The OECD InputOutput database and SupplyUse Tables in SNA 1993 Rev 1 OECDNBS Workshop on National Ac - PowerPoint PPT Presentation

1 / 29
About This Presentation
Title:

The OECD InputOutput database and SupplyUse Tables in SNA 1993 Rev 1 OECDNBS Workshop on National Ac

Description:

... produces 20 units of manufacturing we assume that it produces 20/220 per cent ... that all processing be assumed substantial and therefore gross flows ... – PowerPoint PPT presentation

Number of Views:52
Avg rating:3.0/5.0
Slides: 30
Provided by: Ahm7
Category:

less

Transcript and Presenter's Notes

Title: The OECD InputOutput database and SupplyUse Tables in SNA 1993 Rev 1 OECDNBS Workshop on National Ac


1
The OECD Input-Output database and Supply-Use
Tables in SNA 1993 Rev 1 OECD-NBS Workshop on
National Accounts September 25-28, 2007, Beijing
  • Contact nadim.ahmad_at_oecd.org

2
OECD Input-Output Database
  • History
  • 2006 Edition
  • Creating Symmetric Tables
  • Data Sources
  • Why Industry by Industry?
  • Dissemination

3
History
  • 1995 Edition
  • 10 countries
  • SNA68
  • ISIC Rev 2
  • 36 sectors
  • Up to 1990
  • 2002 Edition
  • 20 Countries (2 non member)
  • SNA 93
  • ISIC Rev 3
  • 42 Sectors
  • Up to 1998

4
2006 Edition
  • 2006 Edition
  • 37 countries (9 non members)
  • SNA93
  • ISIC Rev 3
  • 48 sectors
  • Up to 2003

5
2006 Edition - Country Coverage
  • Country coverage corresponds to over 90 of
    global GDP (80 in 2002 Ed and 70 in 1995).
    Population coverage (66 versus 40 and 10
    respectively)

6
2006 Edition - Tables
7
2006 Edition Industry Coverage
8
2006 Edition - Value-Added Final Demand
  • VA
  • FD

9
Creating Symmetric Tables
  • Requests for Industry by Industry (IxI)
    (preferably 48x48 at BP)
  • Or Supply-Use, or Commodity by Commodity (CxC)
    and Supply, or CxC
  • Conversion Steps
  • S-U at purchasers prices Convert use table to
    BP.
  • S-U at BP (total economy only) Convert Use
    table into separate domestic and import use
    tables
  • Convert S and Domestic Use tables into IxI tables
    using Fixed Product Sales Structures
    assumption.
  • Aggregate to 48x48
  • CxC and Supply (reverse engineer the Use table
    and follow steps above)
  • CxC - aggregate only (Japan, Korea. Chinese
    Taipei and Indonesia)
  • Other e.g. FISIM, c.i.f/f.o.b

10
Transforming Supply Use into Input-Output Tables
  • SU tables are CxI not CxC or IxI
  • So for CxC, its necessary to convert output by
    industries into output by products
  • And for IxI, its necessary to convert demand by
    products into demand by industries.
  • If each industry produced only one product this
    would be trivial.
  • Unfortunately this is rarely the case.

11
(CxC) Input-Output Tables
  • Two assumptions prevail in constructing CxC
    tables, which can be used in isolation or often
    in combination
  • Product technology - Each product is produced in
    its own specific way, irrespective of the
    industry where it is produced
  • Industry technology - Each industry has its own
    specific way of production, irrespective of its
    product mix.

12
Product technology
This illustrates one of the biggest practical
problems with the implementation of the product
technology assumption although it could of
course be used to identify problems with the
original SU balance
Agriculture produces 20 units of manufacturing
we assume the same structure as in manufacturing,
hence (minus) 80/20020-8 etc
13
Industry technology
Note there are no negatives. A strength of the
industry technology assumption.
Agriculture produces 20 units of manufacturing
we assume the same structure as in agriculture,
hence zero for agricultural products and (minus)
60/15020-8 etc
14
(CxC) Input-Output Tables
  • A third assumption is the hybrid technology which
    uses parts of the industry and product technology
    assumptions.

15
(CxC) Input-Output Tables
  • These transformations can be described
    algebraically as
  • Where U is the original SU IC CxI Table VA is
    the VA vector (VA by I) q, the vector of
    domestically produced products and g, the vector
    of the output of industries

16
(IxI) Input-Output Tables
  • Like CxC two assumptions prevail in constructing
    IxI tables
  • Fixed Product Sales Structures Each product has
    its own specific sales structure, irrespective of
    the industry where its produced.
  • Fixed Industry Sales Structures - Each industry
    has its own specific sales structure,
    irrespective of its product mix.

17
Fixed Product Sales Structures
Agriculture produces 20 units of manufacturing
we assume that it produces 20/220 per cent of all
products and that each consumer purchases this
share of manufactured products from the
agriculture industry, so, of the 60 purchased by
agriculture 6020/220 5.5 is from the
agriculture industry.
18
Fixed Industry Sales Structures
Agriculture produces 20 units of manufacturing
we assume that the shares are split equally
between consumers of agricultural products, so
80/1302012.3 goes to manufacturing,
50/130207.7 to final demand etc. This can also
result in negatives.
19
(IxI) Input-Output Tables
  • These transformations can be described
    algebraically as
  • Where U is the original SU IC CxI Table fd is
    the final demand vector q, the vector of
    domestically produced products and g, the vector
    of the output of industries

20
So why do the OECD choose Industry by Industry?
  • Linkages to other OECD industrial database
  • STAN, ANBERD, SDBS, IEA (emissions) etc
  • Policy focus
  • Structure of businesses, Entrepreneurship etc
  • Statistical Quality whether CxC or IxI,
    assumptions are needed
  • Information sources, typically, business
    (industry) based. IxI using Fixed Product Sales
    assumption (FPSA) preserves observed VA
    relationships. CxC does not.
  • Equally the CxC assumption of heterogeneity in
    products is intrinsically linked to empirical
    facts classification systems are too aggregate
    and businesses rarely have the same cost
    structures.
  • Simplicity
  • IxI tables easily produced using FPSS (no
    negatives)

21
Sources
22
Dissemination
  • http//www.oecd.org/std/io-tables/data
  • 2002 Edition available now (on request) from
  • 1995 Edition available on-line
  • 2006 Edition release imminent
  • See also Ahmad Yamano, 2006 for more
    information.

23
Special Issues SNA93 Rev 1 implications
  • Although the SU tables are not in themselves
    subject to change in the SNA revision, a number
    of changes in other areas will have an effect.

24
Special Issues SNA93 Rev 1 implicationsAncill
ary Units
  • The 1993 SNA specifies that units conducting only
    a specified list of activities designated as
    ancillary should not be treated as separate
    units but their costs should be consolidated with
    the units they serve. This means that when
    accounts for a region are compiled, head offices
    and other ancillary units located there are
    excluded if the units they serve are located
    outside the region. This results in a difference
    between ancillary units located abroad, which are
    treated as separate units, and those that are
    resident but distant from their related
    enterprises.

25
Special Issues SNA93 Rev 1 implicationsAncill
ary Units
  • The AEG recommended that ancillary units can be
    establishments in their own right if they satisfy
    the normal requirements of an establishment
    allocated to the main service classification
    provided by the unit.

26
Special Issues SNA93 Rev 1 implicationsGoods
sent abroad for processing
  • The 1993 SNA and BoP treat goods sent abroad for
    processing differently. The SNA records gross
    flows only in the case of substantial processing
    (reclassification of the good at three-digit
    CPC). The Balance of Payments Manual, as a
    practical matter, suggests a convention that all
    processing be assumed substantial and therefore
    gross flows are recorded.
  • Further, the position is that when goods are sent
    abroad for processing, no change in ownership
    takes place and thus there are no actual
    transactions.
  • Does the advent of globalization and the
    increasing amount of goods processed abroad
    suggest a change in practice would be
    appropriate?

27
Special Issues SNA93 Rev 1 implicationsGoods
sent abroad for processing
  • The AEG has decided to resolve this issue by
    never imputing a change of ownership, and, so,
    not recording gross flows. Further the AEG also
    recommended that the same approach should be used
    in dealing with goods processed domestically even
    if between related enterprises,

28
Special Issues SNA93 Rev 1 implicationsMercha
nting
  • Merchanting is defined in BoP as the purchase of
    a good by a resident of country A from a resident
    of country B which is then sold to a resident of
    country C, without the good entering the
    merchants economy. The SNA does not cover this
    topic.
  • There is a need for a clear and precise
    definition of merchanting arising out of this
    there needs to be clear guidance on whether
    merchanting (when redefined) should be recorded
    on a net or a gross basis and under goods or
    services.

29
Special Issues SNA93 Rev 1 implicationsMercha
nting
  • The AEG has recommended that, the acquisition of
    goods by the merchanter should be recorded as an
    import, identified as a negative export, of the
    merchanter. The resale of these goods is then
    shown as an export with the difference in values
    (exclusive of holding gains/losses) allocated to
    wholesale/retail exports.
Write a Comment
User Comments (0)
About PowerShow.com