Entry, Innovation and Productivity Growth in the U.S ...
Transcript of Entry, Innovation and Productivity Growth in the U.S ...
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