Transcript Trade Statistics at UNCTAD - African Centre for Statistics

Experts Group Meeting on the Compendium of
Intra-African and Related Trade Statistics
Addis Ababa, Ethiopia, 5-7 October 2011
Trade Statistics at UNCTAD
Markie Muryawan
Statistician
29 September 2011
[email protected]
Trade Statistics at UNCTAD
Role of UNCTAD statistics
UNCTAD statistics highlight the interconnection
between globalization, trade and development.
Reliable and timely data help developing
countries diagnose and evaluate the state of
economies
Trade Statistics at UNCTAD
Contents of UNCTAD statistics
• International merchandise trade, with long historical time series up
to the most recent 2010 data available, including breakdowns of
exports and imports by partner and products, merchandise trade
indicators
• International trade in services
• Economic trends, with national accounts, balance of payments
statistics, and exchange rates
• Foreign direct investment
• External financial resources, including migrants' remittances,
development assistance, external debt and international reserves
• Commodity prices, updated monthly, including long-term series and
calculations on the instability of prices
• Population and labour force
• Information economy
• Creative economy
• Maritime transport including the liner shipping connectivity index
Trade Statistics at UNCTAD
Insights from UNCTAD statistics
“The food crisis, the climbing cost of oil and minerals until mid 2008,
concerns about inflation, and then the global crisis,
the severely decline of international trade in 2009,
the improved economic activity in 2010,
commodity price indices exceeds its previous 2008 peak in beginning 2011,
the long standing decrease of Official Development Assistance (ODA)
share in financial flows to developing countries,
the increased representation of foreign direct investment (FDI)
and migrants’ remittance in GDP share …”
Trade Statistics at UNCTAD
Use of Trade statistics
Africa: International Reserve including gold
Number of months of imports
12
10
8
6
4
2
0
1971-1980
1981-1990
1991-2000
2001-2010
Source: UNCTADstat
Africa: Contribution to real GDP growth
10.0
8.0
6.0
% point
• One of UNCTAD primary focus
areas is in trade, therefore
trade statistics is indispensible
and is used in various
analytical reports and
publications
• But, in order to have more
meaningful interpretation for
analysis, trade statistics needs
to be (1) reliable, (2) complete,
and (3) coherence
• In addition, it is to be combined
with other statistics domains to
construct new indicators.
• See examples on the right
4.0
2.0
0.0
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
-2.0
-4.0
Net exports
Source: UNCTADstat
Consumption
Investment
GDP (% real change pa)
List of UNCTAD trade series
Series Name
Freq.
Update Calendar
Merchandise Total Trade
Annual
Apr/May, Sep/Oct
Trade by partners[1]
Annual
Apr/May, Sep/Oct
Trade by partners and products (Trade Matrix)
Annual
Apr/May, Sep/Oct
Trade Indices: Volume, Unit Value, etc.
Annual
May, Oct
International Trade Price Indices[2]
Annual
Apr, Oct
Quarterly volume growth rate
Quarterly
May, Aug, Nov, Feb
Concentration Indices / Structural Changes[3]
Annual
May, Oct
Tariff
Annual
Feb
Creative Economy – Creative Goods Trade
Annual
Nov
Information Economy – ICT Trade
Annual
Dec
[1]
Internal use only, not for publication
Idem
[3] Derived indicators from trade matrix
[2]
Trade Statistics at UNCTAD
For trade statistics, the main sources are …
• At total trade level:
– EUROSTAT, IMF - BOP, IMF - DOT, IMF – IFS, Interagency Common Data Set (CDS), OECD, OPEC - Annual
Statistical Bulletin, UN Comtrade, World Bank - WDI, EIU
Online, WTO, and National Sources
• Trade by partner:
– IMF DOTS and UN Comtrade
• Trade by partner and product:
– UN Comtrade
• Conclusion: Data are widely available and easily
accessible even at the most detailed level
Trade Statistics at UNCTAD
But those data may not be …
• Reliable
– Figures vary among sources, sometimes with great
magnitude differences
– Do not follow acceptable trends: outliers?
– Sometimes do not reflect well known fact
– Excessive confidentiality measure on products or partners
• Complete
– Missing major products or partners
– Data availability for non-reported countries
• Coherence
– Cannot directly combined with other statistical domains as
methodologies may differ (ex. trade in services or national
accounts)
Trade Statistics at UNCTAD
And as consequences …
• Users cannot use “raw” trade data directly for economic
analysis to create trade indicators such as territorial
aggregation, intra-trade on specific region or trade
concentration.
• See examples below showing different figures at total
trade level
Gambia exports
Benin exports
Comtrade: Re-exports
included for 2009 only
70.00
1 600.00
60.00
1 400.00
1 200.00
mln usd
mln usd
50.00
40.00
30.00
1 000.00
800.00
600.00
20.00
400.00
10.00
200.00
Break in series:
change the source
0.00
0.00
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2001
2002
2003
2004
2005
2006
2007
2008
2009
VALUEHBS
VALUEIFS
VALUEWTO
VALUEHBS
VALUEIFS
VALUEWTO
VALUECMTR
MBS
VALUECDS
NSO
VALUEBOP
VALUECMTR
VALUEEIU
VALUE COMPENDIUM
MBS
VALUECDS
VALUE COMPENDIUM
2010
Trade Statistics at UNCTAD
Therefore, we need to …
• Reliability
– Choose the best reliable figures preferable the one that
complies to international recommendations
– To the extent possible, detect potential outliers and correct
them
• Completeness
– Detect incomplete partners or products and adjust them
– Estimate non-reporting countries
• Coherence
– Be careful when mixing data from different methodologies,
and put notes as much as possible
Trade Statistics at UNCTAD
Actions in more detail …
• At total level:
– Working closely with
WTO to compile
Interagency CDS,
especially for LDC
countries
– Estimation to the extent
possible (sometimes
estimated data is not to
be published but used as
input to the calculation of
derived indicators)
• At detailed level (SITC 3digit):
– To the extent possible,
adjust data to be in the
acceptance range of
total trade
– Decipher unknown
partners and products
– Estimate non-reporting
countries
Trade Statistics at UNCTAD
Our estimation methodology
1st Step
2nd Step
Partner A
3rd Step
• Start from finding the best total
figures then set the band of
acceptable
• Then estimate partner
distributions based on IMF
DOTS or UN Comtrade mirror
• Then distributes the products
based on historical trends and
UNCTAD price indices
• To the extent possible,
estimate significant unknown
partners and products
SITC 0
SITC 1
…
SITC 9
Partner B
Trade Statistics at UNCTAD
Limitation
• Selected totals may not cover the whole economic territory
• Potential risk of error in mixing IMF DOTS and UN Comtrade
as their methodologies differ
• Unreliability of mirror statistics due to asymmetries in partner
attribution, valuation and trade system
• Price indices may not reflect situation in particular countries or
regions
• Potential break in series between reported data and estimated
data
• And …
• Estimated data may not reflect reality on the ground
Trade Statistics at UNCTAD
Further improvement
• Get the totals right with the assistance from countries (thanks
to their knowledge on national circumstances)
• Avoiding the break in series by respecting the trends for the
combination of partner and product
• Build expert knowledge base on specific country or sector and
integrate it on estimation methodology
• Better sharing of data or methodology among organizations
dealing with data estimation
• And …
• To extend time series coverage and more detailed products
using HS classification
Trade Statistics at UNCTAD
UNCTAD Trade Indices
• Unit value indices of exports and imports
• Terms of trade indices
– The ratio of the export unit value index to the import unit
value index
• Volume indices of exports and imports
• Purchasing power indices of exports
– The value of index of exports deflated by the import unit
value index
For individual countries and trade/economic groups
Trade Statistics at UNCTAD
Example use of trade indices in publication
• UNCTAD Trade and Development Report 2011
Trade Statistics at UNCTAD
Trade indices main data sources
• UNCTAD total trade series and
• Unit value of imports and exports (aggregates)
–
–
–
–
IFS unit value indices of exports and imports
CEPALSTAT data for ECLAC member states
Some national sources
For other countries, UNCTAD calculates unit value indices
on the basis of "International trade price indices" and
UNCTADstat Trade matrix
Trade Statistics at UNCTAD
International trade price indices
• Calculated/Estimated at SITC 3-digit level on the basis of:
– Germany Busdesamt, trade indices
– Japan Customs, unit value indices
– Bank of Japan, import and export price indices
– UK Office of National Statistics, import and export price indices
– U.S. Bureau of Labor Statistics, import and export price indices
– UNCTAD commodity price statistics
– FAO commodity prices.
– IMF-IFS commodity prices
– World Bank commodity prices
Trade Statistics at UNCTAD
Users of trade indices and joint project
• Internal analysis and international organizations
– UNCTAD Trade and Development Report Ch.1, Section
International Trade
– UN WESP World Economic Situation and Prospects
– World Bank WDI
• Joint UNCTAD/WTO project on volume indices
– Quarterly volume indices
– Coherent annual and quarterly volume indices
Interdependence of trade series
Illustration: Simplified UNCTAD Merchandise Trade Series interdependence,
External sources, Tariff, Creative Economy and ICT series are not shown
International Trade Price Indices
Total Trade
Trade by Partners
Trade Matrix
Concentration Indices /
Structural Changes
Quarterly volume growth rate
Trade Indices
Trade Statistics at UNCTAD
Dissemination system
• All UNCTAD statistics are made available at
UNCTADstat
• UNCTAD Tariff database is accessible from WITS
• Single point of contact: [email protected]
Trade Statistics at UNCTAD
Selected UNCTAD publications