Handbook of Methods > Quarterly Census of Employment and Wages

Handbook of Methods Quarterly Census of Employment and Wages History

Quarterly Census of Employment and Wages: History

The Quarterly Census of Employment and Wages (QCEW) is an administrative dataset guided by unemployment insurance (UI) coverage laws.^⁠1 These laws vary across states. They have evolved over time and continue to do so from state to state and year to year. We have more information about UI coverage historically. The following timeline presents important events in the history of the QCEW program.

Hover over the red dot to see historical information.


Year	Event
1935	Social Security Act is passed and establishes the basic framework of the Federal-State system of unemployment insurance (UI).
1938	The Federal Unemployment Tax Act becomes effective and is administered by U.S. Department of Labor (DOL).
1938	The Railroad Retirement Unemployment Insurance Act was approved, providing coverage for railroad workers separate from federal unemployment tax.
1945	The Federal Unemployment Tax Act is amended to exclude service performed for an international organization from coverage.
1946	Extends UI coverage to maritime service. Provides that a state can cover the crew of an American vessel if the operating office is within the state.
1948	The Federal Unemployment Tax Act is amended to exclude services performed by a news agent from coverage.
1948	Supreme Court rules to limit the term "employee" to an employee under the common law rule of "master–servant" relationship, retroactive to 1939.
1950	Employment and Wages is first issued as a semiannual report by the Social Security Administration (SSA).
1950	Social Security Act Amendments of 1950 is passed, redefining exclusion for casual labor not related to the employer’s usual course of business.
1953	The Federal Unemployment Tax Act is amended to permit states to cover services performed by general agents of the Secretary of Commerce in connection with American vessels owned or chartered to the United States under certain conditions.
1954	The Federal Unemployment Tax Act is amended to include organizations that test for public safety in the list of nonprofit organizations for which services are excludable.
1954	Extended coverage to Federal civilian employees employed after December 31, 1954, subject to benefit provisions of state laws, under Title XV of the Social Security Act, the Unemployment Compensation for Federal Employees (UCFE) program.
1956	Effective January 1, 1956, UI coverage was extended to employers of 4 or more workers in 20 weeks in a calendar year.
1958	Ex-servicemen's Unemployment Compensation Act of 1958 (UCX) establishes a permanent program to provide benefits for veterans under the law of the state in which a claim is filed to coincide with the end of special coverage for Korean veterans.
1960	Social Security Amendments of 1960 is passed that takes effect on January 1, 1961, coverage under the UCFE program is extended to certain instrumentalities that were neither wholly nor partially covered by FUTA, such as Federal Reserve banks, land banks, and credit unions.
1960	Puerto Rico is brought into the Federal-State unemployment insurance (UI) system, as are American aircraft employment and certain nonprofit institutions.
1961	Mutual Educational and Cultural Exchange Act of 1961 amends the Federal Unemployment Tax Act by providing an exclusion for services performed by a nonresident alien while temporarily in the United States as a nonimmigrant.
1970	Employment Security Amendments of 1970 extends coverage, effective January 1, 1972, to several groups previously uncovered. For details, see https://www.bls.gov/cew/additional-resources/federal-unemployment-insurance-employment-coverage-timeline.htm#1961-70
1972	The Bureau of Labor Statistics (BLS) assumes responsibility for publication of the now quarterly Employment and Wages. Employment and Wages was first issued as a semiannual report.
1975	Employment and Wages becomes an annual publication.
1976	The Unemployment Compensation Amendments of 1976 that becomes effective January 1, 1978, extends coverage to: agricultural labor for employers with 10 or more workers in 20 weeks or who paid $20,000 or more in cash wages in any calendar quarter; household workers of employers who paid $1,000 or more in any calendar quarter for such services; State and local governments with certain minor exceptions; and employees of nonprofit elementary and secondary schools.
1976	The Virgin Islands is permitted to become part of the Federal-State unemployment insurance system.
1981	Economic Recovery Tax Act of 1981 adds to exclusions from the Federal Unemployment Tax Act coverage services excluded from Federal Insurance Contributions Act under section 3121(b)(2) of the Internal Revenue Code.
1982	Tax Equity and Fiscal Responsibility Act of 1982 excludes from the Federal Unemployment Tax Act (and Federal Insurance Contributions Act) coverage of certain direct sellers and real estate sellers who are licensed direct sellers of real estate or who execute an agreement with an employer in which substantially all remuneration is based on sales or output (commission).
1986	Consolidated Omnibus Budget Reconciliation Act of 1985 creates permanent exemption for summer camp counselors and certain fishing boat crew members, with conditions.
1986	Tax Reform Act of 1986 creates Indian Tribal Government exclusions for services performed: before, on, or after date of enactment (October 22, 1986), but before January 1, 1988, and during a period in which the Indian tribal government is not covered by a state UI program
1988–91	Business Establishment List Improvement project moves from data collection on a reporting unit basis to a worksite basis.
1989	States begin providing data to the Quarterly Census of Employment and Wages (QCEW) at the business establishment level.
1994–97	The Microdata/Macrodata (MIC/MAC) project creates a single file that combines both microdata and macrodata.
1995	The Electronic Data Interchange (EDI) Collection Center is established.
1996	Small Business Job Protection Act of 1996 passes and becomes effective January 1, 1996. Personal Responsibility and Work Opportunity Reconciliation Act of 1996 is passed and the amendments apply to remuneration paid after December 31, 1984, and before January 1, 1995.
1997	Balanced Budget Act of 1997 amends the Federal Unemployment Tax Act to permit states to exclude from coverage services performed by certain election officials for less than $1,000 and also persons committed to a penal institution.
1997	The North American Industry Classification System (NAICS) is developed to replace the Standard Industrial Classification (SIC) system.
2000	Consolidated Appropriations Act, 2001 amends the Federal Unemployment Tax Act to treat Indian tribes (governmental entities and any subdivision, subsidiary, or business enterprise wholly owned by the tribe) similar to state and local governments.
2002	QCEW program publishes data under NAICS 2002 codes and starts publishing data quarterly.
2003	The Business Employment Dynamics program publishes first quarterly release using QCEW establishment data.
2003	BLS begins releasing fully detailed industry data at the county, Metropolitan Statistical Area (MSA), state, and national levels.
2007	QCEW data are published under revised NAICS 2007 codes.
2010	Last printed edition of Employment and Wages, Annual Averages is published. Online versions of this publication are available at https://www.bls.gov/cew/publications/employment-and-wages-annual-averages/home.htm.
2012	QCEW data are published under revised NAICS 2012 codes.
2014	QCEW data are made available publicly as an open data product. QCEW data are published under revised NAICS 2012 codes.
2017	QCEW data are published under revised NAICS 2017 codes.
2018	QCEW quarterly data are separated into two releases every quarter.
2020	California Assembly Bill No. 5 changes the state’s coverage of UI effective on January 1, 2020. This bill includes a three-part test to determine if a worker should be considered an independent contractor or if an employee should be covered by unemployment insurance.
2020	Starting with the November 2020 QCEW news release, QCEW began implementing a new imputation method.

Program identification and data publication

The QCEW program was initially named Employment Security Report 202, after the form on which data were collected. In 1998, to give some context to what the program produced, the name was changed to Covered Employment and Wages (CEW). The program became known as the Quarterly Census of Employment and Wages in 2002, to describe the mission more accurately.

The Standard Industrial Classification system (SIC) was used by the QCEW program to code establishments by industry until the fall of 2002, when the North American Industry Classification System (NAICS) was utilized in the publication of 2001 data.

In the late 1970s and early 1980s, special extracts and aggregations were provided to the public via mainframe computer tape and, later, on floppy disks. Alternatively, special print tables were produced with the information, as requested by customers. Data were subject to basic disclosure limitation procedures.

Following the early 1980s and into the late 1990s, the QCEW program followed a similar data pattern, but with more sophisticated disclosure limitation procedures. Some data requests were filled via CD, while more complicated requests required that BLS be reimbursed. During this timeframe, QCEW established subscription agreements with some of the larger private data companies. These companies would get updated files when they were available. Although data were scheduled to be released quarterly, they were actually released irregularly. On occasion, there were delays that pushed back file creation until more than one added quarter of data could be released at once.

From the late 1990s forward, requests that required reimbursement slowed and then ceased. More sophisticated disclosure limitation procedures were used. Early online tools offered annual average data for 1997 forward. In the early 2000s, a small number of downloadable files were posted on the BLS File Transfer Protocol site.

In 2014, QCEW data were made available as an open data product via comma-separated values (CSV) files on the program's web page. Open data, as described in OMB memorandum M-13-13 (PDF), refers to publicly available data structured in a way that enables the data to be fully discoverable and usable by end users. The CSV files are designed to allow third-party programmers, developers, and organizations to retrieve published QCEW data in CSV format. QCEW CSV files are organized by industry, by area, and by establishment size class.

Beginning with the publication of fourth quarter 2017 QCEW data, published in mid-2018, BLS accelerated the news release 2 weeks ahead of the normal full QCEW data update. BLS has changed this release schedule for two reasons. The first is to share QCEW data for all states faster. The second is to release the data before the Bureau of Economic Analysis (BEA) publishes quarterly gross domestic product (GDP) data, which uses the most recent QCEW wage data. BEA personal income estimates are heavily based on QCEW wages. For their earliest estimates, BEA uses Current Employment Statistics (CES) data to project wages. After BLS shares QCEW data with BEA, the GDP estimates are updated, replacing the CES-based forecast with QCEW wage data.

After reviewing the QCEW program’s procedures, BLS decided with the fourth quarter 2017 QCEW data to begin issuing the County Employment and Wages news release earlier.

The first QCEW release is a traditional news release, with a partial data update. This partial update will consist of the most recent monthly and quarterly levels presented in the County Employment and Wages news release. (See tables 1, 2, and 3.) The second QCEW release, consisting of the full QCEW dataset (current quarter, as well as any current year quarter, QCEW data), is published two weeks after the first QCEW (news) release.

Major projects

The QCEW program has undertaken several major projects in its brief history.

Business Establishment List (BEL) Improvement Project

BLS initiated the BEL Improvement Project to obtain information on multiple-establishment employers on a quarterly basis. The project involved changing the collection of data on employment and wages for multiple-establishment employers from a reporting-unit (county or industry total) basis to an individual-worksite (establishment) basis. As part of this change, the size criterion used to define multiple-establishment employers was lowered to include employers of smaller establishments. As a result, both the number of establishments and the number of multiple-establishment employers increased. The State Statistical Supplement forms that were previously used to collect county and industry summary-level data on multiple-establishment employers were replaced by the Multiple Worksite Report (MWR), a standardized form for use in all states.

The BEL project began in late 1988, using the Annual Refiling Survey (ARS). The survey obtained worksite identification information for existing multiple-establishment employers and identified those multiple-establishment employers which were previously categorized as single-unit employers. To capture ARS physical location address information for single-unit employers, the SIC Refiling Control System (SRCS) was modified in mid-1988 and the new version was provided to the states. The new, standardized MWR received OMB clearance and was mandated for state use beginning with data for the first quarter of 1991. (Note that, although state use of the form was mandatory, depending on state UI laws, it was not mandatory that all employers in the state use the form.) The MWR was intended to collect quarterly information on employment and wages from multiple-establishment employers.

The MWR was a major component of the BEL improvement project launched in 1989. The BEL moved the BLS sampling frame from a reporting-unit to an establishment basis. The Longitudinal Data Base (LDB) was established and later became part of the popular Business Employment Dynamics (BED) program. As mentioned earlier, the MWR replaced State Statistical Supplement forms. Several states had already collected establishment-level data, and their forms served as a model for the MWR. The initial quarter during which MWR establishment-level data were collected was the first quarter of 1991. The MWR enabled the QCEW program to be the first in BLS to develop a standardized magnetic-media reporting format. The MWR was implemented in September 1992.

Electronic Data Interchange (EDI) Collection Center

In 1995, BLS established the EDI Collection Center in Chicago, IL, to improve and to expedite the MWR collection process. Employers who complete the MWR for multiple-location businesses submit information on employment and wages via an electronic medium directly to the collection center, instead of submitting separate forms or files to each individual state agency. The collection center then edits the employers' data and forwards the data to the appropriate state agency. The EDI Collection Center is geared toward firms that meet certain size thresholds.

Electronic and web-based filing has reduced costs considerably for large and midsized employers. MWR web data collection started with data for the first quarter of 2007. The MWR electronic-reporting format is now routinely included in the products of prominent payroll- and tax-reporting software developers. Since 2012, BLS has successfully worked with a contractor to print, mail, receive, and scan returned MWR forms to state agencies, substantially reducing their workloads..


First quarter 2019 (initial enhanced quarterly Unemployment Insurance file)	Number or percentage of employers
Percentage of employers filing the MWR	1.4
Percentage of national worksites	17.3
Percentage of national employment	41.05

It is useful to examine some recent statistics regarding overall MWR data collection. In that regard, the numbers in the following tabulation, for the first quarter of 2019, exhibit a stable pattern that appears to be the usual pattern for MWR statistics:

As the data above show, 1.4 percent of employers completed the MWR for about 17 percent of national worksites and more than 41 percent of national employment. Moreover, 60 percent of all MWR employment came electronically, from either the EDI Center or MWRweb. Of this 60 percent, the EDI Center collected 24 percent of total MWR employment while MWRweb accounted for 36 percent.

Microdata/Macrodata (MIC/MAC) Project

Prior to 1989, states submitted only an enhanced file containing macrolevel employment and wage data. That meant that the highest level of detail received by the QCEW program was aggregate data cells made up of a combination of industry, county, and ownership codes.

In the first quarter of 1989, states began delivering a quarterly microdata file that included individual business establishment data. At the time, states were submitting nine separate deliverables per year: four quarterly microdata files, four quarterly macrodata files, and the annual Control File, providing Code Change Supplement data.

Development of the MIC/MAC project began in 1994. Teams consisting of BLS staff from both the national office and regional offices, as well as state staff members, were established to define the new processing requirements and deliverable standards. State and BLS systems were updated to reflect these major operational changes. The mission was to create a single quarterly file that would make possible the creation of macrolevel totals from the microdata. The project would promote more efficiency by eliminating duplication of work in reviewing and correcting the micro- and macrodata files. The project would also create uniformity in the micro- and macrolevel totals, as well as in the editing standards for the states and BLS. The first MIC/MAC files were submitted to BLS in the third quarter of 1997, providing for a more efficient work process and producing higher quality data.

SIC-to-NAICS conversion

The SIC system was established in the 1930s in order to classify establishments by their primary type of activity. The system was updated multiple times in its history of being utilized by the QCEW (formerly, ES-202) program; significant changes were made in 1967, 1972, and 1987. In July 1992, the Office of Management and Budget (OMB) established the Economic Classification Policy Committee, chaired by the Bureau of Economic Analysis (BEA) of the U.S. Department of Commerce, with representatives from the U.S. Bureau of the Census (now the U.S. Census Bureau) of the U.S. Department of Commerce and from BLS. The OMB charged the committee with conducting a “fresh slate” examination of economic classifications for statistical purposes and determining the desirability of developing a new industry classification system for the United States that would be based on a single economic concept. In 1997, NAICS was introduced, the product of a cooperative effort on the part of the official statistical agencies of the United States, Canada, and Mexico.

In the fall of 2002, the QCEW program published its 2001 data using NAICS 2002 classified industry data for the first time. Over several years prior to that milestone, state agencies had worked to convert more than 8 million establishments to NAICS. Along with the initial 1997 version of NAICS and the 2002 update, versions were released in 2007, 2012, and 2017, to reflect changes to the economy.

One of the major changes to industry coding that was introduced by NAICS was the sector called information (NAICS 51), which includes establishments producing and distributing information and cultural products, establishments providing the means to transmit or distribute these products, and establishments providing data or communications services, including data processing. Another new sector introduced by NAICS was management of companies and enterprises (NAICS 55).

The BLS conversion to NAICS was notable for its detail. An example may be seen in the construction (NAICS 23) sector. Within the specialty trade contractors (NAICS 238) subsector in the BLS implementation of NAICS, each base group in specialty trade contractors—for example, masonry contractors (NAICS 23814) or roofing contractors (NAICS 23816)—is further divided, on the basis of whether the establishment's predominant activity is in residential construction or nonresidential construction.^⁠2

Early BEA finance data

BEA is responsible for producing quarterly estimates of gross domestic product (GDP) and personal income (PI). A key component of the GDP and PI numbers is data on wages, from the QCEW program. Prior to 2008, BEA had difficulty estimating wages in a number of industries. The most difficult industry was finance (NAICS 52), because wage levels for that industry are greatly affected by irregular payments (for example, bonuses and stock options). These irregular payments are usually made in either the fourth or first quarter. In 2008, at BEA's request, BLS agreed to deliver first-quarter macrodata earlier than is customary, in order to enable BEA to adjust its estimates by means of QCEW total wage levels.

In early July of 2008, the 10 states with the largest proportion of national aggregate wages in the finance industry participated in a pilot test of accelerated macrodata delivery to BEA. The early-delivery pilot proved to be highly successful with major national users, such as the U.S. Department of the Treasury and the Council of Economic Advisers, which noted a decrease in revisions. On the basis of this success, BEA asked the QCEW program to provide early macrodata as a deliverable each first quarter, beginning with the first quarter of 2009. The BEA wage estimates that are being improved with QCEW data are national, not state, metro, or local, estimates.

Business Employment Dynamics

The Business Employment Dynamics (BED) program uses QCEW establishment data to produce a quarterly series of gross job gains and gross job losses for the entire economy. The program published its first release in 2003. The data covered the period from 1992 through 2002, with subsequent quarterly updates published since then. (For more on BED, visit the BLS website Business Employment Dynamics and the BLS Handbook of Methods for BED.)

QCEW Imputation Improvement Project (QIIP)

The economic situation in 2020, which reflects the impact of COVID-19, may not have been optimally captured with the old QCEW imputation method. In 2020, BLS explored three imputation (specific techniques of handling for non-response from active establishments) improvements. These improvements were accepted for production. The imputations were first used for the second published version of data for the first quarter of 2020 and for all versions of QCEW data beginning with the first published data for the second quarter of 2020.

The first improvement is a version of the ratio method of imputation described below.

BLS has conducted extensive research on alternative imputation methods for both employment and wages. The findings of the research indicate that current trends reported by similar businesses should be applied to nonrespondents. BLS defines the procedure for doing so as the ratio method. According to this method, the ratio for a particular estimation cell is computed as the sum of a current month’s reported employment (from similar businesses) divided by the sum of the previous month’s reported employment. To impute the current month’s employment for a nonrespondent, the ratio is then multiplied by the nonrespondent’s previous month’s employment. A similar procedure is applied to impute total quarterly wages.

The second improvement immediately identifies missing employers that are likely to have ceased operations. This eliminates imputation for many establishments that are out of business. The improvement uses counts of claims for unemployment benefits, aggregated by employer, to mechanically identify businesses that are likely to have ceased operations.

The third improvement is the use of benefit claims counts by state QCEW staff as a supplement to existing manual data review. Tables presenting the impact of the ratio method and the new identification of out-of-business establishments are available from the 2020 2nd quarter QCEW Improved Imputation Methodology Impact worksheet.

Notes

^⁠1 For the most recent changes to federal and state UI legislation, see "Most recent comparison of state UI laws," (U.S. Department of Labor, March 29, 2004), https://oui.doleta.gov/unemploy/statelaws.asp#Statelaw.

^⁠2 For more information on NAICS, see "North American Industry Classification System (NAICS) at BLS," (U.S. Bureau of Labor Statistics, February 27, 2020), https://www.bls.gov/bls/naics.htm.

For QCEW's industry finder web application, visit https://data.bls.gov/cew/apps/bls_naics/v2/bls_naics_app.htm#tab=hierarchy&naics=2017&hier=default; and David R. H. Hiles, "A first look at employment and wages using NAICS," Monthly Labor Review, December 2001, pp. 22–31, https://www.bls.gov/opub/mlr/2001/12/art3full.pdf.