From the November 2013 South Dakota e-Labor Bulletin
Unemployment rates across the nation have been declining as the recovery from the recession continues. South Dakota's current (October 2013) not seasonally adjusted unemployment rate is 3.4 percent, compared to the national rate of 7.0 percent. South Dakota typically has one of the lowest unemployment rates in the nation. We currently rank second, behind North Dakota, which has typically claimed the lowest rate in the nation since the oil boom hit.
With our low unemployment rates in South Dakota, it is often assumed there are few jobs available in South Dakota. And if there are, workers may not be readily available to fill these positions. However, the unemployment rate in and of itself does not provide the whole story regarding the available workforce; estimates of available workers are heavily influenced by the willingness of workers to change jobs.
Decisions regarding willingness to change jobs are usually related to wages, fringe benefits, hours and working conditions. Most workers willing to change jobs could be referred to as the underemployed. The underemployed includes persons working full-time or part-time jobs who earn below their earning capacity or level of competence. Underemployment has also been defined as "involuntary part-time" employment or employment of a person on a part-time basis when full-time work is desired.
In 2000, the Labor Market Information Center (LMIC) created a data model to gauge the level of underemployed. This data model factors in those without jobs, workers who live in the area and persons who commute into the South Dakota to work. Monthly 'labor supply' estimates by county have since been published to assist in estimating the level of available labor for South Dakota.
When the labor supply model was first developed, the main inputs included a variety of decennial data from the U.S. Census Bureau. Also included was data produced through the Quarterly Census of Employment and Wages (QCEW), Local Area Unemployment Statistics (LAUS) and Current Employment Statistics (CES) programs, which are administered by the LMIC in cooperation with the U.S. Bureau of Labor Statistics (BLS). Over time, the model methodology evolved as more sophisticated data sources became available from the U.S. Census Bureau through its Local Employment Dynamics (LED) program.
The current methodology includes data inputs from the Current Employment Statistics (CES) program (also referred to as the Nonfarm Wage and Salaried Workers data), Local Area Unemployment Statistics (LAUS) program and the Quarterly Workforce Indicators (QWI) data developed by the Census Bureau through the LED program.
A recent review of the various QWI data elements available resulted with questions regarding whether the most appropriate 'new hires' variable was being used within the model to calculate the most representative estimate of the labor supply.
The new hire data is incorporated into the data model to estimate how many workers are willing to change jobs, which combined with the current level of unemployed, provides an estimate of available labor. Because several different 'hire' and 'employment' indicators are available from the LED program, identifying the most appropriate data sets can be a challenge, as they are all useful for different purposes.
The QWI data element most currently used in the model to estimate the number of workers willing to change jobs is 'HirN', which represents an estimate of new hires that includes stable, seasonal and temporary employment. Seasonal and temporary employment components by nature tend to be cyclical. Therefore, use of the 'HirN' in the data model overstates the churning level within the labor market, as the labor supply model was intended to measure the number of workers seeking stable employment, rather than seasonal or temporary employment.
An alternate measure is being incorporated into the data model to better reflect the intent of the labor supply model. The QWI measure 'HirNS' variable estimates the number of stable jobs - workers who started a job they had not held within the past year and the job turned into a job that lasted at least a full quarter with a given employer.
Therefore, the variables HirNS (hires new stable) and EmpS (full-quarter employment stable) are now incorporated in the data model, to estimate new hire rates and provide a more defined indicator of the relative share of 'nonfarm workers who may be willing to change jobs'. This model change is reflected within the October 2013 labor supply estimates published in this issue of the South Dakota e-Labor Bulletin and will be used going forward.