Improved Methods to be Used for Labor Force Estimates

Printer Friendly PDF Version

DES MOINES, IOWA - Iowa Workforce Development (IWD) is notifying data users of an upcoming change in the methods that are used to produce monthly employment and unemployment estimates for the state and substate areas. This change will occur in March 2005, with the release of the January 2005 labor force data. On November 8, 2004, a Federal Register Notice was issued that detailed these revisions.

Since assuming responsibility for developing state and local employment and unemployment estimates in 1972, the Bureau of Labor Statistics has been continually improving and refining the methods that produce these estimates. Beginning in 1989, the Bureau moved to a modeling approach for the 39 smaller states and the District of Columbia. In 1996, due to budgetary reductions and a subsequent decrease in the sample size of the Current Population Survey (household survey), estimates for the other 11 large states also went to that modeling approach. Prior to that, large states had been estimated directly from the results of the Current Population Survey. Since that time, the Bureau has worked on improving the models for the states based on historical and current relationships found within each state’s economy. The Bureau has designed two models for each state — one for the employment-to-population ratio (employment divided by the population 16 years of age and older) and one for the unemployment rate.

The new models, which will be implemented with the calculation of the January 2005 labor force data, will probably have more noticeable effects on the states’ estimates than any previous methodology changes. Iowa Workforce Development, along with other state employment security agencies, tested the new model during 2004 to determine the extent to which the employment and unemployment estimates will be affected by the change. Based on the results of the evaluation period, Iowa’s unemployment rate derived from the new methodology was on average 0.3 percent higher, while employment was about 22,000 lower. (The current estimate levels will vary from state to state.) To facilitate analysis of the data, a new labor force series will be provided back to 1976.

The upcoming methodology change for the statewide labor force estimates is expected to have several benefits:
1) Regional totals will add to the national total,
2) Reliability measures will be made available to users,
3) Historical estimates will be more consistent with current estimates, and
4) There will be smaller end-of-year revisions.

Two key changes will also be built into the method for producing substate labor force estimates:
1) New residency adjustment ratios, based on 2000 Census commuting patterns, will be used to adjust nonfarm employment from a place-of-work to a place-of-residence concept, and
2) Monthly statewide data from the Current Population Survey will be used to improve the estimate for unemployed entrants and re-entrants.

For detailed information on the upcoming Local Area Unemployment Statistics Program redesign, access the Questions and Answers (Q’s & A’s) section from the Bureau of Labor Statistics web site: www.bls.gov/lau/lauschanges2005.htm.

Due to the time needed to implement the changes in methodologies, the January 2005 labor force and nonfarm employment data will be released on Thursday, March 17, 2005.