Divergence Fax

Using Monthly Employment Change to Verify Hires and Separations Data

Over the past year, a number of employers have requested a clarification of the JOLTS definitions and reporting instructions. Specifically: whether a firm’s monthly employment change must equal the reported Hires minus Separations.

Example: Firm A reports:

Month	Employment	Openings	Hires	Separations
December	10	3	0	0
January	14	2	5	1

In this case, Employment from December to January is fully explained by the “5” Hires and “1” Separation – but this may not always be true. In many industries, this relationship will be very close each month, or will average-out across several months; however, there are a number of reasons why it may not be exact, including:

• the reference period for employment is the 12^th of the month, whereas Hires and Separations have an end-of-month reference period;

• part-time employees may not be scheduled to work every pay period, which could cause a temporary employment drop, but no actual separations; or

• employees on an extended non-pay leave status who have not been officially removed from the firm’s personnel roster.

While this is not an exact relationship, respondents should use it as a helpful guide when preparing their monthly data.

Using Monthly Employment Change to Help Verify Hires and Separations Data:

Does Employment Change = Hires – Separations ?

We are requesting our respondents to check this relationship each month as they prepare their data. This will help identify possible Hires/Recalls or Separations which may have been missed. Correspondingly, our data collectors also will periodically request respondents to help clarify their reported data where the employment change appears out-of-line with the reported Hires and Separations data. Again, as noted above, there are valid reasons for differences to occur.

The employment relationship has proven very useful in developing accurate Hires and Separations data; using this relationship, respondents have been able to identify instances where:

• Hires data did not include the “Recalls” who were reported as Layoffs two (2) months earlier; and

• Separations data did not include the subsequent layoff of temporary, seasonal hires (e.g., Christmas build-up/layoff cycle).