Service Sector Productivity in
y Japan: p
An Analysis Based on the JIP Database
October 2010
O
t b 2010
Kyoji Fukao
y j
(Hitotsubashi University and RIETI)
Tsutomu Miyagawa
i
(
(Gakushuin University and RIETI)
y
)
1
1. Motivation
1. Motivation
• As the share of the service sector in the economy i
increases, productivity growth in this sector is d i i
h i hi
i
becoming the key to Japan’s future economic growth. • But productivity in this sector has stagnated.
g
y
y
p
• Using mainly industry level data of factor inputs and productivity, we examine why service sector p
productivity in Japan has stagnated. y
p
g
• We also try to answer what Japan needs to do to improve service sector productivity.
improve service sector productivity.
2
1. Why is the Service Sector is Important for Japan’s Future Growth?
• Historically
Historically, labor productivity in the tertiary sector was not labor productivity in the tertiary sector was not
much lower than in the secondary sector except during the period of rapid industrialization from 1947‐1965.
period of rapid industrialization from 1947
1965.
• But total factor productivity (TFP) growth in the manufacturing sector has been much higher than that in
manufacturing sector has been much higher than that in other sectors. • However, the share of the manufacturing sector is declining However, the share of the manufacturing sector is declining
rapidly.
• Hence, for Japan, which is now experiencing population Hence for Japan which is now experiencing population
decline, productivity growth in the service sector is the key. 3
Comparison of sectoral labor productivity (primary=1)
6
5
4
3
Secondary
T ti
Tertiary
2
1
0
Notes:
Subsidiary occupations are taken account of for periods before the Second World War.
Data before 1965 are on a per worker basis. Data from 1973 are on a man-hour basis.
The primary sector consists of agriculture, fishery and forestry. The secondary sector consists of mining,
manufacturing and construction. The tertiary sector consists of all other sectors except imputed rent.
Sources:
Data for periods between 1885-1940 are from Osamu Saito and Tokihiko Settsu (2009), "Unveiling historical
occupational
p
structures and its implications
p
for sectoral labour p
productivity
y analysis
y in Japan’s
p
economic ggrowth,"
,
paper presented at INCHOS 2009, 28-30 July 2009, King’s College, Cambridge, p. 31, table 11.
Data for periods between 1947-65 are from Ohkawa, K. and M. Shinohara, eds., (1979), Patterns of Japanese
Economic Development: A Quantitative Appraisal (New Haven: Yale University Press), p. 41,Table2.12.
Data for periods after 1970 are from the JIP Database 2009.
4
Figure 5. TFP growth in manufacturing and non-manufacturing (market
economy excluding imputed house rent), 1970-2006 (1970=1)
3
Manufacturing
25
2.5
2
1.5
1
1970
0
1972
2
1974
4
1976
6
1978
8
1980
0
1982
2
1984
4
1986
6
1988
8
1990
0
1992
2
1994
4
1996
6
1998
8
2000
0
2002
2
2004
4
2006
6
0.5
5
Source: JIP Database 2009.
Share of the manufacturing sector in the macro-economy: Japan-US Comparison
Nominal value added of the manufacturing
sector/nominal GDP: Japan
0.4
Number of workers in the manufacturing sector/total
number of workers in the economy: Japan
Nominal value added of the manufacturing
sector/nominal GDP: US
0.35
Number of workers in the manufacturing sector/total
number of workers in the economy: US
0.3
0 25
0.25
0.2
0.15
Sources: JIP 2006 and Council of Economic Advisers, Economic Report of the President , 2005.
2000
1995
1990
1985
1980
1975
1970
1965
1960
0.1
6
2. Why does Productivity in Japan’s Service Sector Stagnate?
• Japan
Japan experienced relatively high TFP growth experienced relatively high TFP growth
in the ICT producing sector. • However, the problem Japan is that TFP growth in ICT‐using sectors, such as distribution services (retail, wholesale and transportation) and in the rest of the
transportation) and in the rest of the manufacturing sector (i.e., excluding electrical machinery) declined substantially after 1995
machinery), declined substantially after 1995. • And these ICT‐using sectors have larger shares in the economy than the ICT‐producing sector.
7
TFP Growth in the Market Sector: by Sector and by Country
Manufacturing, excluding
electrical machinery
Electrical machinery, post
and communication
3.0
3.0
10.0
9.0
8.0
7.0
60
6.0
5.0
4.0
3.0
2.0
1.0
0.0
Other goods-producing
industries
2.5
2.0
2.0
1.0
1.5
-1.0
US
Italy
UK
France
F
Geermany
Korea
US
S
Italy
y
UK
K
Francee
US
Italy
UK
France
Germany
Korea
Japan
-0.5
-1.0
Germany
y
0.0
Japan
0.0
0.5
Korea
a
19952005
1.0
Japan
n
198095
-2.0
-1.5
-3.0
Other goods producing industries
Finance and business services
Distribution services
1.5
4.0
4.0
Personal and social services
1.0
3.0
3.0
0.5
2.0
-2.0
US
Italy
UK
France
F
Geermany
40
-4.0
Korea
-1.0
US
Ittaly
UK
ance
Fra
Germany
-1.5
0.0
-3.0
-5.0
-1.0
Japan
US
Italy
UK
France
Germany
Korea
-2.0
Japan
-1.0
1.0
Ko
orea
-0.5
0.0
Jap
pan
0.0
2.0
1.0
-2.0
-2.5
25
-3.0
Source: Kyoji Fukao, Tsutomu Miyagawa, Hak K. Pyo and Keun Hee Rhee (2009), “Estimates of Total Factor Productivity, ICT Contributions and Resource Reallocation Effects in Japan and Korea.”
8
International comparison of labor productivity by sector
• Labor productivity levels in Japan’s non‐manufacturing sectors are less than 50 percent of the U S levels
are less than 50 percent of the U.S. levels. • In the case of non‐ICT capital service input per man‐hour, Japan uses much more such input than the United States
uses much more such input than the United States.
• With regard to labor quality, there is not much difference between the two countries.
between the two countries. • The differences in labor productivity are mainly caused by p
p
p
Japan’s low level of ICT‐capital service input and low TFP. • Since ICT investment may contribute to innovation in production processes, the difference in ICT‐capital service input between Japan and the other countries is likely one of the major causes of the stagnation of TFP in Japan. 9
10
International comparison of labor productivity by sector
(contd.)
• We should note that it is very difficult to compare differences in the quality of services across countries and to measure PPP for
the quality of services across countries and to measure PPP for services. • The quantity of truck transportation is basically measured by a The quantity of truck transportation is basically measured by a
unit of tonnage times kilometers per man‐hour (how much transportation in tonnage and distance was conducted by one man‐hour) and differences in quality, such as just‐in‐time delivery, are not taken account. • In retail services, the length of business hours is not taken into I
il
i
h l
h fb i
h
i
k i
account. Measured productivity of German shops might be higher than that of Japanese shops because German shops are
higher than that of Japanese shops because German shops are only open until, say, 6 p.m. and customers adjust their lifestyle accordingly. 11
Accumulation of ICT assets in Japan and Korea was very slow in comparison with other countries.
comparison with other countries.
Figure 3-2 ICT Investment/GDP Ratio in the Major Developed Countries
14
12
Japan
Korea
10
US
UK
%
8
6
France
Germany
Italy
4
2
0
Source: Kyoji Fukao, Tsutomu Miyagawa, Hak K. Pyo and Keun Hee Rhee (2009), “Estimates of Total Factor Productivity, ICT 12
Contributions and Resource Reallocation Effects in Japan and Korea.”
Why was Japan left behind in ICT investment? • The ratio of custom software investment to packaged software investment is much larger in Japan than in g
p
the United States. When Japanese firms introduce ICT technology, such
• When Japanese firms introduce ICT technology, such as an ICT system for customer services or the management of information flows within the firm,
management of information flows within the firm, they prefer custom software in order to avoid business reorganization and the training of workers.
business reorganization and the training of workers. • This results in a smaller productivity improvement from ICT investment This example suggests that it is
from ICT investment. This example suggests that it is important to compare intangible investment in Japan with that in other developed economies
with that in other developed economies.
13
Intangible Investment in Japan
• The intangible investment/output ratio in Japan is h
bl
/
smaller than that in the United States. • The intangible investment/gross value added ratio in Japan’s non‐manufacturing sector is much lower than that in Japan’s manufacturing sector.†
• The low level of intangible investment probably is one g
p
y
important reason for the stagnation of TFP growth and the delay of ICT investment in Japan’s service sector.
y
p
† Fukao, Kyoji, Tsutomu Miyagawa, Kentaro
Mukai, Yukio Shinoda, and Konomi
Tonogi (2009), “Intangible Investment in Japan: Measurement and (
) “
bl
d
Contribution to Economic Growth,” Review of Income and Wealth 55(3): 717‐736.
14
Japan invests a Japan
invests a
lot in R&D but very little in
very little in economic competencies
competencies such as brand equity, firm‐
equity, firm
specific human capital, and
capital, and organizational structure.
structure. Intangible investment by category : share in total intangible investment
100%
90%
80%
Computerized
information
70%
60%
Science and
engineering R&D
50%
40%
Other innovative
property
30%
Brand equity
20%
Firm-specific
resources
10%
0%
Japan,
2000-05
US, 19982000
UK, 2004
Sources: Japan: Fukao et al. (2009), US: Corrado, Hulten and Sichel (2006),
UK: Marrano and Haskel (2006).
15
Intangible investment in Japan is relatively small in comparison with the US.
Figure 10-a. Japan
25
%
25
23
23
21
21
19
Figure 10-b. US
17
Intangible
Investment
15
Tangible
Investment
19
Tangible
Investment
17
15
13
13
11
11
9
9
7
7
5
5
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
Intangible
Investment
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
%
Figure 10: Business investment (Percentage of business output): Japan-US comparison
Sources: Fukao, Miyagawa, Mukai, Shinoda, and Tonogi (2009). Original data for the US are taken from Corrado, Hulten and 16
Sichel (2005, 2006).
The intangible investment/gross value added ratio in Japan’s non‐manufacturing sector is much lower than that in Japan’s manufacturing sector
17
Intangible investment in the service sector decreased
drastically after 1995.
When we focus on the retail industry, we find that intangible
assets have contributed negatively to output growth and
negative TFP growth in the retail industry since 2000.
In several non‐manufacturing sectors, we observe large negative reallocation effects and exit effects on industry‐level labor productivity y
p
y
growth. It seems that Japan’s low metabolism also impedes productivity growth in service sectors.
Decomposition of labor productivity increase in non-manufacturing sectors: average value of results for 1997－
99 and results for 2000-02, annual values, %
Ind str name
Industry
Construction
Wholesale
Retail
Real estate
Transportation
Warehousing and other transportation related services
Communication
Electricity
Gas, water and heat supply
Hotels
Broadcasting
Personal services
Business services
Information services
Eating and drinking places
Oth services
Other
i
Advertisement
Amusement
Medical and care services and hygiene
Source: Kim, Kwon and Fukao (2009).
Labor
Reallocation
productivity
prod
cti it Within effect
effect
growth
a=b+c+d
-4.80%
4 05%
4.05%
4.74%
1.78%
-4.84%
-3.48%
36.83%
-1.11%
-3.93%
-0.23%
-10.19%
10.19%
0.19%
-0.32%
-2.62%
0.77%
-0.30%
0 30%
-6.21%
3.69%
0.23%
b
0.07%
1 59%
1.59%
2.53%
1.85%
0.61%
0.72%
8.60%
6.88%
0.56%
1.75%
0.22%
0.36%
1.83%
-0.64%
0.49%
0 12%
0.12%
-2.64%
3.40%
0.94%
c
-5.03%
0 83%
0.83%
1.02%
-0.37%
-3.91%
-4.37%
25.56%
-8.06%
-5.35%
-1.03%
-15.38%
15.38%
-1.74%
-2.78%
-3.70%
-1.15%
0 17%
0.17%
-9.85%
-0.26%
-0.06%
Net entry
y
effect
Entr effect
Entry
E it effect
Exit
d=e+f
0.16%
1 63%
1.63%
1.19%
0.30%
-1.53%
0.17%
2.67%
0.07%
0.86%
-0.95%
4.96%
1.58%
0.63%
1.72%
1.44%
-0.59%
0 59%
6.29%
0.56%
-0.65%
e
1.48%
1 33%
1.33%
0.95%
-0.66%
-1.80%
-1.34%
1.11%
0.03%
0.98%
-2.08%
4.85%
-0.13%
-1.54%
0.49%
-2.33%
0 99%
0.99%
9.81%
-1.73%
-2.54%
f
-1.32%
0 30%
0.30%
0.24%
0.95%
0.27%
1.50%
1.56%
0.04%
-0.13%
1.13%
0.11%
1.70%
2.17%
1.22%
1.74%
0 45%
0.45%
-3.52%
2.28%
20
1.89%
The start‐up and closure rates of establishments in Japan are much lower than those in the US.
Fi
Figure
4.1
4 1 Start-up
S
and
d Cl
Closure R
Rate off E
Establishments:
bli h
J
Japan-US
US C
Comparison
i
Figure 4.2.Panel B. Closure Rate:
Japan-US Comparison %
Figure 4.1.Panel A. Start-up Rate: JapanUS Comparison %
16
16
14
14
12
12
10
All industries: US
All industries:
Japan
10
8
Wholesale, retail
and restaurants:
Japan
6
6
Services: Japan
4
4
2
2
0
0
Manufacturing:
Japan
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
8
Note:
N
t D
Data
t are b
based
d on statistics
t ti ti ffrom th
the employment
l
t iinsurance program off each
h country.
t
Sources: Small Business Administration, US Government (1998), Small and Medium Enterprise Agency, Ministry of
Industry, Trade and Industry, Japanese Government (2001), and Study Group on “Industry Hollowing-out” and Tariff
21
Policy, Ministry of Finance, Japanese Government.
Figure 11. Increase in part-time workers in Japan
70,000,000
2,400
65,000,000
2,300
60,000,000
Self-employed and family
workers (JIP)
55,000,000
Part-time workers (JIP)
50,000,000
Full-time workers
45,000,000
Total number of workers
(JIP)
2,200
2,100
2,000
1,900
40 000 000
40,000,000
Total number of workers
(Labor Force Survey)
1 800
1,800
35,000,000
Average working hours (JIP)
1,700
1,600
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
Jan-09
Feb-09
F
Mar-09
M
Apr-09
A
May-09
M
Jun-09
J
Jul-09
Aug-09
A
Sep-09
Oct-09
Nov-09
N
Dec-09
D
Jan-10
Feb-10
F
Mar-10
M
Apr-10
A
May-10
M
Jun-10
J
Jul-10
Aug-10
A
30,000,000
Source: JIP Database 2010. 22
Figure 11. Increase in part-time workers in Japan
パート比率の推移
Percentage of part‐time workers in total workers by sector: 1970‐2005
35 0
35.0
30.0
電子計算機・同付属装置
通信機器
自動車
25.0
小売業
金融業
20 0
20.0
％
電信・電話業
情報サービス業
15.0
Computers and related devices
C
Communication
i ti equipment
i
t
Automobile
Retail
Finance
Telecommunications
Information services
10.0
5.0
Source: JIP Database 2008. 20
04
20
02
20
00
19
98
19
96
19
94
19
92
19
90
19
88
19
86
19
84
19
82
19
80
19
78
19
76
19
74
19
72
19
70
0.0
23
Increase in part‐time workers in Japan (contd.)
• Firms invest little in on‐the‐job training and off‐the‐job i
i
li l i
h j b i i
d ff h j b
training for part‐time workers. The increase in part‐time workers will slow down human capital accumulation
workers will slow down human‐capital accumulation.
• The productivity gap between part‐time workers and regular workers is larger than the wage gap between part‐
l
k i l
th th
b t
t
time workers and regular workers. It seems that firms pay a premium to part time workers in order to obtain flexibility
premium to part‐time workers in order to obtain flexibility of employment.†
• The increase in part‐time workers is not only a problem of The increase in part time workers is not only a problem of
inequality but also a problem of productivity.
†
Kyoji Fukao, Ryo Kambayashi, Daiji Kawaguchi, Hyeog Ug Kwon, Young Gak Kim, Kyoji
Fukao Ryo Kambayashi Daiji Kawaguchi Hyeog Ug Kwon Young Gak Kim
and Izumi Yokomaya (2007) “Deferred Compensation: Evidence from Employer‐Employee Matched Data from Japan,” Hitotsubashi University.
24
Size and Internationalization of Firms
In the case of commerce and other service sectors, US firms are much ,
larger than Japanese firms on a consolidated basis. It seems that Japanese firms are left behind in internationalization and in enjoying scale economies. Number of listed firms and their average size measured by consolidated workers: US-Japan comparison, consolidated
g
Manufacturing
Wholesale and retail
Other services
Number Average
Average Number Average
Average Number Average
Average
Number
Number
Number
of
number
number
number
number of
number of
number
of US
of US
of US
Japanese of
Japanese of
Japanese of
of
of
of
firms
firms
firms
firms
workers
workers firms
workers
workers firms
workers
workers
2000
909
7,306
1,451
7,145
262
2,554
363 19,620
148
1,866
655
7,183
2001
774
6,611
1,466
6,729
243
1,626
361 20,302
157
1,462
655
7,096
2002
1,079
4,799
1,490
6,501
418
1,093
368 19,857
315
906
691
7,745
2003
1,291
4,546
1,558
6,314
540
1,207
375 21,841
398
855
690
7,418
2004
1,567
4,702
1,575
6,560
630
1,506
390 21,404
496
999
717
7,658
2005
1,622
4,960
1,606
6,376
668
1,597
389 23,175
564
981
715
7,736
Communication
Co
u cat o
Other
Ot
e industries
dust es
2000
17 16,422
96 11,576
234
5,970
763
5,302
2001
13 19,438
102 10,938
214
4,343
827
4,981
2002
23 11,193
108 10,664
307
2,944
848
5,111
2003
32
8 046
8,046
121
9 453
9,453
367
2 858
2,858
894
4 822
4,822
2004
44
6,440
122
9,204
466
3,191
963
5,120
2005
50
5,881
123
9,281
493
3,052
967
5,195
25
Source: Kwon (2010)
Summary
• TFP growth in the manufacturing sector is much higher than that in the other sectors. However, the share of the manufacturing sector is declining rapidly. For Japan, which is now experiencing population decline, productivity growth in the service sector is key to economic growth.
productivity growth in the service sector is key to economic growth.
• TFP growth in ICT‐using sectors declined substantially after 1995.
p
y
• Accumulation of ICT assets in Japan and Korea was very slow in comparison with other countries.
• The low level of intangible investment probably is one important reason f
for the stagnation of TFP growth and the delay of ICT investment in f
f
Japan’s service sector.
• It seems that Japan
It seems that Japan’ss low metabolism also impedes productivity growth.
low metabolism also impedes productivity growth
• Firms invest little in on‐the‐job training and off‐the‐job training for part‐
time workers. The increase in part‐time workers will slow down human‐
p
capital accumulation.
• It seems that Japanese firms are left behind in internationalization and in enjoying scale economies. j i
l
i
26
Summary
• If Japan’s comparatively low investment in ICT indeed is the result of a reluctance to reorganize business structures and retrain workers, this may in turn be linked to the increase in part‐time workers, which at least in part is a consequence of Japan’ss labor market structures. Possible government is a consequence of Japan
labor market structures. Possible government
policies to address these issues would include the provision of structures and incentives that facilitate and reward investment in ICT (and intangible assets) and in the training of part‐time workers and mitigate the ) di h
i i
f
i
k
d ii
h
polarization in the labor market between regular workers and part‐time workers while at the same time providing greater labor market flexibility. p
gg
y
• Slow growth, the low metabolism of the economy, and the fact that Japan’s service sector has fallen behind in terms of internationalization and exploiting economies of scale are all interrelated. Many areas of the d
l iti
i
f
l
ll i t
l t d M
f th
service sector remain heavily regulated, thus preventing competition and g
greater dynamism through the entry, growth, and exit of firms. Obvious y
g
y g
policy measures would be to proceed with deregulation and, for instance, facilitating inward foreign direct investment to encourage the entry of firms that already operate internationally and enjoy economies of scale
firms that already operate internationally and enjoy economies of scale.
27