Entry, Innovation and Productivity Growth in the U.S. Economy: Facts and Open

Questions (i.e., Puzzles)

John HaltiwangerUniversity of Maryland

Dynamics of Entry, Productivity dispersion and Productivity growth

2

-0.01

-0.005

0

0.005

0.01

0.015

Years 4-6 Years 7-9

Changes in Productivity Dispersion and Growth from a 1% (one time) Increase in Entry Rate (Years 1-3), High Tech

Dispersion(High Tech) Growth(High Tech)

Surge in entry in a given 3-yearperiod leads to:• Rise in within industry productivity dispersion and decline in industryproductivity growth in next 3-yearPeriod• Decline in within industryproductivity dispersion and rise

in industry in subsequent 3-yearperiod • Surge in reallocation followingsurge in entry as well (not depicted).• Similar, dampened patterns for Non-Tech

Source: Foster et. al. (2018)Using 4-digit NAICS data for High Tech sectors (ICT in mfg and non-mfgplus sectors such as Bio Tech)

Up or out!

0

2

4

6

8

10

12

14

16

18

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16+

Perc

ent o

f em

ploy

men

t

Firm age

Job destruction from exiting firms

Net job creation of continuing firms

Sour

ce: D

ecke

r et a

l. (2

014)

Note! Age group 0 not shown because they only create jobs (no destruction)

A view of the skew

-60

-40

-20

0

20

40

60

80

100

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16+

Empl

oym

ent g

row

th ra

te (D

HS d

enom

inat

or)

Firm age

Median90th percentile10th percentile

Source: Decker et al. (2014). Employment-weighted distributions.

A view of the skew – High Growth Firms are Disproportionately Young Firms

-60

-40

-20

0

20

40

60

80

100

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16+

Empl

oym

ent g

row

th ra

te (D

HS d

enom

inat

or)

Firm age

Median90th percentile10th percentile

Source: Decker et al. (2014). Employment-weighted distributions.

6

Large Differences in Skewness Across Sectors – High 90-50 in High Tech Driven by Young Firms

0

5

10

15

20

25

30

35

High Tech Retail

90-50

50-10

50-1090-50

Source: Decker et. al. (2016)

7

0.00

0.20

0.40

0.60

0.80

1.00

1.20

High-techyoung

High-techmature

Non-techyoung

Non-techmature

Productivity Dispersion (within industry)

0.00

0.05

0.10

0.15

0.20

0.25

0.30

High-techyoung

High-techmature

Non-techyoung

Non-techmature

Responsiveness to Productivity Differentials

Young Businesses Exhibit More (Labor) Productivity Dispersion and Greater Responsiveness to Productivity Differentials

Source: Decker et. al. (2018)

8

Annual Labor Productivity Growth, High Tech/Non Tech, 1987-2015

Annualized Labor Productivity Growth, Quarterly, 1987:1 to 2018:1, U.S. Private Non Farm

Source: BLS

-6

-4

-2

0

2

4

6

8

10

12

1987

1988

1989

1991

1992

1994

1995

1996

1998

1999

2001

2002

2003

2005

2006

2008

2009

2010

2012

2013

2015

2016

2017

Labor Productivity Growth Smoothed (HP Filter)

-10

-5

0

5

10

15

20

Tech HP Filter Non Tech HP Filter

Tech Actual NonTech Actual

9

Startup and Exit Rates in U.S. Private Sector, 1981-2017

5

7

9

11

13

15

1981

1983

1985

1987

1989

1991

1993

1995

1997

1999

2001

2003

2005

2007

2009

2011

2013

2015

2017

Perc

ent o

f Firm

sStartup Rate

Exit Rate

Source: Business Dynamic Statistics (Census) Spliced with Business Employment Dynamics (BLS)

Entrepreneurship by industryRetail falling gradually, 1980s-present

High tech flat or rising in 1980s-1990s, falling after 2000.

Source: Decker et al. (2018)

Times series patterns of skewness (high growth) vary dramatically across sectors

Retail: dispersion declineequal parts 90-50, 50-10High Tech: Growing Skewness in 1990s, sharpDecline post 2000

15

20

25

30

35

40

45

5019

7919

8119

8319

8519

8719

8919

9119

9319

9519

9719

9920

0120

0320

0520

0720

0920

11

Retail 90-50Retail 50-10Tech 90-50Tech 50-10

High Tech includes (most of) Information but also High Tech Mfg and Services. Source: Decker et. al. (2016)

Skewness primarily accountedfor by Young Firms. In High Tech,Decline in young firms and declineIn High Growth Firms in High Tech

0

10

20

30

40

50

60

70

80 90-50 Old Tech

50-10 Old Tech

90-50 Young Tech

50-10 Young Tech

12

-5

0

5

10

15

20

25

30

35

40

Median

90thPercentile

High Growth vs. Median Growth Firms in High-Tech (Employment-Weighted Distribution)

Source: Tabulations from the Longitudinal Business Database (Census). HP Trends depicted

Rising Productivity Dispersion and Declining Responsiveness

Productivity Dispersion “Responsiveness”

0.00

0.20

0.40

0.60

0.80

1.00

1.20

LP dispersion (within-industry sd)

Tech youngTech matureNon-tech youngNon-tech mature

0.10

0.15

0.20

0.25

0.30

0.35

0.40

Growth differential

-0.14

-0.12

-0.10

-0.08

-0.06

-0.04

-0.02

0.00

Exit differential

Source; Decker et. al. (2018)

14Source: Tabulations from LBD (Census) by Decker et. al. (2018) spliced with Business Employment Dynamics (BLS)

Job Reallocation Rate (Hodrick Prescott Trends) for U.S. Private Sector, High-Tech and Retail Sectors

20

25

30

35

40

45

1979

1981

1983

1985

1987

1989

1991

1993

1995

1997

1999

2001

2003

2005

2007

2009

2011

2013

2015

2017

Retail Economywide High-tech

15

8.0

13.0

18.0

23.0

28.0

33.0

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

2016

2017

Worker Reallocation (H+S)

"Excess" Churning (H-JC=S-JD)

Job Reallocation (JC+JD)

Declining Entrepreneurship and Business Dynamism Part of Broader Decline in Labor Market FluidityHigh Pace of Fluidity, Dynamism and Entreprenenurship Important for Job Ladders of Young Workers

Source: Updated chart from Davis and Haltiwanger (2014)

Facts and Puzzles• Periods of rapid innovation (especially in innovative intensive industries like

High Tech):• First surge of entry• Then experimentation (dispersion)• Then productivity growth• Potentially long (and variable) lags

• Both innovative intensive industries (High Tech) and other industries have seen relatively modest entry and productivity growth post 2000.

• Part of declining entry, dynamism and labor market fluidity post 2000.• Dispersion in Productivity Growth in High Tech and Non Tech has risen

substantially in the post 2000 period• Experimentation that has not yet resolved?• Diminished Dynamism – Slower diffusion or slower adjustment dynamics

16