U. S. Bureau of Labor Statistics Future Developments in Business Employment Dynamics Kristin Fairman, Sheryl Konigsberg Views expressed here are those.

U. S. Bureau of Labor Statistics

Future Developments in Business Employment Dynamics

Kristin Fairman, Sheryl Konigsberg

Views expressed here are those of the authors and not BLS.

2

Agenda• Business Employment Dynamics (BED)

– What are business employment dynamics?– BED components– How are BED data used?– Current BED data– Why BED is so exciting

• Recent BED Developments– Industry data– Size class data– Firm survival

• Future BED Developments– State data

• Ongoing BED Research– Annual data– Birth and Death data

3





• Ongoing BED Research– Annual data– Birth and Death data– Establishment age– Entrepreneurship

4

What are Business Employment Dynamics?

• Set of statistics measuring changes in employment at the establishment level

• BED data demonstrate the idea that jobs are constantly created and destroyed

• Job growth is the net result of jobs being created, or gained, and destroyed, or lost

• Data source for BED is the Quarterly Census of Employment and Wages

5

Quarterly Census of Employment and Wages

• Census of all U.S. establishments– at least one employee – accounts for nearly 98% of all U.S. non-farm

workers

• QCEW is a universe, or complete list, and is not subject to errors such as sampling or estimation error

6

BED Components• Establishment – economic unit that

produces goods or services, usually one physical location

• Gross Job Gains – sum of employment at opening and expanding establishments– Openings – establishments that did not exist

or reported zero employment in the prior quarter and report positive employment in the current quarter

– Expansions –establishments that report employment increases over the quarter

7

BED Components• Gross Job Losses – sum of employment at

closing and contracting establishments– Closings – establishments reporting positive

employment in the prior quarter and report zero employment or do not exist in the current quarter

– Contractions - existing establishments that report employment decreases over the quarter

8

How are BED data used?

• BED data used to assess:– the business cycle– the level of labor market volatility– the effect of establishment employment

changes on aggregate employment

• A decrease in employment stemming from gross job gains decreasing has far different business cycle implications than a decrease due to gross job losses increasing

9

Current BED Data

• National data – first published for fourth quarter 2002

• National data by industry – first released for third quarter 2003

• National data by firm size class – first released for second quarter 2005

• All data is available from third quarter 1992

10

Why BED is so exciting

• Job “churning”, or reallocation, not seen in net job data

• Expansions and contractions are larger than at openings and closings

• Gross job gains and losses have business cycle properties

• Gross job gains remained low in 2006, while gross job losses are on decline

11

12

13






14

Recent Developments: Industry• Service-providing and goods-producing sectors

created more jobs than they destroyed during 2005 and 2006

• Job creation and destruction has slowed in most industries – Construction industry– Education and healthcare industry

• Manufacturing sector – experiencing a decline in the overall number of jobs – the numbers of gross jobs lost has been declining in

recent quarters

15

16

17






18

Recent Developments: Size Class• Largest and smallest firms have largest shares

of gross job gains and gross job losses– Smallest firms tend to have the highest share of jobs

created from openings and jobs lost from closings– Largest firms tend to have the highest share of jobs

created from expansions and jobs lost from contractions

• Firms with 1000 or more employees and firms with 20-49 employees have been the largest contributors to net job growth

19

Share of total quarterly gross job gains and gross job losses by firm size, June 2006 to September 2006, seasonally adjusted

0.0

2.0

4.0

6.0

8.0

10.0

12.0

14.0

16.0

18.0

1-4 5-9 10-19 20-49 50-99 100-249 250-499 500-999 1,000 ormore

Size Class - Number of Employees

Perc

ent

Share of gross job gains Share of gross job losses

20

Recent Developments: Size Class

• Jessica Helfand, Akbar Sadeghi, “Employment Dynamics: small and large firms over the business cycle”, March 2007– Small firms, those with 1 to 499 employees, create about 64

percent of new jobs – Share of growth of small firms is larger than their base share of

employment causing small firms to become large, increasing the employment share of large firms over time

– Firms of different size class behave differently throughout the phases of the business cycle

– Contribution of large firms to net job gains during the current economic recovery appears to come from a fall in gross job losses, rather than increased job creation

– Bulk of net job losses in the 1991 recession occurred in small firms, while large firms generated the majority of job losses during the economic slowdown of 2001

21






22

Firm Survival• Amy Knaup, “Survival and Longevity in the

Business Employment Dynamics Data,” May 2005– 66% of business establishments opening in

second quarter of 1998 were still in existence 2 years later

• Amy Knaup and Merissa Piazza, "Establishment Survival Using the BLS Longitudinal Database," December 2006– Extends original cohort through 2005

23

Firm Survival:Seven Year Survival Study

• Findings from Knaup, Piazza study:– Survival rates across industries tend to stay

consistent over time– Employment patterns still vary more than

survival rates– Surviving establishments continue to increase

employment over their lifetime – A greater percentage of establishments

survive in the fifth year and beyond

24

212,182

81.2%

65.8%

54.3%

44.4%

38.3%34.4%

31.2%

0.0%

10.0%

20.0%

30.0%

40.0%

50.0%

60.0%

70.0%

80.0%

90.0%

100.0%

active births 1 year later 2 years later 3 years later 4 years later 5 years later 6 years later 7 years later

Length of Survival

Figure 1. Survival rates of new establishments from second quarter of 1998

25






26

Future BED Developments: State data

• Expected release date: Summer 2007• Preliminary data for second quarter 2006 shows:

– Gross job gains exceeded gross job losses in 43 states

– Alaska had the highest rates of gross job gains, gross job losses and net change

• Preliminary data for September 1992 to June 2006 shows:– States with high rates of job growth tended to have

higher rates of job churn– Nevada had the largest net change in employment – Alaska had the highest rates of both gross job gains

and gross job losses

27

Rates of Gross Job Gains, Gross Job Losses and Net Change by State, March 2006 - June 2006, seasonally adjusted

4.5

5.5

6.5

7.5

8.5

9.5

10.5

11.5

12.5

Ken

tuck

yS

outh

Car

olin

aM

issi

ssip

piW

est V

irgin

iaK

ansa

sIn

dian

aT

enne

ssee

Mar

ylan

dA

rkan

sas

Ala

bam

aN

ebra

ska

Min

neso

taM

isso

uri

Nor

th D

akot

aO

hio

Okl

ahom

aV

irgin

iaM

ichi

gan

Flo

rida

Pen

nsyl

vani

aW

yom

ing

Nat

iona

lC

alifo

rnia

Ore

gon

Wis

cons

inD

elaw

are

New

Jer

sey

New

Nor

th C

arol

ina

Iow

aT

exas

New

Yor

kH

awai

iM

aine

Nev

ada

DC

Ver

mon

tC

onne

ctic

utG

eorg

iaA

rizon

aM

onta

naS

outh

Dak

ota

Loui

sian

aId

aho

Col

orad

oM

assa

chus

ett

Rho

de Is

land

Illin

ois

Was

hing

ton

Uta

hN

ew M

exic

oA

lask

a

Rat

es o

f G

ross

Jo

b G

ain

s an

d G

ross

Jo

b L

oss

es

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

Rat

e o

f N

et C

han

ge

Gross Job Gains Gross Job Losses Net change

28






29

Ongoing BED Research:Annual data

• Josh Pinkston and Jim Spletzer paper analyzed annual BED data – "Annual Measures of Job Creation and Job

Destruction Created from Quarterly Microdata"

• Openings, expansions, closings and contractions CAN NOT be summed through the year to get an annual number

30

Ongoing BED Research :Annual data

• Preliminary data shows:– Over calendar year 2005, 26.4 million jobs

were created and destroyed– 14.3 million jobs were gained and 12.1 million

jobs were lost – 2.1 million new jobs were added to the U.S.

labor market– 131,000 new establishments were added to

the economy

31






32

Ongoing BED Research: Establishment Births and Deaths

• How are births different from openings?• Opening: any establishment that reports positive

employment in current quarter and did not exist or was reporting zero employment in just the ONE prior quarter

• Birth: any establishment that reports positive employment in current quarter and did not exist or had zero employment for the prior FOUR quarters

• The difference between openings and births are seasonal re-openings

33


• How are deaths different from closings?• Closing: any establishment that was reporting

positive employment in prior quarter but has zero employment or does not exist in the ONE following quarter

• Death: any establishment that was reporting positive employment in prior quarter but has zero employment or does not exist in the FOUR following quarters

• The difference between closings and deaths are temporary shutdowns

34


• Preliminary estimates show• In 2006 first quarter, there were 219,000

births – 368,000 openings over same period

• In 2005 first quarter, there were 185,000 deaths – 348,000 closings over same period

* Data is seasonally adjusted


Measuring Births and Deaths in Business Employment Dynamics

Data Series

Akbar Sadeghi

Preliminary data for research purposes only: do not cite.

36

Separating Births from Openings and Deaths from Closings

• Openings: – Establishments with employment for the first time at the end of the

quarter, or establishments that increase their third month employment from zero.

– Openings include births and re-openings of seasonal businesses.• Closings:

– Establishments with Positive third month employment in the previous quarter and zero employment in the current quarter.

– Closings include deaths and temporary shutdown of seasonal businesses.

• Estimates of births and deaths will lead to other measures of business demographics such as:– Age – Survival rates– Entrepreneurship

37

How to Measure Births and Deaths

• Two main approaches:– First appearance (births) and last appearance

(deaths) in database with positive employment. – Zero or no employment in several quarters followed

by positive employment (births), and positive employment in a quarter followed by zero employment for several quarters (deaths)

• We selected the second approach. Some records appear for the first time with zero employment, not counted as births in the first approach.

38

How to Measure Births and Deaths

• What month of employment?– For the second approach, what month of employment should be used?

Third month or all three months of quarters?• How many months or how many third months should be considered

in identifying a birth in the second approach?• To be consistent with other BED data elements, we selected the

third month as employment reference month.• We selected 4 quarters of zero or no employment for both birth and

death definitions:– Birth: Positive third month employment followed four quarters of zero or

no employment.– Death: four quarters of zero employment followed the last quarter of

positive third month employment• Including the current quarter, five quarters of employment are

compared for each record.• Five quarters eliminate the seasonality effects.

39

How many birth options tested?• Option 1: births are new records that appeared for the first time in

the database with positive third month employment.• Option 2: births are records with positive employment in the third

month of a quarter and zero or no employment in the third month of the last four quarters.

• Option3: births are records with positive employment in the third month of a quarter and zero or no employment in the last five quarters.

• Option 4: births are records with positive employment in the third month of a quarter and zero or no employment in all months over of the last four quarters.

• Option 5: births are records with positive employment in the third month of a quarter and zero or no employment in all months over of the last five quarters.

• Firm level births or entrepreneurial births.

40

Entrepreneurial Births

• Entrepreneurial births are births at the firm level, as opposed to establishment level

• Establishment: an economic unit engaged in one or predominantly one activity at single location

• Firm may consist of several establishments. Firm-level data are compiled on aggregation of establishment under common ownership.

• A new addition to a multi-unit firm is a birth, but not an entrepreneurial birth.

41

Number of Births

140000

150000

160000

170000

180000

190000

200000

210000

220000

230000

95 96 97 98 99 00 01 02 03 04 05

Option 1Option 2Option 3

Option 4Option 5

Chart 1: Number of births, seasonally adjusted

42

Job Gains from births

800000

900000

1000000

1100000

1200000

1300000

1400000

95 96 97 98 99 00 01 02 03 04 05

Option 1Option 2Option 3

Option 4Option 5

Chart 2: Gross jobs gained by births, seasonally adjusted

43

Number of Births is rising

100000

150000

200000

250000

300000

350000

400000

95 96 97 98 99 00 01 02 03 04 05

Firm-level births (statewide)Establishment-level births

Firm-level births (national)Openings

Table 5: Number of openings, establishment and firm level births, seasonally adjusted

44

Declining average size of births and rising productivity

Chart 7: Average size of establishment birth against multifactor productivity

4

4.5

5

5.5

6

6.5

7

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005

Ave

rag

e b

irth

siz

e

85

90

95

100

105

110

115

Pro

du

ctiv

ity

Average size of birth Multifactor productivity index

45

Job growth vs. net establishment growth (birth rate minus death rate)

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

95 96 97 98 99 00 01 02 03 04 05

Birth rate minus death rate Net change in employment

Chart 10: Job growth vs. establishment growth (birth rate minus death rate)

46

Births per 1000 in labor force

.75

.80

.85

.90

.95

95 96 97 98 99 00 01 02 03 04 05

Chart 11: Firm births per 1000 in labor force

47

Conclusion

• Recent developments: currently being released– National BED data by industry– National BED data by size class

• Future developments– BED data by state

• Expected release summer 2007

• Ongoing BED research– Annual BED data– Establishment Births and Deaths– Establishment Age– Entrepreneurship


For any comments or questions, please contact:

[email protected]

U. S. Bureau of Labor Statistics Future Developments in Business Employment Dynamics Kristin Fairman, Sheryl Konigsberg Views expressed here are those.

Documents

current bed data national

establishment level

death data slide

quarter slide

net job data expansions

lost data source

positive employment

aggregate employment