Forest Products Trade Flow Database

Forest Products Statistics and Trade Flows

Search the database

I. Introduction

The construction of the EFI-WFSE Forest Products Trade Flow database has been made possible through collaboration between the World Forests, Society and Environment (WFSE) program and the European Forest Institute. The main researchers involved in the development of the database are Philip Wardle and Bruce Michie.

The EFI-WFSE Trade Flow database uses trade data from the COMTRADE database, provided by the United Nations Statistical Division as a starting point. The data then are processed in order to obtain preciser estimates of the trade flows. The why and how will be explained in the following chapters. Further, this guide contains a step-by-step overview on how to use the database.

The EFI-WFSE Trade Flow database covers annual bilateral trade flows covering all main forest products and all countries for all years from 1962 until 2006. The forest products trade flows can be retrieved from the database by grouping categories, based on the FAO forest products classification. The database is under constant development and improvement and in the nearby future it will be made possible to study forest products trade flows at a more detailed level than the current aggregation of products. Also the possibilities of graphical display of the data is being further developed. The data for following years will be added to the time series as they become available.

At the end of this document are the references for this introduction with more background information, of which some can be downloaded in pdf format.

I.1. Data source

The EFI Forest Products Trade Flow database uses UN-COMTRADE Trade Data compiled by the United Nations Statistics Division (UNSD). The UN COMTRADE database covers annual bilateral commercial trade flows of all main forest products and countries for all years from 1962 until 2006.

The main work on the development of international statistical information has been carried out in United Nations organizations, statistical information on the forest sector being a particular remit of the Food and Agriculture Organization (FAO) in Rome and the joint FAO/ECE (Economic Commission for Europe) division in Geneva. International trade statistics are collected by the United Nations Statistics Division.

Due to the work-intensive process of exchanging manually recorded statistical information, FAO and ECE collected forest product trade statistics directly from the member countries. This effort mainly involved the collection of total imports and exports from each country, although trade flows between a few large traders were collected.

Since the early 1970s, UN organizations enter their statistical data into computer systems and UN and FAO yearbooks are generated from computer databases. By 1980 it has become routine to exchange data in computer readable format. The development of comprehensive databases makes that data become available on forest products in FAO, in particular data on production, total imports and total exports. Data on international trade at the detail of trade flows, are collected at the UN, and the historical data series cover all years since 1962.

I.2. Commodity coding and definitions

The UNSD COMTRADE database records data from national trade recordings on all merchandise trade. Products are classified according to the Standard International Trade Classification (SITC). SITC classifies, codes and defines all products. Since the inception of this classification, its coding and definitions have been revised on a number of occasions (SITC Rev.1 UN 1961, SITC Rev.2 UN 1975 and SITC Rev.3 UN 1985). Revision 3 is made fully consistent in the classification of products with the Harmonized Commodity Description and Coding System 1996 (Harmonized System 1996, HS96) trade classification of the Customs Co-operation Council (CCC, now better known as the World Customs Organization (WCO)).

HS96 is a classification of goods by criteria based on raw materials and the stage of production of commodities (Eurostat Inventory of International Statistical Classifications, 2000). It is a contractual agreement on the classification and definition of items in trade used by 109 countries.

The SITC Rev.3 and HS96 classifications provide an excellent and effective framework for trade statistics, although they do impose a limitation on the detail. While developing the HS, the CCC has taken little notice of the concerns of foresters in the classification for roundwood trade.

II. Preparing UN Comtrade data for the EFI-WFSE Forest Products Trade Flow Database

II.1. Product classification and aggregation

Countries may report data in several SITC revisions but not all countries report in all revisions. Therefore it is important to search the individual revisions to see if any countries have reported only in one particular or in more than one revision. When a country reports in several revisions it is important to check if quantity data exists in all revisions. Data under the latest revision is used unless data in an earlier revision appears.

In the development of the forest products trade flow database, the product classification of FAO has been adopted. International organizations have established standard classifications and definitions for forestry and forest products on the basis of careful consultations with experts from member countries. The FAO Classification and Definition of Forest Products is in accordance with the trade classifications but it goes into greater detail to meet the requirements of forestry, industry and marketing.

II.2. Standardising quantity units

Once the proper SITC codes were selected from each revision for the WFSE Trade Flow Database, it was necessary to convert the units from metric tonnes (the most common unit found in the COMTRADE data) to cubic meters (solid wood products only). Where data are reported in weight units (metric tonnes) a standard conversion factor is applied to convert from weight to quantity (Ref. EFI-WFSE Trade Flow Database - General description, Appendix 3). These conversion factors mainly come from the FAO Yearbook of Forest Products.

II.3. Quantity data preparation process

The next step is the preparation and cleaning of trade flow data. This includes the replacement of missing quantity data and checks to ensure that quantity and value data are consistent. Missing quantity data occurs when the value of a trade flow exists but when there are no corresponding quantity data. Quantity data is considered to be inconsistent when the unit price of a trade flow is more than 5 times the appropriate average unit price or less than 1/5th of the average unit price. The reason for this consistency test is based on experience that the most common quantity errors (or inconsistencies) arise from misplaced decimal points (e.g. data in kg or 100 kg rather than tonnes).

Substituting for missing or inconsistent data is done by first obtaining a world average unit price for the product and trade flow in question. Appropriate world average unit prices are calculated for imports and exports. Once the average unit price has been determined the missing quantity data is substituted by using an estimate calculated from the value of the trade flow and the average unit price. For veneer the procedure was modified to use average unit price of large exporters rather than world average unit price.

For each trade flow 4 estimates of quantity exist. The first estimate (Quantity) is the result of the cleaning process. The second estimate (Qorig) is the original quantity received from the UN Statistics Division (no substitution of missing quantities has been done for this estimate). The third estimate (QWUV) is based on value and world unit value. The fourth estimate (QEBP) is based on value and the average unit price of large exporters. These steps complete the standardization of existing records.

II.4. Estimating missing trade flows

When an entire trade flow (both quantity and value) is missing then trading partner information is substituted for the missing reporting country information. This type of substitution occurs (or does not occur) based on the existance (or in-existance) of trade flow data at the FAO product level.

This is illustrated by the following example: a non-coniferous sawnwood trade flow might be reported by a reporting country under SITC Rev.2 definitions (product code 2483) while the trading partner might report under SITC Rev.3 definitions (product code 2484). The result would be double counting of the trade flow if trading partner substitution is done at the SITC product level.

It should be noted that substitution of trading partner information for missing data introduces a certain bias in that the trading partner data is entered unchanged as the estimate of the missing trade flow. This means that the value of missing imports are underestimated by the use of the exporters value (fob - free on board) while the missing exports are over valued by the use of importers (cif - cost insurance freight) value. The difference between cif and fob values is transportation and insurance costs. An additional, but unfortunately unknown, bias results from the time lag between delivery of exported units to a port and the subsequent retrieval of those units from the destination port (as imports). Additionally, not all exported units actually arrive at the intended destination port.

II.5. Changing national and regional boundaries

Countries have joined together and split apart over the 1962-2006 time period covered by the data. This forces one to aggregate data in one part of the time series so that it can be compared with data in other parts of the time series. This means that the Area of the Former Soviet Union (a country aggregate for data starting in 1993) is necessary to compare current data with historical data for Ex-USSR (data prior to 1993). Since 1993 data is reported for 7 additional countries in Asia (Armenia, Azerbaijan, Georgia, Kazakhstan, Kyrgyzstan, Tajikistan, Turkmenistan), one country (Russian Federation) that extends into both Asia and Europe and 6 countries wholly in Europe (Belarus, Estonia, Latvia, Lithuania, Moldova Republic and Ukraine). While Germany refers to the full country since 1991 and the aggregate of Germany DR (former German Democratic Republic) and Germany FR (former Federal Republic of Germany) prior to unification. Ex-Czechoslovakia is used for data prior to 1993 while Czech Republic and Slovakia are used for data starting in 1993. Ex-Yugoslavia is used for data prior to 1993 while former Yugoslav Republic of Macedonia, Bosnia and Herzegovina, Croatia, Slovenia and Yugoslavia (rump) are used for data starting in 1993. Ex-Ethiopia, Ethiopia (rump) and Eritrea are necessary to follow Ethiopia data. Yemen, Yemen Democratic Republic and Yemen Arab Republic are necessary to follow Yemen data through time. Belgium-Luxembourg data were aggregated prior to 1999. Starting from 1999, data is being made separately available for Belgium and Luxembourg, as the historical series is being maintained.

III. How to query the EFI-WFSE Forest Products Trade Flow Database

Selecting the data variables for tabular output

1. Select a reporting country. Only single selection is possible.
2. Select (a) trading partner(s). Multiple selection is possible.
3. Or select a group of trading partners. Only 1 group at a time. Please note that the selection of a group will make the query to ignore the country selection!
4. Select the data type(s): Exports and/or Imports. Multiple selection is possible.
5. Select the product group(s): Sawnwood; Panels; Woodpulp; Paper and Paperboard and/or All Forest Products. Multiple selection is possible.
6. Select the output unit: Quantity or Value.
7. Press the "Search" button.

IV. Making use of the database

This database serves as a major tool for retrieving actual and historical bilateral forest products trade flow data, and constitutes a well-founded basis for (a) the analysis and projections of forest products trade flows and (b) for studies on markets' development at the national, regional or global scale. The Forest Products Trade Flow Database may also serve as input information for scientifical studies as e.g. the estimation of carbon sink in forest products.

V. Acknowledgements

The EFI/WFSE Forest Products Trade Flow Database is constructed from the trade flow data which are available at the United Nations COMTRADE database.

The construction of the EFI/WFSE Forest Products Trade Flow Database is being made possible by the collaboration between the World Forests, Society and Environment (WFSE) programme and the European Forest Institute.

The main researchers involved in the development of the database are Philip Wardle and Bruce Michie. Philip Wardle is an Associated Researcher at the European Forest Institute and Fellow Researcher in the WFSE programme. Bruce Michie is an expert in trade statistics and Senior Researcher at the European Forest Institute.

VI. References

VI.1. Basic information sources on the EFI/WFSE Forest Products Trade Flow Database:

EFI/WFSE Trade Flow Database General description
Bruce Michie and Philip Wardle
European Forest Institute, Joensuu, Finland, 2000. Updated 2002.
16p.

Includes:

  • Table with countries and regions in database
  • Table with UN product codes making up DataBase products
  • Table with metric ton to cubic meter conversions

Download in PDF format, free of charge

UNSTAT Trade Data as Basis for Analysis and Projections of Forests Products Trade Flows

Bruce Michie and Philip Wardle
Working Paper 17
European Forest Institute, Joensuu, Finland, 1998
ISBN: 952-9844-59-X, ISSN: 1237-5126, 44 p.
Price: 15 EUR
Download in PDF format, free of charge

VI.2. Publications referring to the EFI/WFSE Forest Products Trade Flow Database:

  • Brooks, D.J. (1995). European Timber Trends: Issues and Priorities for Further Research. Working Paper 6. Joensuu, European Forest Institute, 19p.
  • Michie, B., Wardle, P. and Kin, S. (2000). Transport Flows of Forest Products between Global Regions. (in) Logistics in the Forest Sector. Sjöström, K. (ed). Timber Logistics Club (email:Econpap@yahoo.com). 298 p. (ISBN 952-91-1942-9).
  • Michie, B., Wardle, P. (2000). Handel mit Holzprodukten in Europa. AFZ Der Wald, 55, 11: 581.
  • Michie, B., Chandrasekharan, C. and Wardle, P. (1999). Production and Trade in Forest Goods. (in) Palo, M. and J. Uusivuori (eds). World Forests, Society & Environment Volume I. Kluwer Academic Publishing. Amsterdam and New York. 398 p. (ISBN 0-7923-5595-4).
  • Michie B. and Kin, S. (1999). A Global Study of Regional Trade Flows of Five Groups of Forest Products. World Forests, Society and Environment Research Program, Helsinki, 76 p. (ISBN 951-40-1674-2).
  • Pajari, B., Peck, T. & Rametsteiner, E. (Eds.) (1998). Potential Markets for Certified Forest Products in Europe. Joensuu, European Forest Institute, Proceedings 25, 352 p.
  • Rametsteiner et al. (1998). Potential Markets for Certified Forest Products in Europe. Joensuu, European Forest Institute, Discussion Paper 2, 24p.
  • Solberg, B. & Moiseyev, A. (Eds.) (1998). Analyzing Structural Changes in Roundwood and Forest Products Markets in Europe: Empirical Studies and Research Priorities. Joensuu, European Forest Institute, Proceedings 26, 162p.
  • Solberg, B. & Moiseyev, A. (Eds.) (1995). Demand and Supply Analyses of Roundwood and Forest Products Markets in Europe - Overview of Present Studies. Joensuu, European Forest Institute, Proceedings 17, 418p.
  • Wardle P. and B. Michie. (2001) World forest products trade and the changing role of roundwood. World Forests, Society & Environment Volume III. Kluwer Academic Publishing. Amsterdam and New York (forthcoming).
  • Wardle, P. & Michie, B. (1999). ACP Forest Products Trade and the European Union. Joensuu, European Forest Institute, Discussion Paper 4, 28p.
  • Wardle, P. & Michie, B. (1998). Markets for Forest Products in Europe in the Face of Integration and Globalisation. Joensuu, European Forest Institute, Discussion Paper 3, 28p.