© Copyright 2003 by the Wyoming Department of Employment, Research & Planning
Turnover as a Measure of Labor Market Activity
by: Douglas W. Leonard, Research Analyst
"By developing small businesses whose products and services require workers with advanced skills, wages should increase and exit rates decrease."
Where employment trend analysis is concerned, some media outlets and analysts emphasize top-level aggregate effects. Employment growth is nearly always interpreted as being “good” while employment losses receive the opposite interpretation. What often goes unnoticed are the effects of labor market events not visible when performing aggregate analysis. In conducting a more rigorous analysis, we strive to learn not only if the market is growing or declining but also what underlying structural trends may contribute to aggregate changes. This article explores such underlying events by using turnover (exit rates) as a measure of labor market
stability.1 In addition, demographic variables such as age and gender are used to describe how exiting behavior manifests itself in Wyoming’s labor market.
Data Sets and Methods
The primary data set used in this analysis is Wyoming’s Wage Records database.2 In order to simplify the analysis, average annual values are displayed. The analysis also includes data from Wyoming’s Current Employment Statistics Survey,3 Covered Employment and Wages program,4 and Local Area Unemployment Statistics program.5 The span of data utilized is from 1993 to 2001.6
Turnover is defined as an individual exiting a firm.7 We exclude from the analysis those individuals who were hired and left employment with a firm during the same calendar quarter.
The same workers are not necessarily evaluated from year to year in the longitudinal charts due to arriving and exiting workers. The arrival and exit of workers can occur multiple times in a given year or over several years.
Average Annual Exit Rates
Exit rates increased steadily between 1993 and 2001 with only two exceptions, 1997 and 2000 (Figure 1). Overall, unemployment rates decreased during the 1993 to 2001 period. Employment growth was robust from 1993 to 1994, decelerated during 1995 and 1996, and accelerated again beginning in 1997.
The data show exit rates vary not only by year of analysis but also by the age of the individuals employed. Figure 2 shows average exit rates by age for three years: 1994, 1998, and 2001. Once workers reach age 20, a nonlinear relationship exists between age and the average annual exit rate, which means that exit rates are high for younger workers, decrease during middle age, and increase again around retirement age. For the three years of data analyzed, the greatest amount of separation8 occurs between ages 30 and 55. In addition, the change in exit rates for this age range appears to have decelerated somewhat from 1998 to 2001 as compared to the 1994 to 1998 period.
When viewing age versus average exit rates, the curve characteristics do not change appreciably with a change in gender or industry. What we generally observe is a level shift in the curves. To enhance readability in the remaining figures, only the data for those from 30 to 60 years of age are shown.
The same type of nonlinear relationship shown in Figure 2 exists between age and average exit rates by gender (Figure 3). As the figure shows, average exit rates for both men and women increased between 1994 and 2001. However, rates for men appear to have increased more rapidly, as the separation (vertical distance) between the male and female lines in 2001 is more discernable.
One may intuitively assume that larger firms generally have lower exit rates and tend to be more stable than smaller firms due to resource availability and experience. However, do these rates vary significantly for people who are the same age within firms of different sizes? Data in Figure 4 show, for the most part, the size of firm does not significantly affect average exit rates when controlling for age cohort. Significant effects are not evidenced unless the largest firms (those averaging 100 or more employees per quarter) are compared to other firms.
Just as average annual exit rates vary with firm size and gender, we also expect average annual exit rates vary by industry. In Figure 5 we compare one of Wyoming’s largest industries in terms of total employment, Retail Trade, with one of its smallest industries, Finance, Insurance, & Real Estate (FIRE).9 Exit rates are significantly lower in FIRE than in Retail Trade at all age levels. Furthermore, average annual exit rates in FIRE also tend to be much lower than in all other industries combined.
Wyoming’s labor market is becoming increasingly unstable as individuals are changing jobs more frequently, even though employment is steadily increasing. If the individual changes jobs for career advancement, this is seen as a positive step from the individual’s perspective. However, from the employer’s point of view, the individual is taking their experience and knowledge with them to new employers. Moreover, each time a job change occurs, costs are incurred by the firms that lose and hire workers.10 New employees often must be hired or current employees retrained to take the departing employee’s position. There are also job search costs for individuals making job changes. Some costs involve time and others involve the use of funds, such as when job search materials are purchased or an out-of-town interview or change of residence is required. However, these costs are sometimes overlooked by the individual if the change leads to a significant pay increase.
While gender and industry both affect exit rates, age offers one of the larger contrasts in average annual exit rates (Figure 4). Since very large firms (100+ employees) have much lower exit rates than smaller firms, increasing the number of large firms may help reduce turnover. In doing so, the overall stability of the workforce increases, thereby decreasing job switching and its associated costs.
However, one should not view the establishment of large firms as the only solution. A balanced mixture of firms of various sizes may be ideal because if employment becomes more concentrated in large firms (especially in small communities), the closure of a single large firm can have more negative effects on the economy than the closure of small firms. In addition, by developing small businesses whose products and services require workers with advanced skills, wages should increase and exit rates decrease.
Although many, if not most, people focus on overall job growth as a barometer of economic health, it is clear that the undercurrent of turnover may be undermining the state’s long-term economic prospects. To what extent public policy can stem the undesirable effects of exit rates is unclear. This challenge will not be resolved by focusing on aggregate data alone. A more detailed look at underlying factors is required to fully understand the dynamics of Wyoming’s labor market.
1Murray, S. (Ed.). (2002, December). Market dynamics from administrative records: Seven state project report for December 2002. [On-line]. Available: http://doe.state.wy.us/LMI
2Wage Records is an administrative database. Each employer in the state who has employees covered under Unemployment Insurance, by law, must submit quarterly tax reports to the state showing each employee’s Social Security Number and wages earned in the quarter.
3The Bureau of Labor Statistics cooperates with state Employment Security agencies in the Current Employment Statistics (CES) or establishment survey to collect data each month on employment, hours, and earnings from a sample of nonfarm establishments (including government). The national sample includes about 160,000 businesses and government agencies, which cover approximately 400,000 individual worksites drawn from a sampling frame of over eight million Unemployment Insurance tax accounts. The active CES sample includes approximately one-third of all nonfarm payroll workers. From these data, employment, hours, and earnings information in considerable industry and geographic detail are prepared and published each month.
4The Covered Employment and Wages program publishes a quarterly count of employment and wages reported by employers covering 98 percent of U.S. jobs, available at national, state, county, and Metropolitan Statistical Area (MSA) levels by industry at http://www.bls.gov/cew/home.htm
5The Local Area Unemployment Statistics program produces monthly and annual employment, unemployment, and labor force data for U.S. Bureau of the Census regions and divisions, states, counties, metropolitan areas, and many cities by place of residence at http://www.bls.gov/lau/home.htm#overview
6The span of data matches that of the Wage Records database. The Covered Employment and Wages (CEW) program was an exception as only one year of data (2000) is used in the analysis. Year 2000 data are used because it was the last year in which Standard Industrial Classification (SIC) codes were used for CEW. SIC was replaced by the North American Industry Classification System (NAICS).
7Turnover represents the average percentage of a firm’s workforce that exits quarterly, excluding those that left employment with a firm during the same calendar quarter they were hired. Exit rates were aggregated to yearly averages and can range from 0 to 100 percent.
8Separation in this case refers to the vertical distance between exit rate lines (i.e., males vs. females in the same year).
9In 2000, the average monthly employment in Retail Trade was 46,846 and the average monthly employment in Finance, Insurance, & Real Estate was 8,054.
10Spence, M. (1973, August). Job market signaling. Quarterly Journal of Economics, 355-374.
Table of Contents | Labor Market Information | Wyoming Job Network | Send Us Mail
designed by Julie Barnish.
Last modified on by Susan J. Murray.