Robert Stehrer

1
The WIOD database Construction and first
results on the factor content of trade

• Robert Stehrer
• The Vienna Institute for International Economic
  Studies (wiiw)
• Robert.Stehrer_at_wiiw.ac.at
• WPTGS workshop, October 6, 2010 OECD Paris
• Version 2010-10-06

The WIOD-project is funded by the European
Commission, Research Directorate General as part
of the 7th Framework Programme, Theme 8
Socio-Economic Sciences and Humanities, Grant
Agreement no 225 281.
2
WIOD project www.wiod.org

• Project funded within the 7th framework program
  of the EU
• 10OECD partners involved
• Project started in May 2009 and ends in April
  2012
• Construction and applications
• Construction of inter-country SUT/IO tables
• Data publicly available in May 2012
• Should allow for a wide range of applications
• Socioeconomic issues
• Environmental issues
• (Policy) Modelling

3
WIOD project Data coverage

• Inter-country Supply-Use and Input-Output tables
• Benchmarked to NA data
• Period 1995-2006
• 40 countries included
• EU-15 countries
• EU-12 countries
• NAFTA Canada, USA, Mexico
• BRI Brazil, Russia, India
• CHN China
• OTHER Turkey, Japan, South Korea, Taiwan,
  Indonesia, Australia
• Sector and product classifications of SUTs
• 59 products (corresponding to CPA)
• 35 industries (corresponding to NACE rev. 1)

4
WIOD project Data coverage

• Trade data
• Goods trade (HS 6-digit use category CPA)
• Services trade (BoP codes)
• Plus satellite accounts
• Energy and environment
• Socio-economic indicators (capital and labour)
• Deflated tables

5
Construction of International SUTs

• 1. Time-series of SUTs at purchasers prices
• Extrapolation and benchmarking of SUTs to
  National Accounts statistics, based on SUT-RAS
  method
• 2. From SUTs at purchasers prices to basic
  prices
• Construction of net tax, trade and transport
  margin matrices
• 3. From national to inter-country SUTs
• Breakdown of USE table into domestic and imported
  (by supplying country)
• Based on HS6 to end-use categorisation (improving
  on BEC)
• Breakdown by supplying country differentiated by
  use categories
• In later stage use import tables from NSIs if
  available
• Rest of World exports to RoW is calculated as
  residual and can become negative

6
Construction of International WIOT

• 4. From SUTs to inter-country input-output table
• Technology assumptions (on product sales or
  production)
• 5. From current price to constant price tables
• National deflators based on industry gross output
  deflators, and row wise deflation of SUT. At
  later stage add in more information from national
  accounts
• International deflators (PPPs) World Bank ICP
  expenditure PPPs adjusted and allocated to
  industries (for 2005)
• For details see documentation on WIOD webpage
• A.A. Erumbam, R. Gouma, B. Los, R. Stehrer, U.
  Temurshoev, M. Timmer, and G. de Vries (2010)
  World Input-Output Database Construction and
  Applications.
• U. Temurshoev and Timmer (2009) SUT-RAS
  procedure.
• J. Francois and O. Pindyuk (2010) WIOD service
  trade data.
• J. Pöschl and R. Stehrer (2010) WIOD goods trade
  data.

7
Construction of International WIOT
8
The bilateral factor content of
trade(preliminary results)
Accounting for trade in intermediates (following
Reimer, JIE 2006 Trefler and Zhu, JIE 2010) N
Number of countries G Number of industries F
Number of factors Direct plus indirect factor
input A coefficient matrix of dimension NG x
NG D direct factor input matrix of dimension F
x NG Bilateral (NG x N) import matrix
9
The bilateral factor content of trade

• Factor content of trade for country c
• Diagonalization of results in
• Bilateral trade matrix for factor
  allowing for
• Bilateral and sector specific calculations
• Focus on bilateral factor services flows
• NOTE
• ROW not included in calculations
• Results are preliminary as partly relying on
  imputed values

10
Selected results Net exports of EU-15
Trade in goods and services
Trade in value added
in mio US-
in mio US-
11
Selected results Structure of value added trade
Exports
Imports
12
Selected results Structure of value added
tradenot including intra EU-15 trade
Exports
Imports
13
Trade in value added by various factorsnot
including intra EU-15 trade

• Capital
• ICT capital
• Non-ICT capital
• Labour
• Low educated
• Medium educated
• High educated

14
Trade in value added by various factors not
including intra-EU trade
15
Structure of net exports of VA by selected
industries
Textile and textile products (NACE 17 and 18)
Leather and leather products (NACE 19)
16
Structure of net exports of VA by selected
industries
Transport equipment (NACE 34 and 35)
Electrical and optical equ. (NACE 30 to 33)
17
Next steps

• Improving bilateral trade in services and trade
  in goods
• Improving breakdown by use categories (see below)
• Comparing our use-classification of trade flows
  with official import IO tables.
• Improving factor input data for non-OECD
  countries
• as part of work by World KLEMS consortium
• Constant price series
• National deflators
• Purchasing power parities for output and
  intermediate inputs
• Volume masures of labour and capital
• Processing export trade tables for Mexico and
  China
• Testing by users
• Available to consortium members
• Full availability in May 2012

18
Appendix Use categories and proportionality
assumptions

• Breakdown of imports by use categories
• Based on trade data (revised BEC correspondence)
• Based on information from existing import tables
• Breakdown of each cell by country of origin
• Differentiated for intermediates, final
  consumption and gross fixed capital formation

19
Comparisons of use breakdown

• Use existing import USE or SIO tables
• Eurostat
• OECD
• Shares of imported intermediates in total imports
• Compare to BEC or BEC-WIOD categorisation

20
Comparisons to SIOT data
Share of imported intermediates in total
imports NACE 34 - Motor vehicles, trailers and
semi-trailers
2000
2005
21
Comparisons to SIOT data
Share of imported intermediates in total
imports NACE 35 - Other transport equipment
2005
2000
22
Comparisons to SIOT data
Share of imported intermediates in total
imports NACE 18 Wearing apparel, furs
2005
2000
Note different scales
23
Product descriptions

• Product descriptions at detailed level are not
  sufficient to differentiate between use
  categories
• Example HS 6101 and 6105
• 6101  Men's or boys' overcoats, car-coats, capes,
  cloaks, anoraks (including ski-jackets),
  wind-cheaters, wind-jackets and similar articles,
  knitted or crocheted, other than those of heading
  No 6103     6101 10  - Of wool or fine
  animal hair               6101 20  - Of cotton
                6101 30  - Of man-made fibres    
            6101 90  - Of other textile materials
• 6105  Men's or boys' shirts, knitted or crocheted
      6105 10  - Of cotton    
            6105 20  - Of man-made fibres    
            6105 90  - Of other textile materials
• Country-specific adjustments, weights, etc. used
  by NSIs
• Additionally, specific problems for particular
  products (cars, light bulbs, etc.)

24
Conclusions

• Give weights to use categories
• Further improvements by including information
  from existing import tables
• Breakdown into use categories
• Allocation of intermediate product over using
  industries

25
Comments to ...

• Robert Stehrer
• The Vienna Institute for International Economic
  Studies wiiw
• www.wiiw.ac.at
• Robert.Stehrer_at_wiiw.ac.at
• The WIOD project www.wiod.org