Intangible investment 20121104 - Hitotsubashi University
Transcript of Intangible investment 20121104 - Hitotsubashi University
Intangible Investment in Japan
Prepared for the OECD WORKSHOP ON PRODUCTIVITYOECD Conference Centre, Paris, 5‐6 November 2012
Kyoji FUKAO (Hitotsubashi University and RIETI)YoungGak KIM (Senshu University)
HyeogUg KWON (Nihon University, RIETI)
1
Motivation• Available data on intangible investment at the industry‐and firm‐level is increasing rapidly.
• It is a good time to collect stylized facts on intangible investment, which will contribute to further developments in economic theory on this issue.
• It seems that Japan’s data on intangible investment is relatively advanced in the world.
• Another notable feature of Japan is that intangible investment has stagnated in recent years.
• We collect stylized facts on intangible investment in Japan and try to answer some questions, such as why intangible investment in Japan has stagnated and what type of firms invest actively in intangible assets.
2
Structure of This Presentation1. Recent Trends in Macro‐Level Intangible
Investment in JapanNext, using industry‐level and firm‐level data, we examine the following issues for each category of intangible assets.What industries lead intangible investment?What type of firms invest in intangible assets: large or small, young or old, productive or non‐productive?2. Innovative Property3. Economic Competencies4. Computerized InformationAnalysis is very preliminary. We have not examined the impact of intangible assets on productivity yet.
3
Major Data Sources on Intangible Investment in Japan Industry level: JIP Database 2011: http://www.rieti.go.jp/en/database/JIP2011/index.html• Nominal and real investment flows and real stock data for each category of
intangible investment for 108 industries, which cover the whole economy, are available for 1985‐2009. The JIP Database Project collaborates with the EU KLEMS Consortium and the World KLEMS Database Project, and KLEMS‐type data are available.
• Estimation procedures are explained in:Miyagawa, T., and S. Hisa (2012) “Estimates of Intangible Investment by Industry and
Productivity Growth in Japan,” Gakushuin University. Fukao, K., T. Miyagawa, K. Mukai, Y. Shinoda and K. Tonogi (2009) “Intangible
Investment in Japan: Measurement and Contribution to Economic Growth,” Review of Income and Wealth, Vol. 55, No. 3, pp. 717‐736.
Firm level:Basic Survey of Japanese Business Structure and Activities• The survey is conducted annually since 1994 by the Ministry of Economy, Trade and
Industry (METI). The survey covers all firms with at least 50 employees and 30 million yen of paid‐in capital in manufacturing, mining, commerce, and most service sectors.
• The data include R&D expenditures, expenditures for advertisement, stock of software investment (package and order‐made, available only from 2006), Off‐JT (opportunity cost is not included, available only from 2009). 4
5
1. Recent Trends in Macro‐Level Intangible Investment in Japan
According to recent studies, the contribution of intangible investment to labor productivity growth in Japan is the lowest among the major developed countries.
‐1.0
0.0
1.0
2.0
3.0
4.0
5.0
6.0
Austria
Belgium
Czech Repu
blic
Denm
ark
Finland
France
Germ
any
Ireland
Italy
Nethe
rland
s
Sovenia
Spain
Swed
en
Unite
d Kingdo
m
Unite
d States
Japan
TFP
Labor composition
Tangibles
Intangibles
Labor productivity growth
Source: Corrado et al. (2012) and Miyagawa and Hisa (2012).
Contribution to the growth in output per hour: 1995 to 2007 (annual rate, %)
6Source: JIP Database 2011.
Looking at intangible investment in detail shows that, since around 2000, investment in economic competencies and in computerized information has stagnated particularly.
0%
1%
2%
3%
4%
5%
6%
7%
8%
9%
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
Total
Innovativeproperty
Economiccompetencies
Computerizedinformation
Intangible Investment‐GDP Ratio
7
In the 2000s, intangible assets in economic competencies recorded negative growth in Japan.
Source: JIP Database 2011.
‐5%
0%
5%
10%
15%
20%
1986
1988
1990
1992
1994
1996
1998
2000
2002
2004
2006
2008
Total
Computerizedinformation
Innovativeproperty
Economiccompetencies
Growth Rate of Intangible Assets
In contrast with Japan’s experience, all three categories of intangible assets made a positive contribution to labor productivity growth in the US and European countries (Corrado et al. 2012).
8
9
2. Innovative Property Investment in Japan
Source: JIP Database 2011.
• Innovative property investment is concentrated in a small number of industries.
• Four industries (broadcasting, communication equipment, motor vehicle parts and accessories, and pharmaceutical products), which produce 3% of GDP, conduct 25% of Japan’s total innovative property investment.
• The stagnation in innovative property investment could be due to a decline in the value added share of innovative property intensive industries.
0.0%
0.5%
1.0%
1.5%
2.0%
2.5%
3.0%
3.5%
4.0%
4.5%
0% 20% 40% 60% 80% 100%
Cum
ulat
ive
contr
ibution
to in
nova
tive
pro
pert
y
inve
stm
ent/
GD
P
Cumulative industry value added percentile in GDP
Cumulative Contribution of Industries
to Innovative Property Investment:
2000-08
Using the following identity, we decompose the slowdown of the macro‐level increase in the innovative property investment‐GDP ratio into the inter‐industry effect and the intra‐industry effect.
10
The decline of the value added share of innovative property intensive industries has contributed to the stagnation in investment.
Did the Value Added Share of Innovative Property Intensive Industries Decline in Japan?
i ii
i
i
i
i
ii
i
iT
iT
ii
i
i
iT
iT
ii
iT
iT
iT
iT
ii
ii
iiT
iiT
YY
YZ
YY
YZ
YY
YZ
YY
YZ
Y
Z
Y
Z
,0
,0
,0
,0
,0
,0
,
,
,0
,0
,
,
,
,
,
,
,0
,0
,
,
where Zt,i denotes innovative property investment by industry i in year t and Yt,i denotes the nominal value added of industry i in year t.
Period Total effect Inter‐industry effect Intra‐industry effect (percentage points)
1990‐2000 0.69% ‐0.23% 0.92%2000‐2008 0.40% ‐0.60% 1.00%
In Japan’s manufacturing sector, R&D investment is concentrated in large, old and productive firms. But recently, the importance of smaller, young and less‐productive firms has been increasing.
11
Who Conducts R&D Investment in the Manufacturing Sector?
01
23
Cum
ulat
ive
inte
nsity
ove
r tot
al sa
les o
f the
sect
or (%
)
0 .2 .4 .6 .8 1Cumulative share of sales
1995200020052009
Manufacturing
R&D
LargeSmall
01
23
Cum
ulat
ive
inte
nsity
ove
r tot
al sa
les o
f the
sect
or (%
)
0 .2 .4 .6 .8 1Cumulative share of sales
1995200020052009
Manufacturing
R&D
12
Who Conducts R&D Investment in the Manufacturing Sector – Young or Old Firms?
OldYoung
01
23
Cum
ulat
ive
inte
nsity
ove
r tot
al sa
les o
f the
sect
or (%
)
0 .2 .4 .6 .8 1Cumulative share of sales
1995200020052008
Manufacturing
R&D
13
Who Conducts R&D Investment in the Manufacturing Sector ‐ Productive or Unproductive Firms?
With High TFPWith low TFPNote: The relative TFP level is measured for each industry of the JIP industry classification. Intangible investment is not treated as a factor input.
0.2
.4.6
.81
Cum
ulat
ive
inte
nsity
ove
r tot
al sa
les o
f the
sect
or (%
)
0 .2 .4 .6 .8 1Cumulative share of sales
1995200020052009
Non-manufacturing
R&D
14
Who Conducts R&D Investment in the Non‐manufacturing Sector?
OldYoung
As in the case of the manufacturing sector, smaller, young and less‐productive firms have tended to conduct more R&D than other firms in recent years.
R&D by Small, Young, Less Productive Firms
15
• In the 1990s, the TFP growth of large firms accelerated in both the manufacturing and the non‐manufacturing sector. Small and medium‐sized firms (SMEs) were left behind.
※工業統計表の甲票と乙票を分析対象にしているため、 分析期間を1999年までにしている。
TFP Growth by Factory Size (Annual Growth Rate)
0.0
0.5
1.0
1.5
2.0
2.5
3.0
1980‐1985 1985‐1990 1990‐1995 1995‐1999
Average TFP growth
Top quartile in terms of salesSecond quartile
Third quartile
Bottom quartile
Possible reasons: decrease in technology spillovers from large firms.
Probably, we can explain the recent increase in R&D by small, young and less productive firms as their response to the decline in spillovers.
In the case of investment in economic competencies, the investment‐sales ratio is quite similar across industries.
16
3. Investment in Economic Competencies in Japan
Source: JIP Database 2011.
0%
1%
1%
2%
2%
3%
0% 20% 40% 60% 80% 100%
Cum
ula
tive
contr
ibution t
o inve
stm
ent
in e
conom
ic
com
pete
ncie
s/G
DP
Cumulative industry value added percentile in GDP
Cumulative Contribution of Industries to Investment in
Economic Competencies: 2000-08
In the 2000s, the stock of economic competencies declined in many industries, including most manufacturing industries. However, it increased rapidly in high growth industries such as social insurance and social welfare (non‐profit), information services and internet‐based services.
17Source: JIP Database 2011.
‐20.00%
‐15.00%
‐10.00%
‐5.00%
0.00%
5.00%
10.00%
15.00%
Rice, w
heat produ
ction
Misc
ellaneou
s crop farm
ing
Livestock and sericulture farm
ing
Agricultural services
Forestry
Fisheries
Mining
Livestock prod
ucts
Seafoo
d prod
ucts
Flou
r and
grain mill produ
cts
Misc
ellane
ous foo
ds and
related prod
ucts
Prepared
animal fo
ods and organic fertilizers
Beverages
Tobacco
Textile produ
cts
Lumbe
r and
woo
d prod
ucts
Furnitu
re and
fixtures
Pulp, paper, and
coated an
d glazed
paper
Paper p
rodu
cts
Printin
g, plate making for p
rintin
g and bo
okbind
ing
Leathe
r and
leather p
rodu
cts
Rubb
er produ
cts
Chem
ical fertilizers
Basic
inorganic chem
icals
Basic
organic chemicals
Organic che
micals
Chem
ical fibe
rsMisc
ellane
ous c
hemical produ
cts
Pharmaceu
tical produ
cts
Petroleum produ
cts
Coal produ
cts
Glass and
its prod
ucts
Cemen
t and
its p
rodu
cts
Pottery
Misc
ellaneou
s ceramic, stone
and
clay prod
ucts
Pig iro
n and crud
e steel
Misc
ellaneou
s iro
n and steel
Smeltin
g and refin
ing of non
‐ferrou
s metals
Non
‐ferrou
s metal produ
cts
Fabricated
con
structional and
architectural m
etal produ
cts
Misc
ellaneou
s fabricated
metal produ
cts
Gen
eral indu
stry m
achine
rySpecial ind
ustry machinery
Misc
ellane
ous m
achinery
Office an
d service indu
stry machines
Electrical gen
erating, transm
ission, distrib
ution and indu
stria
l app
aratus
Househ
old electric app
liances
Electron
ic data processin
g machines, digita
l and
analog compu
ter e
quipmen
t and
…Co
mmun
ication eq
uipm
ent
Electron
ic equ
ipmen
t and
electric
measurin
g instruments
Semicon
ductor devices and
integrated
circuits
Electron
ic parts
Misc
ellaneou
s electrical machinery equ
ipment
Motor veh
icles
Motor veh
icle parts and
accessorie
sOther transportatio
n eq
uipm
ent
Precision
machine
ry & equ
ipment
Plastic
produ
cts
Misc
ellane
ous manufacturin
g indu
strie
sCo
nstructio
nCivil engineerin
gElectricity
Gas, heat sup
ply
Waterworks
Water su
pply fo
r ind
ustrial use
Waste disp
osal
Who
lesale
Retail
Fina
nce
Insurance
Real estate
Railw
ayRo
ad transportatio
nWater transportatio
nAir transpo
rtation
Other transportatio
n and packing
Telegraph an
d teleph
one
Education (private and no
n‐profit)
Research (p
rivate)
Medical (p
rivate)
Hygien
e (private and no
n‐profit)
Other pub
lic se
rvices
Advertising
Rental of o
ffice equ
ipmen
t and
goo
dsAu
tomob
ile m
ainten
ance services
Other se
rvices fo
r businesses
Entertainm
ent
Broadcastin
gInform
ation services and
internet‐based
services
Publish
ing
Video picture, sou
nd inform
ation, character inform
ation prod
uctio
n and distrib
ution
Eatin
g and drinking
places
Accommod
ation
Laun
dry, beauty and bath se
rvices
Other services for individu
als
Education (pub
lic)
Research (p
ublic)
Medical (p
ublic)
Hygiene
(pub
lic)
Social insurance and social welfare (p
ublic)
Public adm
inistratio
nMedical (n
on‐profit)
Social insurance and social welfare (n
on‐profit)
Research (n
on‐profit)
Other (n
on‐profit)
1990‐2000 2000‐08Growth Rate of Economic Competencies by Industry
There is a high cross‐industry correlation between the growth rate in economic competencies and the growth rate of other factor inputs.
It seems that the reason Japanese firms do not invest in economic competencies is simply that they expect low sales and factor input growth.
18
Investment in Economic Competencies: Industry‐Level Analysis
Source: JIP Database 2011
y = 0.8826x ‐ 0.0259R² = 0.2822
‐20%
‐15%
‐10%
‐5%
0%
5%
10%
‐10% ‐5% 0% 5% 10% 15%
Growth ra
te of e
cono
mic co
mpe
tencies: 2000‐2008
Growth rate of other factor inputs (capital+labor, a divisia index): 2000‐2008
Cross‐industry Relationship between Growth of Economic Competencies and Growth of Other
Factor Inputs: 2000‐2008
0
10
20
30
40
50
60
1970 1975 1980 1985 1990 1995 2000 2005 2008
Social insurance andsocial welfare (public)
Eating and drinkingplaces
Retail
Miscellaneous foodsand related products
Semiconductor devicesand integrated circuits
Household electricappliances
Finance
Telegraph andtelephone
Motor vehicles
Share of part‐time workers in total workers by sector: 1970‐2008 (in %)
Source: JIP Database 2011
19
It seems that the decline in the accumulation of economic competencies was partly caused by the harsh restructuring resulting from the long‐term economic stagnation. For example, many firms increased the percentage of part time workers in total workers and did not provide intensive training in the case of part time workers. This change reduced training expenditure substantially.
In Japan’s non‐manufacturing sector, younger and less productive firms tend to conduct more advertising than other firms.
20
Who Conducts Advertising in the Non‐manufacturing Sector?
Non-manufacturing
Advertising / Sales (%)lnTFP (-1) -1.164 *** -1.066 *** -0.037 * -0.041 **
[0.029] [0.028] [0.020] [0.020]
lnEMP (-1) 0.309 *** 0.325 *** -0.114 *** -0.118 ***
[0.007] [0.007] [0.013] [0.013]
lnAge -0.255 *** -0.292 *** -0.064 *** -0.063 ***
[0.010] [0.010] [0.015] [0.015]
Indutry dummies Yes Yes Yes Yes Yes Yes Yes Yes
Year dummies Yes Yes Yes Yes Yes Yes Yes Yes
Observations 163,264 169,926 168,924 162,321 163,264 169,926 168,924 162,321Adj. R-Squared 0.242 0.241 0.234 0.255 -0.176 -0.17 -0.172 -0.176
OLS Fixed effect estimation
21
Who Conducts Advertising in the Non‐manufacturing Sector?
With low TFP With high TFP
Advertising
In Japan’s non‐manufacturing sector, large and productive firms tend to have a higher off‐JT expenditure/sales ratio.
22
Who Conducts Off‐JT in the Non‐manufacturing Sector?
Note: We used cross‐section data for 2009.
Non-manufacturing
Off-JT expenditure / Sales (%)lnTFP (-1) 0.064 *** 0.067 ***
[0.012] [0.012]lnEMP (-1) 0.011 *** 0.011 ***
[0.002] [0.002]lnAge 0.004 0.001
[0.003] [0.003]Indutry dummies Yes Yes Yes Yes
Observations 5,611 5,885 5,870 5,597Adj. R-Squared 0.08 0.075 0.072 0.083
OLS
Investment in computerized information is more concentrated in certain sectors than in the case of investment in economic competencies, but less concentrated than in the case of innovative property investment.
Industries such as information services and internet‐based services, finance and insurance, and telephone and telegraphlead investment in computerized information.
23
4. Investment in Computerized Information in Japan
0.0%
0.5%
1.0%
1.5%
2.0%
2.5%
0% 20% 40% 60% 80% 100%
Cum
ula
tive
contr
ibution t
o inve
stm
ent
in c
om
pute
rize
d
info
rmat
ion/G
DP
Cumulative industry value added percentile in GDP
Cumulative Contribution of Industries to
Investment in Computerized Information:
2000-08
In Japan, large and productive (reverse causality?) firms tend to have a higher software stock/sales ratio. In the case of the non‐manufacturing sector, younger firms tend to have a higher software stock/sales ratio.
24
Who Conducts Software Investment?
Note: We used pooled data for 2006‐09.
Software stock / Sales (%)lnTFP(-1) 0.624 *** 0.392 *** 0.101 *** 0.172 ***
[0.037] [0.037] [0.038] [0.037]lnEMP(-1) 0.126 *** 0.118 *** 0.169 *** 0.179 ***
[0.004] [0.004] [0.008] [0.008]lnAge 0.007 -0.008 -0.127 *** -0.158 ***
[0.006] [0.006] [0.012] [0.012]
Indutry dummies Yes Yes Yes Yes Yes Yes Yes YesYear dummies Yes Yes Yes Yes Yes Yes Yes YesObservations 34,930 35,751 35,694 34,874 36,032 37,544 37,434 35,927
Adj. R-Squared 0.051 0.072 0.042 0.076 0.078 0.087 0.079 0.093
Manufacturing Non-manufacturingOLS OLS
In the case of the non‐manufacturing sector, younger firms tend to have a higher software stock/sales ratio.
25
Who Conducts Software Investment in the Non‐manufacturing Sector?
Young Old
Nominal Stock of Software
Conclusion• Intangible investment in Japan has stagnated in recent years.
The stagnation is particularly serious in investment in economic competencies.
• Using industry and firm‐level data, we examined this trend and also investigated what industries and what type of firms led intangible investment in Japan.
Findings from Industry‐Level Analysis• In the case of innovative property and computerized
information, intangible investment is concentrated in a small number of industries. Innovative property: broadcasting, communication equipment, motor vehicle parts and accessories, and pharmaceutical products.Computerized information: information services and internet‐based services, finance and insurance, and telephone and telegraph services.
26
Findings from Industry‐Level Analysis (Contd.)
• Our decomposition analysis at the industry level showed that a decline in the value added share of innovative property intensive industries has substantially contributed to the stagnation in investment.
• In the case of economic competencies, the investment‐sales ratio is quite similar across industries. There is a high cross‐industry correlation between the growth rate of economic competencies and the growth rate of other factor inputs. It seems that the reason Japanese firms do not invest in economic competencies is simply that they expect low sales and factor input growth.
• Many firms increased the percentage of part time workers in total workers and did not provide intensive training in the case of part time workers. This change reduced training expenditure substantially.
27
Findings from Firm‐Level Analysis• In Japan’s manufacturing sector, R&D investment is
concentrated in large, old and productive firms. However, recently the importance of smaller, young and less‐productive firms has been increasing. Similarly, in the non‐manufacturing sector, small, young and less‐productive firms have tended to conduct more R&D in recent years.
• One promising candidate to explain these recent trends probably is that small, young and less‐productive firms have been increasing their R&D in response to the decline in technology spillovers from large firms.
• In the case of the non‐manufacturing sector, younger firms tend to conduct more advertising and have a higher software stock/sales ratio.
• It has been argued that because of low economic growth, high entry and exit cost, the zombie firm problem, etc., the sales share of young firms has been small in Japan. Since young firms are active in several types of intangible investment, the low share of young firms may have contributed to the stagnation of intangible investment. 28
29
91.1 91.083.1 80.3
90.698.2
75.966.2
59.6
73.6
40.533.3
20.3
57.4 60.5
6.7 3.27.9 12.5
8.91.3
16.226.6
34.2
23.3
56.9
53.7
58.3
33.3 31.0
1.61.6 5.2 3.7
0.2 0.43.6 3.2 5.1 3.1 0.6
8.8
10.0
5.6 4.0
0.6 4.2 3.7 3.4 0.3 0.1 4.3 4.1 1.1 2.0 4.111.3
3.7 4.5
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Prim
ary indu
stry
Chem
ical, m
etal and
petroleu
m
Machine
ry
Other m
anufacturin
g
Constructio
n
Electricity
and
gas
Who
lesale
Retail
Fina
nce an
d insurance
Real estate
Tran
sportatio
n
Commun
ication
Public services
Other business services
Other hou
seho
ld services
Sales Share by Firm Age Group
Established after2002
Established in1997‐2001
Established in1975‐1996
Established before1974