![]() |
ECE/TIM/DP/5 UN-ECE/FAO TIMBER AND FOREST DISCUSSION PAPERS ETTS V WORKING PAPER MODELING FOREST PRODUCTS DEMAND, SUPPLY AND TRADE |
United Nations Economic Commission for Europe
Food and Agriculture Organization
of the United Nations
Timber Section
Geneva, Switzerland
by David Brooks, Anders Baudin and Peter Schwarzbauer
UNITED NATIONS
New York and Geneva, 1995
ETTS V WORKING PAPERS
This is one of a series of papers to accompany the fifth study of European timber trends and prospects (ETTS V). The ETTS V Working Papers contain material which is too technical, too detailed or too voluminous for inclusion in the main study, such as results and projections by country or methodological discussion.
ETTS V was prepared in a two stage process:
sector-by-sector analysis, based on the work reported in the Working Papers;
creation of unified scenarios, after a consistency analysis.
It was sometimes necessary to modify the sector-by-sector data in order to get a comprehensive but internally consistent scenario for the forest and forest products sector as a whole. For this reason, although the ETTS V scenarios are based on the analysis in the Working papers, not all the results correspond exactly. Readers are referred to ETTS V itself for an explanation of the methods used.
Although prepared as part of ETTS V, the Working Papers are the responsibility of their authors.
Further information about ETTS V, and other documentation is available on request from the ETTS V coordinator,
Mr. C. Prins | Telephone | +41 22 917 2874 |
Chief, Timber Section | Fax: | +41 22 917 0041 |
ECE Trade Division | E-mail: | [email protected] |
Palais des Nations | ||
CH 1211 GENEVA 10 |
UN-ECE/FAO TIMBER AND FOREST DISCUSSION PAPERS
The objective of the Discussion Papers is to make available to a wider audience work carried out, usually by national experts, in the course of ECE/FAO activities. They do not represent the final official output of the activity (which is normally issued as a UN-ECE/FAO Timber and Forest Study Paper), but a contribution which because of its subject matter, or quality, or for other reasons, deserves to be disseminated more widely than the restricted official circles from whose work it emerged, or which is not suitable (e.g. because of technical content, narrow focus, specialised audience) for distribution as a Study Paper.
In all cases, the author(s) of the discussion paper are identified, and the paper is their responsibility. The ECE Timber Committee, the FAO European Forestry Commission, the governments of the authors' country and the FAO/ECE secretariat, are in no way responsible for the opinions expressed and the facts presented in the discussion paper.
In the interests of economy, Discussion Papers are issued in the original language only. They are free of charge and available on request from the secretariat while stocks last. They are distributed automatically to nominated forestry libraries and information centres in member countries. Those interested in receiving these Discussion Papers on a continuing basis should contact the secretariat.
Abother objective of the Discussion Papers is to stimulate dialogue and contacts among specialists. Comments or questions should be sent to the secretariat, who will transmit them to the authors.
Preface by the secretariat
Understanding and projecting the forces which will influence production and consumption of forest products is one of the most important parts of ETTS V, along with forecasting developments for the forest resource. ETTS V has relied on advanced econometric analysis to carry out this task for Western European market economies. This document presents the results of the modeling work carried out in the context of ETTS V to understand the relationship between consumption and production f forest products and other parameters, notably GDP, activity in user sectors, prices and costs. ECE/TIM/DP/6 contains the projections for consumption and production of forest products for the period to 2020.
This pathbreaking work has been carried out by Messrs. David Brooks (USA), Anders Baudin (Sweden) and Peter Schwarzbauer (Austria). Their work was essentially financed by their national governments and home institutions, with a relatively minor contribution of resources from FAO and ECE. The secretariat takes this opportunity to express its thanks to the three experts, their institutions and the governments of Austria, Sweden and the USA for their contribution to this international cooperative study.
Hyperlinks to non-FAO Internet sites do not imply any official endorsement of or responsibility for the opinions, ideas, data or products presented at these locations, or guarantee the validity of the information provided. The sole purpose of links to non-FAO sites is to indicate further information available on related topics.
A multiple-equation model of demand and supply
Models of forest products demand in smaller markets
Some facts about the Timber Committee
Modeling forest products demand, supply, and trade1
David J. Brooks
Research Forester
Forest Service--USDA
Pacific Northwest Research Station
3200 SW Jefferson Way
Corvallis, OR 97331 USA
Anders Baudin
Associate Professor
SIMS, Statistical Institute
University of Umeå
S-901 87 Umeå SWEDEN
Peter Schwarzbauer
Associate Professor
Univ. f. Bodenkultur
Gregor-Mendel Strasse 33
A-1180 Wien AUSTRIA
Forest products demand models were estimated for 18 European countries; supply models were estimated for 9 European countries. Elasticities were estimated for five groups of solid wood products (coniferous sawn wood, non-coniferous sawn wood, plywood, particleboard, and fiberboard), and three groups of paper and board products (newsprint, printing and writing paper, and other paper and board). A multiple equation framework was developed to model demand, supply, and trade of solid wood forest products in the 9 European countries that account for the majority of consumption and production of forest products. For these countries, demand models are similar to those developed in previous Timber Trends Studies; price and a composite indicator of end use activity are used to explain consumption. However, separate demand models are estimated for consumption from domestic markets and consumption of imports; where available, both import price and domestic market price were used as explanatory variables for the two demand equations.
The supply models for the major markets and producers represent an expansion of the scope of modeling done for Timber Trends Studies (TTS). For this study, forest products supply is explained using prices and costs; separate supply models are estimated for export supply, and supply to domestic markets. Using this approach to modeling supply and demand, elasticities can vary by market (for supply equations) and by source of supply (for demand equations). As a result, substitution behavior in both consumption and production can be examined.
For the 9 countries that are smaller markets for forest products in Europe, a model of total demand (apparent consumption) was estimated using gross domestic product and price as explanatory factors. For these countries, a time-series, cross-section approach was used to estimate elasticities; countries were grouped by per capita income.
For all countries and products, estimated elasticities are generally consistent with those reported in previous TTS, and those reported in the scientific literature. However, trade equations and elasticities (import demand and export supply) add new information to the TTS. Because trade is modeled, additional information is available regarding the outlook for European forest products consumption, and demand on European forests. Where complete data sets were available, results generally indicate that there is substitution in consumption between imports and domestic production, for most products, in most countries. In addition, import elasticities often are higher than elasticities estimated for consumption from domestic sources. Substitution between export markets and domestic markets also is evident in export supply equations.
There is a long tradition of quantitative studies of the demand for forest products. Holland (1960), Gregory (1966), and Hair (1967) are among the early studies. Multi-country studies by Buongiorno (1977, 1978) provided the foundation on which subsequent large-scale studies, such as those done by the Food and Agriculture Organization (1986, 1991), have built. McKillop and Wibe (1987) review the development of demand models for sawnwood and panel products, and Uutela (1987) reviews the development of demand models for paper and board products.
Modeling the supply of forest products is less common than demand modeling, especially in large-scale, multi-country studies. Kumar (1985) and Cardellichio and Kirjasniemi (1987) describe the development of supply models, and provide useful background information; McKillop (1967), Robinson (1974), Adams and Haynes (1980), Singh and Nautiyal (1986), and Lönner (1991), provide examples of approaches to supply modeling in the forest sector.
This working paper reports the results of statistical analyses of demand and production of forest products that follow in the tradition of this earlier work. In particular, we build on and extend the approaches reported in the previous Timber Trends Study (hereafter cited as ETTS IV) (ECE/FAO 1986), for models of the demand for forest products. We add to this an analysis of the supply of forest products for selected countries, focusing on export supply. For the countries that are the larger producers and markets for forest products in Europe we develop a multiple equation approach to estimating demand and supply elasticities. Our approach has some similarities to the model described by Singh and Nautiyal (1986); the trade equations we estimate are similar to import demand and export supply models estimated by Buongiorno and Manurung (1992). However, we develop our modeling framework with the explicit objectives of (a) supporting projections of total demand and supply, country-by-country, and (b) allowing for scenario analysis of key factors (such as economic growth, prices, and costs) in the outlook for European forests. Our work also builds on exploratory analyses done as follow-up to the review of ETTS IV.2
Two approaches to the analysis of demand and supply of forest products are described in this working paper:
a multiple equation model of demand (two equations) and supply (one equation), estimated for the countries that are the larger markets and/or major producers of forest products in Europe; and
a single-equation model of demand (apparent consumption), estimated for the countries that are the smaller markets for forest products in Europe.
Table 1 displays the countries and country groupings used in this analysis. The decision regarding the intensity of the statistical analysis for each country was initially based on the relative importance of the country in western European forest products markets. In some cases, data availability was an important factor in our decision. Each of the countries in Group I was examined separately, using an expansion of the core data set developed for ETTS IV.3 Modified versions of the “end-use elasticities” models reported in ETTS IV were estimated for the 9 countries in Group I (table 1). The modifications to the demand models are described in detail below.
Supply models also were estimated for the countries in Group I. We focused our attention on models of export supply that, combined with the two demand models provide the basis for projections of total demand and supply for these countries.
A single-equation demand model was estimated for the countries in Group II of table 1. Countries were sorted by per capita income (as shown), and a variant of the “GDP elasticities” model described in ETTS IV was estimated. The differences between the model estimated here, and that estimated in ETTS IV are primarily in data handling and in the specification of the price term.
Table 1--Country groupings
Group I: Demand and supply models estimated | |
---|---|
Austria | Norway |
Finland | Spain |
France | Sweden |
Germany | United Kingdom |
Italy |
Group II: Demand models estimated | |
---|---|
Higher income | Lower income |
Belgium-Luxembourg | Greece |
Denmark | Ireland |
Netherlands | Portugal |
Switzerland | Turkey |
As in previous Timber Trends Studies (TTS)4, the demand and supply analysis was done for 5 mechanical wood products, and 3 paper and board products. Product groupings are listed in table 2.
Table 2-Product groupings
Mechanical wood products | Paper and board products |
---|---|
coniferous sawnwood | newsprint |
nonconiferous sawnwood | printing and writing paper |
plywood | other paper and board |
particleboard | |
fiberboard |
4 Previous Timber Trends Studies were published in 1953, 1964, 1976, and 1986.