the birth/death offset must be accounted for by using a model-based approach.
(16) How are the birth/death adjustment amounts calculated?
During the net birth/death modeling process, simulated monthly probability estimates containing continuous and imputed employment over a 5-year period are created and compared with population employment levels that contain actual business births and deaths along with the continuous units. Moving from a simulated benchmark, the differences between the series across time represent a cumulative error component. Those residuals are converted to month-to-month differences and are used as input series to the modeling process.
Models are fit using X-12 ARIMA. Outliers, level shifts, and temporary ramps are automatically identified. Five models are tested, and the model exhibiting the lowest average forecast error is selected for each series.
(17) How do strikes affect CES estimates?
Anyone paid for working any portion of the reference pay period (pay period that includes the 12th of the month) is counted as employed. Therefore, to be counted as not employed for purposes of the CES survey, a person on strike or strike-related layoff must not receive pay for the entire reference pay period.
Average Weekly Hours (AWH) and Average Hourly Earnings (AHE)
These are hours for which employees are paid for work or on paid leave for the reference pay period (including paid vacation, holidays, sick leave, or other paid leave).
When strikers or laid off employees work part but not all of the reference pay period, then they are counted as employed according to the CES survey but with reduced hours. The magnitude of the reduction on average weekly hours depends on the proportion of employees in the industry's sample with reduced hours and the number of hours they worked.
Employees who are on strike or layoff for the entire reference pay period do not have any effect on the average weekly hours estimate unless their normal hours differ significantly from the average for the industry. Similarly, average hourly earnings estimates will be little affected unless the normal hourly earnings of those on strike or layoff differ significantly from the average for the industry.
January with reference week Sunday 1/6 to Saturday 1/12
February with reference week Sunday 2/10 to Saturday 2/16
Company A strike: Strike/layoff activity. (This company has a weekly pay period.)
|1/4||1||strike||2,000||on strike the whole reference pay period|
|1/8||2||strike||1,500||on strike part of the reference pay period|
|1/13||3||layoffs||3,000||laid off after the reference pay period|
The strike is settled February 19. All employees are called back to work February 20.
Effect on January employment:
over-the-month change lowered by 2,000
Effect on February employment:
over-the-month change lowered by 4,500
Effect on January AWH:
reduced slightly by the 1,500 on strike part of the reference pay period*
Effect on February AWH:
the January effect is reversed because the employees with shorter hours in that month are off payrolls*
* Both the January and February AWH and AHE also could be affected if strikers' normal hours and/or hourly earnings differ significantly from industry average.
Note: For confidentiality reasons, CES staff cannot provide company-specific information, including dates or employees involved in strike/layoffs, other than what is already publicly available at the time of the strike. Contact the company or news sources for more specific information.
Last Modified Date: February 1, 2013