Recent changes in the State and Metropolitan Area CES survey: like its national counterpart, the State and Metropolitan Area Current Employment Statistics survey has converted to the North American Industry Classification System ; two other changes are a minimum guaranteed publication structure and the use of a Small Domain Model for guaranteed series that do not meet publication criteria

Monthly Labor Review, June, 2003 by Molly E. Barth

With the release of the January 2003 preliminary State and Area data on March 20, 2003, the Current Employment Statistics (CES) survey introduced a number of important changes. Most notable is the conversion from the 1987 Standard Industrial Classification (SIC) system to the 2002 North American Industry Classification System (NAICS). All State and Area CES series, including historical data as well as current estimates, are now reported on a NAICS basis; SIC data will no longer be published. The CES program also completed a redesign of its sample methodology, moving from a quota-based to a probability-based sampling technique.

The new changes apply to State and Area, as well as national, CES series. The national CES program implemented NAICS and the final phase of the CES redesign with the release of the preliminary May 2003 estimates on June 6, 2003. (1) Other important State and Area CES changes include the development of a minimum guaranteed publication structure for all States and metropolitan areas and the CES Small Domain Model for guaranteed series that do not meet publication criteria under sampling methodology. These changes and others are described in this article.

Overview of the CES program

The CES program is a Federal-State cooperative program that produces monthly estimates of employment, hours, and earnings based on nonfarm establishment payrolls for the Nation, the 50 States, the District of Columbia, Puerto Rico, the Virgin Islands, and more than 270 Metropolitan Statistical Areas (MSA's). Information for these estimates is derived from a sample of 300,000-plus business establishments. CES estimates are closely followed and widely used economic indicators, offering timely data with an abundance of industry and geographic detail.

State and Area CES estimates are derived by the States, as provided for in the Federal-State cooperative agreement; sample and estimation procedures are designed to produce accurate data for each State. The Bureau of Labor Statistics estimates national employment series independently, using sample data from all States and the District of Columbia. State estimates are not forced to sum to national totals and vice versa.

Estimates from the CES survey are benchmarked once a year to universe counts from State Unemployment Insurance (UI) tax records, provided by the Covered Employment and Wages program. For State and Area CES series, the most recent UI universe counts available are used to benchmark employment estimates; the monthly estimates are replaced with universe data back to the previous benchmark. Estimates for months subsequent to the most currently available UI universe counts are recomputed by using sample-based links applied to the new UI benchmark levels.

NAICS replaces SIC

After 60 years of use by U.S. statistical agencies, the SIC system was retired, and NAICS was adopted as the standard means of industry classification. NAICS was devised by the statistical agencies of the United States, Mexico, and Canada to enhance comparability of economic data across the North American Free Trade Association (NAFTA) trade region. (2)

Conversion of State and Area series to NAICS

The CES program has reconstructed State and Area all-employees series on a NAICS basis back to January 1990 for most industries. Converting the data from SIC to NAICS involved several steps. First, the CES reporting units were assigned NAICS codes. Then, publication standards were developed, estimating cells were defined, and historical data were reconstructed.

Reconstruction methodology. Before the CES program could begin the reconstruction of historical employment series, NAICS codes needed to be assigned to all the records of the BLS Longitudinal Database (LDB). (3) The NAICS and ownership codes for each establishment in the LDB as of the first quarter of 2001 were then carried back through the entire history of the database. For statewide series, the microdata for all establishments within the scope of the CES survey were first added to six-digit NAICS levels. These totals were then aggregated to all official NAICS-based statewide CES publication levels. For MSA series, the LDB microdata were added to six-digit NAICS employment levels on the basis of the county code as of the first quarter of each year. These levels were then aggregated to all official NAICS-based CES publication levels in a process to be described shortly.

One of the strengths of this reconstruction methodology is its universal applicability: because series were summed from information on individual establishments, all the industry and geographic detail needed to re-create employment data for even small MSA'S was available. (4)

Keeping the NAICS codes constant over the historical data of the LDB gave consistency to the data by eliminating all noneconomic code changes caused by coding error. However, any true economic code changes caused by a change in primary business practice were removed as well. Similarly, the ownership codes remained constant. This again eliminated any noneconomic changes, but also erased all true changes in ownership, such as the privatization of an establishment. The location of establishments was not held constant; rather, the county code from the first quarter of each year was used to assign a given firm to an MSA. Consequently, both economic and noneconomic location changes were present in the data.