The establishment payroll survey, known as the Current Employment Statistics (CES) survey, is based on
a survey of approximately 147,000 businesses and government agencies representing approximately 634,000 worksites throughout the United States. The primary statistics derived from the survey are monthly estimates
of employment, hours, and earnings for the Nation, States, and major metropolitan areas. Preliminary National
estimates for a given reference month are typically released on the third Friday after the conclusion of the
reference period in conjunction with data derived from a separate survey of households, the Current Population
Survey (CPS). The reference period for the CES survey is the pay period which includes the 12th of
The Current Employment Statistics (CES) program is a monthly survey of business establishments (or jobs). CES
produces estimates on the number of employees on nonfarm payrolls, average hourly earnings, average weekly earnings,
and average weekly hours. The Current Population Survey (CPS) is a monthly survey of households (or people). The
household survey produces estimates about the labor force, the employed, the unemployed, the unemployment rate, and
demographic information about the employed and unemployed. More information about the differences between the CPS
and the CES surveys is available at www.bls.gov/web/empsit/ces_cps_trends.pdf.
The CES and Quarterly Census of Employment and Wages (QCEW) programs are related but do not report the exact same
information at the same frequency. The QCEW program publishes a quarterly count of employment and wages covering 98
percent of U.S. jobs, available at the county, Metropolitan Statistical Area (MSA), State, and National levels by industry. The CES program surveys
about 147,000 businesses and government agencies, representing approximately 634,000 worksites, in order
to provide detailed industry data on employment, hours, and earnings of workers on nonfarm payrolls on a monthly basis.
Unemployment Insurance (UI) tax reports, submitted by nearly all businesses in the U.S., are used as both the input
data for QCEW data and as the majority of the sample frame for the CES survey and cover almost all private industries
and government agencies. CES employment figures are benchmarked each year in large part using data from the QCEW program
because both programs use the pay period including the 12th of the month as the reference period for employment.
QCEW quarterly wages include total compensation paid during the calendar quarter to all workers;
CES hours and earnings data are reported for all employees and for production or nonsupervisory
employees in private industry who received pay (whether they worked or not) during any part of the pay period that includes the 12th day of the month. CES earnings do not include irregular bonuses or retroactive pay.
CES data are published 3 weeks after the week that includes the 12th of the month, typically the first Friday of the
following month. QCEW data are published much later, approximately 6 months after the end of the reference period.
The household survey administered by the Current Population Survey (CPS) program and establishment survey
administered by the Current Employment Statistics (CES) program both produce sample-based estimates of employment
and both have strengths and limitations. The establishment survey employment series has a smaller margin of error
on the measurement of month-to-month change than the household survey because of its much larger sample size. An
over-the-month employment change of about 100,000 is statistically significant in the establishment survey, while
the threshold for a statistically significant change in the household survey is about 400,000. However, the household
survey has a more expansive scope than the establishment survey because it includes self-employed workers whose
businesses are unincorporated, unpaid family workers, agricultural workers, and private household workers, who are
excluded by the establishment survey. The household survey also provides estimates of employment for demographic groups.
The Current Employment Statistics program administers the establishment survey and uses the data collected to produce monthly nonfarm payroll estimates.
The types of data produced include the following:
Production or nonsupervisory employees (depending on industry)
Average weekly hours
Average hourly earnings (constant dollar and current dollar)
Average weekly earnings
Average overtime hours in manufacturing
Indexes of aggregate hours and payrolls
All data are available not seasonally adjusted, and some data are available seasonally adjusted.
The Employment Situation report is released monthly, typically on the third Friday after the conclusion of the
reference week, which is the week that includes the 12th of the month. A table of upcoming
Employment Situation release dates is available at www.bls.gov/ces/ces_tabl.htm.
Current Employment Statistics employment data date back to 1939 at the highest levels of aggregation. Most detailed
industries only date back to 1990, and all employee hours and earnings series began in 2006. For a complete list of start
dates for NAICS-based all employee series, see www.bls.gov/web/empsit/cesseriespub.htm.
Additionally, discontinued SIC-based employment estimates are available from 2003 going back to 1964 and in some cases
as far back as 1939 or 1919. These series are not comparable with current CES NAICS-based series. To access
these discontinued CES data, go to www.bls.gov/ces/data.htm. For more information about SIC coding of industries, visit www.bls.gov/ces/cesnaics.htm#3.2.3.
Manufacturing and other goods-producing industries were the primary focus of early industry data produced by the BLS.
Therefore, hours and earnings time series for these industries have a relatively long history. Data for manufacturing
and its broadest industry categories begin in 1939, and for mining and construction, in 1947. Prior to 1964, however,
the collection of hours and earnings data for the service-providing sector was limited to a few select industries. Given
this incomplete industry coverage along with the size of the services portion of the economy, it was not possible to
compute hours and earnings estimates for all of the private industries. Beginning in 1964, the collection of hours and
earnings data for services was expanded to an extent sufficient to compute total private hours and earnings estimates.
These estimates were first published in 1967.
Discontinued SIC-based hours & earnings estimates are available from 2003 going back to 1964 and in some cases as far
back as 1947 or 1939. These series are not comparable with current CES NAICS-based series. To access these
discontinued CES data, go to www.bls.gov/ces/data.htm. For more information about SIC coding of industries, visit www.bls.gov/ces/cesnaics.htm#3.2.3.
CES draws its sample and sets its benchmark employment level from the business establishment list maintained by the Quarterly Census of Employment and Wages (QCEW) program. This universe for business establishments is based on Unemployment Insurance (UI)
administrative records, so workers who are not covered by UI will not be captured. In Agriculture there are numerous exemptions
to requirements for UI coverage, making the sample frame for Agriculture insufficient for calculating statistically sound
estimates. In addition, a substantial number of Agricultural enterprises are self-proprietorships, which are out of scope for
the CES survey.
Historically, the U.S. Department of Agriculture's Census of Agriculture has been the primary survey used to measure farm
labor. The Census of Agriculture is available at www.agcensus.usda.gov/index.php.
The Current Employment Statistics (CES) program is examining the feasibility of publishing monthly CES employment,
hours, and earnings estimates by firm size. Currently, BLS publishes the first preliminary CES employment estimates
for a given month at selected industry detail. Subsequent estimates for that month are published in more industry
detail with the following month's first estimates. Research suggests that the available sample may make it feasible
to publish monthly size-class employment estimates by major industry sector together with the first preliminary
estimates. Employment change by firm size would add a valuable dimension of detail to understanding current employment
trends. More information about the experimental CES size class series is available at
Some data by size of establishment are also available from the Quarterly Census of Employment and Wages
CES estimates are categorized by ownership and industry. Respondents are assigned an ownership code — private or public
with public ownership further divided into Federal, State, or local. Respondents are then assigned a North American Industry
Classification System (NAICS) code. NAICS codes group establishments into industries based on the activity in which they are primarily
engaged. Establishments using similar raw material inputs, similar capital equipment, and similar labor are classified in the same
industry. More information about NAICS codes in general is available at www.bls.gov/bls/naics.htm.
More information about NAICS codes in the CES program is available at www.bls.gov/ces/cesnaics.htm.
CES survey respondents are categorized by Unemployment Insurance (UI) accounting code, location, ownership,
size, and reporting unit. Each business has also been categorized into a certain industry by
North American Industry Classification System (NAICS) code. Establishments are stratified by UI account number for the
purpose of sample allocation and selection. The sample strata, or subpopulations, are defined by State, metropolitan statistical area, industry,
and employment size, yielding a state-based design. Thirteen
industries (treating manufacturing as one industry and not including government) and 8 size classes result in
104 total allocation cells per state.
After 60 years of use, CES retired the SIC system and replaced
it with the NAICS. NAICS is the product of a collaborative effort between the United States (U.S.), Canada, and
Mexico. A classification system shared across the three countries allows direct comparison of economic data across
borders in North America.
NAICS codes are not related to SIC codes; rather NAICS is a completely redesigned way of coding industries. NAICS
recognizes hundreds more businesses than SIC did, largely in the fast-growing service sector.
The U.S. Census Bureau issued a notice, available at
www.census.gov/eos/www/naics/federal_register_notices/notices/fr09ap97.pdf, making NAICS effective in the U.S. in
April 1997 and published the first NAICS U.S. manual in mid-1998. The Bureau of Labor Statistics (BLS) did not transition
to the first version of NAICS, NAICS 1997. Instead, NAICS 2002 was the first version implemented by BLS, and the CES
program converted from SIC to NAICS in June 2003. Reviews of NAICS are scheduled every 5 years; NAICS 2012 is the
most current version.
The U.S. Census Bureau reviews and updates NAICS codes every 5 years. Once these updates are available to BLS, CES converts all estimates to
these revised NAICS codes. The most current version is NAICS 2012. More information about NAICS 2002, NAICS 2007, and NAICS 2012 is described and linked below.
NAICS 2007 to NAICS 2012 Conversion
With the release of January 2012 data on February 3, 2012, CES updated the National nonfarm payroll series to the NAICS 2012 from the NAICS 2007 basis.
The conversion to NAICS 2012 resulted in minor content changes within the manufacturing and the retail trade sectors, as well as minor coding changes within
the utilities and the leisure and hospitality sectors. Several industry titles and descriptions also were updated. All employee (AE) series are published
at a more detailed level than all employee hours and earnings, production employee, women employee, or production employee hours and earnings series,
collectively called non-AE series. The non-AE series were sometime unaffected or affected at a less-detailed level than the AE series.
With the release of January 2008 data on February 1, 2008, the CES national nonfarm payroll series updated to the NAICS 2007
from the NAICS 2002 basis. The conversion to NAICS 2007 resulted in minor definitional changes within manufacturing, telecommunications,
financial activities, and professional and technical services. Several industry titles and descriptions also were updated.
With the release of May 2003 data on June 6, 2003, the CES national nonfarm payroll series underwent a number of changes. The basis for
industry classification changed from the SIC 1987 to the NAICS 2002. NAICS replaced the SIC system. The CES survey published national data
on a NAICS 2002 basis with the release of May 2003 data on June 6, 2003. SIC-based data is no longer produced or published; it is still be
available but not updated past April 2003.
CES data are published monthly, but are also available as historical time series. The data are available
as part of a monthly news release, as a searchable database, and in text format. The table below lists the
ways to download CES data from www.bls.gov/ces/home.htm. More information about retrieving CES data is available at www.bls.gov/web/empsit/cestips.htm. Archived editions of
Employment and Earnings Online are available from April 2007 forward here:
www.bls.gov/opub/ee/archive.htm. Earlier back issues are kept at federal depository libraries.
To join the BLS e-mail subscription service that provides excerpts from and links to The Employment
Situation, Real Earnings, and other BLS news releases of interest, visit the BLS News Service
Subscription E-mail page, available at subscriptions.bls.gov/accounts/USDOLBLS/subscriber/new.
JAVA must be installed on your computer and pop-up blockers must be turned off. Choose each option in order of the red numbers. Time periods can be adjusted to include more than 10 years, both seasonally and not seasonally adjusted data are included, graphs can be automatically generated, and the table format can be changed.
If you do not have JAVA, this tool will search the same data base as the One Screen Data Search. Time periods can be adjusted to include more than 10 years, both seasonally and not seasonally adjusted data are included, graphs can be automatically generated, and the table format can be changed.
BLS requires that all published Current Employment Statistics (CES) estimates meet strict quality and privacy
guidelines. These guidelines are designed to ensure that there is adequate sample to produce statistically sound
estimates and protect the confidentiality of our survey respondents. CES estimates are subject to annual review
to determine if they meet BLS publication and disclosure standards. Failure to meet the standards may stem from
inadequate sample size, inadequate sample response rates, or dominance of the sample by a few reporters. A list of the most recent changes to published CES series is available at www.bls.gov/web/empsit/cesnewseries.htm, and a complete list of currently published CES series is available at www.bls.gov/web/empsit/cesseriespub.htm.
The mission of the Bureau of Labor Statistics (BLS) is to collect, process, analyze, and disseminate essential statistical data to
the American public, the U.S. Congress, other federal agencies, state and local governments, business, and labor. In order to maintain
credibility and trust with our survey respondents, confidentiality protections for our data are essential. Protecting the
confidentiality of data is central to accomplishing the BLS mission.
When collecting data, the BLS makes a pledge of confidentiality to its respondents. This pledge varies depending on the context of
each survey, but the standard BLS confidentiality pledge promises that data collected are used for statistical purposes only.
Information about the BLS confidentiality policy and the laws that protect reporters to BLS surveys can be found here:
Researchers can get access to BLS microdata under certain circumstances. Information about qualifying for
the program and the application process through which access may be granted can be found here:
The Current Employment Statistics (CES) survey collects earnings data for private nonfarm business establishments,
excluding government. To calculate the real earnings series, these nominal dollar earnings estimates for all employees
(AE) back to 2006 and for production and nonsupervisory employees (PE) starting as early as 1964 (depending on the
industry) are adjusted for inflation using 1982-1984 dollars. The data are available on our website at
www.bls.gov/ces/data.htm. Click on "CES Data Access Tips" for further instructions.
CES does not collect or publish government hours or earnings data. Government earnings data are
available from the Quarterly Census of Employment and Wages (QCEW) program. QCEW provides
annual, quarterly, and weekly wage data for various private and government industries based on Unemployment Insurance tax
reports. QCEW wage information can be found at
www.bls.gov/cew/home.htm. CPI price index information can be found at www.bls.gov/cpi/home.htm.
The BLS does not publish average payroll frequency of establishments. However, the CES survey does adjust for
different payroll frequencies which can affect seasonally adjusted hours and earnings estimates. More information
about CES seasonal adjustment methodology can be found at www.bls.gov/web/empsit/cestn.htm#section6e.
The Current Employment Statistics (CES) sample is a stratified, simple random sample of worksites, clustered by Unemployment
Insurance (UI) account number. The UI account number is a major identifier on the Bureau of Labor Statistics (BLS) Longitudinal
Database (LDB) of employer records, which serves as both the sampling frame and the benchmark source for the CES employment
estimates. The sample strata, or subpopulations, are defined by state, industry, and employment size, yielding a state-based design.
The sampling rates for each stratum are determined through a method known as optimum allocation, which distributes a fixed number of
sample units across a set of strata to minimize the overall variance, or sampling error, on the primary estimate of interest. The
total nonfarm employment level is the primary estimate of interest, and the CES sample design gives top priority to measuring it as
precisely as possible, or minimizing the statistical error around the national total nonfarm employment estimates.
The establishment survey, like other sample surveys, is subject to two types of error, sampling and nonsampling
error. The magnitude of sampling error, or variance, is directly related to the size of the sample and the percentage
of universe coverage achieved by the sample. The establishment survey sample covers over one-third of total universe
employment; this yields a very small variance on the total nonfarm estimates. More information about error in the CES
survey and measurements of the error associated with sample estimates are available in the CES Technical Notes at
Yes, about 42 percent of the establishment survey sample is composed of business establishments with fewer than
20 employees. The establishment survey sample is designed to maximize the reliability of the total nonfarm employment
estimates for each State; firms from all size classes and industries are appropriately sampled to achieve this goal.
Sample data are weighted to represent other establishments in the same state, industry, and size class. More information
about the sampling methods used for the CES survey is available at
The CES government sample is not part of the probability-based survey design. CES is able to achieve a very high percent
of universe employment coverage (74 percent) by obtaining full payroll employment counts for many government agencies, thus
a probability-based sample design is not necessary for government. The high coverage rate virtually assures a high degree of
reliability for the government employment estimates. The large government sample does not bias the total nonfarm employment
estimates because it is used to estimate only the government portion of total nonfarm employment. The probability sample is
used to estimate employment for all industries in the private sector. Total private and government estimates are summed to
derive total nonfarm employment estimates.
Each month, BLS collects data on employment, payroll, and paid hours from a sample of establishments. To
encourage participation in this voluntary survey, BLS uses a variety of collection techniques, tailored to
individual firm preferences. Data collection centers perform initial enrollment of each firm via telephone,
collect the data for several months via Computer Assisted Telephone Interviewing (CATI), and where possible
transfer respondents to a self-reporting mode such as touch-tone data entry, fax, or web collection.
Very large, multi-establishment firms' ongoing reporting is established via Electronic Data Interchange
(EDI). Firms using EDI provide electronic files to BLS that include data from all their worksites.
CES tracks collection rates for the CES sample on a monthly basis for each release of estimates. Collection rates
are the percent of reports received for a monthly estimate compared to the total number of actively-reporting sample
units on the sample registry.
CES requests data for the pay period that includes the 12th of the month. CES survey respondents report earnings data for pay periods (which are specific to each establishment and may or may not reflect the Sunday-to-Saturday calendar week) that include the 12th of the month, regardless of the length of the pay period. Establishments have a range of payroll frequencies (weekly, bi-weekly, semi-monthly, and monthly). Respondents report pay period length to CES, but they do not report the start and end dates of their pay periods. Therefore, an establishment’s payroll might include the 15th of the month, but CES has no way of knowing if it does. CES normalizes all hours and earnings data to a weekly basis so that data from establishments with pay periods of differing lengths reflect a weekly value. About 30 percent of survey respondents have a weekly pay period and about 35 percent have a bi-weekly pay period, while semi-monthly and monthly salaries tend to be in industries such as finance and professional and technical industries.
CES continues to research calendar-related effects on our data series.
The Current Employment Statistics (CES) program uses a matched sample concept and weighted link relative estimator to produce
employment, hours, and earnings estimates. A matched sample is defined to be all sample members that have reported data for the
reference month and the month prior. Excluded from the matched sample is any sample unit that reports that it has zero employees in the current or previous month.
CES employment is an estimate of the number of nonfarm, payroll jobs in the U.S. economy. Employment is the
total number of persons on establishment payrolls employed full- or part-time who received pay (whether they worked or not) for any part of
the pay period that includes the 12th day of the month. Temporary and intermittent employees are included, as
are any employees who are on paid sick leave, on paid holiday, or who work during only part of the specified
pay period. A striking employee who only works a small portion of the survey period, and is paid, would be
included as employed under the CES definitions. Persons on the payroll of more than one establishment are counted
in each establishment. Data exclude proprietors, self-employed, unpaid family or volunteer workers, farm workers,
and household workers. Persons on layoff the entire pay period, on leave without pay, on strike for the entire period,
or who have a pending job but have not yet reported for work are not counted as employed. Government employment
covers only civilian employees; it excludes uniformed members of the armed services.
CES draws the survey sample from roughly 9.7 million U.S. business establishments covered by the Unemployment Insurance (UI) tax
system representing 97 percent of all employment within the scope of CES in the 50 states, the District of Columbia, Puerto Rico,
and the U.S. Virgin Islands. Excluded from the CES scope, although they are included in the list of establishments covered by UI
taxes, are private households and agricultural businesses.
The remaining 3 percent of establishments included in the CES scope that are not covered by UI laws are students paid by their school as part of a work study program, interns of hospitals paid by the hospital for which they work, employees paid by state and local government and elected officials, independent or contract insurance agents, employees of non-profits and religious organizations (this is the largest group of employees not covered), and railroad employees covered under a different system of UI administered by the Railroad Retirement Board (RRB). More information about noncovered employment and the
methodology used to include this employment in the CES estimates can be found in the CES Technical Notes at www.bls.gov/web/empsit/cestn.htm#NCE.
The CES data also exclude proprietors, the unincorporated self-employed, unpaid volunteer or family employees, farm employees, and domestic employees. Government employment covers only civilian employees; military personnel are excluded. Employees of the Central Intelligence Agency, the National Security Agency, the National Imagery and Mapping Agency, and the Defense Intelligence Agency also are excluded.
The production and nonsupervisory employee groups vary by industry. In service-providing industries, these data are
collected for nonsupervisory employees — those who are not owners or who are not primarily employed to direct, supervise,
or plan the work of others.
In goods-producing industries, the data are collected for production employees in mining and logging and in manufacturing,
and for construction employees in construction. Production and construction employees include working supervisors or group
leaders who may be "in charge" of some employees, but whose supervisory functions are only incidental to their regular work.
The production employee/construction employee categories in goods-producing industries exclude employees not directly involved
in production, such as managers, sales, or accounting personnel.
Yes, the CES survey captures counts of all employees on the payroll, including part-time employees. However,
part-time employees are not counted separately from full-time employees, so CES data does not include separate
estimates of part- and full-time employment.
It is likely that the CES survey includes at least some undocumented immigrants. However, the establishment survey
is not designed to identify the legal status of workers. Therefore, it is not possible to determine how many are counted
in the survey.
The Current Population Survey (CPS), also known as the household survey, does include questions which identify the foreign
and native born employees, but it does not include questions about the legal status of the foreign born employees. More
information about foreign born employees in the CPS survey is available at
BLS is unable to quantify the impact of reservists being called to active duty in CES employment figures.
In concept, persons on active military duty for the entire survey reference period are not included on employer
payrolls. Some reservists hold jobs not covered by the payroll survey — such as the self-employed or those in agriculture
— and others may not hold jobs at all. Any reservist who worked for or received pay from their regular employer
during the survey reference period is counted on the employer's payroll. If reservists are replaced by new employees
on an employer's payroll during the pay period including the 12th of the month, there is no net change in the number of
jobs counted. If reservists are not replaced, a net decline in the employer's job count results. If a reservist and a
replacement employee for the reservist each worked at any time during the same reference pay period, they are counted
as two employees.
Government employment includes only civilian employees. Military personnel on active duty are excluded. Employees
of the Central Intelligence Agency, the National Security Agency, the National Imagery and Mapping Agency, and the
Defense Intelligence Agency also are excluded.
Establishments report the number of persons on payroll during the pay period that includes the 12th of the month. A
person working multiple jobs at different establishments is counted once at each establishment. A person working different
jobs at the same business establishment is counted once.
Unusually severe weather, natural disasters, government shutdowns, and other catastrophic events are more likely
to have an impact on CES estimates of average weekly hours than on employment. In the establishment survey, the
reference period is the pay period that includes the 12th of the month. Average weekly hours are estimated for paid
time during the pay period, including pay for holidays, sick leave, or other time off. Any event in which employees
are prevented from working a normal schedule typically results in a reduction in average weekly hours. For example,
some employees may be off work for part of the pay period and not receive pay for the time missed, while some workers,
such as those dealing with cleanup or repair, may work extra hours.
In order for catastrophic events to reduce the estimate of payroll employment, employees have to be out of work
without pay for the entire pay period. About two-thirds of all employees in the payroll survey have a 2-week, semi-monthly,
or monthly pay period. Employees who receive pay for any part of the pay period, even 1 hour, are counted in the payroll
employment figures. Because the hours that employees work can be impacted by these special circumstances, but those
employees might still be counted as employed by an establishment if they were paid for work done during a portion of the
pay period, it is not possible to quantify the effect of catastrophic events on estimates of employment from the
In addition to their direct impact, these events sometimes have secondary effects. When the magnitude of significant
secondary effects are known these secondary effects are discussed in the monthly
Employment Situation news release and other BLS publications.
BLS independently develops a national employment series; state estimates are not forced to sum to national totals. 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 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
volatile error structure.
The Current Employment Statistics (CES) program produces earnings, but not wage data. CES average earnings are a measure of gross
payrolls divided by total hours paid during the pay period that includes the 12th day of the month. Averages of hourly earnings differ
from wage rates. Earnings are the return to an employee for a stated period on average in an industry; rates are the amount stipulated
for a given unit of work or time in a specific job. Average hourly earnings do not represent employers' total compensation costs because
they exclude items such as employee benefits, irregular bonuses and commissions, retroactive payments, and the employer's share of payroll
taxes. A more comprehensive explanation is available at www.bls.gov/opub/hom/homch2.htm.
The Quarterly Census of Employment and Wages (QCEW) program produces wages by industry, available at
www.bls.gov/cew/home.htm. The Occupational Employment Survey produces wages by occupation (instead of industry), available at
Yes, employers report total gross pay earned during the entire pay period, including overtime pay but excluding irregular payments,
and the total number of hours for which employees received pay during the entire pay period including overtime. Overtime hours are
published for manufacturing industries only. Respondents in manufacturing report the total number of hours for which employees received
overtime premiums because they worked more than their regularly scheduled hours.
BLS recommends that CES earnings series not be used in contract escalation clauses. Instead, BLS recommends that you use
the Employment Cost Index (ECI), which measures changes in labor costs free from the influence of employment shifts among industries
and occupations. For help on how to use the ECI for contract adjustments, visit
The CES sample alone is not sufficient for estimating the total employment level because each month new firms generate
employment that cannot be captured through the sample. There is an unavoidable lag between a firm opening for business and
its appearance on the CES sample frame. The sample frame is built from Unemployment Insurance (UI) quarterly tax records.
These records cover virtually all U.S. employers and include business births, but they only become available for updating
the CES sampling frame 7-9 months after the reference month. After the births appear on the frame, there is also time
required for sampling, contacting, and soliciting cooperation from the firm, and verifying the initial data provided. In
practice, CES cannot sample and begin to collect data from new firms until they are at least a year old.
There is a parallel though somewhat different issue in capturing employment loss from business deaths through monthly
sample collection. Businesses that have closed are unlikely to respond to the survey, and data collectors may not be able
to ascertain until after the monthly collection period that firms have in fact gone out of business. As with business births,
hard information about business deaths eventually becomes available from the lagged UI tax records.
Difficulty in capturing information from business birth and death units is not unique to the CES; virtually all current
business surveys face these limitations. Unlike many surveys, CES adjusts for these limitations explicitly, using a statistical
modeling technique. Because the goal of the CES program is to estimate an employment total each month
and business births and deaths are important components contributing to these totals, CES uses a model-based adjustment in
conjunction with the sample. Without the net birth/death model-based adjustment, the CES nonfarm payroll employment estimates
would be considerably less accurate.
Most series published by the Current Employment Statistics (CES) program show a regularly recurring seasonal movement
that can be measured from past experience. By eliminating that part of the change attributable to the normal seasonal
variation, it is possible to observe the cyclical and other nonseasonal movements in these series. Seasonal adjustment is
the process by which these normal seasonal patterns are removed from the estimates leaving behind only non-seasonal trends
and irregular movements. Seasonally adjusted estimates of employment and other series are generated using the X-13ARIMA-SEATS
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. This holiday change in retail trade is seasonal and is removed by seasonally adjusting the
series. Seasonally adjusted series are published monthly for selected employment, hours, and earnings estimates.
BLS published employment on a seasonally adjusted basis beginning in April 1955. Before this period, the Federal Reserve
seasonally adjusted CES employment; those series are available on the St. Louis FED website at
research.stlouisfed.org/fred2/categories/11. Hours and earnings were first seasonally adjusted by CES in July 1960.
The 4/5 week adjustment used in the Current Employment Statistics (CES) program's seasonal adjustment procedures adjusts for inconsistencies
in the CES series that arise because of variations of 4 or 5 weeks between reference periods in any given
pair of months. In highly seasonal months and industries, this variation can be an important determinant of the magnitude
of seasonal hires or layoffs that have occurred at the time the survey is taken, thereby complicating seasonal adjustment.
The CES program first incorporated a 4/5 week adjustment with the release of May 1996 data. At that time, historical estimates
were revised for the 4/5 week differences back to January 1986. Furthermore, historical data was again re-seasonally adjusted
using the 4/5 week adjustment back to January 1986 when CES switched from SIC-based industry definitions to NAICS-based industry
definitions in 2003.
A research paper discussing these 4 to 5 week inconsistencies in the CES estimates called Adjusting for a Calendar Effect
in Employment Time Series (1996) is available at www.bls.gov/ore/pdf/st960190.pdf. More
information about the CES seasonal adjustment process and special model adjustments such as the 4 to 5 week calendar effect are
available at www.bls.gov/web/empsit/cestn.htm#section6e.
The CES Small Domain Model (SDM) is used for national and state estimation of a small number of series with sampling
limitations. The CES SDM is a weighted least squares model with two employment inputs:
an estimate based on available CES sample for that series, and
an Autoregressive Integrated Moving Average (ARIMA) projection based on trend from ten years of historical QCEW data.
These two over-the-month change estimates are then weighted based on the variance of each of the estimates.
The shipyard index tracks the relative change in the dollar amount of average hourly earnings over a given period of time for
certain shipyards contracted to do work for the Navy. In calculating the index, each yard's current straight-time hourly earnings
are weighted according to its share of aggregate hours for all yards over the past 12 months. Then, this current weighted earnings
measure is compared with the equivalent weighted earnings measure for a given base period (May 1987 for the index that includes
lump sum payments and September 1980 for the one that does not). The index is scaled so that the base value equals 100. The values
for the other months are expressed as a percentage of the base value.
For the purposes of the CES shipyard index, a lump-sum payment is defined as a payment made to all production workers in lieu of
all or part of a wage increase; no other lump-sum bonuses are included. When a payment is reported, it is prorated forward based on
the number of weeks in each month of the lump-sum period.
CES revises published estimates to improve its data series by incorporating additional information
that was not available at the time of the initial publication of the estimates. CES revises its initial
monthly estimates twice, in the immediately succeeding 2 months, to incorporate additional sample receipts from respondents
in the survey. More information about the monthly revisions is available at
On an annual basis, CES incorporates a benchmark revision that reanchors estimates to nearly complete
employment counts available from Quarterly Census of Employment and Wages (QCEW) data, County Business Pattern data, and other
state-collected data. The benchmark helps to control for sampling error in the estimates. More information about the annual
benchmark revision is available at www.bls.gov/web/empsit/cesbmart.htm. All versions of the estimates through the month preceding the most recent benchmark are available at www.bls.gov/web/empsit/cesvininfo.htm.
It can be nearly 2 years before not seasonally adjusted Current Employment Statistics (CES) estimates are considered final. CES first preliminary
estimates of employment, hours, and earnings are published each month approximately 3 weeks after the reference period.
Estimates are then revised twice before being held constant until the annual benchmark release. Second preliminary estimates
for a given month are published the month following the initial release, and final sample-based estimates are published 2 months
after the initial release. The annual benchmark revisions affect nearly 2 years of data, so most months are subject to revisions
during 2 separate benchmark periods.
Seasonally adjusted CES estimates are generally subject to revisions for 5 years after their initial publication. Current Employment Statistics (CES) first preliminary
seasonally adjusted estimates of employment, hours, and earnings are published each month approximately 3 weeks after the reference period.
Estimates are then revised twice before being held constant until the annual benchmark release. Second preliminary estimates
for a given month are published the month following the initial release, and final sample-based estimates are published 2 months
after the initial release. Once a year with the benchmark release, 5 years of seasonally adjusted CES estimates are re-seasonally adjusted.
For most series, CES uses 10 years of not seasonally adjusted data as an input to seasonal adjustment. However, the all employee hours and earnings series begins in 2006. Until CES has a full 10 years of input data for the AE hours and earnings series, CES will use the entire history of the not seasonally adjusted series as inputs and replace the entire history of the seasonally adjusted data. Continuing these updates until all years have been adjusted using a full 10 years of input data ensures that all data are adjusted using the same methodology.
Further revisions may occur after the final estimates have been produced due to changes in scope, NAICS revisions, data errors, or other circumstances
that require the reconstruction of historical CES estimates.
The benchmark adjustment, a standard part of the CES 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 labor market information agencies. The difference between the March population counts
and the March sample-based employment estimates is referred to as the benchmark revision. A preliminary estimate of the
benchmark revision is published in late September, and the final benchmark revision, affecting 21 months of previously
published data and anchored to March of the previous year, is published with the January preliminary estimates in early
The Quarterly Census of Employment and Wages (QCEW) program maintains a quarterly tabulation from administrative records
of the number of employees covered by Unemployment Insurance (UI) laws, including Unemployment Compensation for Federal Employees
(UCFE). UI universe counts, available on a lagged basis, contain individual employer records for approximately 9.7 million
establishments and cover nearly 97 percent of total nonfarm employment; these records provide most of the benchmark levels
for the sample-based estimates. For the small segment of the population not covered by UI, BLS develops employment benchmarks
from several alternative sources, primarily records from the Railroad Retirement Board and County Business Patterns.
Following standard BLS methodology for national estimates, the March UI-based benchmark employment level replaces the March sample-based employment estimate, and the estimates for the 11 months prior to the benchmark month are adjusted using a wedge procedure. In this process, the difference, or error, between the benchmark level and the previously published March estimate for each estimating cell is computed. This difference is linearly distributed across the 11 months of estimates subsequent to the previous benchmark. For example, the benchmark revision that was released in February 2017 replaced the March 2016 estimate with the benchmark level, decreasing the employment level for that month by 81,000. To wedge this adjustment over the prior year, 1/12 of the difference was added to April 2015, 2/12s to May and so forth, through February 2016 which received 11/12s of the difference. Employment for March 2015 had been set to a benchmark amount in the prior year and was not revised with the March 2016 benchmark. The wedge procedure assumes that the total estimation error accumulated at a steady rate since the last benchmark.
Employment benchmarks are applied to not seasonally adjusted estimates. On a seasonally-adjusted basis, 5 years of historical data may revise, because new models for seasonal adjustment are selected and seasonal factors based on the new models are updated with each year’s benchmark release.
CES benchmark revisions only affect April of the previous year to October of the benchmark year and do not affect changes to earlier employment levels. However, earlier months of employment history are subject to change due to reconstructions to CES series. Reconstructions can result in revisions to both the not seasonally adjusted data and the seasonally adjusted data at both the detailed and aggregate levels.
Estimates for the period after the benchmark is applied, called the post-benchmark period, are calculated for each month by applying
previously derived over-the-month sample change ratios to the revised March levels. New net birth/death model estimates also are
calculated and included in post-benchmark estimation. Additionally, new sample from the annual sample update is introduced starting
with the third release of estimates for November following the benchmark month.
Employment benchmarks, including the post-benchmark period, are applied to not seasonally adjusted estimates. On a
seasonally-adjusted basis, 5 years of historical data may revise, because new models for seasonal adjustment are selected and
seasonal factors based on the new models are updated with each year's benchmark release.
To begin reporting your CES data or if you have any questions while reporting, please contact the CES Help Desk or call 1-800-827-2005. Include your CES report number(s) in your request or have them available when you call.
The CES report is voluntary under federal law and is mandatory under state law in North Carolina, Oregon, and South Carolina. The South Carolina requirement applies to firms with more than 20 employees. Reporting is also mandatory in the Commonwealth of Puerto Rico. Legal citations for the state requirements are listed on the
CES report form.