OLMIS

by Tracy A Morrissette
Published Aug-21-2009

UI Claims Data as Inputs to Unemployment Estimation Program to Measure Insured Unemployment Statistics (PROMIS) Continued Claims Data by Industry Continued Claims Data by Education Level How "Unemployment" Differs From "Unemployment Insurance" Conclusion

In the current recession, the number of unemployed in Oregon and the nation has increased rapidly. But what do we know about the workers who are currently unemployed?

There are two general sources of "unemployment" data for Oregon. One is the Local Area Unemployment Statistics (LAUS) program. LAUS is a Bureau of Labor Statistics (BLS) Federal-State cooperative program responsible for monthly estimates of the labor force and unemployment rate. Another source is the Employment and Training Administration (ETA). ETA issues reports related to the administration of Oregon's Unemployment Insurance (UI) system. Both LAUS and ETA produce reports labeled as "unemployment." Although there is some overlap in definition, there are important differences in how the term "unemployment" is used in each program. For more information on these distinctions, see the text box "How Unemployment Differs From Unemployment Insurance."

It is possible to examine details about those unemployed workers covered by UI using the Program to Measure Insured Unemployment Statistics (PROMIS). This is a software query tool developed by BLS that exploits the overlap in the two sources of unemployment data to produce information about the unemployed covered by UI.

Although definitional differences make LAUS and ETA separate counts of unemployment, there is some overlap between the two versions. A special tabulation of UI claims matching key concepts from the Current Population Survey (CPS) regarding reference periods, residency status, and labor force categorization serve as data inputs to the unemployment estimation algorithms used in LAUS. The UI continued claims data used in LAUS estimation are a subset of total unemployment.

In Graph 1, UI claims data are lower than total unemployment since some groups in the CPS definition of unemployment do not qualify for UI benefits. Examples of groups that do not qualify for UI benefits are unemployed entrants to the labor force and people who have voluntarily quit their jobs. Entrants are individuals who have entered the labor force either for the first time or after an absence and have begun a job search. They do not qualify since they do not have wages to file a claim against.

The special tabulations of unemployed UI claimants used in LAUS data processing algorithms come from PROMIS. PROMIS is a database and query tool that extracts continued claims records matching the CPS concept of unemployment. PROMIS also provides initial claims data inputs for the Mass Layoff Statistics (MLS) program, another BLS Fed-State program. The four general criteria used to identify continued claimants matching the CPS concept are (1) UI claimants that reside in Oregon, (2) claimants that match the CPS definition of unemployment, (3) claims for benefits during the CPS reference week, and (4) individuals.

PROMIS can also provide detailed information about the UI data used in LAUS and MLS programs. Some "statistical purposes" information is collected during the UI benefit filing process and registration in iMatchSkills. This is information about age, gender, race, ethnicity, education, industry of previous employer, and occupation of claimants. PROMIS breaks UI claims out by the "statistical purposes" information to provide reports about the demographic, occupation and industry composition of the claims inputs for LAUS and MLS.

Table 1 contains 2008 and 2009 first quarter averages of continued claimants by industry from PROMIS. These counts reflect quarterly averages of individuals residing in Oregon who have claimed a week of benefit payments during the CPS reference weeks and match the CPS definition of unemployment.

Total PROMIS continued claimants more than doubled (+105.3%) between the first quarters of 2008 and 2009. This increase is consistent with total unemployment, which increased by 101.2 percent during the same time period. Each industry experienced an increase in claims between 2008 and 2009. The largest percentage increases in 2009 occurred in the manufacturing (182.2%), professional and technical services (164.4%), and wholesale trade (153.4%) industries. The smallest year-to-year increases occurred in utilities (23.4%), finance and insurance (28.5%), and public administration (44.0%).

In the first quarter of 2009, nearly half (49.6%) of UI claims came from three of the 20 major industries: manufacturing (21.0%), construction (17.9%), and retail trade (10.7%). These three industries have comprised the majority of total claims for each of the first quarters from 2006 through 2009. However, their share has steadily risen each year since comprising 39.4 percent of the total in the first quarter of 2006. This increasing share isn't too surprising given that payroll jobs in manufacturing and construction have declined by 11.8 percent and 16.6 percent, respectively, during the same time period, compared to a decline of 4.5 percent for nonfarm payroll jobs in all industries.

The distribution of claimants by education level has remained fairly constant over the first quarters of the last three years; nearly 70 percent of the total has completed 0 through 12 years of education, and roughly one-quarter have attended one or more years of college.

There was a slight drop in the percentage with an education of grade 0 to 12 between the first quarter of 2008 and the first quarter of 2009, and a slight increase in claimants with two and four years of college during the same time period. It's not clear if these changes in percentages are related to the business cycle or instead reflect more typical year-to-year fluctuations.

LAUS estimates of unemployment are based on the labor force concepts used in the Current Population Survey (CPS). The CPS is a survey of households conducted by the U.S. Census Bureau. In LAUS, Oregon and all states' unemployment data are estimated using statistical methods (a mathematical model of CPS results) based on CPS definitions. In CPS, unemployment is defined as "all persons not employed during the CPS reference week but who were available for work and had made specific efforts to find employment during the last four weeks."

The ETA produces reports on the "insured unemployed" that contain data on activity in Oregon's unemployment insurance (UI) system. UI reports from ETA contain counts of Initial Claims (first time filers for UI benefits), Continued Claims (weeks claimed for benefit payments), and other information related to the administration of the UI program. The concept of "insured unemployment" is based on Oregon's UI laws; UI counts are based on claims and eligibility for benefit payments under the rules and laws that administer the UI system.

The fact that there are two sources of unemployment information for Oregon (like the U.S. and all states) has caused some misconceptions about the unemployment rate that is published each month. The LAUS program is responsible for Oregon's unemployment rate, which is based on the CPS concept of unemployment. The CPS interview does not consider eligibility for UI claims in determining who is unemployed. The basic monthly CPS questionnaire can be found online at:  www.bls.census.gov/cps/bqestair.htm. Likewise, Oregon's UI laws do not consider whether a person is unemployed according to the CPS definition in the determination of eligibility for UI benefit payments.

Data from PROMIS provide a way to dig deeper into the composition of claimants that are used in LAUS estimation algorithms. Since the sources of these claims reports, the UI system and iMatchSkills, are not designed with the intention of collecting statistics about the labor market (but rather to pay UI benefits and match job skills to employment opportunities), some care should be taken in using the UI data as labor market information. But if the claims data can be reconciled with other economic data, a fuller picture of Oregon's labor market emerges.

The PROMIS and ETA UI reports differ in how each source defines and measures "unemployment," since the two programs have different purposes for their reports. ETA reports are designed to measure the performance and administration of the UI system, while PROMIS reports are designed to provide labor market information. For this reason, the counts of claimants in each report are different. The PROMIS reports should not be interpreted as measures of the performance of Oregon's UI system, since they are not designed for this purpose.