Bureau of Labor Statistics > Employment Projections > Methods

Occupational Separations Methodology Frequently Asked Questions

Summary

BLS implemented a separations method for measuring occupational separations with the 2016–26 projections that replaced the replacements method of measuring replacement needs. The following responses address common questions about the new method.

What do projections of separations measure?
Why does BLS create projections of both employment change and separations?
Why are openings from separations higher than openings from growth?
Should I pay more attention to projections of employment change or occupational openings?
How do projections of occupational openings relate to job openings measured by the Job Openings and Labor Turnover Survey (JOLTS)?
Are openings the same as training needs for an occupation?
Why don’t occupations with many older workers approaching retirement have the highest separation rates?
Does the separations method measure all separations from an occupation?
How should projections of openings be used?
Are there openings in declining occupations?
Why are the current projections of openings so much higher than what BLS has previously published?
Why did BLS change methods for measuring openings?

Top

What do projections of separations measure?

Separations measure the number of workers who are projected to leave an occupation. This includes individuals who retire, leave the labor force for reasons other than retirement, or transfer to a different occupation. It does not include workers who change jobs but stay employed in the same occupation. In all of these cases, the separation of the worker reduces the number of people employed in the occupation. For growing occupations, the separation creates an opening for a new worker to enter the occupation. For declining occupations, not all workers who separate need to be replaced.

Why does BLS create projections of both employment change and separations?

Opportunities for new workers to enter an occupation come from two sources – newly created jobs and existing jobs where the incumbent worker has left the occupation. BLS projections of employment change measure the first source and show the net change in employment in an occupation (growth or decline). BLS projections of separations measure the second source and show opportunities that exist in occupations with flat or declining employment. For career guidance and workforce development purposes, the cause of an opening – whether from growth or separations – is not important. In fact, many newly hired workers may not even know if they are filling a job created through growth or from a separation.

Why are openings from separations higher than openings from growth?

For many occupations, the number of openings that arise from the need to replace workers who separate from an occupation is much larger than the number of openings that arise from growth in the occupation. Openings from separations are determined by the rate at which people leave an occupation, while openings from growth are determined by the rate at which new jobs are created. There is no explicit connection between these rates; either can be higher than the other, but in practice, most occupations have growth rates that are lower than the rate of separation of existing workers, meaning more openings are created due to separations than due to growth.

Should I pay more attention to projections of employment change or occupational openings?

Projected employment change by occupation provides information on how the workforce is changing – which occupations are becoming more common, and which are declining. These variations in the occupational composition of the workforce are driven by structural changes in the economy, such as increases in demand for particular goods or services, or the impact of new technology on labor productivity. Employment change also provides information on the relative demand for workers in an occupation – those with accelerating growth may need more new workers than they did in the past, while declining occupations may need fewer. Occupational openings, however, provide the most complete picture of demand for new workers in an occupation as it encompasses all sources of demand, not just demand from new jobs that the projections of employment change represent. Both metrics represent demand only, however, and do not take into account the supply of available workers, which impacts whether a labor shortage or surplus exists. BLS does not create projections of supply.

How do projections of occupational openings relate to job openings measured by the Job Openings and Labor Turnover Survey (JOLTS)?

BLS data on job openings, hires, and separations from the JOLTS program measure different concepts than projections of occupational openings and separations. The JOLTS job openings metric is a stock measure of the number of job openings, or positions that are open, not filled, on the last business day of a month. In contrast, occupational openings are projections of the number of positions that are created and filled on an annual basis. Between the last business days of two consecutive months, some existing job openings are filled while new job openings are created, so there is no way to determine how many of the job openings in the first month are the same as the job openings in the second. There is therefore no way to create an annual average metric of job openings from JOLTS that could be compared with the annual average number of occupational openings.

Two other JOLTS metrics, however, are flow metrics that can be compared with occupational openings and separations. Hires, or new employees added to the payroll, and separations, or employees removed from the payroll, are measured on a monthly basis, but can be summed over all months to get an annual total. In 2018, there were 68.9 million hires and 66.1 million separations. The 68.9 million hires are comparable to the 19.7 million occupational openings, with the difference in magnitude being that JOLTS hires cover all workers who start new jobs with a different employer, while occupational openings covers all workers who start new occupations. Therefore the reason the JOLTS hires metric is several times larger than the occupational openings metric is that JOLTS hires data capture workers who change employers but not occupations, in addition to those who change occupations, which is what is estimated by occupational openings. Notwithstanding the differences in time periods, this implies that workers are much more likely to change jobs than change occupations.

Similarly, the JOLTS separations estimate of 66.1 million and the occupational separations estimate of 18.9 million separations differ because the JOLTS separations metric measures movements out of jobs, while the occupational separations metric measures movements out of occupations.

With both sets of metrics, the differences between JOLTS hires and separations, and between occupational openings and separations, are a proxy for net employment change.

Are openings the same as training needs for an occupation?

Some occupations require workers to have specific education or training before they can enter. Prospective entrants may or may not need training depending on their prior work experience, education, and training. For example, a worker entering the labor force for the first time will generally need to earn a bachelor’s degree before starting work as a budget analyst; however, a worker who was employed for several years as an economist will not generally need additional education before starting work as a budget analyst. The number of entrants who need training depends on the characteristics and source of the supply of workers, and BLS does not produce projections of supply. As a result, the BLS projections of occupational openings are not direct estimates of training needs for an occupation, as not all of the workers who will fill openings require training.

Why don’t occupations with many older workers approaching retirement have the highest separation rates?

As workers approach traditional retirement ages, they become more likely to leave the labor force. However, older workers are less likely to transfer to different occupations than younger workers, since older workers generally have more experience in an occupation, which can correlate with higher pay and greater responsibilities, some of which may not be transferrable to other occupations. As a result, occupations with many older workers tend to have higher labor force exit rates, but lower occupational transfer rates, than occupations with a younger workforce. As a result, occupations with many older workers do not consistently have the highest overall separations rates.

Does the separations method measure all separations from an occupation?

The separations method is designed to estimate the number of workers who leave their occupation and need to be replaced by new entrants into the occupation. It is not a measure of all movement in and out of occupations, but instead an estimate of workers who permanently leave an occupation.

The separations method identifies a worker as having left an occupation permanently if they either exit the labor force for at least 4 consecutive months, or if they transfer to an occupation within a different Standard Occupational Classification (SOC) major occupational group. This significantly limits the number of separations identified as generating job openings. The number of individuals who leave the labor force for at least 4 consecutive months is less than half the number who leave for 1 month, and only a portion of occupational transfers are to different SOC major groups.

How should projections of openings be used?

BLS publishes an annual average number of openings for an occupation over the 10-year projections period. While BLS releases a specific number, this number should not be considered a precise target for the number of openings expected in any given year, for a number of reasons:

The number of openings is an annual average over 10 years, not a specific estimate for each year.
Projections cannot account for unexpected events during the 10–year horizon that will affect the number of openings in an occupation.
The methodology for projecting separations requires assumptions that may not apply equally well to all occupations.

As a result, projections of openings are best used to give a general sense of the scale of expected openings, especially in relation to other occupations. Using rankings or percentiles of occupations based on occupational openings is an appropriate use of the data that conveys the scale of results without undue focus on the precise number of openings. For example, ordering occupations by number of occupational openings and grouping by decile (example: Top 10 percent) provides context to the number of openings by allowing identification of occupations with large, moderate, or small numbers of openings.

Are there openings in declining occupations?

In most declining occupations, the number of workers projected to separate from the occupation is larger than the projected decline in employment, meaning there will be occupational openings for new workers. In rare cases, the decline in employment may be equal to the projected number of separations, which would result in no openings for an occupation, but these cases are limited to those occupations that are declining very rapidly.

Why are the current projections of openings so much higher than what BLS has previously published?

BLS instituted a new method to measure openings from separations with the 2016–26 projections. This method does a better job of capturing all the movements of workers out of an occupation that result in opportunities for new entrants. The prior method undercounted openings because it only accurately measures workers who follow the traditional concept of a career path–entering an occupation at a young age, working in the same occupation for many years, then retiring–which is not the case for many workers in most occupations. The current method measures more of the separations that are actually occurring and therefore this method yields higher values of projected separations and openings for occupations.

The larger values for projected openings by occupation do not reflect actual increases in opportunities for workers but rather a better estimate of the opportunities that were already occurring. Published estimates from the 2014–24 projections and prior years should not be compared with the 2016–26 and later projections to estimate how the number of openings has changed over time.

Why did BLS change methods for measuring openings?

BLS used a cohort–component method for estimating openings due to replacement needs from 1991 through the 2014–24 projections. BLS identified statistical and conceptual issues with the implementation of this method that compromised the accuracy and validity of the resulting estimates, and therefore developed the current separations method to measure this concept.

On the statistical side, the old replacements method required using historical CPS data broken down by detailed occupation and by age cohort. Estimates derived from small sample sizes result in large margins of error, which contributed to the large observed variability of published estimates from projections cycle to projections cycle.

On the conceptual side, in order for the cohort–component method to accurately estimate replacement needs, it requires a strong assumption: that workers who need to be replaced are those who are replaced by workers from different (usually younger) cohorts. In a traditional conception of a worker’s career, this assumption is valid: workers enter at a young age, work in their field until they are old, and then retire, creating opportunities for the next generation of young workers. In this framework, occupation is fixed throughout a worker’s career. However, many workers do not remain in the same occupation throughout their career, and BLS determined that a newer, more robust and more statistically sound model was necessary.

The separations method has the following advantages:

More clarity about what is being measured

The separations method directly measures workers who leave an occupation, taking advantage of the longitudinal aspects of the CPS monthly survey and supplements. The replacements method indirectly measured leavers by measuring employment change by age group.
In addition to measuring total separations, the separations method provides discrete estimates of labor force leavers and occupational transfers for each occupation. The replacements method provided just a single, total, measure for each occupation.

More robust methods

The separations method uses a regression to model projected separations, while the replacements method extrapolated from historical trends. The regression approach allows the separations method to incorporate many different demographic characteristics, where the replacements method only accounts for age.
The separations method better accounts for future growth or declines by incorporating the results of the projections of employment change for occupations. The replacements method assumed historical patterns of growth or decline of occupations would continue.

More reliable results

The separations method allows for estimates for occupations newly added to classification systems, while the replacements method required 10 years of comparable historical data to create estimates for an occupation.
The separations method more accurately estimates replacement needs for small occupations, while the replacements method required proxies for many small occupations.
The replacements method generated results for some occupations that varied considerably from year to year, due to survey error. The regression approach used in the separations method reduces survey error resulting in more consistent results.

Last Modified Date: September 6, 2023