CES Birth-Death Model Frequently Asked Questions

1. Do the net birth-death forecasts vary from first preliminary to final estimate?

   No, the net birth-death forecast for a given month does not change between first preliminary, second preliminary, and final sample-based estimates. However, the net birth-death forecasts used in the re-estimated values from April of the benchmark year to December of the benchmark year are updated during the benchmarking process to reflect newly forecasted birth-death values.

2. Are net birth-death forecasts seasonally adjusted?

   No, they are calculated using population data that is not seasonally adjusted, and the forecasts are applied to the sample-based not seasonally adjusted estimates. Months with generally strong seasonal increases such as April, May, and June generally have a relatively large positive forecast. Conversely, months with overall strong seasonal decreases, such as January, generally have a relatively large negative forecast.

3. Can I subtract the net birth-death forecast from the seasonally adjusted over-the-month change to determine what it is adding to employment?

   No, net birth-death forecasts are a component of the not seasonally adjusted estimate and, therefore, are not directly comparable to the seasonally adjusted monthly changes.

   The not seasonally adjusted employment estimate for each month comprises the sample-based component plus the birth-death model forecast. Employment estimates are subsequently seasonally adjusted. Seasonal adjustment smooths the employment series by removing normal seasonal variations due to factors such as weather and holidays.

   Users who wish to compare the model's contribution to overall employment change reported for a month should compare against the unadjusted estimates, not the seasonally adjusted series. Comparing the model amounts to seasonally adjusted estimates misrepresents the model-based component's contribution to over-the-month employment change.

   The birth-death model component generally shows the same overall seasonal patterns as the unadjusted sample-based component. For example, total nonfarm employment records a large drop in employment each January and the model forecasts a substantial drop in net birth-death employment each January. Table 1 shows the net birth-death component and the over-the-month change in not seasonally adjusted total private employment.

   Table 1. Net birth-death forecasts and over-the-month change in total nonfarm employment (in thousands)

   	April 22	May	June	July	August	September	October	November	December	January	February	March 23
   Birth-death model amount	323	254	93	327	120	-153	511	21	-37	-144	176	-29
   Not seasonally adjusted total employment change	1,038	842	926	-342	410	524	1,232	601	-298	-2,505	1,144	457

4. Can BLS provide an estimate of the contribution of the net birth-death forecast to the seasonally adjusted monthly payroll change?

   BLS does not calculate an estimate of the seasonally adjusted contribution of the birth-death model. The sample, the imputation of business births using deaths, and the net birth-death forecast are all necessary components for obtaining an accurate total employment estimate. The components are not seasonally adjusted separately because they do not have any particular economic meaning in and of themselves.

5. How frequently are the net birth-death forecasts revised, and why do the values change?

   Prior to the release of preliminary January 2011 employment estimates in February 2011, net birth-death forecasts were calculated on an annual basis and then applied each month during development of monthly estimates. With the release of the January 2011 preliminary estimates, CES began updating the net birth-death model component of the estimation process on a quarterly basis instead of annually. This change allows for the incorporation of QCEW data into the birth-death model as soon as it becomes available and reduces the post-benchmark revision in the CES series. This change does not impact the timing or frequency of CES monthly and annual releases or when benchmarking is done. For more information on the timing and updates to the inputs for the net birth-death forecasts, please see Introduction of Quarterly Birth-Death Model Updates in the Establishment Survey.

6. How long has CES been using the net birth-death forecasts, and is the history of birth-death available?

   Implementation of the birth-death model was associated with the implementation of a probability-based sample design and estimator. The methodology was phased in gradually under the following schedule:

   Table 2. Schedule for Implementation of the Birth-Death Model

   Release Date	Benchmark Date	Industry Updated to Birth-Death Model
   June 2000	March 1999	Wholesale trade on an SIC-basis
   June 2001	March 2000	Mining, construction, and manufacturing on an SIC-basis
   June 2002	March 2001	All of total private with the exception of services on an SIC-basis
   June 2003	March 2002	All total private industries with the conversion to NAICS

   The net birth-death forecasts used in previously published estimates can be found on the Historical Net Birth-Death Forecasts webpage. More information about the conversion from SIC to NAICS is available in Notice of Conversion to the North American Industry Classification System 2002 and Other Series Changes in June 2003.

7. Were estimates purely sample-based before 2003?

   No, prior to the implementation of birth-death modeling, CES used a technique known as bias adjustment to account for business births, business deaths, and other limitations of the survey. More information about the bias adjustment model and the CES transition to the birth-death model is available in the 2002 Benchmark Article.

8. Were changes made to the net birth-death model to account for the effects of the COVID-19 pandemic on employment estimates?

   Yes, changes were made to the model for the period from March 2020 to September 2021. The widespread disruption to labor markets due to the COVID-19 pandemic and the potential impact to the birth-death model prompted BLS to both revisit research conducted in the aftermath of the Great Recession (2008-2009) and incorporate new ideas to account for changes in the number of business openings and closings. Two areas of research were implemented to improve the accuracy of our birth-death model in the CES estimates. These adjustments better reflected the net effect of the contribution of business births and deaths to the estimates. These two methodological changes were the following:

   • A portion of both reported zeros and returns from zero in a given month from the sample were used in estimation to better account for the fact that business births and deaths do not offset.
   • Each month's sample growth rates were included in the net birth-death forecasting model to better account for the changing relationships between business openings and closings.

   Firstly, a proportion of reports that fell to zero employment and reports that returned from zero employment in each month were used to adjust the weighted contribution of each report used in the calculation of the over-the-month change of the sample-based estimates. Typically, reports with zero employment in either of two adjacent months are not included in estimation. To account for an excess amount of reports going to zero employment as well as reports returning from zero employment, BLS calculated the likelihood that either a reported zero or a return from zero exceeded what would be expected for the month. These “excess declines to zero” and “excess returns from zero” (collectively called excess reported zeroes) partially account for drops in employment (when more business deaths than are usually observed in historical population data occur) and for increases in employment (when there are more business births than normal). More specifically, “excess declines to zero” were used in March 2020 final and subsequent months' first preliminary, second preliminary, and final estimates. “Excess returns from zero” were used in first, second, and final estimates from May 2020 to September 2021.

   Secondly, BLS adjusted the portion of business births and deaths that cannot be accounted for using our sample data by including more recent information. Net birth-death forecasts are normally modeled using an auto-regressive integrated moving average (ARIMA) based on 5 years of historical birth-death residual values that end 9 months before the forecast of a given month. Instead of using only historical data—data that would not accurately account for how the labor market has changed due to COVID-19—a regression variable that includes data up to the given month was included in the model. The regression variable is the CES sample-based ratio of over-the-month change, known as the sample link, for each of the major industry sectors. Each major industry sector sample link was used for the basic-level industries only within that sector.

   The original birth-death model is described in detail in the CES Handbook of Methods in the Business births and deaths section. Additional information is available in a research paper called Impact of business births and deaths in the payroll survey by Kirk Mueller. A paper called Back to the Future: Using Current Regression Variables to Forecast Forward from Historical Net Birth-Death Employment by Victoria Battista details previous research into using a regressor as part of the net birth-death model.

 

Last Modified Date: February 2, 2024