April 24, 2012 Comparing China Input-Output Tables from ... · historical Input-Output survey results, underlying materials and even the confidential policy files from the early-year

Comparing China Input-Output Tables from Two Different

Time Series: Major Differences and Potential Bias

Leona Li, Xiaoqin Li

April 24, 2012

www.conferenceboard.org © 2012 The Conference Board, Inc. | 2

What are the two different IO table series?

One is the 1995-2009 SUTs and IOTs from World Input-Output Database (WIOD),

which adopted a SUT-RAS method to jointly estimate SUTs based on all publicly

available statistics;

the other is 1981-2005 Use Table series and estimated IOTs from China KLEMS

project, led by Professor Ren Ruoen, in corporation with Professor Dale Jorgenson

and China’s National Bureau of Statistics (NBS).

China KLEMS project established close collaboration with China NBS, thus the

compilation of their Use table time series involves a thorough inspection of all the

historical Input-Output survey results, underlying materials and even the

confidential policy files from the early-year planned economy system, which is also

the initial reason why China compiled its first IOT under MPS framework.

www.conferenceboard.org © 2012 The Conference Board, Inc. | 3

What is the major difference in the compilation

methodology?

WIOD’s compilation is mainly based on China’s public benchmark tables and

aggregate control totals from national accounts , assuming:

1. China’s public benchmark tables are consistent;

2. China’s IO account is consistent with GDP accounts;

while the above assumptions are somehow problematic, because

1. China’s statistical system is still improving over time,

2. 5-year intervals is a relatively long enough time to introduce some real

structure change;

3. The Value-added from IO accounts are always slightly bigger than GDP

accounts, because of the import tariffs and adjustments for consumption of

financial services, etc.

China KLEMS has included some internal information to make the necessary

adjustments according to the above inconsistency issues.

www.conferenceboard.org © 2012 The Conference Board, Inc. | 4

What are the major additional information provided by

NBS?

For the industrial enterprises’ survey, there’s scale-based system in China. For the

enterprises with more than 5 million RMB’s annual revenue, they supposed to

follow a much detailed and systematic survey system. While for the enterprises

under this designed scale, the cost and expense accounting is usually less detailed.

Specific supply structure for SMEs in provided.

China’s IOTs’ total output by products is at producer price. The value-added tax

expenditure by products is provided

For import data in IOT, value creation for the import goods by the transportation

enterprises, import tariff , consumption tax for imported goods, and adjustments for

processing trade are all provided;

For export data in IOT, the FOB price excluding turnover cost, and adjustments for

processing trade are all provided;

Most of underlying survey material comes from the use sectors, which accounting

price is purchaser price. Turnover cost matrix (including business additional

expenditure and transportation) is provided, which could be subtracted from the

purchaser price table and construct the tables at producer price.

www.conferenceboard.org © 2012 The Conference Board, Inc. | 5

One example for adjustment details

Table 2.1 The adjustments for processing trade in 2007 benchmark table Unit: billion RMB

(1) (2) (3) (4) = (1) - (2)＋(3) (5) (6) (7) = (5) - (6)

1 Agriculture 66.6 0 0 66.6 236.87 4.08 232.8

2 Mining 72.14 12.85 4.72 64.01 1,086.90 53.01 1,033.89

3 Construction 40.89 0 0 40.89 22.13 0 22.13

4 Food and related 200.38 11.23 2.06 191.21 164.99 6.84 158.15

5 Textiles, Leather, Apparel 1,475.14 113.33 27.04 1,388.85 200.53 57.82 142.71

6 Wood and related 248.72 7.68 1.41 242.45 29.78 2.73 27.05

7 Paper, printing and publishing 268.72 49.96 7.69 226.44 96.16 13.3 82.86

8 Petroleum 97.55 24.12 3.36 76.78 145.24 0.23 145.01

9 Chemicals & Rubber 751.13 35.02 7.68 723.79 973.66 63.14 910.52

10 Other mineral 149.55 1.64 0.46 148.37 40.26 2.53 37.73

11 Metals 892.74 25.83 4.48 871.4 534.78 44.26 490.52

12 Machinery 587.09 17.79 4.38 573.69 712.38 8.05 704.33

13 Electrical & Electronics 3,564.18 510.37 90.24 3,144.06 2,767.84 401.46 2,366.38

14 Transport equipment 331.63 4.27 0.85 328.22 301.35 1.03 300.32

15 Misc. manufacturing 152.83 23.24 4.56 134.14 182.81 19.76 163.05

16 Utility 6.51 0 0 6.51 1.8 0 1.8

17 Trade 400.76 40.1 40.1 400.76 0 0 0

18 Transportation services 398.3 5.29 5.29 398.3 106.32 0 106.32

19 Communications 49.51 0 0 49.51 43.97 0 43.97

20 Finance, insurance and real estate 8.62 0 0 8.62 12.92 0 12.92

21 Other private services 465.31 0 0 465.31 413.1 0 413.1

22 Public services 4.2 0 0 4.2 6.51 0 6.51

Total 10,232.50 882.72 204.32 9,554.10 8,080.31 678.25 7,402.06

Notes:

(1): All exports

(2):

Exports of processing with

foreign supplied material

(3):

Charges for processing with

foreign supplied material

(4): Exports for IOT

(5): All imports

(6):

Imports for processing with

foreign supplied material

(7): Imports for IOT

www.conferenceboard.org © 2012 The Conference Board, Inc. | 6

What are differences in the aggregate level?

Table 2.2 Percentage difference of GDPs between China KLEMS/WIOT with National Accounts and 2002 Benchmark IO

Deviation in % GDP, total Agriculture Industry Construction Service Deviation in % GDP, total Agriculture Industry Construction Service

1995 2.67% 1.73% 3.33% 1.89% 2.57% 1995 0.00% 0.00% 0.00% 0.00% 0.00%

1996 2.66% 1.88% 3.19% 1.91% 2.59% 1996 0.00% 0.00% 0.00% 0.00% 0.00%

1997 2.60% 2.08% 2.91% 1.92% 2.60% 1997 0.00% 0.00% 0.00% 0.00% 0.00%

1998 2.11% 1.09% 2.14% 1.90% 2.60% 1998 0.00% 0.00% 0.00% 0.00% 0.00%

1999 1.68% 0.52% 1.43% 1.75% 2.44% 1999 0.00% 0.00% 0.00% 0.00% 0.00%

2000 1.24% 0.71% 0.02% 1.99% 2.60% 2000 0.00% 0.00% 0.00% 0.00% 0.00%

2001 1.28% 0.71% -0.06% 1.99% 2.69% 2001 0.00% 0.00% 0.00% 0.00% 0.00%

2002 1.27% 0.71% -0.12% 1.99% 2.68% 2002 0.00% 0.00% 0.00% 0.00% 0.00%

2003 1.10% 0.71% -0.38% 1.77% 2.58% 2003 0.00% 0.00% 0.00% 0.00% 0.00%

2004 0.88% 0.27% 0.30% 0.98% 1.66% 2004 0.00% 0.00% 0.00% 0.00% 0.00%

2005 0.71% 3.00% 0.92% 4.98% -0.78% 2005 0.00% 0.00% 0.00% 0.00% 0.00%

2002 Benchmark IO 0.00% 0.00% -2.29% 0.00% 2.22% 2002 Benchmark IO -1.25% -0.71% -2.17% -1.95% -0.46%

China KLEMS WIOD

www.conferenceboard.org © 2012 The Conference Board, Inc. | 7

What are differences with public benchmark tables?

Table 2.3 Weight Absolute Percentage Error between Benchmark IOT, WIOD and China KLEMS

BU2002

vs.

BU2007

BU2002

vs.

WIOT2002

BU2007

vs.

WIOT2007

WIOT2002

vs.

WIOT2007

WIOT2002

vs.

WIOT2005

Total Intermediate % 9.46 0.92 0.08 8.57 5.72

Total Intermediate % by industry 8.39 1.15 1.32 8.82 5.91

Direct consumption coefficient - all 30.58 13.49 14.68 31.18 26.15

Final consumption coefficient - all 22.96 9.69 9.95 20.64 16.38

Final household consumption 25.56 14.06 12.22 31.57 20.65

Final government consumption 18.72 12.74 9.17 6.09 12.63

Gross capital formation 17.69 7.26 9.61 11.57 10.20

Export 28.09 13.84 17.91 29.18 20.45

Import 24.74 0.54 0.83 24.76 17.98

WIOT2005

vs.

WIOT2007

BU2002

vs.

CNUSE2002

CNUSE2005

vs.

BU2002

CNUSE2005

vs.

BU2007

CNUSE2002

vs.

CNUSE2005

Total Intermediate % 3.02 0.00 7.88 2.33 7.88

Total Intermediate % by industry 3.01 3.22 9.00 6.26 7.55

Direct consumption coefficient - all 9.08 14.10 25.74 33.89 15.22

Final consumption coefficient - all 10.00 7.54 20.94 18.07 17.00

Final household consumption 12.39 5.96 19.20 24.41 14.11

Final government consumption 7.92 8.54 10.44 8.46 1.91

Gross capital formation 8.96 1.47 27.85 22.64 27.05

Export 9.47 12.74 26.59 20.77 22.07

Import 11.25 8.98 20.64 14.06 19.88

www.conferenceboard.org © 2012 The Conference Board, Inc. | 8

What are differences in the Input Coefficient Matrix? Table 2.4 Number of discrepancies between WIOD and China KLEMS in key input coefficient and zero input coefficient

Industry (#discrepancy) 95 96 97 98 99 00 01 02 03 04 05 Total 95 96 97 98 99 00 01 02 03 04 05 Total

1 Agriculture 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 0 0 0 1 1 0 5

2 Mining 3 3 1 4 4 2 2 2 2 2 2 27 0 0 0 0 0 0 0 0 1 1 0 2

3 Construction 0 2 2 2 4 2 2 4 3 3 4 28 0 0 0 0 0 0 0 0 1 1 0 2

4 Food and related 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 3

5 Textiles and related 0 0 0 0 2 0 0 0 0 0 1 3 0 0 1 1 1 1 1 1 2 2 1 11

6 Leather 0 0 0 0 1 1 1 1 1 0 0 5 1 1 2 2 2 2 2 2 2 2 1 19

7 Wood and related 3 2 2 4 3 2 2 1 2 2 2 25 1 1 2 2 2 2 2 2 2 2 1 19

8 Paper related 0 0 1 1 1 2 2 2 1 3 3 16 0 0 1 1 1 1 1 1 2 2 2 12

9 Petroleum 1 1 1 1 1 0 0 1 0 0 0 6 0 0 2 2 2 2 2 2 1 1 2 16

10 Chemicals 1 0 2 2 0 2 1 1 3 2 1 15 0 0 0 0 0 0 0 0 1 1 2 4

11 Rubber 0 0 0 0 1 0 0 0 0 0 0 1 0 0 0 0 0 1 1 1 2 2 2 9

12 Other mineral 3 2 3 0 1 1 3 3 2 3 1 22 0 0 0 0 0 0 0 0 1 1 1 3

13 Metals 0 2 1 1 2 2 4 2 1 2 2 19 0 0 1 1 1 1 1 1 1 1 1 9

14 Machinery 2 0 0 0 2 2 2 2 1 1 1 13 0 0 1 1 1 1 1 1 1 1 2 10

15 Electrical & Electronics 1 0 1 0 1 0 0 0 1 1 1 6 0 0 0 0 0 0 0 0 1 1 1 3

16 Transport equipment 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 9

17 Misc. manufacturing 3 4 4 4 4 3 3 3 3 5 3 39 0 0 0 0 0 0 0 0 1 1 1 3

18 Utility 2 4 2 4 4 1 0 0 1 2 2 22 0 0 1 1 1 1 1 1 1 1 1 9

19 Trade 1 3 1 3 4 1 3 1 2 2 3 24 0 0 1 1 1 0 0 0 1 1 1 6

20 Transportation services 2 0 0 0 2 1 1 1 1 0 1 9 0 0 0 0 0 0 0 0 1 1 1 3

21 Communications 4 1 4 5 4 6 4 4 8 8 7 55 1 1 1 3 3 1 1 1 2 2 1 17

22 Finance related 2 1 2 1 0 0 0 1 2 1 2 12 0 0 1 0 1 0 0 0 1 1 1 5

23 Other private services 1 1 2 2 1 3 0 0 1 2 2 15 0 0 0 0 0 0 0 0 1 1 1 3

24 Public services 4 2 1 1 1 3 3 4 5 3 3 30 1 1 0 1 1 1 2 2 2 2 2 15

Key Input Coefficient Zero Input Coefficient

www.conferenceboard.org © 2012 The Conference Board, Inc. | 9

How do we measure the differences? (1)

Table 2.5.1 Weighted average percentage error between two Use table series by different level of aggregations (current price series)

30 X 24 - WAPE (%) 95 96 97 98 99 00 01 02 03 04 05 30 X 19 - WAPE (%) 95 96 97 98 99 00 01 02 03 04 05

Whole matrix 9 9 9 10 10 11 10 9 10 9 10 Whole matrix 8 8 8 10 10 11 10 9 9 9 9

Intermediate use matrix 21 21 21 28 30 26 23 21 23 26 28 Intermediate use matrix 20 20 20 27 29 26 23 20 22 25 27

Direct consumption coefficient matrix 21 20 19 23 25 24 22 21 24 28 31 Direct consumption coefficient matrix 22 21 20 25 26 22 20 20 23 27 30

Final use matrix 24 20 17 19 18 14 15 12 15 18 21 Final use matrix 24 20 17 19 18 14 15 12 15 18 21

Final consumption coefficient matrix 32 28 25 38 40 72 35 11 37 52 75 Final consumption coefficient matrix 32 28 25 38 40 72 35 11 37 52 75

Value-added 8 8 10 11 11 15 13 13 12 8 10 Value-added 7 7 9 10 10 14 13 12 12 8 9

Total output by industry 10 10 11 14 14 14 13 13 11 8 9 Total output by industry 8 8 9 12 12 14 12 11 11 7 7

30 X 9 - WAPE (%) 95 96 97 98 99 00 01 02 03 04 05 19 X 19 - WAPE (%) 95 96 97 98 99 00 01 02 03 04 05

Whole matrix 7 7 7 8 9 9 8 7 8 8 8 Whole matrix 8 8 8 9 9 11 10 9 9 8 8

Intermediate use matrix 17 17 17 23 26 21 19 17 18 19 21 Intermediate use matrix 19 18 18 24 27 24 22 19 20 22 24

Direct consumption coefficient matrix 19 19 17 22 25 23 21 20 23 27 30 Direct consumption coefficient matrix 19 19 18 22 24 20 19 18 21 24 27

Final use matrix 24 20 17 19 18 14 15 12 15 18 21 Final use matrix 22 19 16 17 16 13 13 12 13 13 17

Final consumption coefficient matrix 32 28 25 38 40 72 35 11 37 52 75 Final consumption coefficient matrix 29 26 24 35 32 55 24 10 30 29 54

Value-added 3 3 3 4 3 7 6 5 5 2 2 Value-added 7 7 9 10 10 14 13 12 12 8 9

Total output by industry 3 4 3 9 10 10 7 5 6 5 4 Total output by industry 8 8 9 12 12 14 12 11 11 7 7

9 X 9 - WAPE (%) 95 96 97 98 99 00 01 02 03 04 05

Whole matrix 5 5 5 7 7 8 6 5 6 6 6

Intermediate use matrix 9 10 9 17 19 15 12 9 13 13 15

Direct consumption coefficient matrix 10 11 10 15 17 15 13 10 15 19 21

Final use matrix 15 13 12 14 14 8 8 7 8 6 8

Final consumption coefficient matrix 21 16 16 23 26 36 18 7 22 14 35

Value-added 3 3 3 4 3 7 6 5 5 2 2

Total output by industry 3 4 3 9 10 10 7 5 6 5 4

www.conferenceboard.org © 2012 The Conference Board, Inc. | 10

How do we measure the differences? (2)

Table 2.5.2 Weighted average percentage error between two Use table series by different level of aggregations (constant price series)

30 X 24 - WAPE (%) 95 96 97 98 99 00 01 02 03 04 05 30 X 19 - WAPE (%) 95 96 97 98 99 00 01 02 03 04 05

Whole matrix 9 9 9 10 10 11 10 9 10 9 10 Whole matrix 8 8 8 10 10 11 10 9 9 9 9

Intermediate use matrix 21 21 21 28 30 26 23 21 23 26 28 Intermediate use matrix 20 20 20 27 29 26 23 20 22 25 27

Direct consumption coefficient matrix 21 20 19 23 25 24 22 21 24 28 31 Direct consumption coefficient matrix 22 21 20 25 26 22 20 20 23 27 30

Final use matrix 24 20 17 19 18 14 15 12 15 18 21 Final use matrix 24 20 17 19 18 14 15 12 15 18 21

Final consumption coefficient matrix 32 28 25 38 40 72 35 11 37 52 75 Final consumption coefficient matrix 32 28 25 38 40 72 35 11 37 52 75

Value-added 8 8 10 11 11 15 13 13 12 8 10 Value-added 7 7 9 10 10 14 13 12 12 8 9

Total output by industry 10 10 11 14 14 14 13 13 11 8 9 Total output by industry 8 8 9 12 12 14 12 11 11 7 7

30 X 9 - WAPE (%) 95 96 97 98 99 00 01 02 03 04 05 19 X 19 - WAPE (%) 95 96 97 98 99 00 01 02 03 04 05

Whole matrix 7 7 7 8 9 9 8 7 8 8 8 Whole matrix 8 8 8 9 9 11 10 9 9 8 8

Intermediate use matrix 17 17 17 23 26 21 19 17 18 19 21 Intermediate use matrix 19 18 18 24 27 24 22 19 20 22 24

Direct consumption coefficient matrix 19 19 17 22 25 23 21 20 23 27 30 Direct consumption coefficient matrix 19 19 18 22 24 20 19 18 21 24 27

Final use matrix 24 20 17 19 18 14 15 12 15 18 21 Final use matrix 22 19 16 17 16 13 13 12 13 13 17

Final consumption coefficient matrix 32 28 25 38 40 72 35 11 37 52 75 Final consumption coefficient matrix 29 26 24 35 32 55 24 10 30 29 54

Value-added 3 3 3 4 3 7 6 5 5 2 2 Value-added 7 7 9 10 10 14 13 12 12 8 9

Total output by industry 3 4 3 9 10 10 7 5 6 5 4 Total output by industry 8 8 9 12 12 14 12 11 11 7 7

9 X 9 - WAPE (%) 95 96 97 98 99 00 01 02 03 04 05

Whole matrix 5 5 5 7 7 8 6 5 6 6 6

Intermediate use matrix 9 10 9 17 19 15 12 9 13 13 15

Direct consumption coefficient matrix 10 11 10 15 17 15 13 10 15 19 21

Final use matrix 15 13 12 14 14 8 8 7 8 6 8

Final consumption coefficient matrix 21 16 16 23 26 36 18 7 22 14 35

Value-added 3 3 3 4 3 7 6 5 5 2 2

Total output by industry 3 4 3 9 10 10 7 5 6 5 4

www.conferenceboard.org © 2012 The Conference Board, Inc. | 11

Is there any bias in IPD/ISD analysis? (1)

China KLEMS 1995

0.0

0.5

1.0

1.5

2.0

2.5

0.0 0.5 1.0 1.5

Ind

ex o

f S

ensitiv

ity o

f D

isp

ers

ion

Index of Power of Dispersion

WIOD1995

0.0

0.5

1.0

1.5

2.0

2.5

0.0 0.5 1.0 1.5

Ind

ex o

f S

ensitiv

ity o

f D

isp

ers

ion

Index of Power of Dispersion

www.conferenceboard.org © 2012 The Conference Board, Inc. | 12

Is there any bias in IPD/ISD analysis? (2)

China KLEMS 2000

WIOD 2000

0.0

0.5

1.0

1.5

2.0

2.5

0.0 0.5 1.0 1.5

Ind

ex o

f S

ensitiv

ity o

f D

isp

ers

ion

Index of Power of Dispersion

0.0

0.5

1.0

1.5

2.0

2.5

0.0 0.5 1.0 1.5

Ind

ex o

f S

ensitiv

ity o

f D

isp

ers

ion

Index of Power of Dispersion

www.conferenceboard.org © 2012 The Conference Board, Inc. | 13

Is there any bias in IPD/ISD analysis? (3)

China KLEMS 2005

WIOD 2005

0.0

0.5

1.0

1.5

2.0

2.5

0.0 0.5 1.0 1.5

Ind

ex o

f S

ensitiv

ity o

f D

isp

ers

ion

Index of Power of Dispersion

0.0

0.5

1.0

1.5

2.0

2.5

0.0 0.5 1.0 1.5

Ind

ex o

f S

ensitiv

ity o

f D

isp

ers

ion

Index of Power of Dispersion

www.conferenceboard.org © 2012 The Conference Board, Inc. | 14

Is there any bias in DPIC analysis? (1) Table 3.2.1 DPIC by Final household consumption

1995-1997 1998-2001 2002-2005 1995-1997 1998-2001 2002-2005

1 Agriculture 0.4574 0.3936 0.3163 0.5074 0.4192 0.3500

2 Mining 0.0436 0.0588 0.0602 0.0427 0.0432 0.0407

3 Construction 0.0099 0.0143 0.0144 0.0075 0.0110 0.0119

4 Food and related 0.2591 0.2284 0.2007 0.2825 0.2510 0.2415

5 Textiles, Leather, Apparel 0.1435 0.1158 0.0998 0.1143 0.1265 0.1159

6 Wood and related 0.0190 0.0189 0.0177 0.0142 0.0139 0.0112

7 Paper, printing and publishing 0.0277 0.0316 0.0331 0.0355 0.0353 0.0354

8 Petroleum 0.0240 0.0367 0.0408 0.0275 0.0307 0.0362

9 Chemicals & Rubber 0.1438 0.1424 0.1281 0.1381 0.1357 0.1339

10 Other mineral 0.0432 0.0343 0.0248 0.0462 0.0444 0.0265

11 Metals 0.0736 0.0733 0.0645 0.0830 0.0701 0.0523

12 Machinery 0.0344 0.0340 0.0295 0.0310 0.0303 0.0251

13 Electrical & Electronics 0.0648 0.0887 0.0681 0.0609 0.0770 0.0633

14 Transport equipment 0.0368 0.0399 0.0419 0.0296 0.0350 0.0432

15 Misc. manufacturing 0.0279 0.0236 0.0152 0.0092 0.0082 0.0038

16 Utility 0.0415 0.0597 0.0721 0.0377 0.0489 0.0647

17 Trade 0.1153 0.1183 0.1110 0.1421 0.1376 0.1270

18 Transportation services 0.0480 0.0694 0.0932 0.0771 0.0966 0.0879

19 Communications 0.0159 0.0147 0.0047 0.0116 0.0241 0.0396

20 Finance, insurance and real estate 0.1778 0.1704 0.1559 0.1547 0.1548 0.1489

21 Other private services 0.1566 0.2493 0.3407 0.1378 0.2051 0.2802

22 Public services 0.0000 0.0000 0.0000 0.0000 0.0000 0.0002

Total 1.9634 2.0160 1.9328 1.9905 1.9985 1.9393

China KLEMS WIOD

www.conferenceboard.org © 2012 The Conference Board, Inc. | 15

Is there any bias in DPIC analysis? (2) Table 3.2.2 DPIC by Gross Capital Formation (GCF)

1995-1997 1998-2001 2002-2005 1995-1997 1998-2001 2002-2005

1 Agriculture 0.0999 0.0954 0.1106 0.1112 0.1077 0.0934

2 Mining 0.0944 0.1104 0.0932 0.0843 0.0762 0.0751

3 Construction 0.5429 0.6077 0.5645 0.5060 0.5732 0.5428

4 Food and related 0.0485 0.0291 0.0407 0.0472 0.0330 0.0303

5 Textiles, Leather, Apparel 0.0669 0.0440 0.0256 0.0743 0.0423 0.0221

6 Wood and related 0.0270 0.0283 0.0268 0.0335 0.0414 0.0363

7 Paper, printing and publishing 0.0317 0.0285 0.0244 0.0396 0.0316 0.0257

8 Petroleum 0.0429 0.0598 0.0509 0.0416 0.0428 0.0466

9 Chemicals & Rubber 0.1420 0.1398 0.1083 0.1328 0.1248 0.1187

10 Other mineral 0.2010 0.1385 0.1045 0.1929 0.1663 0.1300

11 Metals 0.2387 0.2452 0.2365 0.2650 0.2500 0.2360

12 Machinery 0.1634 0.1747 0.1825 0.1636 0.1738 0.1776

13 Electrical & Electronics 0.1114 0.1546 0.1249 0.1043 0.1445 0.1317

14 Transport equipment 0.0983 0.0995 0.1105 0.1068 0.1086 0.1300

15 Misc. manufacturing 0.0270 0.0191 0.0088 0.0112 0.0087 0.0049

16 Utility 0.0468 0.0603 0.0671 0.0433 0.0506 0.0677

17 Trade 0.1440 0.1255 0.0993 0.1388 0.1258 0.0990

18 Transportation services 0.0648 0.0775 0.0935 0.0697 0.0880 0.0867

19 Communications 0.0143 0.0154 0.0012 0.0133 0.0272 0.0369

20 Finance, insurance and real estate 0.0574 0.0690 0.0660 0.0807 0.0717 0.0670

21 Other private services 0.0684 0.1134 0.1411 0.0566 0.0703 0.0823

22 Public services 0.0000 0.0000 0.0000 0.0000 0.0000 0.0002

Total 2.3315 2.4358 2.2807 2.3166 2.3587 2.2410

China KLEMS WIOD

www.conferenceboard.org © 2012 The Conference Board, Inc. | 16

Is there any bias in DPIC analysis? (3) Table 3.2.3 DPIC by Exports (EX)

1995-1997 1998-2001 2002-2005 1995-1997 1998-2001 2002-2005

1 Agriculture 0.1610 0.1344 0.1020 0.1733 0.1259 0.1018

2 Mining 0.0972 0.1104 0.1062 0.0919 0.0812 0.0847

3 Construction 0.0083 0.0116 0.0119 0.0090 0.0099 0.0089

4 Food and related 0.0958 0.0792 0.0594 0.1016 0.0807 0.0663

5 Textiles, Leather, Apparel 0.4197 0.3491 0.2600 0.4216 0.3598 0.2858

6 Wood and related 0.0373 0.0354 0.0381 0.0360 0.0284 0.0260

7 Paper, printing and publishing 0.0391 0.0421 0.0438 0.0519 0.0489 0.0435

8 Petroleum 0.0443 0.0612 0.0648 0.0437 0.0475 0.0592

9 Chemicals & Rubber 0.2735 0.2760 0.2362 0.2563 0.2607 0.2444

10 Other mineral 0.0710 0.0510 0.0403 0.0732 0.0666 0.0460

11 Metals 0.2429 0.2246 0.2277 0.2609 0.2285 0.2285

12 Machinery 0.1192 0.0972 0.0958 0.0770 0.0821 0.0946

13 Electrical & Electronics 0.2418 0.3273 0.3963 0.2722 0.3475 0.4391

14 Transport equipment 0.0689 0.0622 0.0637 0.0456 0.0536 0.0633

15 Misc. manufacturing 0.0777 0.0802 0.0835 0.0350 0.0412 0.0383

16 Utility 0.0492 0.0629 0.0721 0.0463 0.0524 0.0709

17 Trade 0.1353 0.1623 0.1493 0.1380 0.1751 0.1388

18 Transportation services 0.0764 0.0997 0.1241 0.1136 0.1159 0.1167

19 Communications 0.0160 0.0128 0.0024 0.0150 0.0214 0.0240

20 Finance, insurance and real estate 0.0472 0.0508 0.0476 0.0712 0.0580 0.0464

21 Other private services 0.1080 0.1572 0.1856 0.0995 0.1346 0.1462

22 Public services 0.0008 0.0005 0.0005 0.0007 0.0005 0.0010

Total 2.4307 2.4880 2.4114 2.4335 2.4202 2.3742

China KLEMS WIOD

www.conferenceboard.org © 2012 The Conference Board, Inc. | 17

What are the sources of Japan-China IO tables?

National Bureau of Statistics of China and Japan International Cooperation Agency

compiled a Japan-China IOT 2007 (NBS-JICA)

WIOT 40-country IOT 2007 transformed into two-country IOT 2007

Examples of adjustment:

Enterprises’ consumption expenditure

Office facilities

R&D activity

www.conferenceboard.org © 2012 The Conference Board, Inc. | 18

How large is the difference between the two Japan-China

tables? (1)

Total Output

Japan China Japan China Japan/China

Japan 30.4 71.1 24.8 62 4.2

China 83.3 17.7 44.8 10.2 1

ROW 62.1 52.5 53.3 45.3

VA share 10.6 2

Output 4.2 1

Table 4.1.2 WAPE of Total Output, Value-added Share, Direct Consumption Coefficient

and Final Consumption Coefficient

Intermediate Use Final Use

Aggregate level

China-China

Japan-Japan

Intercountry linkage

www.conferenceboard.org © 2012 The Conference Board, Inc. | 19

How large is the difference between the two Japan-China

tables? (2)

Sectoral breakdown of export

China Export to Japan

(in mil. RMB)

0 50,000 100,000 150,000 200,000 250,000 300,000

Agriculture

Construction

Textiles, Leather, Apparel

Paper, printing and publishing

Chemicals & Rubber

Metals

Electrical & Instrument

Misc. manufacturing

Trade

Communications

Other private services

NBS-JICA WIOTJapan Export to China

(in mil. RMB)

0 50,000 100,000 150,000 200,000 250,000 300,000 350,000 400,000

Agriculture

Construction

Textiles, Leather, Apparel

Paper, printing and publishing

Chemicals & Rubber

Metals

Electrical & Instrument

Misc. manufacturing

Trade

Communications

Other private services

NBS-JICA WIOT

www.conferenceboard.org © 2012 The Conference Board, Inc. | 20

How large is the difference between the two Japan-China

tables? (3)

Final Use% of export

Japan Export to China

( Final Use %)

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

Agriculture

Construction

Textiles, Leather, Apparel

Paper, printing and publishing

Chemicals & Rubber

Metals

Electrical & Instrument

Misc. manufacturing

Trade

Communications

Other private services

NBS-JICA WIOTChina Export to Japan

( Final Use %)

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

Agriculture

Construction

Textiles, Leather, Apparel

Paper, printing and publishing

Chemicals & Rubber

Metals

Electrical & Instrument

Misc. manufacturing

Trade

Communications

Other private services

NBS-JICA WIOT

www.conferenceboard.org © 2012 The Conference Board, Inc. | 21

Is there discrepancy in IPD/ISD analysis?

Rank correlation of IPD and ISD between the two series are both at 0.94 with less

than 1% significance;

Country Industry* IPD IPD ISD ISD

NBS-JICA WIOT NBS-JICA WIOT

Japan Mining 0.8620 0.7945 0.5059 0.5830

Japan Textiles, Leather, Apparel 0.9080 1.0250 0.6208 0.5848

Japan Petroleum 0.5365 0.5988 0.9235 0.8356

Japan Misc. manufacturing 0.8001 1.0691 0.5671 0.5065

Japan Utility 0.7278 0.8049 0.8221 0.8400

Japan Trade 0.6741 0.7334 1.1414 1.4963

Japan Communications 0.7681 0.7904 0.8043 0.6341

Japan Public services 0.7631 0.6914 0.5331 0.4958

China Mining 1.0688 0.9995 1.6731 1.5132

China Wood and related 1.3014 1.2926 0.8131 1.0226

China Petroleum 1.0490 0.9888 1.1687 0.9142

China Metals 1.3252 1.2733 2.2743 2.0240

China Machinery 1.3386 1.2824 1.1577 1.0380

China Misc. manufacturing 1.0248 1.1412 0.7039 0.4933

China Utility 1.2771 1.1984 1.7681 1.4916

China Trade 0.8660 0.8323 0.8881 0.9966

China Other private services 1.1016 1.0312 1.2398 1.4966

* selected industries with high discrepancies

www.conferenceboard.org © 2012 The Conference Board, Inc. | 22

Is there discrepancy in inducement analysis? (1)

Country Final Demand Items NBS-JICA WIOT NBS-JICA WIOT NBS-JICA WIOT NBS-JICA WIOT

Japan Household consumption 26,905,299 28,506,064 1,291,146 976,219 1.4667 1.5582 0.0704 0.0534

Japan Government consumption 8,368,342 9,182,963 81,857 117,070 1.6253 1.5739 0.0159 0.0201

Japan Gross capital formation 14,336,791 14,443,499 774,506 823,219 1.7648 1.841 0.0953 0.1049

Japan Export to ROW 10,482,602 10,449,090 308,733 298,005 2.0435 2.1234 0.0602 0.0606

China Household consumption 327,492 351,739 20,835,671 19,840,343 0.0339 0.0371 2.1597 2.0911

China Government consumption 86,033 111,082 7,695,792 7,359,642 0.0244 0.031 2.1869 2.056

China Gross capital formation 913,450 876,632 28,120,993 27,994,044 0.0827 0.079 2.545 2.5214

China Export to ROW 698,064 640,426 23,582,552 24,277,631 0.0745 0.0686 2.5172 2.6002

(mil. RMB) Domestic Production Inducement Coefficient

Japan China Japan China

Japan’s domestic production is mainly driven by final household consumption (43%)

and gross capital formation (23%); while that of China is driven by gross capital

formation (34%) and export to RoW (29%);

Domestic inducement structure very similar between two tables;

Large variation observed at China’s inducement of Japanese household and

government consumption, and Japan’s inducement of Chinese government

consumption.

www.conferenceboard.org © 2012 The Conference Board, Inc. | 23

Is there discrepancy in inducement analysis? (2)

For both countries, around 97% of gross value-added is induced by domestic final

demand;

Pattern of discrepancy is identical to that of production inducement.

Country Final Demand Items NBS-JICA WIOT NBS-JICA WIOT NBS-JICA WIOT NBS-JICA WIOT

Japan Household consumption 15,562,551 16,078,873 373,202 289,726 0.8484 0.8789 0.0203 0.0158

Japan Government consumption 4,753,078 5,482,960 24,713 34,792 0.9231 0.9398 0.0048 0.006

Japan Gross capital formation 6,522,918 6,368,595 205,115 214,498 0.803 0.8118 0.0252 0.0273

Japan Export to ROW 4,103,538 4,104,814 82,124 79,893 0.7999 0.8341 0.016 0.0162

China Household consumption 134,720 135,003 8,020,191 7,919,978 0.014 0.0142 0.8313 0.8347

China Government consumption 35,207 43,071 3,116,921 3,066,423 0.01 0.012 0.8857 0.8566

China Gross capital formation 361,846 328,322 8,094,469 8,180,736 0.0327 0.0296 0.7326 0.7368

China Export to ROW 275,023 241,897 6,536,891 6,794,985 0.0294 0.0259 0.6977 0.7278

(mil. RMB) VA Inducement Coefficient

Japan China Japan China

www.conferenceboard.org © 2012 The Conference Board, Inc. | 24

Is there discrepancy in decomposition of energy

consumption?

Following Miller and Blair (2009) and Zhang and Lahr (2011), we decompose

change in gross energy consumption between 1995 and 2005 into

1) change in energy efficiency, defined as energy input per unit of gross output;

2) change in production structure, described by Leontief inverse matrix;

3) change in final demand structure;

4) change in final demand level.

Caveat on interpretation: China KLEMS uses double deflation to construct constant

price IOT series

www.conferenceboard.org © 2012 The Conference Board, Inc. | 25

Is there discrepancy in decomposition of energy

consumption?

China KLEMS WIOD

Changes in energy consumption and its determinants

China KLEMS IOT, 1995-2005

-60%

-40%

-20%

0%

20%

40%

60%

80%

1995 1997 2001 2005

∆ total ∆ energy efficiency

∆ Production s tructure ∆ FD structure

∆ FD level

Changes in energy consumption and its determinants

WIOT, 1995-2005

-60%

-40%

-20%

0%

20%

40%

60%

80%

1995 1997 2001 2005

∆ total ∆ energy efficiency

∆ Production s tructure ∆ FD structure

∆ FD level

Energy efficiency;

Final demand structure;

www.conferenceboard.org © 2012 The Conference Board, Inc. | 26

Is there discrepancy in decomposition of CO2 emission?

China KLEMS

WIOD

Changes in CO2 emission and its determinants

China KLEMS IOT, 1995-2005

-60%

-40%

-20%

0%

20%

40%

60%

80%

1995 1997 2001 2005

∆ total ∆ energy efficiency

∆ Production s tructure ∆ FD structure

∆ FD level

Changes in CO2 emission and its determinants

WIOT, 1995-2005

-60%

-40%

-20%

0%

20%

40%

60%

80%

1995 1997 2001 2005

∆ total ∆ energy efficiency

∆ Production s tructure ∆ FD structure

∆ FD level

www.conferenceboard.org © 2012 The Conference Board, Inc. | 27

Is there discrepancy in labor IPD/ISD analysis ?

Relative position and ranking of IPD and ISD resembles each other closely;

Larger variation in ISD than in IPD;

Industry* 1995-1997 1998-2001 2002-2005 1995-1997 1998-2001 2002-2005 1995-1997 1998-2001 2002-2005 1995-1997 1998-2001 2002-2005

Mining 0.7369 0.5911 0.5376 0.7234 0.6005 0.4957 1.0937 0.9574 0.7321 1.0723 0.8128 0.6248

Textiles, Leather, Apparel 1.0841 1.2022 1.3162 1.0977 1.2003 1.3833 0.6206 0.609 0.5628 0.5203 0.5994 0.6092

Wood and related 0.9845 1.2107 1.4515 1.2048 1.3585 1.6483 0.4038 0.5049 0.6305 0.4788 0.6444 0.8034

Misc. manufacturing 1.9037 1.9519 1.6559 1.8255 1.9348 1.7049 1.7415 1.6934 1.1171 1.3729 1.4523 1.0645

Utility 0.5235 0.4437 0.4194 0.4859 0.429 0.3659 0.2615 0.2465 0.2103 0.2562 0.2355 0.2234

Transportation services 0.732 0.6725 0.6961 0.747 0.6943 0.6829 0.8325 0.7918 0.9396 1.0084 0.951 0.8643

Communications 0.88 0.7656 0.6975 0.8932 0.612 0.5005 0.7636 0.404 0.2433 0.7427 0.4611 0.346

Finance, insurance and real estate 0.3123 0.3488 0.3377 0.3166 0.3111 0.3168 0.1436 0.17 0.1938 0.1865 0.1937 0.1946

Other private services 1.645 1.4772 1.4238 1.6902 1.5184 1.4928 2.7361 2.9096 3.2819 2.4795 2.3808 2.6138

* selected industries with high discrepancies

WIOT

ISD

China KLEMS WIOT China KLEMS

IPD

www.conferenceboard.org © 2012 The Conference Board, Inc. | 28

Is there discrepancy in aggregate labor inducement

analysis?

Spearman's Rho HH con. GCF Export

1995 1.00 0.40* 0.99

1996 0.92 0.97 1.00

1997 0.83 0.99 0.48*

1998 0.95 0.74 0.98

1999 0.90 0.79 1.00

2000 0.94 0.80 0.91

2001 0.52* 0.92 0.71

2002 0.85 0.85 0.82

2003 0.93 0.91 0.92

2004 0.97 0.85 1.00

2005 0.99 1.00 1.00

* NOT significant at 10%

Increase in household consumption in Agriculture has a predominant yet declining

effect on aggregate labor employment, followed by Other private services and

Trade;

Portrait by two IOT series similar: except for household consumption in 2001, GCF

in 1995, and export in 1997, all rank correlations are significant at less than 1%.

www.conferenceboard.org © 2012 The Conference Board, Inc. | 29

Is there discrepancy in labor compensation inducement

analysis?

HH con. GFC Export HH con. GFC Export HH con. GFC Export

1995 0.96 0.49* 0.91 0.92 0.75 0.95 0.94 0.93 0.95

1996 0.90 0.75 0.85 0.60 -0.04* 0.40* -0.09* 0.55 0.35*

1997 0.86 -0.15* 0.50* 0.64 0.87 0.01* 0.84 0.93 0.60

1998 0.87 0.79 0.90 0.60 -0.37* 0.95 0.92 0.75 0.95

1999 -0.08* 0.16* 0.56 0.34* 0.06* 0.94 0.83 0.14* 0.97

2000 -0.01* 0.65 0.25* 0.95 0.19* 0.40* 0.99 0.17* 0.95

2001 0.69 0.43* 0.17* 0.44* 0.21* 0.64 0.92 0.58 0.98

2002 0.97 0.84 0.95 0.62 0.34* 0.73 0.34* 0.15* 0.17*

2003 0.75 0.54 0.65 0.45* 0.87 0.81 0.85 0.97 0.92

2004 -0.40* 0.40* -0.48* 0.22* 0.62 0.53 0.74 0.71 0.75

2005 -0.41* -0.46* 0.81 0.80 0.80 0.87 0.92 0.99 0.93

* NOT significant at 10%

Medium-skilled Low-skilledHigh-skilledSpearman's

Rho

www.conferenceboard.org © 2012 The Conference Board, Inc. | 30

Conclusion

Overall comparable and generate similar patterns in inducement analysis;

From sectoral perspective, service industries and heavily state-owned sectors are

more problematic;

From year perspective, 1999-2000 and 2003-2004 seem to be with higher variation;

Implication is research-question-driven: e.g. energy and utility analysis.

High (6) Medium (5) Low (11)

Mining Construction Agriculture

Wood and related Other mineral Food and related

Other private services* Machinery Textiles, Leather, Apparel

Misc. manufacturing* Utility* Paper, printing and publishing*

Trade Transportation services* Petroleum

Communications Chemicals & Rubber

Metals

Electrical & Electronics*

Transport equipment

Finance, insurance and real estate

Public service

* with concordance issue betn. WIOT and China KLEMS

Sectoral level discrepancy

www.conferenceboard.org © 2012 The Conference Board, Inc. | 31

Q & A

Thank you!