BLS “Fudge Factor” Pushes Wyoming’s Unemployment Rate to 9.7 Percent in January.

Monthly, each state’s employment agency provides the Bureau of Labor Statistics (BLS) with an estimate of the number of nonagricultural wage and salary jobs worked, and a count of unemployment insurance (UI) claims filed. These data, along with data on employment and unemployment from the Current Population Survey (CPS) of households, are factored into equations based on historic relationships to estimate the “true” labor force. State labor force estimates are a product of information supplied by the states to BLS. Since the sample of households in the CPS is too small to reliably estimate Wyoming’s labor force, it is necessary to offset the lack of reliability by including information on nonagricultural employment and claims data in the equations.

In January 1994 labor force estimates began using a new household survey questionnaire in the monthly interviews, and portable computers replaced paper and pencil for recording answers. In a memo released February 22, 1994, BLS technicians announced that, since the equations could not account for these two changes, it would be necessary to estimate their impacts. They developed fudge factors in an attempt to compensate for sudden shifts in equation outcomes. These adjustment numbers were implemented suddenly, without comments from states, and without a policy change from BLS's administrators

States have not been informed as to when and under what conditions the fudge factors will be removed from the equations. BLS, however, indicates that the fudge factors “...can provide insurance against possible large shifts in the CPS and then later (can) be dropped if no shift is detected. Thus, for shifts of unknown magnitude, they may help -- and are unlikely to harm”. Obviously in Wyoming’s case the fudge factor did not work as intended.

The net effect of these unverified fudge factors differs from state to state a you can see here. The fudge factor applied to Wyoming’s equations raised the unemployment rate from a revised December 1993 rate of 5.9% to 9.7 percent (preliminary rate) in January 1994.

Essentially by using the fudge factor, BLS technicians caused the estimated number of persons newly unemployed in Wyoming to increase from December to January. That increase is roughly equivalent to the population of Big Horn county. These artificially high numbers may or may not continue into the future.

Until the unemployment rate statistics return to the realm of “normality,” we should focus on the best statistical series available: UI claims filed and the Wyoming Nonagricultural Wage and Salary Employment information published in Wyoming Labor Force Trends.

On a year to date basis, initial UI claims filed in January 1994 continued to run below last years level. In fact, initial claims are down 7.0 percent from January 1993.

Wyoming’s Nonagricultural Wage and Salary Employment grew by 2.7 percent from January 1993 to January 1994. Total goods producing grew by 3.0 percent and total service producing industries grew by 2.6 percent in the same time period. Based on the fact that UI claims are down and employment is up for goods and service producing sectors, Wyoming's unemployment rate is no where near the 9.7 percent as calculated by BLS. In reality it is most likely in the range of 5.5 to 6.5 percent.

Only the next few months will tell if Wyoming's unemployment rate will return to the range of reasonableness. As things change we will keep our readers informed of the situation.

Tom Gallagher is the Manager of Research & Planning.

Table of	Labor Market	Employment	Send Us
Contents	Information	Resources	Mail