Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

An econometric analysis of the demand for wood products in Japan by product type, species, and source Gaston, Christopher Willem 1997

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1997-250512.pdf [ 8.77MB ]
JSON: 831-1.0075170.json
JSON-LD: 831-1.0075170-ld.json
RDF/XML (Pretty): 831-1.0075170-rdf.xml
RDF/JSON: 831-1.0075170-rdf.json
Turtle: 831-1.0075170-turtle.txt
N-Triples: 831-1.0075170-rdf-ntriples.txt
Original Record: 831-1.0075170-source.json
Full Text

Full Text

A N E C O N O M E T R I C ANALYSIS OF THE D E M A N D F O R W O O D P R O D U C T S IN J A P A N BY PRODUCT TYPE, SPECIES, AND S O U R C E  by  CHRISTOPHER WILLEM GASTON B . S c , The University of British Columbia, 1979 M . S c , The University of Guelph, 1982  A T H E S I S SUBMITTED IN PARTIAL FULFILMENT O F THE REQUIREMENTS FOR THE D E G R E E OF DOCTOR OF PHILOSOPHY in T H E F A C U L T Y O F G R A D U A T E STUDIES Department of Forest Resources Management  We accept this thesis as conforming to the requjred standard  THE UNIVERSITY OF BRITISH COLUMBIA July, 1997 © Christopher W. Gaston, 1997  In  presenting this  degree at the  thesis  in  University of  partial  fulfilment  of  of  department  this thesis for or  by  his  or  requirements  British Columbia, I agree that the  freely available for reference and study. I further copying  the  representatives.  It  publication of this thesis for financial gain shall not  is  of  r^Jo^^-  ^f^r^h  The University of British Columbia Vancouver, Canada  Date  DE-6 (2/88)  py  t  1  A »- y±-^e-J^~ e  e  Library shall make it  granted  by the  understood  that  head of copying  my or  be allowed without my written  permission.  Department  an advanced  agree that permission for extensive  scholarly purposes may be her  for  Page ii  Gaston Abstract  Abstract This thesis investigates the Japanese demand for wood by product type, by country of origin, and by species, over the period 1965 to 1993. The product types include softwood and hardwood logs, softwood and hardwood lumber, and wood-based panel products (plywood, fibreboard and particle board). In addition to estimating the own-price effects on quantity demanded for individual wood product imports, substitution effects within these product categories are documented to the degree possible, including Japanese substitution with domestic product and non-wood alternatives.  The research makes two important contributions. The first is to offer Japanese demand descriptors at a level of wood product detail which is not found in the existing literature. The second is to review and critique the existing methodologies available for investigating substitution effects among disaggregated products (such as softwood lumber by species).  A s it was discovered that the available approaches are inadequate for  properly dealing with product detail, strong recommendations for further research are made for improving our ability to document cross-price effects.  The primary conclusion of the study is that individual wood products, by product type, by country of origin, or by species, behave as distinct economic units. This suggests that studies which aggregate wood products into broad categories such as "softwood lumber" risk obscuring important dimensions of both forest products' trade and forest policy.  Gaston  Page iii  Table of Contents  TABLE OF CONTENTS  1.0  ABSTRACT  ii  T A B L E OF CONTENTS  iii  LIST O F T A B L E S  v  LIST O F F I G U R E S  vi  ACKNOWLEDGEMENTS  vii  INTRODUCTION  1.1 Motivation for the Research 1.2 Background 1.3 Scope of the Research 1.3.1 The Research Problem 1.3.2 Objectives 1.3.3 Hypotheses 1.4 Organization of Thesis 2.0  LITERATURE REVIEW ON L O G A N D L U M B E R SUBSTITUTIONS A N D IMPLICATIONS F O R FURTHER R E S E A R C H  2.1 The Econometric Estimates of the Price Elasticity of Demand 2.1.1 Estimates of Wood for Wood Substitutes 2.1.2 Estimates of Non-Wood for Wood Substitutes 2.2 Parametric Demand Elasticity Estimation Techniques 2.3 Implications for the Present Study 3.0  12 12 16 20 37  T H E M A R K E T F O R W O O D P R O D U C T S IN J A P A N  3.1 3.2 3.3 3.4 3.5  4.0  1 3 6 8 10 10 11  The Japanese Domestic Timber Resource The Use of Japanese Domestic Timber Production Imports of Wood Products into Japan Japanese Processing of Domestic and Imported Logs Summary  T H E O R E T I C A L F O U N D A T I O N S A N D IMPLICATIONS F O R T H E E M P I R I C A L  40 44 49 57 60  Gaston  Table of Contents  Page iv  ANALYSIS OF THE J A P A N E S E DEMAND FOR WOOD PRODUCTS  4.1 Theoretical Foundations 4.2 The Empirical Model 4.3 The Data Sources Used in the Empirical Analysis  5.0  EMPIRICAL RESULTS  5.1 Direct Estimation of Japanese Price Elasticities of Demand for Wood Imports 5.2 Estimation of the Armington Two-Stage Model of the Japanese Demand for Total Wood Imports 5.3 Comparison of the Two-Stage and Direct Estimates of the Own-Price Elasticities of Demand for Wood Product Imports in Japan 5.4 Non-Wood Substitution in Japan 5.5 Summary  6.0  . . 85 93 100 103 105  DISCUSSION OF R E S U L T S  6.1 6.2 6.3 6.4 6.5 6.6 6.7 6.8  7.0  63 71 73  Japanese Wood Product Imports, Aggregated by Product Type Japanese Softwood Lumber Imports, Aggregated by Source Japanese Softwood Lumber Imports from Canada, Aggregated by Species . . . Japanese Softwood Lumber Imports from Non-Canadian Sources, Aggregated by Species Japanese Softwood Log, Aggregated by Source Japanese Softwood Log, Aggregated by Species Japanese Hardwood Lumber and Log Imports, Aggregated by Source Japanese Panel Product Imports, Aggregated by Source  108 112 116 122 127 130 134 136  C O N T R I B U T I O N S , LIMITATIONS A N D IMPLICATIONS F O R F U R T H E R R E S E A R C H  7.1 Research Contributions and Implications for the BC Forest Industry 7.1.1 Summary of the Results 7.1.2 Implications of the Research for BC Wood Product Marketing 7.1.3 Implications of the Research for BC Forest Policy 7.2 Limitations 7.3 Implications for Further Research  141 142 144 150 151 154  BIBLIOGRAPHY  158  Gaston  List of Tables  Page v  L I S T  Table Table Table Table  2.1 2.2 2.3 2.4  Table 2.5 Table 2.6 Table 2.7 Table 2.8 Table 2.9 Table Table Table Table  4.1 4.2 5.1 5.2  Table 5.3 Table 5 . 4 Table 5.5 Table 5.6 Table 5.7 Table 5.8 Table 5.9 Table 7.1  O F  T A B L E S  Own-Price Elasticity of Demand for Softwood Lumber in N.A 13 Cross-Price Elasticity of Demand for Similar Lumber in Different Regions . . . . 15 Cross-Price Elasticity of Demand for Different Lumber 16 Own-Price and Cross-Price Demand Elasticities for Construction Materials: McKillop, etal 17 Own-Price and Cross-Price Demand Elasticities for Construction Materials: Rockel and Buongiorno 18 Own-Price and Cross-Price Demand Elasticities for US Softwood Lumber . . . 19 Own-Price and Cross-Price Demand Elasticities for Selected Canadian Construction Materials 19 Elasticities of Demand of US Hardwood Plywood Imports by Country of Origin 33 Elasticities of Demand of US Softwood Lumber Imports from Canada By Species 37 Japan Tariff Association Data, Converted Codes 75 B.C. Offshore Lumber Exports Relative to the Whole of Canada (000s m ) . . . 83 Estimates of the Japanese Demand for Aggregated Wood Imports 86 Estimates of the Japanese Demand for Selected Disaggregated Wood Products 90 Estimates of the Constant Elasticity of Substitution Over Varying Degrees of Wood Import Aggregation, Correcting for Serial Correlation 94 Calculated Constant Elasticity of Substitution Weights from Table 5.3 96 Cochrane-Orcutt Estimates of the Japanese Demand for Selected Aggregations of Wood Imports, Utilizing CES Quantity and Price Indices 96 Calculated Own- and Cross-Price Elasticities of Demand for the Japanese Imports of all Wood Products by Type 98 Calculated Own- and Cross-Price Elasticities of Demand for the Japanese Imports of Softwood Lumber by Country of Origin 99 Calculated Own- and Cross-Price Elasticities of Demand for the Japanese Imports of Canadian Softwood Lumber by Species 100 Cochrane-Orcutt Estimates of the Japanese Demand for Aggregated Wood Imports, With the Inclusion of a Non-Wood Regressor 105 Destination of Canadian Softwood Lumber and Log Exports, 1992 146 3  Gaston  List of Figures  Page vi  L I S T  Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure  1.1 1.2 3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8 3.9 3.10 3.11 3.12 3.13 3.14 3.15 4.1  O F  F I G U R E S  Random Lengths S-P-F Lumber Futures PNW Douglas Fir Lumber Prices Distribution of Man-Made Forest by Age Class (Japan) Japanese Domestic Log Production by Species Japanese Domestic Log Production by Ownership Japanese Domestic Log Supply by Utilization Japanese Housing Starts by Number Japanese Housing Starts by Area Japanese Industrial Wood Supply Japanese Self-Sufficiency in Logs Japanese Self-Sufficiency in Lumber Japanese Self-Sufficiency in Panel Products Japanese Imports of Softwood Lumber and Logs, 1993 Japanese Imports of Softwood Lumber and Logs, 1965 Japanese Imports of Hardwood Lumber and Logs, 1993 Japanese Imports of Hardwood Lumber and Logs, 1965 Japanese Lumber Shipments by Use Nominal Price of Japanese Imports of Canadian Sitka Spruce Lumber, By Size Figure 4.2 Nominal Price of Japanese Imports of Canadian Yellow Cedar Lumber, By Size Figure 4.3 Non-wood Housing Starts in Japan Figure 5.1 Observed versus Predicted Values of Quantity Demanded of Aggregated Softwood Lumber Imports by Japan Figure 5.2 Number of Non-Wood Housing Starts in Japan Figure 6.1 Japanese Imports of Wood Products by Major Product Types Figure 6.2 Japanese Imports of Softwood Lumber by Source Figure 6.3 Japanese Imports of Canadian Softwood Lumber by Species Figure 6.4 Japanese Imports of US Softwood Lumber by Species Figure 6.5 Japanese Imports of Former USSR Softwood Lumber by Species Figure 6.6 Japanese Imports of NZ/Chile Softwood Lumber by Species Figure 6.7 Japanese Imports of "Other" Softwood Lumber by Species Figure 6.8 Japanese Imports of Softwood Logs by Source Figure 6.9 Japanese Imports of US Softwood Logs by Species Figure 6.10 Japanese Imports of Former USSR Softwood Logs by Species Figure 6.11 Japanese Imports of Hardwood Lumber by Source Figure 6.12 Japanese Imports of Hardwood Logs by Source Figure 6.13 Japanese Imports of Veneer Sheets by Source Figure 6.14 Japanese Imports of Plywood by Source Figure 6.15 Japanese Imports of Particle Board and Fibreboard  4 7 42 43 44 46 48 48 49 50 53 53 55 55 56 56 58 80 80 53 88 104 108 113 117 123 125 126 128 129 131 133 135 137 138 139 140  Gaston  Acknowledgments  P a g e vii  A C K N O W L E D G M E N T S  I gratefully acknowledge Dr. David Haley as my supervisor, for his continued guidance, encouragement and support. A s a side, special appreciation is extended for Dr. Haley's faith in my teaching abilities, arranging for me to instruct F R S T 319 (Forestry Economics) while he was on sabbatical. I would also like to express my appreciation to the members of my advisory committee, being Dr. Clark Binkley, Dr. David Cohen and Dr. Russell Uhler. Special thanks are also due to Dr. lian Vertinsky and Dr. Casey V a n Koonten at the Forest Economics and Policy Research Unit (FEPA), Dr. William Stanbury at the Faculty of Commerce and Business Administration, and Dr. Bill Wilson at the Canadian Forest Service for their comments and advice on many aspects of my research. As for the many friends and colleagues that have helped make the long process of doing a Ph.D. bearable, I can only say that I could never have done it without your moral support. Although the people that I have had the pleasure to get to know over the years are too numerous to mention, I wish to single out two individuals which have made particularly strong impressions on me, both in a professional and a friendship capacity. Ramvir Singh and Paul Mitchell-Banks, I thank you! I am more than grateful for the financial assistance I have received to reduce the burden of doing a graduate degree. Appreciation goes to the University of B C for the Donald S. McPhee Fellowship, to F E P A for the Forest Economics and Policy Analysis Research Grant, and for the Canadian Forest Service for the F R D A Research Grant. Finally, I wish to express my strongest appreciation of all to my parents, Lloyd and Suzanne. Without their love, support and encouragement, I would never have dreamed of such an ambitious undertaking.  Gaston  Page 1  Chapter One  Chapter 1 Introduction  This thesis investigates an aspect of Pacific Rim log and lumber trade which has received surprisingly little attention to date: factor demand estimation with recognition that wood inputs are imperfect substitutes in production. While there have been many studies which have estimated demand parameters for wood inputs, virtually all of them have used highly aggregated trade data (such as "softwood lumber"). The present study investigates demand substitution by product, by region, and by species.  1.1  Motivation for the Research Hiding wood characteristics through data aggregation tends to obscure important  dimensions of both forest products' trade and forest policy. A good example of this problem is illustrated by U S allegations that B C export restraints on softwood logs constitute a subsidy for B C sawmills. By aggregating all softwood logs, there is a danger of obscuring the log export pattern which might exist in the absence of export restrictions. Kalt's (1994) submission to the U S Department of Commerce in the Canada/US softwood lumber countervail case offers a good example of how to improve trade analysis with less aggregated data. He argues that British Columbia (BC) export restraints, which primarily affect coastal logs, including a significant proportion of logs from which clear and merchantable grade lumber can be extracted, do not constitute a subsidy for interior sawmills producing mostly lower grade construction lumber. Another example is offered by the determination of allowable annual cuts within the  Gaston  Chapter One  Page 2  context of BC's sustained yield policy. Haley and Luckert (1994) and van Kooten (1993), for example, argue that meeting the objectives of sustained yield does not simultaneously meet the objectives of sustainable development. In short, choosing forest rotations and/or silvicultural regimes which maximize volume, without any reference to value, does not necessarily promote a strong, forest-based economy. If one adds goals to incorporate non-timber values in forest management, the inherent problems in focusing on physical volume alone become further amplified. There are a number of important questions which require an investigation of trade related to wood species and sources of origin. For example, will B C ' s second growth Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco) be able to compete with New Zealand's plantation produced clear radiata pine (Pinus radiata D. Don)? More generally, where have BC's comparative advantages lain in the past, and where are they likely to lie in the future? What will be the economic consequence of BC's transition to a second growth resource, particularly in light of increased environmental pressures to reduce the forest land base? To what degree will non-wood materials substitute for existing forest products produced in B C , and what will be the economic and environmental consequences of such substitutions? How does the emergence of engineered wood products affect demand substitution for B C timber resources? There are two potential situations which will have to be faced as B C makes the transition toward a forest industry that is wholly dependent on second-growth and subsequent forest crops: 1)  according to the most recent B C Ministry of Forests timber supply reviews,  Gaston Chapter One  Page 3  B C is going to witness a significant reduction in the volume of available timber over the next couple of decades; 2)  according to Constantino (1986) and Constantino and Haley (1988), without appropriate changes in B C forest policy, the quality^ of timber is going to be significantly lower.  If the forest industry in B C is to minimize these potentially negative impacts on the provincial economy, it will be necessary to examine marketing opportunities for the future, and translate these into appropriate land use plans, levels of silvicultural activities, and forest rotations. In other words, it is time for the forest sector to make the transition from a production-oriented to a market-oriented industry. This can only be accomplished by a detailed analysis of which B C wood products have historically contributed most to net revenues, and which are most likely to do so in the future.  This need will become  increasingly important as old-growth timber becomes scarcer and price increases lead to accelerated substitution.  1.2  Background A s can be seen in Figure 1.1, cash prices for lumber more than doubled in the first  couple of months of 1993 (following decades of limited price growth). Since then, prices have been extremely volatile, making any forecast of future price trends difficult. There is some debate as to the significance of this price spike. While some believe  lt is not easy to define quality in a general way. For example, one definition of quality might be the presence of attributes in wood that are related to appearance. In the case of softwood lumber, this would include such characteristics as clear grain, large dimensions, and narrow ring width. Another definition of quality might be structural strength. In other words, quality must be related to the intended purpose of the lumber. Constantino gets around defining quality in terms of specific wood characteristics by using a price index, where quality is related to the buyer's aggregate willingness to pay. 1  Gaston  Page 4  Chapter One  500  1989  Figure 1.1  1990  1991  1992  1993  1994  Random Lengths S - P - F Lumber Futures, Chicago Mercantile Exchange, Spot Contract*. Compiled from various issues of The Financial Post. * A s this chart always tracks the nearest delivery month, prices are analogous to the cash market.  that this occurrence was not all that unusual (see Sohngen and Haynes, 1994), others suggest that the market is displaying a structural change (see Sutton, 1994; and Michaelis, 1994). The latter opinion would suggest that prices will either stabilize at a new plateau at some point in the future or continue to demonstrate real price growth. Historically, the demand for construction lumber has been price inelastic . This can 2  be explained by one or more of the following: there have been few substitutes (this has not likely been the case); price has not been an issue (e.g., lumber has represented a small  A review of the literature which reports historical lumber elasticities, both own-price and cross-price, is offered in Chapter 2. 2  Gaston Chapter One  Page 5  portion of the cost of a home); or, there have historically been no inexpensive available substitutes relative to the price of lumber. However, demand for construction lumber may become price elastic (i.e. a structural change) if an increased price level leads to reduced wood consumption through the building of smaller homes and/or lumber substitution. The economic implications of the potential for such wood product substitutes translates into the central theme of this thesis. Substitutes for logs, lumber or further processed wood products can take many forms. The most obvious is substitution with the same basic product, but from a different location. In the literature review offered in Chapter 2, it will be seen that such cross-price elasticities of demand for lumber are significantly higher (even elastic) as compared to the own-price elasticities. In other words, while the quantity of local lumber demanded is not very price responsive (such as the demand for U S Midwest lumber given the price of Midwest lumber), the quantity demanded is responsive to the price of similar wood from a different area, such as imports from Canada. For example, a 1% decrease in the price of Canadian lumber may cause the quantity of US Midwest lumber demanded to decrease by more than 1%. Further, the review in Chapter 2 illustrates that cross-price elasticities may be high even for dissimilar types of wood, such as imports of hardwood from Indonesia to replace U S consumption of local softwood. Some studies also show that non-wood materials may substitute for logs and lumber. While the apparent willingness to substitute seems rather straight forward, it must be noted that no mention has been made of the specific characteristics of "similar" products. Due to an apparent lack of trade data broken down by grade, little can be found  Chapter One  Gaston  Page 6  in the literature to document this potentially important aspect of substitution. Figure 1.2, showing the prices of three grades of P N W Douglas-fir lumber over the past two decades, illustrates the danger of describing lumber (or logs) as a single homogeneous commodity. Note that these lumber prices are in real terms, not nominal. Over the time period indicated, clear grade Douglas-fir export prices rose roughly 3.5% per annum, the merchantable grade price trend was virtually flat, and the price of the structural grade fell. Given these distinct differences in price trends, it is obviously not rational to expect that construction grade lumber, for example, can fully substitute for clear grades. Prices can also vary tremendously within a grade. For example, prices for clear grade coastal B C lumber of certain species, when the timber from which it is cut is "hand picked" by Japanese buyers, have been reported to exceed $15,000 C D N per thousand board feet (Currie, 1994 ). 3  1.3  Scope of the Research Analysis of silvicultural regimes, forest practices, land-use and trade policies are all  negatively affected by the lack of information on wood product demand by some measure of disaggregation. However, before research can be carried out which addresses the policy and trade ramifications of using aggregated wood product data, significant background research is needed. This must begin with a quantification of the uniqueness of individual product types (logs, lumber and further processed products), species and source as distinct economic goods.  Personal communication, Valuation Branch, Timber Pricing Section, BC Ministry of Forests, Victoria.  3  Gaston  Page 7  Chapter One  2500  2000 .  £1500  ihooo  500 .  1972 '  ' 1975 '  ' 1978'  —*— Clear  Figure 1.2  ' 1981 '  ' 19'84'  ' 19lB7 '  ' 19'90'  ' 1993'  —•— Merchantable —*— Structural  PNW Douglas Fir Lumber Prices ($US per thousand board-feet, real, PPI adjusted, 1992=100)  Source:  Complied from Random Lengths, Various Yearbooks.  Grade Definitions: - Clear  Douglas-Fir, green, #2 Clear, 15% #3; 2% X 6 and wider; export price, f.o.b. dock, Or. and Wash, (prior to 1985 prices f.a.s.).  - Merch.  Douglas-Fir, Merch., #1, 15% #2; 6 X 12 and wider; export price, as above (prior to 1985 prices f.a.s., based on #1, 25% #2).  - Struct.  Douglas-Fir, green, #1 and better, random 10/20; 2 X 4 ; domestic price, f.o.b. mill.  Gaston Chapter One  Page 8  A s the central theme of this thesis is to quantify the degree of substitution of wood products, the sole focus will be on demand descriptors. Further, to keep the analysis manageable, the thesis focuses on a single market—Japan. The Japanese market was chosen as 1) it represents the largest importer of forest products today (Sedjo, 1994); 2) it has an interesting history of evolving from reliance on domestic production, then importation of whole logs, and most recently importation of lumber (which allows for quantification of the substitution between these alternative inputs); and 3) it has been a significant buyer of both construction grade and appearance grade wood products. The balance of this chapter is devoted to defining the problem to be investigated, leading to the research hypotheses.  1.3.1  The Research Problem Figure 1.2 (page 7) helps put the research problem into perspective. By comparing  the three grades of Douglas-fir lumber over the past two decades, the growing market premium for the clear grade (and, to a lesser extent, the merchantable grade) is obvious. Although international trade data do not offer such grade detail, they do offer species detail, from which grade measures can often be deduced. For example, the species mix spruce-pine-fir (S-P-F) lumber, which is exported primarily from North America, is known as a construction or structural grade commodity. North American lumber exports to Japan of such species as yellow cedar (Chamaecyparis nootkatensis (D. Don) Spach) and Sitka spruce (Picea sitchensis (Bong.) Carr.), on the other hand, are primarily of clear and merchantable grades. Further, the source of the wood also offers an association with  Page 9  Gaston Chapter One  grade. New Zealand log and lumber exports, for example, have historically been known to provide sub-structural grades, which have been used in Japan primarily as packaging materials. The research problem is best addressed through a number of questions. For instance, how do the own-price elasticities of demand differ in Japan by product type, source and by species? How do the cross-price elasticities for these wood products differ, both with other wood (type, species and source) and non-wood substitutes? Will scarcity in North American S - P - F lumber lead to real price growth as evidenced in higher grade lumber, or will price rises be met with reduced demand through substitution—both wood and non-wood (Perez-Garcia, 1993; Prins, 1993a and 1993b).  Will there be less  substitution in the future in species and/or source typical of clear grades as compared to structural? The second ramification of the price trend distinctions shown in the figure is that coastal B C relies on old-growth timber stands for the vast majority of its present lumber production. Timber yielding clear grades is exploited in such stands. Given existing silvicultural efforts and harvest rotations, the supply of clear timber will be significantly reduced as old-growth availability declines. This leads to the question of whether it is possible to generate this high-grade material economically from second growth stands (although this will largely be an implication for further study).  Finally, related to the  substitution questions above, it must be asked to what extent clear lumber from second growth can compete with clear lumber from old-growth timber (again, this question is posed as a motivation for the present research; the answer can only come from research which  Gaston  Chapter One  Page 10  extends beyond the scope of this study).  1.3.2  Objectives The research problem described in the previous section can be translated into the  following three objectives:  1.3.3  1.  To determine own-price and cross-price demand elasticities in Japan for logs, lumber and other wood products by region and species. Cross-price demand elasticities include substitution with Japanese domestic logs, and substitution with non-wood products.  2.  To qualitatively extend objective one to explore Japan's demand for broad grade categories (construction versus appearance).  3.  To explore the implications of the above for B C forest industry strategy and public forest policy.  Hypotheses The research hypotheses to be tested are as follows: 1.  B C wood species, in the form of logs, lumber or further processed products, behave as distinct economic goods, as measured by own-price and crossprice elasticities of demand, in the Japanese market.  2.  The market share of B C wood species in Japan, in the form of logs, lumber or further processed products, is dependent on the individual prices relative to other species and wood products in Canada and around the world.  3.  Japan's wood product import mix is affected by Japan's domestic log supply and non-wood alternatives.  4.  Structural changes in international markets for logs, lumber and panels have affected price levels and trends for these products.  Gaston Chapter One  1.4  Page 11  Organization of T h e s i s The objectives/hypotheses of the previous sections are addressed and presented  in this thesis as follows. Chapter 2 offers a literature review of North American studies which have estimated log or lumber elasticities of demand, as well as a discussion of the implications of such studies for the methodological approach to be used in the present study.  In Chapter 3 the Japanese wood products market is described.  Chapter 4  describes the theoretical foundation, data and the empirical model in detail, while Chapter 5 presents the results of this quantitative analysis. Finally, Chapter 6 offers a discussion of the results, followed by a summary, including limitations and implications for further research in Chapter 7.  Gaston  Chapter Two  Page 12  Chapter 2 Literature Review on Log and Lumber Substitutions and Implications for Further Research  This chapter reviews the literature on log and lumber demand elasticity estimates, including a cross-section of the methodological techniques used. The chapter concludes with methodological implications for the present study.  2.1  The Econometric Estimates of the Price Elasticity of Demand There are a number of potential responses to a rise in the price of domestic lumber.  These responses can be placed in one or more of the following categories: 1) increased efficiency of wood use and/or reduced consumption of finished products; 2) substitution by wood from another location, different form (for example, lumber for logs) or different species; and 3) substitution of non-wood inputs for wood inputs. In this section, these possibilities are examined separately in Sections 2.1.1 through 2.1.3, respectively.  2.1.1  Estimates of Wood for Wood Substitutes There has been a considerable amount of research which investigates demand  elasticities for timber products, primarily for U S lumber. Tables 2.1 through 2.3 highlight some of this research. The summary information is largely adapted from the review by Phelps (1993). The most obvious consistency found in Table 2.1, which presents own-price demand elasticity studies, is that softwood lumber demand is shown to be inelastic, with  Gaston  Page 13  Chapter Two  TABLE 2.1 Elasticity  Own-price Elasticity of Demand for Softwood Lumber in N.A. Time Frame  Author  Comments  -0.173  1947-1974  McKillopefa/. (1980)  US softwood lumber wholesale price index  -0.35  1950-1974  Waggener et al. (1978)  US softwood lumber price  -0.38  07/79-12/84  Gellner etal. (1991)  US softwood lumber price  -0.075  1947-1974  Adams (1977)  U S softwood lumber price index, 1 year lag  -0.285 -0.162 -0.130 -0.111  1950 (Point) 1960 ( " ) 1970( " ) 1980( " )  Spelter (1985)  US softwood lumber price  -0.88 -0.39  1950-1954 1970-1974  Spelter (1985)  US softwood lumber price  -0.91  01/68-12/77  Rockel and Buongiorno (1982)  US Douglas-fir wholesale price index  -0.88  1947-1967  Robinson (1974)  Douglas-fir  -0.667 -0.149  01/77-12/87  Lewandrowski (1992)  <- Southern pine *- Douglas-fir  -0.55 -1.15  1950-1987  Adams etal. (1992)  <- Residential construction <- Non-residential construction  Canada .0,.  i  -0.29  01/71-02/82  Jacques etal. (1982)  Domestic softwood lumber purchases  -0.023  1970-1982  Sharma (1986)  Softwood lumber for residential construction  -0.05  01/71-02/92  Prins(1993)  Total shipments of softwood lumber less exports  a range of-0.023 to -1.15, and an average of roughly -0.4. In other words, these studies suggest that a 10% increase in the domestic price of lumber will decrease the local quantity  Gaston  Page 14  Chapter Two  demanded by an average of 4%. Not surprisingly, demand elasticities which are smaller in value are from studies which estimate short-run demand responses. In the very short run substitutes do not, for all practical purposes, exist. Measurements of elasticities must also consider the specific years being investigated. Spelter's (1985) results illustrate this by showing that elasticities (in this case share elasticities) have fallen over time.  This trend, according to Spelter, may be  attributable to improved technology and utilization, both of which have helped to alleviate scarcity. One must also note that elasticity estimates are affected by the price range over which they are measured. For example, elasticity estimates derived from data where prices tended to be low will likely be significantly different from a comparable study over a different time period where prices tended to be high. The range of elasticity estimates shown can also be partially explained by what is being measured.  For example, Adams, et a/.'s (1992) results show that the demand for  lumber used in residential construction is less elastic than for non-residential construction, supporting the point that home buyers are not greatly influenced by the price of an input which makes up a relatively small portion of the total purchase price, as well as the fact that in non-residential construction more substitutes in the form of non-wood materials are available and acceptable. Lewandrowski's (1992) results show that elasticities can vary by species, here suggesting that southern pines (e.g., Pinus taeda) have more substitutes (i.e., the quantity demanded is more price sensitive) than Douglas-fir.  This point is  paramount to the main theme of this thesis, and will be further explored throughout much of this chapter.  Gaston  Page 15  Chapter Two  T A B L E 2.2 Elasticity  Cross-price Elasticities of Demand for Similar Lumber in Different Regions Time Frame  Author  Comments  1.283  1947-1974  Adams (1977)  0.81  01/71-02/82  Jacques, et al. (1982) I Demand for Canadian shipments; U S I lumber price index.  1.48  01/74-01/86  Buongiorno et al. (1988)  j Demand for imports from Canada; prices j of softwood lumber in the U S .  0.56  1950-1982  Singh and Nautiyal (1986)  | Demand for Canadian lumber; U S price | index for all lumber.  -0.80  1963-85  Flora etal. (1991)  i <-  United States  -1.95 -3.088  j <-  1965-1985  Chen et al. (1988)  2.27 4.39 Sawnwood  j Imports from Canada; ratio of U S to I Canadian import prices.  I <j <-  1975-1985  Constantino (1988)  Offshore demand facing the U S in 1987; performance grade, Off shore demand facing the U S in 1987; construction grade. Demand for Canadian softwood lumber; import price from B C . Price of U S softwood lumber.  i World imports of hardwood from | Indonesia; importing country's price of | hardwood.  12.30 Plywood  Note:  When interpreting the sign of the elasticity values, it must be noted which price is being considered. With C h e n , ef a/.'s (1988) results, for example, a . 10% decrease in the B C price of lumber will increase U S import demand by over 30%. Conversely, a 10% decrease in the U S price of lumber will decrease U S import demand from Canada by over 22%.  Table 2.2 shows some estimates of cross-price elasticities of demand for softwood lumber in different regions, in this case primarily the U S demand for imports from Canada. The most obvious point is that the demand response is now elastic (greater than 1), or at least more elastic than for own-price. This clearly shows, as would be expected, a willingness to substitute for a similar commodity from another geographic area. One might expect  Gaston  Chapter Two  T A B L E 2.3  Page 16  Cross-price Elasticity of Demand for Different Lumber  Elasticity  Time Frame  \ Author  | Comments  1975-1985  | Constantino (1988)  j World imports of hardwood from j Indonesia relative to importing j country's price of softwood.  1.45  1971-1991  j Brooks (1993)  | US imports of tropical lumber relative I to US price of softwood lumber.  1.06  1970-1989  I Youn and Yum I (1992)  I Korean imports of hardwood logs ! relative to the import price of softwood I logs.  1.30 Sawnwood 0.75 Plywood  these elasticity values to be higher than indicated. That they are not suggests that while the imports are good substitutes for domestic production, they are far from being perfect substitutes.  This may be partially due to the fact that aggregate imports in these studies  are not equivalent to aggregate domestic production, neither by species nor other characteristics. Finally, Table 2.3 shows some estimates of cross-price elasticities of demand for different types of lumber. Although the estimates are mostly greater than unit, the demand is generally less elastic than for similar wood products in different regions. This suggests that consumers are less willing to substitute hardwoods for softwoods than Canadian S-P-F for U S southern pine.  2.1.2 Estimates of N o n - W o o d for W o o d Substitutes Another possible reaction to higher lumber prices is, of course, substitution for wood products with non-wood commodities.  For construction lumber, this includes steel,  Gaston  Page 17  Chapter Two  concrete, bricks, plastics, etc. Recent price increases in lumber have already initiated an extensive program by the American Iron and Steel Institute to promote the replacement of wood structural and non-structural members in construction with steel members (Haws, 1994). Surveys undertaken in 1993 indicate that 4 5 % of builders in California would consider switching to steel due to the high and unstable price of wood products. There are a number of studies which have examined the cross-price demand elasticities between wood and non-wood materials.  A better understanding of the  substitution impact is achieved when the extent to which non-wood materials substitute for wood is examined, and the extent to which wood substitutes for non-wood materials. Table 2.4 summarizes the results of McKillop, et al. (1980) for the U S . First, it should be noted that all of the own-price elasticities (the diagonal from topleft to bottom-right) have the expected negative signs and are all less than 1, indicating inelastic demands. Second, the table indicates that the price of lumber influences the quantity demanded of non-wood materials. By contrast, however, the price of substitutes  T A B L E 2.4  Own-price and Cross-price Demand Elasticities for Construction Materials Q Lumber  P Lumber  -0.17  P Plywood  0.14  P Steel P Aluminum  Q Plywood  Q Aluminum  Q Concrete  0.79  0.24  -0.54  -0.4  0.54  0.37  -0.93  0.74  0.02  0.47  P Concrete Source: McKillop, etal. (1980)  -0.67  Q Steel  -0.83 -0.51  Gaston  Page 18  Chapter Two  T A B L E 2.5  Own-price and Cross-price Demand Elasticities for Construction Materials Q Lumber  Q Plywood  Q Non-Wood  P Lumber  -0.91  0.05  0.12  P Plywood  0.09  -0.95  0.12  P Non-Wood  0.09  0.05  -0.88  |  Source: Rockel and Buongiorno (1982)  does not influence the quantity demanded of lumber to the same degree. For example, a 10% increase in the price of steel will increase the demand for lumber by 3.7%. Conversely, a 10% increase in the price of lumber will increase the demand for steel by 7.9%. Rockel and Buongiorno (1982) specifically examined the demand for wood products for residential construction (as opposed to total U S demand) (Table 2.5). The non-wood substitutes included in the study were structural steel, cement, bricks, plumbing and heating fixtures, and selected fabricated metal products. The extremely low cross-price elasticities indicated in the table are the result of aggregating all these inputs into a single basket of goods. In Table 2.6, the result of time on elasticities (primarily technological change) is demonstrated by Spelter for the U S . A s can be seen, both the own-price and the crossprice elasticities fell (with the exception of concrete) from the 1950s to the 1980s. Finally, Prins (1993) examined wood/non-wood substitution in Canada (Table 2.7). Note that while the results suggest that a 100% increase in the price of lumber will cause  Gaston  Page 19  Chapter Two  T A B L E 2.6  Own-price and Cross-price Demand Elasticities for U S Softwood Lumber 1950  1960  1970  1980  P Lumber  -0.285  -0.162  -0.13  -0.111  P Plywood  0.109  0.04  0.009  0.004  P Steel  0.026  0.017  0.012  0.005  P Concrete  0.006  0.006  0.006  0.006  Source: Spelter (1985)  T A B L E 2.7  Own-price and Cross-price Demand Elasticities for Selected Canadian Construction Materials Q Lumber  Q Brick  Q Cement  Q Steel  P Lumber  -0.05  0.51  0.15  0.32  P Brick  0.49  -0.3  0.73  P Cement  0.78  0.71  -0.5  0.55  0.56  P Steel P Gypsum  -0.31  P Panels  0.09  -2.09  0.08  Source: Prins (1993)  only a 5% reduction in the demand for lumber, it also creates a 5 1 % increase in the demand for bricks, a 15% increase in cement and a 32% increase in steel.  This  demonstrates the dominance of wood use in construction: 5% of all lumber used in construction is a significant volume of material relative to 32% of all steel used. Also, note that price increases in non-wood substitutes have a greater impact on the demand for wood than suggested by the previous studies.  Page 20  Gaston Chapter Two  2.2 Parametric Demand Elasticity Estimation Techniques From the outset, it should be emphasized that only a small number of the demand elasticity studies listed in the previous tables made any attempt to disaggregate the data beyond softwood logs or lumber. A s mentioned in the previous chapter, the likely reason for this is the lack of published disaggregated data. All of the studies reported to this point have involved econometric techniques in estimating the price elasticities of demand. While the estimation techniques employed were not unusual (normally ordinary least squares, two or three stage least squares, nonlinear least squares or generalized least squares), a few studies utilized an approach of interest to the present study. Flora and his colleagues at the U S D A Forest Service, Pacific Northwest Research Station in Seattle have conducted North American wood product demand studies which have made direct reference to quality or grade (Flora and Lane, 1994; Flora, et al., 1993; Flora, 1993; Flora, 1992; Flora, etal., 1991-a; Flora, etal., 1991-b; Flora, 1991; Flora, et al., 1990; Flora and McGinnis, 1989; Flora, 1986). In one of the studies (Flora, et al., 1991-b), the researchers developed export supply functions for Pacific Rim log suppliers (US, Canada, Chile, New Zealand and the Soviet Union), and import demand functions for Pacific Rim buyers (Japan, Korea, China and Taiwan).  These supply and demand  functions were then summed across quantities to yield aggregate demand and supply functions. To estimate trade flows pertinent to an individual region, the demand or supply facing that region is developed by netting out all of the other regions' demand and supply functions. The individual equations used by Flora tend to be very simple, with quantity  Gaston Chapter Two  Page 21  demanded typically a function of price, G D P and housing starts, and quantity supplied a function of plantation area, timber harvest, and possibly a sawmilling cost index. Flora's methodology suffers from three specific limitations.  First, where price  projections are made, projections of all variables except price and volume are done outside of the model (by making assumptions relative to a present-day "base case"). This means either heavy reliance on other studies, use of other modelling techniques, or significant personal judgement. The second limitation is Flora's method of dealing with disaggregated trade data. Notes Flora, etal. (1991-b, page 6): Because log-trade data are rarely reported by grade, quality class, or economic category, it was necessary to judge the proportions and relative values. ...Future volume-share shifts among grades were assumed... This, unfortunately, offers little guidance for dealing with such data over a wide range of applications. Finally, Flora's models are limited to a single grade at a time. This does not allow for the measurement of cross-price elasticities of demand across grades or species. However, the own-price elasticities which resulted do suggest that elasticities of demand are negatively correlated with quality (as quality increases, demand becomes more inelastic). Flora (1991-b) categorized logs into one of four grades: Select  logs whose value derives from "appearance" grade lumber;  Performance  Coast and Cascade Grade No. 2 sawlogs; second- and oldgrowth logs with scaling diameters between 12 and 24 inches;  Gaston  Chapter Two  Page 22  Construction  Coast Grade No. 3 sawlogs; second-growth logs with scaling diameters between 6 and 12 inches;  Utility  submerchantable in the export market.  Flora's conclusions suggest that the performance grade will have rising real price growth through the turn of the century, and that the construction grade will see declines due to international competition. A s shown in Table 2.2, the authors pegged the price elasticity of demand facing the U S in 1987 for the performance grade at -0.80 as compared to -1.95 for construction (the author did not analyse the two extremes in grades, reasoning that selects will always be scarce and that the utility grade is unimportant for the export market). Haynes and Fight (1992), also working with the U S D A Forest Service in the P N W , conducted a study on projecting prices of selected grades of Douglas-fir, Coast Hem-fir, Inland Hem-fir and ponderosa pine lumber. Working with historical volumes and prices from representative invoices submitted to the Western Wood Products Association for the P N W region , the authors estimated the relationships between the prices of the selected 4  lumber grades and the price of the dominant lumber grade for each species in the general form: S  jt  where:  = b  + b  y  S S W  2j  jt  d t  jt  = = =  S  dt  + b  Zj  W  (2.1)  fi  regional lumber price for the j species and grade in year t; price of the dominant species and grade in year t, and; the proportion of total lumber production in year t that comes from j species and grade. t h  th  "Unlike the other methodologies reported in this section, Haynes and Fight (1992) are using domestic market data (as opposed to export data). Given the noted premium for export markets (see Flora, et al., 1993), this will underestimate the price premium for higher grades.  Gaston  Page 23  Chapter Two  With the values of the by's estimated, the authors predicted the price for each grade by independently projecting the price of the dominant grade (S ) and grade production dt  proportions.  Their results supported the notion that increasing scarcity of high-grade  material will result in higher prices.  However, given the confines of their analysis,  projections out to the year 2040 showed that the relative price spread for each grade remains virtually unchanged. This is in contrast to historical changes in price spreads over grades. A s noted by Flora, for example, in Japan in 1978, Alaska Prime Spruce cants were worth 3 times as much as U S #3 hemlock logs; by 1992, the multiple increased to 20. Sedjo, et al. (1994) offer a study which specifically investigates cross-price elasticities of wood inputs. Noting the effect of the price of U S logs on the log import behaviour of Japan, the authors reason that imports from any region will be a function of that region's timber price to Japan, the price of Japanese domestic timber, the level of construction activity in Japan, the price of U S timber to Japan, and the price of timber to Japan from any other source that may substitute for the region in question's timber. A multiple regression analysis was used, with the quantity of timber demanded from region "A" as the dependent variable, and each of the above factors taken as independent variables, over the period 1970-1991: QA = b., + b PA + b PUS + b PJ + b HS + £ c,P . 2  where:  QA PA PUS PJ HS Pi  z  = = = = = =  A  5  (  <-) 2 2  quantity imported from region "A"; price of region A's timber in Japan; price of U S logs to Japan; Japanese domestic price of logs; number of Japanese new wooden housing starts; prices of timber to Japan from regions other that "A" or the U S .  Page 24  Chapter Two  Gaston  Interestingly, the results showed no significance for the domestic price in Japan ("sugi" conifer logs), suggesting either that import decisions were made independently of domestic production (possibly suggesting different end uses), or that perhaps there was too little variation in domestic production to estimate the effect. The price parameter on imports from Canada (Douglas-fir lumber) was also found to be insignificant, with the authors reasoning a high degree of multicollinearity of Canadian lumber exports (roundwood equivalent) to Japan and U S Douglas-fir logs exported to Japan . The other 5  regions investigated, all of which were found significant, included the Philippines, Indonesia, New Guinea, and Malaysia, collectively called "tropical" (lauan veneer tropical logs); Russia (larch logs); and Chile/New Zealand, collectively called "radiata". The elasticities of Japanese imports from these three regions/wood type with respect to the independent variables were then calculated. Due to the objective of the study, the authors focused on the cross-price elasticity of Japanese imports with respect to the price of U S logs. This cross-price elasticity of wood quantity from region "A" (QA) with respect to the U S log price (PUS) is the percentage change in Japanese imports from A for each unit percentage change in the U S price, given by: E,QA.PUS = b  where:  3  (PUS/QA)  b (PUS/QA) 3  = =  (2.3)  the regression coefficient of P U S in the previous equation; value using the mean over the study period.  As will become apparent in the following review of Armington (1969) and applications by Chou and Buongiorno (1983) and Hseu and Buongiorno (1992), such multicollinearity is a major limitation of trade studies which have many price regressors of "similar" products in the same equation. 5  Gaston  Page 25  Chapter Two  Using ordinary least squares, the recovered cross-price elasticities for Japanese imports with respect to the U S log price were 0.58, 5.0, and 0.84 for tropical, radiata and Russia, respectively. Unfortunately, none of the methodologies reviewed to this point are totally adequate for addressing the main objective of this thesis, that is to estimate the substitution possibilities among a significant number of disaggregated wood inputs in the Japanese market. The problem which needs to be addressed is how to estimate own- and crossprice elasticities for a number of potentially unique, yet similar, price series. Armington (1969) offers a potential solution. Recognizing the heterogenous nature of products, even when of a similar "kind", Armington relaxes the assumption that a particular "good" produced in a particular country is a perfect substitute for the "same" good produced in another country (relaxing the assumption that the elasticity of substitution between, say, softwood lumber from Canada and softwood lumber from New Zealand is infinite).  This is accomplished by assuming that an importer performs a two-stage  optimization process.  In stage one, the importer decides the total amount of the  commodity "kind" to import from all sources (say, softwood lumber, or even all wood products in aggregate). In the second stage, the importer determines the optimal levels of "good" imports (say, softwood lumber from a number of different sources). Armington makes four assumptions, "systematically simplifying the product demand functions to the point where they are relevant to the practical purposes of estimation and forecasting" (Armington, 1969; page 160). The first is that importer preferences are homogeneously separable.  This assumption is necessary to incorporate two-stage  Gaston  Page 26  Chapter Two  optimization. It is next assumed that market shares depend only on relative prices of the products in the market, not on the size of the market itself.  In the absence of these  assumptions, a country would have mn demand functions in the general form:  ij  X  where:  ••• 'P-inr ^21 ' ^ 2 2 ' ^ 2 m ' ••• ' ^n1'^ n2'  Xjff),P^ P^  =  >  v  m n Xjj D P.  = = = = =  •••^nrr)  number of supplying countries (specific product); number of goods (groups of products); specific product demand; income; specific product price.  With Armington's first two assumptions, this is reduced to m+n demand functions:  ( ij  X  ~  p.. \  p.. p..  (2.5)  ij  X  im  where X, = X ( D , P , P /  and:  1  2  X; is any good, and  P„)  (2-6)  is any product.  From Armington's first two assumptions, Equation 2.5 requires that a linear, homogeneous quantity index function, cp, be utilized for each market, such that X; = cp (X , M  X  i 2  ,X  i m  ) . The reason for this is that if imports of products of the same kind from different  countries are considered to be imperfect substitutes, the arithmetic sum of various imports would not be an appropriate index of total imports. Recognizing that equation 2.5 is still too complicated to be of practical use when more than a few countries are identified, Armington proposes two further simplifying  Gaston  Chapter Two  Page 27  assumptions. These are that the elasticities of substitution are constant for each market, and that the elasticity of substitution between any two products competing in a market is the same as that between any other pair of products competing in the same market. These assumptions are equivalent to specifying that the cp's above are constant elasticity of substitution (CES) functions, having the general form : 6  =- <P, (*„.  =  x,2 W  +  xj  '  (2  +  -  +  7)  b X ] pi  im  pi  im  With these added assumptions, it can be shown that equations 2.5 have the general form:  ( P  - bp  XIJ  where:  X  a, b,  (2.8)  '  I  = =  the constant elasticity of substitution in the i market; a constant. th  Armington notes that equation 2.8 can also be written to express the market share as the dependent variable:  pyIJ  (  —- = b X. 9  (2.9)  IJ  The advantages of Armington's assumptions are obvious. A s stated by the author, "these assumptions yield a specific form for the relation between demand for a product, the size of the corresponding market, and relative prices; and the only price parameter in this  B  Note that this function is of the needed linear and homogenous form.  Gaston  Chapter Two  Page 28  function [equation 2.8 or 2.9] is the (single) elasticity of substitution in that market" (Armington, 1969; page 161).  The problem of multicollinear data has been greatly  reduced. The real advantages to Armington's assumptions come- in the analysis of price changes (determining the elasticities of demand).  Total differentiation of the market  demand equation (2.6) and the product demand equation (2.8) yields the relationship between changes in  and changes in the explanatory variables. Beginning with the  latter : 7  5X, ClX, =  "  5X,  1 dX: +  5X,.  5X, dP:: +  '  bP..  BP,.  IJ  dP: (2.10)  Dividing both sides by X,:  d X , _ 5X^X . dX, (  X,  QXft  dP.  ;  X  t]  o—'I  i  ' P  dP,  + a.—'-  'P.,  i}  (2.11)  Given that the partial elasticity of X , with respect to X, equals unity:  dX  dX  ( dP  x.  X,  p..  p  IJ  1  a  :  tj  \  dPj (2.12) J  The first term reflects the growth in the X, market, while the second term reflects the  7  This derivation follows Armington (1969), Appendix II.  Gaston  Chapter Two  Page 29  percent change in Xy's share of the market. The next step is to differentiate equation 2.6 in terms of dXJX: dX,,  _ dD  X,  ' D  where:  e, Hi n  i/k  A  (2.13)  dP„ w.  dP  j  k  ' P,  = = =  the income elasticity of demand for the direct price elasticity of demand for X the cross elasticity of demand for X, with respect to P . i;  k  Substituting into equation 2.12: dX  :  X„  '  -  e' D  dD  dP, n,-  ( dP  ::  \  dP,)  (2.14)  IJ  It should be noted that there is no P term in equation 2.12, without which it is not possible ik  to derive an expression for the cross price elasticity of Xy. Armington shows that: dP;  dP.ik  , where  S  ik  =  (2.15) iX,  where:  S  i k  =  the market share of X in value terms. i k  This substitution for dP; IP, is possible due to Armington's assumption of a linear and homogeneous indexing function, cp. To sum, it is shown that the effects on Xy of changes in prices of products competing in the i market depend not only on 0^ and f\ but also on market shares. By substituting th  t  2.15 into 2.14, Armington arrives at:  Gaston  Page 30  Chapter Two  +  £ [S„a . - s,/),]  dP  ( ) 216  k  (  with the first bracketed coefficient being the own-price elasticity of demand for X , and the y  second bracketed coefficient being the cross-price elasticity of demand for X with the y  respect to any other product price in the i market. Both of these bracketed expressions th  contain two terms; the first notes that a price change alters relative prices, creating a substitution effect. The second notes that a price change also alters the price level in the market, creating a market "expansion" effect. The Armington model has been applied in a number of studies reported in the literature; its popularity is undoubtably supported by its simplicity and its ability to deal with multicollinearity problems. Agricultural applications are the most prevalent, a cross-section of which can be found in Babula (1987), Grennes, et al. (1977), Penson, et al. (1988), Webb, etal. (1989), and Adelman, etal. (1989). Extensive use of the Armington model is made in the equilibrium modelling of the International Monetary Fund (see Grennes, etal., 1977). Although all of these studies stress the advantages of the model, Armington is not without his critics. Alston, et al. (1990), for example, find that the restrictive assumptions used by the Armington model lead to underestimated price elasticities. While applications of the Armington model in forest products' trade studies are scarce, Chou and Buongiorno (1983) use this approach in their estimation of the US demand for hardwood plywood imports by country of origin (Taiwan, Korea, Japan, Philippines and "rest of world").  Following Armington's stepwise approach to import  Gaston  Page 31  Chapter Two  demand specification, the authors begin by estimating the constant elasticity of substitution : 8  ( Q / Q ; = ( D / D / {PM/PM)°  where:  Qj.Qj = Pi, Pj =  ( 2  1 7 )  quantity imported from country I and j, respectively (I*]); CIF price of Qj and Qj, respectively.  The values of the b, / bj and a coefficients are estimated from four simultaneous equations, setting I = 1 (Taiwan) and j = 2,  5 (Korea, Japan, Philippines and "rest of world"),  restricting the o to be constant and linearizing equation 2.16 by converting it to a logarithmic form: ln(CyQ.) = a  where:  u,  \x\(pjb) + - a \n{PMJPM) + u  ( Z 1 8 ) }  = a random residual.  Their estimate for the constant elasticity of substitution between hardwood plywood imports of different origin, a, is -1.74 and highly significant. The fact that this value is significantly different from zero shows that the two different sources of plywood are not considered complements; the fact that the value is not very high suggests that plywood imports from different sources are also far from being perfect substitutes (requiring that o = °°). The next step followed by Chou and Buongiorno is to estimate demand for total hardwood plywood imports in the US. Expanding an earlier paper (Chou and Buongiorno, 1982), they utilize a demand function derived from residential construction:  As Chou and Buongiorno use monthly data, they utilize lags in their estimation. For the purposes of describing their overall approach, however, the lags are omitted here. 8  Gaston  Page 32  Chapter Two  Q = a where:  Q PM PDA Y  (2.19)  -P  PM PDA = = = =  total quantity of plywood imported by the U S ; the price of total imports; U S producer price index for all commodities; U S housing starts.  However, to link this total demand to the elasticity of substitution from above, the researchers estimate equation 2.19 by replacing Q and P M by their equivalent C E S quantity and price indices, defined as:  Q =  Ib.Q  0  o-1  and,  (2.20)  1 PM = ( I b? P M , where:  I  1 0  y-°  = sum i from 1 to 5.  The b; weights needed to calculate these indices can be determined from the estimated parameters on b, / b, in equation 2.18, adding that the sum of the b;'s must sum to one; this translates into five equations and five unknowns, allowing for the recovery of each b,. The resulting value of the (3 parameter, being the price elasticity of demand for U S hardwood plywood imports in aggregate (equation 2.19) is -2.20 and significant. This compares to a value of -1.98 when the researchers utilize arithmetic indices (Chou and Buongiorno, 1982) as opposed to the C E S indices. With value of the aggregate elasticity of demand, 3, the constant elasticity of substitution, a, and knowledge of the average market share by value for each of the noted supplying regions, the researchers are now  Page 33  Gaston Chapter Two  Table 2.8  Elasticities of Demand of U S Hardwood Plywood Imports by Country of Origin Market Share  Korea  0.45  s X  Taiwan  0.24  s X  Japan  0.14  s X  Philippines  0.07  s X  Rest of World  0.10  s X  Korea  Taiwan  Japan  Philippines  Rest of World  -0.96 -0.99  0.42 -0.53  0.24 -0.31  0.12 -0.15  0.17 -0.22  0.78 -0.99  -1.32 -0.53  0.24 -0.31  0.12 -0.15  0.17 -0.22  0.78 -0.99  0.42 -0.53  -1.50 -0.31  0.12 -0.15  0.17 -0.22  0.78 -0.99  0.42 -0.53  0.24 -0.31  -1.62 -0.15  0.17 -0.22  0.78 -0.99  0.42 -0.53  0.24 -0.31  0.12 -0.15  -1.57 -0.22  s=substitution effect; x=market expansion effect Source: Chou and Buongiorno (1983)  in the position to calculate individual own- and cross-price elasticities as outlined by Armington. The results are shown in Table 2.8, breaking out the substitution and market expansion effects. The net own-price elasticities reported in Table 2.8 (the highlighted diagonal values) and the net cross price elasticities are determined by adding the substitution and market expansion values in any one cell. Beginning with the own-price, note that the net effects have their expected minus sign, and that the substitution effect varies somewhat, with Korean plywood standing out. Also note that the market expansion effect reduces with the supplying country's market share, which is what one would expect. Finally, it can be noted that the net own-price elasticities do not vary much by country of origin, ranging from -1.77  Gaston Chapter Two  Page 34  to -1.95. Keeping in mind that a constant elasticity of substitution is "forced", this outcome should not come as a surprise. Turning to the cross-price elasticities of demand reported in the table, note that while the substitution effects show the expected positive signs, the net effect is negative. A s stated by the authors, "a rise in the price from country j causes a reduction in the total United States hardwood plywood imports which more than offsets the gains of country / arising from the substitution of country I's plywood for that of country /'. It can be seen that the net effect depends on the relative values of the elasticity of substitution (o) and the aggregated import price elasticity of demand (B); when the latter is larger in value as compared to the former, cross-price elasticities will be negative. It can be noted that the net cross-price elasticities are all highly inelastic, with the highest value being -0.21, the import quantity response in the U S for non-Korean hardwood plywood against a price increase in Korean product. In all cases, the market expansion effect counteracts any substitution effect. Recognizing the potentially restrictive nature of Armington's assumptions, a study by Hseu and Buongiorno (1993) attempts to use a more "realistic" approach by allowing the elasticity of substitution to vary. Further, rather than disaggregating the second stage demand by country of origin, the researchers investigate the U S demand for Canadian softwood lumber by species. Starting with a production function for the construction of houses in the U S , the researchers begin by specifying the derived demand for total lumber imported from Canada:  Page 35  Gaston Chapter Two  Q = where:  (2.21)  hP-*P P Y'» e  d  Q  a  aggregated quantity of Canadian softwood lumber imported; average price of Q; price of U S domestic softwood lumber; price of all other inputs; U S housing starts.  Q  P  Y  Next, the share of a particular species is assumed to depend on the price of that species relative to the price of all imports:  (2.22)  where:  Pi  =  is the quantity of softwood lumber of species / imported from Canada; the average price of species I; constants specific to species /.  The authors note: "equation [2.22] differs slightly from Armington's (1969) theory in that the elasticity of substitution O; varies by species". Equations 2.21 and 2.22 are estimated separately in log-log form, and own- and cross-price elasticities calculated from the total differential as in Armington (1969) and Chou and Buongiorno (1983). While the hypothesis that the elasticities of substitution by species are significantly different from one another is worthy of investigation, it is not at all clear how the researchers link such an hypothesis back to Armington's theory.  In fact, there are a  number of major problems with this methodology. First, getting separate estimates for the single (3 and the numerous o-s does not constitute a two-stage optimization; the Q in equation 2.21 must somehow be linked to the  Gaston  Page 36  Chapter Two  Q in equation 2.22. With Armington (1969) and Chou and Buongiorno (1983) this is accomplished by using the C E S indices. Second, estimating equation 2.22, while shown to be possible under Armington's assumptions (see equation 2.9), becomes problematic without some way of "collapsing" all the Qj's into Q (as is done with Armington's tp function). A s is, Hseu and Buongiorno's equation 2.22 violates one of the assumptions of the classical linear regression model. The right-hand-side contains a component which is not independent of a left-hand-side variable: P, being the price of the aggregate imports of softwood lumber from Canada, is determined as the arithmetic mean, I P ; X Q; / Q. While the second problem could have been avoided by defining Q as the total aggregate quantity less Q and P the average u  price of this new Q, other problems with the application of Armington's theory remain. Most importantly, without the linear, homogeneous aggregation function, the substitution for dP; / P, in Armington's equation 2.15 cannot be made. These problems notwithstanding, Hseu and Buongiorno do get interesting empirical results that are worth exploring.  A s shown in Table 2.9, own-price elasticities vary  considerably across species and cross-price elasticities exceed unity in some cases. While the market expansion effect remains constant over species in both the calculations of ownand cross-price elasticities, note that by allowing the "share elasticity" to vary by species, the substitution effect changes. In calculating the share elasticities, the researchers employ a seemingly unrelated regression estimation (SURE).  "To test the validity of  [Armington's] simplification, [equation 2.22] was re-estimated with the restriction that... o, ... [was] the same across equations" (Hseu and Buongiorno, 1992, p. 595). This was  Page 37  Gaston Chapter Two  Table 2.9  Elasticities of Demand of U S Softwood Lumber imports from Canada by Species Market Share  Spruce Pine  0.57  Spruce | s  I  -1-63  I  X  !  - -  !  0...09 | s ! x !  Fir  0.11  1  1 3  3.85 - 1  1 3  | s | x  0.06  !  \ |  Pine  Fir  Hemlock  Red Cedar  Others  0.34 -0.18  0.42 -0.22  0.27 -0.14  0.23 -0.12  0.42 -0.22  -6.15 -0.18  0.74 -0.22  |  0.47 -0.14  0.41 -0.12  0.74 -0.22  0.01 -0.18  -0.09 -0.22  |  0.01 -0.14  0.01 -0.12  0.01 -0.22  Hemlock  0.07  | s I x |  2.27 -1.13  | |  0.36 -0.18  0.44 -0.22  -3.70 -0.14  0.24 -0.12  0.44 -0.22  Red Cedar  0.06  | s | 0.55 x ] -1.13  ; |  0.09 -0.18  0.11 -0.22  0.07 -0.14  -0.91 -0.12  0.11 -0.22  Other  0.11  j  I  0.08 -0.18  0.10 -0.22  0.06 -0.14  0.05 -0.12  -0.80 -0.22  s | x I  0.51 -1-13  !  s=substitution effect; x=market expansion effect Source: Hseu Buongiorno (1992)  rejected with a computed X statistic which was significantly higher than the critical value. 2  2.3  Implications for the Present Study While a detailed description of the data and modelling procedure to be utilized in this  thesis is reserved for Chapter 4, a few comments are offered here in order to provide a link between the existing research with the stated thesis objectives. Following an exhaustive search in B C , the U S and Japan (including trade associations, government agencies and the academic profession), it is clear that the highest level of detail for secondary data on wood products trade for the Pacific Rim is a  Page 38  Chapter Two  Gaston  breakdown by country, species (in some cases, only genus), and product (logs, chips, lumber, or further level of processing). This suggests that it is not possible to improve on many of the noted limitations of the studies reported in the previous section without gathering primary data . Investigation into this possibility also proved futile, both in terms 9  of the confidential nature of individual exporters' data and individual Japanese importers' data. Given the stated research objectives of this thesis, being primarily to estimate ownprice and cross-price elasticities of the Japanese demand for wood products across product type, species and country of origin, the only applicable methodology found is that suggested by Armington (1969). The strength of this methodology lies in the fact that it allows the researcher to estimate an unlimited number of own- and cross-price demand elasticities without sacrificing degrees of freedom or being burdened with multicollinearity over a large number of explanatory price variables. A s has already been suggested, however, this methodology is not without its problems. This is due to Armington's rather restrictive assumptions, and the difficulties involved in circumventing these assumptions. While the methodology employed by Sedjo, et al., also estimates cross-price demand elasticities between alternative wood inputs, such estimates are necessarily limited to a very few substitution possibilities. Finally, in both the Flora, et al., and the Haynes and Fight studies, the methodologies are more appropriate for price forecasting by grade than for estimating  As noted in the previous section, this is not true for the US domestic softwood log market (Haynes and Fight, 1992). However, there is no secondary quantity/price data in Canada broken down to even species (let alone grade), for either logs or lumber. 9  Gaston  Chapter Two  P a g e 39  wood input substitution. The methodologies employed in these studies are, therefore, beyond the scope of the present study, and will be returned to in the final chapter when discussing recommendations for further research. In the following chapter, the complexity of the Japanese demand for wood products is discussed. This is followed by the theoretical foundations and implications for empirically analysing this demand in Chapter 4. An Armington-type, C E S model will be developed, using a similar approach to that suggested by Chou and Buongirno (1983). Given the problems associated with the further application offered by Hseu and Buongirno (1992), this approach is not incorporated into the methodology employed in the present research. However, the practical intent of Hseu and Buongirno's research will be discussed in later chapters.  Page 40  Gaston Chapter Three  Chapter 3 The Market for Wood Products in Japan  This chapter investigates the market for wood products in Japan, including its evolution over the past several decades. This investigation starts with a synopsis of Japan's timber resources and log production, followed by a description of the demand for this production along with the demand for imports. Finally, a discussion of Japan's lumber and panel production, comparing domestic with imported timber inputs as well as competition with lumber and panel imports, is presented. This is an important lead-in to Chapter 4, which describes the economic model developed in this thesis to represent the derived Japanese demand for logs, lumber and other wood products.  3.1  The Japanese Domestic Timber Resource Although approximately 70 percent of Japan's 37 million hectares is covered by  forests, the population is high (approximately 125 million, Canadian Global Almanac, 1995), resulting in a per capita forest area which, at less than a fifth of a hectare, is less than half the world average (Japan Forestry Agency, 1991). Of the total forest land in Japan, there are 13.67 million hectares of natural forests (54%), 10.22 million hectares of man-made forests (40%), with the balance being either unstocked land or bamboo groves (1.37 million hectares, or 6%). The natural forest area consists of 75% deciduous species, 13% coniferous species and 12% mixed forests. The deciduous species include oak (Quercus mongolica and Quercus dentata), elm (Ulmus davidiana), ash (Fraxinus mandshuica), and beech (Fagus crenata). The coniferous  Page 41  Gaston Chapter Three  species include fir (Abies veitchii, Abies mariesii, and Abies sachalinensis), spruce {Picea hondoensis and Picea jezoensis), hemlock {Tsuga diversifolia), larch (Larix leptolepis), pine (Pinus densiflora, or red pine; Pinus pentapphylla, or white pine; Pinus Thunbergii, or black pine), Hinoki (Chamaecyparis obtusa), and Sugi (Cryptomeria japonica). In contrast, 98% of the man-made forest is coniferous. Ten percent of the area is larch, 11% is pine, 24% is Hinoki, 45%Sugi, 9% is spruce and fir, and 1% miscellaneous. The growing stock of the natural forests is roughly 1.5 billion cubic metres compared to roughly 1.4 billion cubic metres in the man-made forests (Japan Forestry Agency, 1991). The man-made forests are primarily a product of extensive planting which took place in the one to two decades following World War II. Annual growth of the plantations is roughly 76 million cubic metres, or a mean annual increment (MAI) of 7.5 cubic metres per hectare. Figure 3.1 shows the age distribution of these trees. In spite of this impressive growth rate on their man-made forests, it has been suggested that the quality of the timber from these stands is not high, particularly in terms of inadequate log diameters due to a noted lack of tree spacing (Iwai, 1986). This point will be returned to later in this chapter. Japan's annual harvest from 1990 to 1993 ranged from about 27 million cubic metres to 30 million cubic metres, with roughly 60% being harvested from the man-made forests. A s can be seen from Figure 3.2, which shows log production by species, this has dropped considerably since the 1960s. However, as made evident by Figure 3.1, the volume  harvested  can be expected to  increase significantly  over the  coming  Page 42  Gaston Chapter Three  6000  5000 II) 1 1 , II  » 4000  <u CO  o  CD X  I111  o 3000  iii  «  c  CD <0  if 2000  i  1000 1-15 Yrs  16-30 Yrs  31-45 Yrs  46-60 Yrs  60 + Yrs  0  Figure 3.1  Distribution of Man-made Forest by Age Class, 1986 (Japan Forestry Agency, 1991)  decades . 10  In terms of forest land ownership, as of 1986 58% was private, 3 1 % national and the remaining 11% was under prefecture and community control (Japan Forestry Agency, 1991). The private forests (14.68 million hectares) are owned by nearly 3 million entities (individuals, corporations and other organizations). Approximately 9 0 % is owned by individuals, with an average holding of only 2.6 hectares (60% of the private owners own less than 1 hectare of land). The national forest land base (7.89 million hectares) is under the jurisdiction of the Forestry Agency. The national forests tend to be located in the steep mountainous areas, unlike the private, prefecture and community forests, which tend to be  Given the reforestation effort in Japan post World War II, and given that forest rotations in Japan are typically 40 to 60 years, depending on species, site quality, etc. (Iwai, 1986), this would suggest a significant increase in the potential availability of domestic supplies early in the next century. However, these rotations will also depend on desired log quality. 10  Gaston Chapter Three  Page 43  120  100  80  60  40  20  1955 J  Cedar  Figure 3.2  1960  1965  ~J Cypress  1970 Pine  1975  1980 Other Soft H  1985  1990  Total Hard  Japanese Domestic Log Production by Species (Japan Forestry Agency Data, Provided by Dr. Y. Mori, Kyoto University, Japan) Note: The volume scale on Figures 3.2, 3.3, 3.4, and 3.7 are identical for ease of comparison.  located in more economically accessible areas (Japan Forestry Agency, 1991). Roughly 33% of the national forests are replanted by area (as of 1989), the vast majority of which is in softwood (Otsuka, 1992). Of the total annual log production in Japan in 1990, 20.5 million cubic metres came from private lands, 1.9 million cubic metres from the prefecture and community lands, and 8.6 million cubic metres from national forests. Figure 3.3 shows how this relationship has varied since 1960. Over 4 0 % by area of the private lands are replanted (as of 1989) (Otsuka, 1992).  Gaston  Page 44  Chapter Three  120 100 80 60  Figure 3.3  3.2  Japanese Domestic Log Production by Ownership (Japan Forest Agency Data, Provided by Dr. Y. Mori, Kyoto University, Japan)  The U s e of J a p a n e s e Domestic Timber Production Japan has a very long history of using wood, particularly for housing. The country's  familiar post and beam construction has created a demand for wood which has both structural strength and appearance qualities. The importance of this combination lies in the visually exposed vertical pillars joined to exposed horizontal beams and girders . 11  The domestic timber of choice for the pillars, due to their combined strength and high appearance characteristics are Hinoki (Japanese cypress) and Sugi (Japanese cedar, or Cryptomeria). For beams, Akamatsu (Japanese red pine) is often used for its ability to  lt is noted that this is only true for traditional post and beam construction. Personal communication with Bob Holm, Executive Director of the BC Wood Specialties Group, reveals that this is changing. Mr. Holm suggests that post and beam construction in Japan today is sometimes similar to North American platformframe construction in-so-far as the posts are hidden from sight with panelling. 11  Page 45  Gaston Chapter Three  handle large shearing stress. Foundations are often made from Hinoki due to natural rotresistant properties.  Sugi heartwood is often used for panelling due to its decorative  colour. Sugi is also used for ceiling boards due to its light weight. Overall, it is the Hinoki that has been the species most valued by the Japanese, even in ancient times (Japan Forestry Agency, 1991). Aside from the Japanese demand for wood used in post and beam construction, there has been a growing demand for wood suitable for platform-frame (PFC) and prefabricated housing construction. The wood imports most suitable for these housing types include North American S-P-F and other dried, planed, dimension softwood lumber, as well as panel products. Largely due to the marketing efforts of the Canadian Council of Forest Industries in the mid 1970s, P F C starts rose from zero in 1975 to 56,299 in 1993, representing 4.8% of all housing starts, or just over 8% of wood housing starts (INTEREX, 1995). Prefabricated housing construction was introduced in the 1950s, and by 1993 represented just over 20% of total housing starts (246,108 starts). However, most of the prefabricated homes are made from steel and concrete, with wooden starts being roughly 30,000 in 1991 (Pesonen, 1993).  Small amounts of wood are required in steel  prefabricated houses. Once again, planed, dimension lumber and panel products are demanded for this housing type. Interestingly, hardwoods in Japan do not typically get used for decorative purposes. In fact, hardwoods are primarily used as pulp furnish or other woodchip products, followed by their use in plywood manufacture. An estimated 35% of Japan's plywood consumption is for the manufacture of concrete forms, with most of the balance being used as sheathing  Page 46  Gaston Chapter Three  in construction for walls, floors and roofing (Sedjo, et al., 1994). Combined with softwood, over 60% of the domestic industrial timber in Japan has historically been used for lumber, followed by roughly 30% for pulp and wood chips combined, and a small percentage for veneer sheets for plywood and other miscellaneous products (Japan Forestry Agency, 1991). The historical context is presented in Figure 3.4. One of the distinctive features of lumber demand in Japan is that in addition to the quality of the wood, dimensions are critical.  Not only are the dimensions demanded  inconsistent with North American standards, for example, but they can change from one region in Japan to the next, or even from one building project to the next (Sedjo, et al., 1994; Japan Forestry Agency, 1991). This partially explains the existence of thousands  F i g u r e 3.4  Japanese Domestic Log Supply by Utilization (Japan Forest Agency Data, Provided by Y. Mori, Kyoto University, Japan)  Page 47  Gaston Chapter Three  of local mills producing to meet highly localized demand conditions (to be discussed in Section 3.4). In 1992, 79% of domestic lumber shipments went into housing construction (Japan Forestry Agency).  For this reason, a look at housing starts over time can be quite  instructive; Figures 3.5 and 3.6 show both wood and non-wood housing starts since 1965, by number and by area respectively. The rapid growth in housing starts throughout the 1960s and early 1970s reflects the rapid economic growth Japan was enjoying over this period. Referring back to either Figure 3.2 or 3.3, domestic wood production also increased through much of this time, peaking at 53 million cubic metres in 1967, and remaining over 40 million cubic metres through the early 1970s.  (As will be made clear in the next section, this domestic  production was, of course, supplemented by growing levels of imports.) From the mid-1970s to the early 1980s this economic growth slowed and ultimately declined, spurred by the "oil crisis" of 1973. After 1986, economic activity stabilized and began to show modest growth (as did housing starts). It is interesting to note that the population in Japan has not changed significantly over the past several decades. During the 1960s, population increased by an average of 1 % per year, peaking at roughly 2% in 1970. Ever since then the rate of population growth declined, being roughly 0.5% per year in 1989 (Yu, et a/., 1990). This suggests that economic activity (or per capita G N P ) is a better indicator of housing starts than is population growth. Before turning to a description of the Japanese lumber processing sector, imports  Page 48  Gaston Chapter Three  2,000,000  1,500,000  1,000,000  500,000  1980 iWood  Figure 3.5  1985  1990  I Non-Wood  Japanese Housing Starts by Number {Japanese Ministry of Construction Data, Provided by Y. Mori, Kyoto University, Japan; 2x4 Data from INTEREX)  160 140 120 CM  E  100 80 60 40 20  1965  1970  1975 Wood  Figure 3.6  1980  1985  1990  I Non-Wood  Japanese Housing Starts by Area {Japanese Ministry of Construction Data, Provided by Y. Mori, Kyoto University, Japan)  Gaston  Page 49  Chapter Three  of wood products are summarized in the following section.  3.3  Imports of W o o d P r o d u c t s into J a p a n A s can be seen in Figure 3.7, Japan has moved from a situation of almost total self-  sufficiency in wood products (95% of Japan's total industrial wood supply in 1955 came from domestic production) to a very strong reliance on imports (only 2 5 % of Japan's total industrial wood supply in 1992 came from domestic production).  120  100  £  80  tu T> O O ==  60  o  40  20  1955  Figure 3.7  1960  1965  1970  1975  1980  1985  Domestic L o g *  Imported Logs  Imported Pulpwood  Imported Lumber  Imported Chips  Imported Misc.  1990  Japanese Industrial Wood Supply (Japan Forestry Agency Data, Provided by Dr. Y. Mori, Kyoto University, Japan)  Page 50  Gaston Chapter Three  100%  80%  £  •  60%  3 CO *CO  £  40%  o  a. 20%  0% 1962  I II  -f-H-  1967  1972  •  Figure 3.8  I  1977  1982  II  IIIIII I  1987  1992  Softwood Logs - A - Hardwood Logs  Japanese Self-Sufficiency in Logs (FAO Yearbook, Various Years)  Figure 3.8 shows that Japan's self-sufficiency in logs alone has declined by a similar magnitude for both softwood and hardwood.  Domestic softwood sawlog production  declined from roughly 26 million cubic metres in the early 1960s to less than 16 million cubic metres in the 1990s, while imports increased from 3 to over 16 million cubic metres over the same period. The drop in domestic hardwood sawlog production is even more dramatic, moving from over 6 million cubic metres to less than 2 million cubic metres over these three decades. Unlike softwood, imports have been an important component of hardwood log supply over the whole period, ranging around the 10 million cubic metre mark both in the 1960s and early 1990s. In fact, in the early 1960s, Japan imported  Page 51  Gaston Chapter Three  almost twice as much hardwood logs by volume than softwood. By the 1990s this turned around, with imports of softwood logs being almost twice that of hardwood logs (1962 and 1993 percentages of softwood log imports were 32% and 62%, respectively, with the total volume of log imports almost tripling). It is also obvious from Figure 3.7 that Japan has shown a strong preference for importing logs rather than lumber, although lumber imports increased throughout this period. Lumber imports increased from 112,000 cubic metres in 1955 to 12,424,000 cubic metres in 1992, in roundwood equivalents. Log imports increased only up to 1973, and have been decreasing ever since. It is widely accepted that this decline in log imports, relative to lumber and other wood product imports, is less a function of Japanese demand and more a function of world supply (Cartwright, 1995; Pesonen, 1993; Sedjo, etal, 1994; Robertson and Waggener, 1995). Indonesia, for example, traditionally a major hardwood log supplier to Japan, adopted log export restrictions in the early 1980s, and by 1985 banned log exports all together.  Presently, parts of Malaysia are considering similar  policies (Sedjo, et al., 1994). This trend toward log export restrictions is due in part to the producing countries' desire to develop their own wood processing industry, and in part to growing local economies (particularly Malaysia and Indonesia), which have increased domestic demand for their wood products.  U S log supplies have also decreased,  particularly logs cut from old growth timber. The drop in the availability of such logs has become most pronounced over the past few years, exacerbated by land withdrawals from the public forests for non-timber uses. A s one Japanese forest economist concludes, it is difficult for second growth P N W logs or domestic Japanese logs to compete with lumber  Gaston Chapter Three  Page 52  imported from B C , which has been produced from old growth timber (Kato, 1982). A s a result of decreasing log supplies, Japan has, however unwillingly, been substituting lumber for log imports. In 1965, 84% of all solid wood imports by value (logs, lumber, panel and other further manufactured products) were in the form of logs. By 1993, log imports as a percentage of total solid wood products had decreased to 44% (Japan Tariff Association).  Lumber imports increased from 15% to 33% of the value of total  imports over the same period, while panel product imports increased from 1% to 2 3 % of the total. Figures 3.9 and 3.10 illustrate how Japanese self-sufficiency in lumber and other panel products changed from 1962 to 1992. Keeping in mind that the domestic lumber or panel product production in these figures includes production from imported logs, the fact that Japan's self-sufficiency in lumber and veneer has dropped significantly since the 1960s confirms its gradual substitution away from logs. Lumber imports, like logs, have also seen a shift away from hardwoods in favour of softwoods. In 1962, only 11% of lumber imports were softwood, as compared to 85% in 1993, and with roughly 15 times the volume of lumber imports in 1993 as compared to 1962. Although Japan exports insignificant amounts of logs and lumber, the country has historically been a major exporter of hardwood plywood. In the 1950s, Japan was the second largest plywood producer (after the US) and the largest exporter (primarily to the U S and Europe).  Due to increased domestic demands, export volumes became  insignificant by the 1970s. Indonesia replaced Japan in this market, exporting large  Page 53  Gaston Chapter Three  100%  80%  70%  eoy.  1962  I I I I I I I 1987 1972  m  Figure 3.9  - • - Veneer  1977  t o f t w a a d Lumbar  I I I I I I I I I I I I 19B2 1987  1992  Hardwood lumbar  Japanese Self-Sufficiency in Lumber (FAO Yearbook, Various Years)  Plywood  - A - Particle Board  Flbreboard  Figure 3.10 Japanese Self-Sufficiency in Selected Panel Products [FAO Yearbook, Various Years)  Gaston Chapter Three  Page 54  volumes of tropical plywood by the early 1980s. By 1988, Indonesia was the largest tropical plywood exporter in the world (FAO). Japan, however, is still the second largest tropical plywood producer in the world, utilizing mostly imported logs and veneer (primarily from Malaysia). In spite of this, Japan's plywood imports have increased from negligible amounts in the early 1960s to over 4 million cubic metres in 1993. Indonesia provides about 9 5 % of these imports. Finally, Japan has also been producing softwood plywood in increasing quantities in recent years (Sedjo, et al., 1994), although imports have remained sluggish (less than 220,000 cubic metres in 1993; Japan Tariff Association). To date, both domestic production and imports of other panel products have been minor compared to plywood.  In 1992, Japanese production of particle board was  estimated at just over 1 million cubic metres (roundwood equivalent; FAO) compared to imports of only 125,000 cubic metres. Equivalent values for fibreboard are 929,000 cubic metres and 155,000 cubic metres, respectively. In neither case are Japanese exports significant. Figures 3.11 through 3.14 demonstrate the rather consistent sources of Japanese imports, especially for softwood logs and lumber. In the case of hardwood logs, the major imports have shifted away from the Philippines, first to Indonesia and then to Malaysia. This has not, however, been the case for further processed products, particularly wood panels. In 1965, the major panel imports into Japan (primarily plywood) came from North America and Europe, whereas in 1993, the biggest plywood imports came from Indonesia; this was also supplemented with fibreboard imports (with New Zealand being the largest supplier), and particle board (with Canada being the largest supplier).  Page 55  Gaston Chapter Three  Canada  Figure 3.11 Japanese Imports of Softwood Lumber and Logs, 1993 (Japan Tariff Association)  Other  Logs  Figure 3.12 Japanese Imports of Softwood Lumber and Logs, 1965 (Japan Tariff Association)  Page 56  G a s t o n Chapter Three  Figure 3.13 Japanese Imports of Hardwood Lumber and Logs, 1993 (Japan Tariff Association)  Other  Logs  Figure 3.14 Japanese Imports of Hardwood Lumber and Logs, 1965 (Japan Tariff Association)  Gaston Chapter Three  Page 57  While the sources of Japanese wood product imports have been rather consistent, the market share that each exporting country enjoys has not. A s already discussed earlier in this section, the biggest change over this time period is the change from imports of hardwood to softwood, followed by the less pronounced change in imports from logs to lumber. This largely translates into an increase in Japanese market share enjoyed by North America (mostly the P N W for logs and B C for lumber).  3.4  Japanese Processing of Domestic and Imported Logs A s made evident in the previous section, Japan has shown a strong preference for  the import of unprocessed logs over lumber and other products. This section explores the processing of logs in Japan, both domestic and imported, in the hope of shedding some light on input preferences. It was shown in Figure 3.4 that approximately 60-70% of Japan's domestic logs are used for lumber, with the balance being used primarily for pulp and wood chips, and small amounts for veneer, fibreboard, scaffolding, and other miscellaneous uses. It was also shown that this rough breakdown has remained relatively consistent over the time period shown. Virtually 100% of the imported logs from North America, the former Soviet Union and New Zealand, by contrast, are used for lumber production.  On the other hand,  imported logs from the South S e a countries are primarily used for veneer, with less than 20% being used for lumber (Japan Forest Agency).  Figure 3.15 shows the breakdown  of Japanese lumber production, from combined domestic and imported logs, over time. An average of 70-75% of the lumber production over the period of this graph was used in  Page 58  Gaston Chapter Three  50  Miscellaneous  40  Constuction: Squares Thickness and Width>7.5 c m .  30  Constuction: Strips Thickness<7.6 cm Width<4 times thickness  20  Constuction: Boards  10  Thickness<7.5 cm Width>4 times thickness  1960  1965  1970  1975  1980  1985  1990  Figure 3.15 Japanese Lumber Shipments by Use (Japan Forestry Agency Data, Provided by Y. Mori, Kyoto University, Japan)  construction, with the remainder being used primarily for packaging and furniture. A s of 1989, roughly 80% of lumber made from domestic logs went into housing construction, as was the case for lumber processed from North American and U S S R log imports. In the case of lumber produced from South S e a logs, however, only 4 5 % went into housing, while an additional 20% went into packaging and the remaining 35% into furniture and other miscellaneous uses. The vast majority (over 80%) of the lumber made from New Zealand logs went into packaging (Otsuka, 1992). 12  A s of 1989, there where over 17,000 sawmills in Japan. Given lumber shipments of approximately 30 million cubic metres at that time, this gives an overall average output  lt must be kept in mind that in 1989, New Zealand had not begun shipping significant quantities of its pruned pine. The percentage of their wood products going into Japanese housing can, therefore, be expected to change quite dramatically. 12  Page 59  Gaston Chapter Three  of only 1,750 cubic metres (roundwood equivalent) per mill.  Of these sawmills,  approximately 4 0 % process domestic logs only, approximately 15% process imported logs only, and the balance process both (Otsuka, 1992). Although the average sawmill size is small, especially by B C standards, they range significantly in size. Those mills that utilize predominantly (or only) domestic logs are generally small family operations employing less than 10 people. These tend to be inland mills located close to the timber source. While this makes up the majority of sawmills by number in Japan, those that process predominately imported logs tend to be considerably larger. These are usually either coastal mills, or mills located close to large urban centres. Logging and sawmilling typically are not integrated in Japan. Sawmills purchase domestic logs directly from the landowners, independent  log producers,  forestry  cooperatives or sawlog markets (which number in the hundreds). Imported logs are mostly purchased from trading companies and wholesalers. Logs from the national forests in Japan are sold competitively (Kato, 1982). There are many sources in the literature which state that Japan's domestic log suppliers have difficulty competing with imports on price (see, for example, Sedjo, etal., 1994; Otsuka, 1992; Iwai, 1986).  In the case of Japanese cypress, for example, the  stumpage cost alone averaged over 40,000 Y e n / m in 1990 (approximately $325 US), or 3  a total cost of producing a domestic log of roughly 55,000 Yen/m (approximately $425 US) 3  (Otsuka, 1992).  This compared to the average import price for U S hemlock logs (a  competing species) of less than 27,000/m Yen in 1990 (or less than $210 US) (Japan 3  Tariff Association). Reporting Japanese cypress log costs a decade earlier, Mochida  Page 60  Gaston Chapter Three  (1984) quotes an even higher price of over 53,000 Y e n / m stumpage and a total log cost 3  of over 65,000 Y e n / m  3 1 3  .  While it appears that imported lumber is also less expensive than domestically produced lumber, the higher cost of shipping lumber reduces the price differential. Otsuka (1992), quotes three price comparisons for lumber from a study done by the Forest Products Research Institute in Japan. These comparisons are for the production of posts in 1990. The final cost of these posts in Japan when domestic cedar logs were used was 54,800 Yen/m ; when North American hemlock logs were used, 48,400 Yen/m ; and finally, 3  3  the cost of imported North American hemlock lumber was 43,000 Yen/m . In terms of U S 3  dollars, these values are approximately $420, $375 and $330/m , respectively. 3  When looking at the price of lumber produced in Japan relative to imports, however, it is not clear, due to inconsistencies in dimension and other measures of quality, that the two are as comparable as is the case with logs.  3.5  Summary This chapter has highlighted the significant trends in Japan's demands for domestic  and imported wood products over the past three decades:  1)  Japan has moved from almost total self-sufficiency in log inputs to a heavy reliance on log imports. Although Japan has undertaken considerable reforestation, this was mostly done after the war, leaving little opportunity to change this trend before the turn of the century.  13  This higher price is likely due to the lower level of competing log imports.  Gaston Chapter Three  Page 61  2)  At present, roughly 60% of Japan's domestic log production is harvested from manmade forests. Given the lower quality and higher cost of this source, compared to domestic old growth, this lends support to the previous point.  3)  Although the harvest from Forest Agency land has remained rather constant over this period, it has shown a gradual decline as a percent of total harvests; the significance of this lies in the fact that it is these lands that contain the largest percentages of old growth, albeit on less accessible lands than the private forests.  4)  Domestic softwood log production dropped roughly 4 0 % over this period while domestic hardwood log production dropped more than 65%. Softwood log imports increased 400% while hardwood log imports stayed relatively constant.  5)  While log imports have shown the noted overall increase over this period, imports actually peaked in 1973 and have been dropping ever since. Against this trend, lumber imports have been increasing throughout. A s was the case with logs, Japan has shifted its lumber imports away from hardwoods in favour of softwoods.  6)  The primary reason for the drop in log imports seems to be less a function of Japanese demand and more a function of the declining availability of logs on the world market (due to growing export restrictions and declining old growth supplies world-wide). The former Soviet Union seems to offer the major potential for added future supply.  7)  While the sources of softwood log and lumber imports have not changed significantly over this time period, total log and lumber market shares have.  8)  North America and the former Soviet Union are the only significant suppliers of imported logs for housing in Japan. Logs from these two countries are used almost exclusively for lumber production. Only 20% of South S e a logs are used for lumber production, and New Zealand logs have been used almost exclusively for packaging material.  9)  Japan has not shown a great acceptance for either softwood plywood or other panel products as substitutes for its traditional hardwood plywood (of which Japan is a significant producer-initially using domestic hardwood logs and later substituting imported hardwood logs and veneer).  Japan clearly enjoys a significant forest resource, with forested land as a percentage of the total land base being very similar to that of B C , and growing conditions  Gaston Chapter Three  Page 62  which allow for significantly greater annual increments in their volume of timber. Given the high population density, however, it is likely that domestic supply will continue to fall far short of demand. Further, given that Japan has already exploited most of its accessible old growth stocks, coupled with a heavy reliance on relatively inexpensive imports, much of Japan's timber resources lie outside of the country's extensive margin (the existence of the imports putting downward pressure on domestic prices). It could be hypothesised, in fact, that much of Japan's man-made timber stocks are also outside of the extensive margin, due to the lower cost of imports relative to the cost of intensive silviculture and harvesting. The question now, however, is will the Japanese substitute lower quality logs, imported lumber and other wood products for the high quality logs, which currently comprise a major proportion of their total imports. Or, will they begin to rely increasingly on domestic wood supplies?. In terms of BC, as a past and potential future supplier of wood products, Japan has shown a strong demand for high quality lumber from B C ' s old growth, in spite of its preference for logs. Given the higher cost of log production in Japan, it is likely that this demand will continue for as long as B C has old-growth stocks to mill; this could last for several decades if the expansion of BC's extensive margin outpaces that of Japan's. At some point, however, it is possible (or even likely given B C ' s present level of silvicultural efforts), that Japan's domestic stocks will be more valuable (revenue minus cost) than B C imports. In addition, of course, is the possibility of increased lumber imports from sources other than B C . It is to these substitution possibilities that this thesis will now turn.  Gaston Chapter Four  Page 63  Chapter 4 Theoretical Foundations and Implications for the Empirical Analysis of the Japanese Demand for Wood Products  This chapter develops the methodology for quantitatively investigating Japan's willingness to substitute alternative wood inputs in their production of wood products, primarily housing. The possible wood input substitutions include: 1) imported wood versus domestic wood; 2) wood from one region versus wood from another region; 3) one type of wood versus another type of wood (for example, softwood versus hardwood; lumber versus logs); and 4) non-wood inputs versus wood inputs. It was stated in Chapter 2 that the most appropriate methodology found in the literature for investigating derived factor demands for disaggregated wood imports comes from the Armington two-stage approach (1969), and its subsequent applications. This chapter starts with a discussion of the theoretical foundation of derived demand, describes the empirical analysis used in the present study, and concludes with a description of the data.  4.1  Theoretical Foundations Japanese home builders, being the primary end-users of wood products, have a  number of choices among possible material inputs. They can use domestic lumber milled from domestic logs, domestic lumber milled from imported logs, or imported lumber. Builders can also choose between hardwood and softwood, and between domestic and imported further processed wood products (such as wood panels). Finally, the builders can change the proportion of wood to non-wood materials used. It must also be noted that the  Page 64  Gaston Chapter Four  builders' choice of input is dependent on the type of housing construction, such as post and beam versus platform-frame, with the former requiring larger dimensions, higher quality, and different species (see Chapter 3). In making these decisions, economic theory suggests that the builders (or the building industry) will combine these alternative wood products with other inputs needed in the production of houses (such as labour, energy, machinery and capital) in such a way as to maximize profits or, equivalently, to minimize costs subject to the production function. To illustrate, consider a Cobb-Douglas production function as follows : 14  GNP = cpQ" Q , 1  where:  GNP Q k  tp,a,  a 2  Q f  ( 4 1 )  = gross national product in Japan; = the quantities of domestic wood, imported wood, and the quantity of "everything else" that goes into building a house; = parameters.  The costs of building these homes can be represented as: C = PQ D  where:  P  D  k  +  P,Q,  +  PQ E  (4.2)  E  = the price of Q  k  If builders are assumed to adjust their input mix in such a way as to minimize costs subject to the technology described by equation 4.1, this will yield the derived demand for  ln this discussion, as in the empirical analysis used in the present study, output is taken to be per capita gross national product (GNP) as opposed to housing starts. This more general measure of output is considered appropriate as various forms of the wood input demand are being investigated, from logs to wood panels. Logs are not directly demanded by house builders, rather the lumber from which the logs are transformed (not to mention the portion of logs which go into veneer for plywood, for example). Further, lumber imports themselves may not be directly demanded by house builders, much of which is remanufactured in Japan. 14  The use of GNP as the output in measuring wood product input demand is not unique to the present study. Further, unlike North America, there exists a strong relationship between per capita GNP and housing starts in Japan, shown by Yu and Mori (1990) to have a historical correlation coefficient of 96 percent.  Page 65  Gaston Chapter Four  input materials, as a function of only input prices and output quantity. The constrained minimization problem is: C = P,Q,+ P Q  Minimize  D  GNP  s.t.  + PQ  D  = cpQ"1 QDa2  E  E  Qf  (4.3) The Lagrangian expression associated with this constrained minimization is:  S£ = P Q D  P,Q,+ PQ  +  D  E  - A  E  ^  If one assumes that the house builder's input choices do not affect the input prices (i.e. that they are price takers, facing a perfectly elastic supply curve for their inputs), the first-order conditions for a minimum are:  5Q  P  - X[chQ,a1QEa3a2QDa2-1]  D  = 0  D  P, -  P  5Q 5A  \  a2  '  5Q,  ^  WQ Q D  E  =  0  ( 4 5 )  -  E  A[cbQ/a1QDa2a3QEa3-1]  = GA/P -  (J)Q  A 1 ;  Q  A 2 D  Q  = 0  = 0  A 3 E  The first three of these equations can be written as:  A[4>Q, Q/ a Q - ]  =  AfoQ^Q^Q,  - ]  =  AMP  - ]  =  A/WP  a1  3  a2  2  A[cbQ  A 1 /  Q  D  a 2  a Q 3  1  D  0 1 1  A 3 £  1  1  AMP  Q Q  Q ;  =  P  =  P,  =  D  P  (4.6)  E  Gaston  Page 66  Chapter Four  where M P is marginal product. Given that the Lagrangian multiplier, A, can be interpreted as marginal cost (MC), as it reflects the change in the objective (costs) given a change in the constraint (output), this can be re-written a s : 15  MC • MP  = P  MC • MP,  = P,  MC • MP  = P  D  E  D  (4.7)  E  Given that M C must equal marginal revenue (MR) under the assumption of profit maximization:  MR • MP  D  = P  D  or  MRP  D  = P  D  (4.8)  MR • MP, = P,  MRP! = P,  MR • MP  MRP  E  = P  E  E  = P  E  where M R P is the marginal revenue product, or the additional revenue obtained from selling what an additional unit of the input produces. Choice of any one input, then, involves equating its marginal revenue product with its price (once again, given the assumption that the buyer of the input is a price taker). In effect, the marginal revenue product is the derived demand for the input. Unfortunately, calculation of the derived demands for multiple inputs is not so straight-forward. For example, given a drop in the price of domestic lumber, there should not only be an increase in the quantity of domestic lumber demanded, ceteris paribus, but  15  See W. Nicholson (1989).  Page 67  Gaston Chapter Four  there will also be a change in the quantity demanded of other wood alternatives as builders adjust to a new cost-minimizing combination of inputs. This can be broken down into a substitution effect (here, substituting domestic lumber for, say, imported lumber from BC) and an output effect (if building a house is now cheaper due to the drop in lumber cost, and if this saving is passed on to the ultimate consumer, the builder will likely see an increase in the demand for houses). Obviously, both of these effects cause the quantity demanded of an input to move in the same direction, and opposite to a price move. This concept will be returned to shortly, when describing the analogous situation of deriving the demand for a specific wood import (such as Douglas-fir lumber from Canada) from the aggregated demand for all imports combined. Returning now to equations 4.5, it is possible to determine the derived demand for each input, expressed in terms of input prices and output. Solving for the derived demand for imported lumber, the final result becomes:  Q = (J) -[a,  a  i  +  a  2  +  a  3  -a,  r  [a  1 1  a  -a,! a,  -cu 2 2  a  ]  3 3  a,  1  + a,]  (4.9)  1 D  x + a 2  3  O  a  1  E  +  a  2  +  «3  GNP  h  2  + a  +  a  3  In short, the derived demand takes the form of (compare to equation 2.19):  Q, =  \vP~^Pl Pl GNP^ 2  3  (4.10)  Page 68  Gaston Chapter Four where the parameters IJJ and the B's are functions of the a's in equation 4.9:  a +a + a +a 2  a  a  1  1  3  2  +  a a a  «1  +  3  (4.11)  2  2  3  a +a 2  3  1 a  1  +  a +a 2  3  By empirically estimating equation 4.10 when Q , , for example, is the quantity of all imported lumber by Japan, the value of B., would represent the own-price demand elasticity of imported lumber (thus the expected minus sign), B and B the cross-price elasticities for 2  3  changes in the quantity of imported lumber demanded given changes in the prices of Japanese domestic lumber and "everything else", respectively, and B the change in 4  quantity of imported lumber demanded given a change in production output. Note that constant returns-to-scale can be tested for and/or imposed by equating the value of B with 4  one (in the constant returns-to-scale case in a Cobb-Douglas production function, a  1  + a  2  + a  3  = 1; see the constraint in equation 4.3). Note as well that - B = B + B , 1  2  3  or that the own-price and the sum of all the cross-price elasticities of the inputs should equal zero. Once again, this can be tested for and/or imposed. A s discussed in Chapter 2, given the stated objectives of this thesis, rather than including P, (the average price of imported lumber in aggregate) in equation 4.10, it would  Gaston Chapter Four  Page 69  be more desireable to disaggregate the P, into its component parts (such as Douglas-fir lumber from Canada, radiata pine lumber from New Zealand, etc.). This would allow for the estimation of a number of potentially relevant cross-price elasticities between country of origin, species or even product type . 16  The problem in obtaining cross-price elasticity of demand estimates in this manner, however, is two-fold. First, every price variable used as a regressor will use up one degree of freedom, which is particularly problematic when the researcher is limited in the number of available observations (such is usually the case when one must rely on annual data). It is not hard to imagine the number of desired price regressors exceeding the number of observations when one considers the possible permutations of the sources of the import, species, and product type. The second problem is that even a few prices used as explanatory variables can lead to problems of multicollinearity. In spite of the fact that it would be expected that there are differences in the prices histories of, say, B C Sitka spruce lumber and New Zealand radiata pine lumber (or logs), it would not be surprising to see that the overall pattern is similar. An example of this problem was offered in Chapter 2 in the description of Sedjo and other's study (1994). Using Japan as the demander for imported timber products (see equation 2.22), the researchers regressed the import price from region "A's" timber on the quantity demanded of this import, along with the price of U S timber, the price of Japanese  ln order to break down the import prices by product type, the dependent variable would have to be the sum total of log, lumber and product quantities. 16  Page 70  Gaston Chapter Four  domestic timber, and the average import price of timber from regions other than "A" or the US. A s the parameters on the Japanese domestic timber price and the "other" timber price were found to be insignificant, they were suppressed. It is possible that this insignificance stemmed from multicollinearity among the price series; if this is the case, it is not clear that the variables should have been omitted, due to the potential for bias . 17  In the present study, the demand for a specific wood product is being sought (such as the Japanese demand for Canadian Douglas-fir lumber), derived from the "output" of aggregated wood imports. The derived factor demand function can be written as (see equation 4.9): -(«  Q  where:  = UJ p  0 , 1  Q  +  =  :  P., = P , P = Q = 2  3  a  2  2  g  + "3) +  °3 p ^ i  +  «2  +  «  2 +  1  3  «3 p <*i «2 a +  3  Q a, a +  2+  a  3  ( . ) 4  13  the quantity demanded of a specific wood product (for example, Canadian Douglas-fir lumber); the price of this product; the prices of other products which make up the total imports, Q; the aggregate quantity of imports (the "output").  This equation could, at least theoretically, be estimated generating the own-price elasticity of product " 1 " as well as the cross-price elasticities of the quantity demanded of "1" (say, Japanese demand for Canadian cedar) relative to the price of either "2" or "3" (say, the price of U S cedar and New Zealand radiata pine). Note, however, that this equation disaggregates total imports into only three component parts; while this in itself could lead  "Multicollinearity and omitted variable bias is well documented in most texts on econometric theory. Multicollinearity tends to exaggerate the variance of the affected regressor, which leads to the likelihood of omitting a variable that should not be omitted. Johnston (1984), pages 259-264 offers a good account of the consequences of omitted variable bias. These potential problems are explored in greater detail when discussing the results in the following chapter.  Page 71  Gaston Chapter Four  to problems of multicollinearity, the elasticity estimates of a greater number of component parts are desired in the present study. A s was shown in Chapter 2, Armington (1969) offers a potential solution. The approach, illustrated empirically by Chou and Buongiorno (1983), allows for a highly aggregated form of equation 4.10 to be used, such as the quantity of all lumber imports by Japan, yet still leads to highly disaggregated price elasticity information. To recap, this is accomplished by a two step process. First, the demand for total imports of a kind (such as plywood) is estimated, derived from the demand for some measure of output. Second, the demand for individual sources of the product, derived from the demand for the product in aggregate is estimated.  A s shown in Chapter 2, this process can be reversed  empirically for the ease of calculating C E S quantity and price indices (i.e., first estimating the constant elasticity of substitution, then using this to estimate the aggregate demand).  4.2  The E m p i r i c a l Model The following chapter will report the results from two sets of analysis . The first of  these will be individual factor demand equations (such as equation 4.10): (4.23)  where Q, and the corresponding P, represent various levels of aggregate quantity and price, respectively, of imported Japanese wood products. When estimating the aggregate derived demand for all imports, P is the real domestic price of Japanese logs. When D  estimating a less aggregated form of the derived demand, such as softwood lumber from  Page 72  Gaston Chapter Four  Canada, P is the average of both the domestic Japanese log price and the price of all D  other imports (everything other that softwood lumber from Canada).  While these  regressions will not allow for the determination of disaggregated cross-price elasticity estimates, the analysis is done for comparison of results with the Armington, two-stage approach. P , being the price of everything else, is taken in this study to be a wage index E  due to its significance in Japan. The second set of analyses will largely utilize the Chou and Buongiorno (1983) application of the Armington model, allowing for a greater range of own- and cross-price elasticity estimation. Specifically, the following equations are estimated, using a linear log-log transformation:  ln(Q„) = AQ - B1 ln(P„)  +  (32 ln(P ) Df  +  B3 l n ( P ) +(34 \n(GNP ) + u a  t  f  (4.25)  and In  ( Q\ Q  ( P) cp  0/  - a In  t  + v  f  (4.26)  t  Equation 4.26 is estimated as system of equations, repeating the process for various levels of aggregation. For example, in the first case the system varies / over major categories of wood (softwood logs, hardwood logs, softwood lumber, hardwood lumber, and wood-based panel products). Subsequent systems focus in on greater detail, beginning with softwood lumber imports by Japan from various countries of origin, then on the imports of softwood lumber from Canada by species.  With the estimates of o (the constant elasticity of  substitution) and the constant terms in hand, the relative weights can be recovered for creating the C E S quantity and price indices (as explained in Chapter 2). With these  Gaston Chapter Four  Page 73  indices, equation 4.25 can be estimated in a manner consistent with Armington's two-stage assumptions.  4.3  The Data S o u r c e s U s e d in the Empirical A n a l y s i s The main source of data for the quantity and price relationships is the Japan Tariff  Association, Imports of Commodity by Country. This publication offers data on the annual import of all commodities in considerable detail, giving volume and value by country of origin. The data set used in this study covers the period 1965 to 1993, with the recognition that the level of detail diminishes as one goes further back in time. The reason that the study does not utilize data prior to 1965 is that the level of detail was considered to be inadequate. In 1965, this publication reported imports from over 50 countries, broken down into a total of 80 categories of wood products. This included 10 categories of softwood logs, 14 categories of hardwood logs, 16 categories of softwood lumber, 10 categories of hardwood lumber, 14 categories of panel products, and 16 categories of further manufactured products. Unlike later years, the wood product detail is primarily in the form of species.  At the other extreme, 1993 showed considerably more detail. The total  number of wood product categories increased to 145, broken down to 10, 20, 21, 23, 41, and 30, respectively, as above. The detail on country of origin also increased, exceeding 80. In order to run time series regressions over the entire 29 year period, it was necessary to aggregate much of the detail offered by these data in the later years in order  Gaston Chapter Four  Page  74  to obtain consistent product categories through all years. While this task was considerable in itself, it was further complicated by the fact that the Japan Tariff Association utilized three different commodity classification systems over this time period. This meant that the data had to once again be aggregated to the common denominator found over the three commodity classification systems. A s this requires a degree of subjective judgement, the end result of the aggregations used in this study is reported in detail in Table 4.1. A s seen in the table, in spite of the need for aggregation due to changes in the data series over time, considerable detail was retained. There are 10 categories of softwood logs (SLG-X) and 10 categories of hardwood logs (HLG-X), the breakdown being purely by species. The first category in each, which is treated as lumber in this study, is logs which are roughly squared or half squared (cants). Although it might have been expected that prices within this category would be significantly higher than prices of dimension lumber, such is not the case; further, the reported volumes in this category are very low. For this reason, SLG-1 and HLG-1 were aggregated with SLM-9 and HLM-9, respectively, these latter categories being softwood and hardwood lumber, not elsewhere specified (n.e.s.), respectively. (For definitions of S L G - 1 , etc., see Table 4.1.) There are 20 final categories of softwood lumber (SLM-X), along with 8 categories of hardwood lumber (HLM-X). In the case of softwood lumber the categories include detail on size breakdown (greater or less than 160 mm) as well as species. Unfortunately, this breakdown was not consistently offered throughout the entire period, starting only in 1974 and 1977 in most cases. It is possible that there was no lumber imported by Japan less than 160 mm previous to these years. Further, just as was the case with cants, although  Gaston Chapter Four  Page 75  Table 4.1: Japan Tariff Association Data, Converted Codes New  65-75  76-87  88-93  Description  SLG-1  242-299  44.04-310  4403.20-010  Coniferous Logs, roughly squared or half squared  SLG-2  242-210  44.03-321  4403.20-091  Sawlogs & veneer logs, Pinus  SLG-3  242-221  44.03-322  4403.20-092  Sawlogs & veneer logs, Sitka spruce  SLG-4  242-229  44.03-323  4403.20-093  Sawlogs & veneer logs, A b i e s and P i c e a , excluding  SLG-5  242-230  44.03-324  4403.20-094  Sawlogs & veneer logs, Larix  SLG-6  242-240  44.03-325  4403.20-095  Sawlogs & veneer logs, white cedar, yellow cedar,  SLG-7  242-250  44.03-326  4403.20-096  Sawlogs & veneer logs, hemlock & other T s u g a  SLG-8  242-260  44.03-327  4403.20-097  Sawlogs & veneer logs, red cedar & other Thuja  SLG-9  242-270  44.03-328  4403.20-098  Sawlogs & veneer logs, Douglas-fir & other Pseudotsuga  SLG-10  242-298  44.03-329  4403.20-099  Sawlogs & veneer logs, conifer, n.e.s.  HLG-1  242-391 +  44.04-100 +  4403.99-210 +  Non-coniferous logs, roughly squared or half squared  242-399  44.04-390  4403.99-311 +  Sitka spruce  & other Chamaecyparis  4403.32-010 + 4403.33-011 HLG-2  242-310 +  44.03-331 +  4403.31-090 +  Sawlogs & veneer logs, lauans and apitons to '75; lauan,  242-381  44.03-336  4403.32-090 +  kruimg mersawa and other Dipterocarpaceae family '76-'87;  4403.33-019 +  All Meranti, Keruing, and kapur, '88 onward, plus mahogany  4403.99-290 HLG-3  242-320  44.03-100  4403.99-319 +  Sawlogs & veneer logs, Kwarin, T s u g e or boxwood, Tagayasan  4403.99-310 +  (Cassia siamea Lam.), red sandal wood, rosewood, or  4403.33-099  ebonywood (excl. ebony w / w h i t e streaks)  HLG-4  242-340  44.03-333  4403.99-391  Sawlogs & veneer logs, cottonwood and aspens  HLG-5  242-350  44.03-200  4403.99-190  Sawlogs & veneer logs, kiri (Paulownia)  HLG-6  242-360  44.03-334  HLG-7  242-370  44.03-335  4403.33-091  Sawlogs & veneer logs, teak  HLG-8  342-382  44.03-337  4403.99-392  Sawlogs & veneer logs, American black walnut  HLG-9  242-383 +  44.03-338  Sawlogs & veneer logs, lignum vitae  Sawlogs & veneer logs, sandalwood  242-384  HLG-10 242-389 + 242-330  44.03-339 +  4403.99-399 +  44.03-390 +  4403.91-000 +  44.03-332  4403.92-000 + 4403.34-000 + 4403.35-000  Sawlogs & veneer logs, non-coniferous, n.e.s. (incl. oak and beech post 1987)  Page 76  Gaston Chapter Four  Table 4.1: Continued New  65-75  76-87  SLM-Oa SLM-Ob SLM-1a  243-211  44.05-310  88-93  Description  4407.10-330  Lumber, S P F , not more than 160 m m in thickness  4407.10-320  Lumber, planed or sanded, n.e.s.  4407.10-121 +  Lumber, Pinus, not exceeding 160 m m in thickness  4407.10-330  SLM-1b SLM-2a  243-212  44.05-510/511  243-221  44.05-512/521  Lumber, Pinus, exceeding 160 mm in thickness 4407.10-341  Lumber, Sitka spruce (combined; post 1977, not exceeding 160 mm)  SLM-2b SLM-3a  243-222  44.05-522  4407.10-349  Lumber, Sitka spruce exceeding 160 m m (after 1977)  44.05-320  4407.10-129 +  Lumber, A b i e s (excluding Calif, red fir, grand fir, noble  4407.10-350  SLM-3b  243-223  44.05-513/530  fir, etc.) & P i c e a , not exceeding 160 m m Lumber, A b i e s (excluding Calif, red fir, grand fir, noble fir, etc.) & P i c e a , exceeding 160 m m  SLM-4a  243-231  44.05-330  4407.10-210 +  Lumber, Larix, not exceeding 160 m m  4407.10-290  SLM-4b SLM-5a  243-232  44.05-540  243-240  44.05-515/551  Lumber, Larix, exceeding 160 mm 4407.10-361  Lumber, white and yellow cedar and other Chamaecyparis (post 1977, not exceeding 160 mm)  SLM-5b  44.05-552  4407.10-369  Lumber, white and yellow cedar and other Chamaecyparis, exceeding 160 m m (post 1977)  SLM-6a  243-250/251  44.05-516/561  4407.10-371  Lumber, hemlock and other T s u g a (post 1974, not exceeding 160 mm)  SLM-6b  243-252  44.05-517/562  4407.10-379  Lumber, hemlock and other T s u g a , exceeding 160 mm (post 1974)  SLM-7a  243-260  44.05-518/571  4407.10-381  Lumber, Douglas-fir and other Pseudotsuga (post 1977, not exceeding 160 mm)  SLM-7b  44.05-572  4407.10-389  Lumber, Douglas-fir and other Pseudotsuga, exceeding 160 m m (post 1977)  SLM-8 SLM-9a  243-271 +  44.05-521/581 + 4407.10-310  243-279  44.05-522/589  243-280  44.05-529/591  4407.10-391  Lumber, conifer, n.e.s. (post 1977, not exceeding 160 mm)  44.05-592  4407.10-399  Lumber, conifer, n.e.s., exceeding 160 m m (post 1977)  SLM-9b  Lumber, incense cedar  SLM-10  243-291  44.13-300  4409.10-310  Planed, grooved ortongued; Pinus, A b i e s , P i c e a and Larix  SLM-11  243-299  44.13-510  4409.10-320  Planed, grooved or tongued, conifer, n.e.s.  Page 77  Gaston Chapter Four  Table 4.1: Continued New  65-75  HLM-1 243-310  76-87  88-93  Description  44.05-100  4407.99-110 +  Lumber, Kwarin, Tsuge or boxwood, Tagayasan (Cassia siamea Lam.), red sandal wood, rosewood, or ebonywood  4407.99-190 HLM-2 243-320  44.05-200  4407.99-210 +  Lumber, Kiri  4407.99-290 HLM-3 243-330  44.05-531/593 4407.21-110 +  HLM-4 243-340  44.05-532/594 4407.99-410 +  HLM-5 243-350  44.05-400  Lumber, teak  4407.21-190 Lumber, lignum vitae  4407.99-490 4407.21-210 +  Lumber, lauan, kruing, mersawa and other Dipterocarpacea  4407.21-290 + 4407.21-300 + 4407.99-310 + 4407.99-390 + 4407.23-000 HLM-6 243-360  44.05-539/599 4407.99-500 +  Lumber, non-conifer, n.e.s.  4407.91-000 +  (oak)  4407.92-000 +  (beech)  4407.22-000 HLM-7 243-393  44.13-400  4409.20-330  Lumber, planed, grooved ortongued, lauan, kruing, mersawa and other Dipterocarpaceae  HLM-8 243-399 +  44.13-590 +  4409.20-350 +  243-391 +  44.13-100 +  4409.20-310 +  Lumber, planed, grooved or tongued, non-conifer, n.e.s.  4409.20-340 243-392  44.13-200  4409.20-320  VS-1  631-111  44.14-100  4408.90-100  Veneer sheets, Kwarin, Tsuge or boxwood, Tagcyasan, red  VS-2  631-112 +  44.14-220  4408.90-200  Veneer sheets, Teak  631-119 +  44.14-210 +  4408.10-010 +  Veneer sheets, n.e.s.  631-120  44.14-230 +  4408.10-021 +  44.14-290  4408.10-029 +  sandalwood, rosewood and ebony wood. 631-113 . VS-3  4408.20-010 + 4408.20-090 + 4408.90-300 + 4408-90-410 + 4408.90-490  *  Page 78  Gaston Chapter Four  Table 4.1: Continued New  65-75  76-87  88-93  Description  PLY  631-210 +  44.15-191 +  4412.19-011 +  Plywood; "cellular"; "blockboard"  631-211 +  44.15-111 +  4412.19-019 +  631-212 +  44.15-119 +  4412.19-021 +  631-213 +  44.15-192 +  4412.19-022 +  631-214 +  44.15-193 +  4412.11-011 +  631-219 +  44.15-194 +  4412.11-019 +  631-220  44.15-195 +  4412.11-021 +  44.15-196 +  4412.11-022 +  44.15-300 +  4412.11-023 +  44.16-000  4412.11-024 + 4412.11-029 + 4412.12-011 + 4412.12-019 + 4412.12-021 + 4412.12-022 + 4412.21-010 + 4412.21-090 + 4412.29-010 + 4412.91-090 + 4412.99-010 + 4412.99-090 + 4418.90-100  PB  631-420  44.18-100 +  4410.10-010 +  44.18-200  4413.00-000 +  Particle board; "reconstituted"; "densified"  4410.90-010 FB LAM  631-410  44.11-100 +  4411.11-000 ->  44.11-900  4411.99-000  44.15-200 +  4412.29-010 +  44.17-000  4412.99-010 +  Fibreboard; "hardboard"; "building board" Laminated; "improved"  4412.21-010 + 4418.90-222 MISC  631-870 —>  44.19-000—>  4409.10-200 +  632-899  44.28-290  4409.20-200 + 4414-00-000 -> 4421.90-099  Misc.; incl. wood beading/moulding, boxes, casks, barrels, wood for decorative use, etc.  Gaston Chapter Four  Page 79  there is a price spread between the two size categories for most species, this spread is not as dramatic as one might expect. This is illustrated in Figures 4.1 and 4.2, which show size comparisons for Japanese imports of selected Canadian lumber species. A s a result, the size detail was not maintained in the regressions reported in this study; volumes and total values were summed within each category. It can also be noted in Table 4.1 that spruce-pine-fir (S-P-F) lumber and "planed or sanded", n.e.s. imports were not reported individually until 1988.  It is assumed that  previous to this date such imports were reported in the category of "planed, grooved or tongued"; the regressions reported in this study, therefore, aggregate SLM-0, SLM-10 and SLM-11 into a single category. In the case of hardwood, the two categories of planed, grooved or tongued lumber, HLM-7 and HLM-8, are aggregated into one category for the regressions. It is with the panel products that the Japan Tariff Association data loses the greatest level of detail in the earlier years. While the data from 1988 onward includes breakdown by species (or at least hardwood versus softwood), and often other characteristics such as density and thickness, previous years lose detail progressively.  For the regressions  reported in this study, a single category is used for each of the following: veneer sheets (VS-1 through V S - 3 are aggregated), plywood, particle board, fibreboard, and laminated lumber. Other assumptions regarding panel descriptions found in the data series are noted in Table 4.1 (for example, including "cellular" and "blockboard" in the plywood category). Finally, the Japan Tariff Association data include considerable detail under the category of miscellaneous, such as mouldings, wood headings, and other further  Page 80  Gaston Chapter Four  •  Figure 4.1  less than 160 mm  —^— greater than 160 mm  Nominal Price of Japanese Imports of Canadian Sitka Spruce Lumber By Size Category (Japan Tariff Association)  120 -,  less than 160 mm  Figure 4.2  — ^ _ greater than 160  mm  Nominal Price of Japanese Imports of White and Yellow Cedar Lumber By Size Category (Japan Tariff Association)  Page 81  Gaston Chapter Four  manufactured wood products. Imports in this category were not included in the present analysis due to their comparatively insignificant volumes, and the problems associated with converting volumes to a common unit. All log and lumber volumes are reported in cubic metres, roundwood equivalent. In the aggregate regressions, where log imports are included with lumber and/or wood panel imports, consideration was given to using a lumber recovery factor. However, this was not done for two main reasons. First, lumber recovery can vary widely, not only due to processing differences among sawmills, but also because of differences in the fibre source, species, log characteristics and product being produced. Secondly, one might just as rightly state that the lumber imports by Japan should also include a "lumber recovery factor" insofar as Japan remanufactures a significant quantity of its raw imports. This is, of course, particularly true for cants or other large dimensions. A s assigning a value to this recovery would be highly arbitrary, it was decided to leave all volumes as reported. Panel products are reported in a number of different units, which makes conversions more difficult. While 1000 square metres of panel products which are 1 mm thick equal one cubic metre, the product categories used in this study span a great variety of thicknesses, not all of which are reported. To get around this problem, it was noticed that the value of panel imports reported by the Japan Tariff Association very closely parallel those quoted by the F A O over the same time period (once converted to a common currency)(Foresf Products  Yearbook, various issues). A s a result, the panel product  volumes in this study were converted using the average values suggested by the F A O data.  Gaston Chapter Four  Page 82  A s particle board and laminated lumber are quoted in cubic metres (solid wood equivalent), they required no conversions for the present study. Plywood and veneer sheets are quoted in square metres and were converted to cubic metres by dividing by 135 (FAO). Fibreboard is quoted in kilograms and was converted to cubic metres by dividing by 300 (FAO). A s the country of origin detail was much too high for the purposes of this study, six country groups were created and data aggregated accordingly: Canada, the United States , the former Soviet Union, the combined imports from New Zealand and Chile 18  (representing the major plantation producers of radiata pine), the combined imports from the South Seas, and the combined imports from the "rest-of-world" (all other countries exporting wood products to Japan). While The Japan Tariff Association does not break down Canadian imports by province, as Table 4.2 demonstrates for lumber, the vast majority of Canadian off-shore solid wood product exports originate from British Columbia. A number of additional time-series data were needed for the present study. A s noted in Chapter 3, domestic log and lumber production were obtained from the Japan Forestry Agency, Table of Demand and Supply. Prices for domestic logs and lumber were obtained from the Japan Wood Products Information & Research Centre; from 1984 on these prices were received directly from their in-house publication (various issues), while prices before this date were obtained from the Global Trade Model data bank at the  Given the similarity of wood products imported from Canada and the United States, it would have been reasonable to combine these two into "North America". However, this disaggregation was desired due to the emphasis on Canada in the results and discussion in Chapters 5 through 7. 18  Page 83  Gaston Chapter Four  Table 4.2  B.C. Offshore Lumber Exports Relative to the Whole of Canada (000s m ) 3  Year  B.C.  Canada  B.C./Canada  1982  4377  5025  87.1%  1985  3810  4004  95.2%  1986  3822  4153  92.0%  1987  5739  6181  92.8%  1988  6141  6854  89.6%  1989  6039  6951  86.9%  1990  6093  7406  82.3%  Source: Selected Forestry Statistics Canada, Information Report E-X-47, Natural Resources Canada, 1993.  University of Washington . 19  To convert the price series used in the estimation of aggregate imports to real values, the deflator employed was the Japanese producer price index for imported "wood, lumber & related products" (Bank of Japan). For the "cost of everything else", noted as P in Sections 4.1 and 4.2, a monthly E  wage earnings index for Japan was used, obtained from FAO's International Financial Statistics Yearbook (1994). In those regressions where it was desirable to investigate cross-price elasticities between the quantity of a wood product demanded and a non-wood import, individual wholesale price indices for iron & steel products, ceramics, stone & clay products, and for plastic products were used, obtained from the Bank of Japan. Finally, the per-capita G N P values were obtained from FAO's International Financial  19  Centerfor International Trade in Forest Products (CINTRAFOR).  Gaston Chapter Four  Statistics Yearbook (1994).  Page 84  Gaston  Chapter Five  Page 85  Chapter 5 Empirical Results  This chapter reports the results of the econometric analysis. Estimates of Japanese aggregate derived demand of all wood product imports are presented in Section 5.1, as well as the derived demands of selected disaggregated products. The disaggregated product own-price elasticities are desired for comparison with the Armington two-stage model, following the empirical approach suggested by Chou and Buongiorno (1983). The two-stage results are presented in Section 5.2. First, the constant elasticity of substitution is estimated over three individual systems, with each system disaggregating Japanese demand for wood product imports by varying degrees of product detail. This is followed by the estimation of the C E S demand function and the resulting calculation of the own- and cross-price elasticities.  In Section 5.3, the results from the direct own-price elasticity  estimations are compared to the calculated values from the two-stage approach, leading to a brief discussion of the appropriateness of Armington's assumptions. Finally, Section 5.4 offers the estimates of non-wood substitution by the Japanese importer.  5.1  Direct Estimation of Japanese Price Elasticities of Demand for Wood Imports Table 5.1 shows the ordinary least square (OLS) and the Cochrane-Orcutt estimates  of the Japanese demand for aggregated wood product imports (including logs, lumber and panel products, both softwood and hardwood).  In log-log form, the quantity of import  demand is regressed on the real unit price of the aggregate wood products import (P,), the domestic price of logs in Japan (P ), an index of wage rates in Japan (P ) and the D  E  Gaston  Page 86  Chapter Five  Table 5.1 Estimates of the Japanese demand for aggregated wood imports. Constant  Pi  PD  PE  GNP  R -Adj.  DW  1.208 (3.366)  -0.802 (-5.347)  0.842 (6.935)  0.646  0.590  2  Ordinary Least Square 12.741 (11.630)  -0.1607 (-0.642)  Durbin's h  Cochrane-Orcutt 13.625 (12.280) Note:  -0.0373 (-0.194)  0.862 (2.324)  -0.719 (-3.341)  0.8253 (4.919)  0.816  0.233  P;, P , P , and GNP, are the logarithms of the price of aggregate wood imports (logs, lumber and panels), the price of Japanese domestic logs, the monthly average wage index in Japan, and per capita GNP in Japan, respectively. Numbers in parentheses are t-values. D  E  per capita gross national product in Japan (GNP). The reason for including the Cochrane-Orcutt estimate is apparent from the investigation of the Durbin-Watson (DW) statistic in the O L S regression. The lower and upper bound critical values for the D W t e s t for five parameters and 29 observations are 1.124 and 1.743, respectively . A s the DW statistic from the O L S regression lies below 20  the lower bound, suggesting that this regression is potentially serially correlated, a Cochrane-Orcutt regression is used. The value of Durbin's h statistic on the Cochrane-Orcutt regression is 0.233, indicating that there is no evidence of higher order autocorrelation.  Under the null  hypothesis of no higher order autocorrelation, Durbin's h is asymptotically normal with zero mean and unit variance. This null hypothesis is rejected if the statistic is greater than 1.645  Durbin-Watson critical values are quoted from Judge, et al. (1988), reproduced from Savin and White (1977). 20  Gaston  Chapter Five  Page 87  at the 5 percent level of significance (1.645 being the t-value at 5 percent significance with °° degrees of freedom) (see Judge, et al., 1988; and Whit, 1993). In the previous chapter, it was shown that the parameters on the aggregate import price, the Japanese log price and the price of "everything else" should theoretically sum to zero (see equation 4.11). The appropriate test on the regression reported in Table 5.1 yields an F-value of 1.1491, which is well below the critical value of 4.26 with 1 and 24 degrees of freedom. A s a result, there was no reason to impose this restriction. It was also shown in the previous chapter that the returns to scale can be determined by taking the reciprocal of the Japanese per capita G N P parameter. The appropriate test to determine whether this regression demonstrates constant returns to scale is to equate the G N P parameter with 1.0. A s this yields an F-value of 1.0843, once again being below the critical value indicated above, it can be concluded that this model demonstrates a return to scale which is not significantly different from being constant. A s a final test on the direct factor demand regression reported in Table 4.1, evidence of structural change was sought in two ways. First, a sequential Chow test was performed. Second, the regression was re-run with the inclusion of dummy variables, both on the intercept and on the import price. The dummy variable was defined as being zero for the first x years and one otherwise. This was done with two different values of x. The first period is defined as being from 1965 to 1973 due to the rapid growth in Japanese housing starts as compared to 1974 to 1993. Further, Japan utilized fixed exchange rates based on the gold standard until the early 1970s. The second period is defined as being 1965 to 1980, due to the introduction of log export controls by two major South S e a  Page 88  Gaston Chapter Five  60,000,000  —ca— Observed  Figure 5.1  _ o — Predicted  Observed versus predicted values of quantity demanded of aggregate softwood lumber imports by Japan when regressed on the average import price, possible wood substitute price (other wood imports and domestic logs), the Japanese wage index, and per capita GNP.  hardwood log producers in the early 1980s. The Chow test rejects structural change, and there are insignificant t-values on the intercept and import price dummy variables. In addition to these tests, Table 5.1 demonstrates that this regression offers a reasonably good fit, with an adjusted correlation coefficient of 0.816. This is graphically illustrated in Figure 5.1. Further, all of the parameter values meet a priori expectations in terms of sign, and all parameter values are significant with the exception of that on the imported wood price. That this parameter is insignificant comes as no great surprise; one would expect the own-price elasticity of aggregate imports of all wood products to be highly  Gaston  Chapter Five  Page 89  inelastic (few available substitutes). The parameter shown can be interpreted as not being significantly different from zero. Table 5.2 shows the results of estimating the derived demand for selected disaggregations of Japanese wood product imports. Cochrane-Orcutt regressions are again employed due to evidence of serial correlation. Each row in the table represents an individual regression, changing the dependent variable Q, and independent variable P, in each case, with the / representing the specific product indicated.  The individual  regressions also include a second price variable, being the average real unit price of all other possible wood substitutes. This includes the price of imported products other than Q, and the price of domestic logs in Japan. The average real price per unit of these two substitution possibilities is indicted as P in the table. The other independent variables, 0  being the wage rate index in Japan (P ) and the Japanese per capita gross national E  product (GNP) are common to all of the regressions. A s was the case with the regression on the aggregate demand for all wood imports by Japan (Table 5.1), it can be noted that the parameters on the specific import price, the price of all possible wood substitutes, and the wage index should theoretically sum to zero. Individual tests yield F-values which are below the critical values in each case, again making it unnecessary to impose this restriction. A s regards returns to scale, while the G N P parameter in some of the regressions would suggest decreasing returns to scale, the appropriate tests also yield F-values which are below the critical value. Over all, the regressions offer reasonably good fits, with adjusted correlation coefficients typically over 0.8. Further, almost all of the parameter values meet a priori  Gaston  Page 90  Chapter Five  Table 5.2  Estimates of the Japanese demand for selected disaggregated wood imports.  Constant  |  P,  Po  P  E  |  GNP  R -Adj.  Durbin's h  2  Softwood lumber from all sources 11.109 (9.313)  -0.650 (-2.340)  1.469 (3.351)  -0.710 (-4.105)  1.562 (8.839)  0.940  0.499  2.104 (2.965)  -0.547 (-1.855)  1.639 (5.561)  0.880  0.290  0.599 (3.833)  -0.308 (-0.961)  1.415 (4.223)  0.835  0.776  1.553 (2.543)  -0.366 (-1.259)  1.222 (3.111)  0.676  0.119  1.993 (2.265)  -0.993 (-0.831)  1.422 (6.619)  0.911  0.259  0.439 (0.630)  -0.577 (-3.169)  1.198 (2.999)  0.618  0.926  2.003 (2.931)  -0.387 (-1.553)  1.500 (8.803)  0.809  0.456  3.596 (2.302)  -0.470 (-0.446)  1.338 (6.172)  0.956  0.188  2.531 (2.013)  -0.891 (-0.809)  1.626 (4.233)  0.788  0.557  0.983 (1.689)  -1.202. (-1.098)  1.577 (2.633)  0.713  0.707  Softwood lumber from Canada 9.897 (5.371)  -0.951 (-2.473)  Yellow cedar lumber from Canada 3.743 (1.415)  -0.398 (-3.300)  Sitka spruce from Canada 8.124 (2.222)  -1.109 (-2.287)  Douglas fir lumber from Canada 5.197 (8.814)  -1.255 (-3.012)  Hemlock lumber from Canada 4.593 (3.770)  -0.209 (-0.255)  Other lumber from Canada 12.282 (6.991)  -1.536 (-2.558)  "Planed" lumber from Canada 3.730 (0.694)  -3.591 (-3.216)  Softwood lumber from the US 3.619 (1.922)  -1.329 (-1.529)  Red cedar lumber from the US 8.808 (6.811) Note:  -0.040 (-1.893)  Pj, P , P and GNP, are the logarithms of the price of the indicated wood product, the average price of other imported wood products and Japanese domestic logs combined, the monthly average wage index in Japan, and per capita GNP in Japan, respectively. Numbers in parentheses are t-values. 0  B  Gaston  Page 91  Chapter Five  I Table 5.2 (Cont.)  Constant  Estimates of the Japanese demand for selected disaggregated wood imports. Pi  Po  PE  GNP  R -Adj.  Durbin's h  4.863 (2.887)  -0.239 (-0.883)  1.207 (4.454)  0.866  0.122  -0.408 (-1.255)  0.967 (4.833)  0.777  0.142  2  "Planed" lumber from the US 9.331 (3.167)  -5.309 (-2.981)  Softwood lumber from New Zealand 6.118 (4.298)  -3.217 (-2.009)  2.881 (2.514)  Softwood lumber from the former Soviet Union 1.181 (2.645)  0.045 (0.893)  0.454 (1.563)  -0.611 (-0.607)  0.983 (3.398)  0.606  0.983  2.691 (3.109)  -0.436 (-3.162)  1.262 (5.331)  0.903  0.188  3.139 (3.222)  -0.228 (-1.936)  1.403 (8.278)  0.897  0.558  1.404 (2.300)  -0.655 (-2.300)  1.239 (5.309)  0.770  0.812  0.174 (0.703)  -0.372 (-1.345)  0.651 (2.475)  0.752  0.774  1.146 (1.943)  -0.653 (-2.318)  0.777 (3.653)  0.634  0.0387  0.399 (2.166)  -0.481 (-1.563)  1.293 (2.599)  0.701  0.836  1.190 (2.203)  -0.936 (-0.816)  0.880 (3.636)  0.793  0.254  0.525 (1.927)  -0.267 (1.193)  1.193 (3.101)  0.660  0.779  Softwood lumber from Other 9.910 (4.111)  -1.417 (-4.253)  Hemlock lumber from Other 14.232 (7.319)  -3.417 (-2.999)  "Planed" lumber from Other 6.689 (5.557)  -0.517 (-1.807)  Softwood logs from all sources 15.609 (20.99)  -0.0901 (-0.294)  Softwood logs from the US 11.712 (6.843)  -0.209 (-1.660)  Sitka spruce logs from the US 5.177 (4.008)  -0.104 (-1.809)  Abies/picea logs from the US 3.696 (1.990)  -0.339 (-1.458)  Yellow cedar logs from the US 8.294 (5.073) Note:  -0.400 (-1.826)  P P , P , and GNP, are the logarithms of the price of the indicated wood product, the average price of other imported wood products and Japanese domestic logs combined, the monthly average wage index in Japan, and per capita GNP in Japan, respectively. Numbers in parentheses are t-values. h  0  E  Gaston  Page 92  Chapter Five  Table 5.2 (Cont.)  Constant  Estimates of the Japanese demand for selected disaggregated wood imports. Pi  GNP  R -Adj.  Durbin's h  -0.622 (-2.456)  0.655 (2.709)  0.819  0.903  0.906 (2.631)  -0.355 (-1.853)  0.773 (1.985)  0.791  0.127  1.939 (1.881)  -0.800 (-2.067)  1.283 (3.929)  0.613  0.505  -0.553 (-1.637)  1.098 (1.986)  0.788  0.816  Po  P  1.121 (2.301)  E  2  Hemlock logs from the US 10.363 (4.468)  -0.388 (-2.473)  Douglas fir logs from the US 3.290 (6.309)  -0.554 (-2.936)  Softwood logs from Canada 2.333 (8.300)  -2.318 (-1.709)  Softwood logs from the NZ/Chile 11.255 (3.901)  -3.015 (-2.190)  3.093 (2.361)  Softwood logs from the former Soviet Union 5.100 (1.994)  -0.109 (-0.029)  0.666 (1.101)  -0.839 (-0.934)  1.459 (2.069)  0.512  1.233  2.447 (4.938)  -1.143 (-5.046)  0.769 (2.349)  0.876  0.311  -0.672 (-1.632)  1.693 (5.519)  0.954  0.421  -0.604 (-2.310)  0.723 (2.419)  0.820  0.808  -0.887 (-0.632)  1.481 (1.195)  0.883  0.866  Softwood logs from "Other" 9.112 (4.373)  -1.747 (-7.419)  Hardwood lumber from all sources 13.989 (9.526)  -1.158 (-4.477)  1.685 (3.685)  Hardwood logs from all sources 12.875 (9.709)  -0.0827 (-0.631)  0.675 (1.913)  Panel products from all sources 14.139 (-3.400) Note:  -2.531 (-3.965)  2.334 (2.360)  Pj, P , P , and GNP, are the logarithms of the price of the indicated wood product, the average price of other imported wood products and Japanese domestic logs combined, the monthly average wage index in Japan, and per capita GNP in Japan, respectively. Numbers in parentheses are t-values. 0  E  Gaston Chapter Five  Page 93  expectations in terms of sign, and the majority of the parameter values are significant. Finally, tests for structural change were again rejected in each case. It is interesting to note the differences in own-price elasticities based on product (with imported log demand being highly inelastic, panels being rather elastic and lumber lying somewhere in-between), source (for example, demand for U S softwood logs being highly inelastic and softwood logs from "other" being elastic), and species (for example, the demand for yellow cedar lumber from Canada being highly inelastic and "planned" lumber being quite elastic). The significance of this variation will be returned to in Section 5.3, as well as in Chapter 6.  5.2  Estimation of the Armington Two-Stage Model of the Japanese Demand for Total Wood Imports Following Chou and Buongiorno's (1983) approach for estimating a two-stage  demand system, the constant elasticity of substitution of the product imports is determined first.  Table 5.3 shows the result of three models (equation 2.18), utilizing various  aggregates of Japanese imports. In the first case, wood products are broken down by major product type (softwood logs, softwood lumber, hardwood logs, hardwood lumber and aggregated wood-based panels).  In the second case, the investigation is limited to  softwood lumber imports by country of origin (Canada, United States, New Zealand, the former Soviet Union, and "other" countries). Finally, in the third case the investigation is limited to softwood lumber imports from Canada by species (yellow cedar, Sitka spruce, hemlock, Douglas fir, "planed lumber", and "other" species).  In all three systems of  Page 94  Gaston Chapter Five  Table 5.3  Estimates of the Constant Elasticity of Substitution over Varying Degrees of Wood Import Aggregation, Correcting for Serial Correlation.  Wood Prod uct by Ty pe  ln( °  = o In  S L M  Q.  ( b  \  - o In  SLM  U  ) (  a  j-SLG,HLG,HLM,PAN  V  < j  b  ^SLM 1 k  -1.456 (-2.3211)  +u  S L M  S  L  k s L M 1 ^PAN  G  -0.1553 (-2.583)  -0.1484 (-2.233)  0.5342 (2.622)  0.3705 (3.380)  Sof twood L umber by Sourc e Q  In  StM  = o In  u s  SLM  B  - o In  US  SLM  P  + u  US  j=SLM ,SLM ,SLM ,SLM CAN  FSU  NZ  0TH  J  P  o  k s L M _ U S ' ^SLM_CAN  bsLM_US 1 ksLM_FSU  -0.633 (-3.953)  -0.4619 (-4.3465)  2.4511 (3.9532)  bsLM_US 1 b  S  L  1.7920 (2.888)  M  N  Z  k s L M J J S ' ^SLM_OTH  2.0612 (14.976)  Car ladian Softwood Lumber b y Specii3S ( 0  In  - o In  4  Q.  o -0.4772 (-2.1768)  /  ( h  \  °SLM  0TH  {  i  b  J  - o In  ( PSLMQJH ] {  j  p  j  + u  j=sitka,cedar,hemlock,Doug-fir,planed  ^SLM.Other 1  ksLM_Other '  bsi_M_Other '  bsLM_Other '  ksLM_Other '  ^SLM_sitka  ^SLM_Y. cedar  bsLM_hemlock  b s L M _ D . Fir  ^SLM_"planed-  -1.5719 (-2.6992)  -1.8048 (-4.3174)  -0.9725 (-1.1131)  -2.7911 (2.6315  -3.0696 (-9.1517)  equations, the estimates are corrected for first-order autocorrelation, and the values of the elasticity of substitution (o) are constrained to be equal across equations. The systems are estimated using Zellner's seemingly unrelated regression method (Judge, et al., 1988). In each case, the elasticity of substitution has the expected sign and is significant. A s these values are all significantly different from zero, the component parts of the Japanese imports are indeed substitutes rather than complements. A s the values are not  Page 95  Gaston Chapter Five  very large, however, the alternative imports are also far from being perfect substitutes for one another. While there are no expectations on the sign of the constant terms, it can be noted that, with one exception, all of the parameters are significant. It should be noted at this point that the appropriate test on the equality on the elasticities of substitution is rejected in all three cases.  restriction  The consequences of  employing this restriction will be returned to in the next section. A s shown in Chapter 2, the b, weights needed to calculate the C E S quantity and price indices can be recovered from the constant values reported in Table 5.3, combined with the knowledge that the sum of the b.'s must sum to one (five equations with five unknowns for the first two systems, and six equations with six unknowns for the third system). Table 5.4 shows the resulting weights for each of the three systems. Combining these weights with the constant elasticity of substitution (o) yields the C E S quantity and price indices needed to estimate the aggregate Japanese demand for wood imports, repeating the process for each of the aggregation levels reported in Table 5.3. The results of the regressions appear in Table 5.5. A s is the case in the regressions reported in Tables 5.1 and 5.2, the explanatory P  0  variable represents the price of all possible alternatives to the independent variable. In the first regression reported in Table 5.5, this is the average real price of Japanese domestic logs. In the second two regressions, P is the average real price of combined Japanese 0  domestic logs and all other imports besides aggregate softwood lumber and softwood lumber from Canada, respectively. A s before, the Japanese wage index (P ) and the per E  Page 96  Gaston Chapter Five  Table 5.4  Calculated CES Weights {b's) from Table 5.3  Wood product by type D  SLM  0.213  bsLG  bhtLG  b|HLM  bpAN  0.238  0.236  0.165  0.148  bcAN  bpsu  b  boTH  0.364  0.100  0.120  0.121  Softwood lumber by source b  u  s  0.295  N  Z  Canadian softwood lumber by species boTHER  bsiTKA  by-CEDAR  bhlEMLOCK  bnouG_FlR  bpLANED  0.001  0.026  0.042  0.600  0.001  0.330  Table 5.5  Cochrane-Orcutt Estimates of the Japanese Demand for Selected Aggregations of Wood Imports, Utilizing CES Quantity and Price Indices.  Constant  P| (CES)  Po  PE  GNP  R -Adj.  Durbin's h  -0.973 (-4.091)  1.185 (7.717)  0.951  0.436  1.569 (6.063)  0.948  0.871  0.831  0.629  2  Aggregate imports of all wood products 8.733 (4.844)  -0.122 (-0.896)  1.390 (2.701)  Aggregate imports of softwood lumber form all sources 11.909 (5.868)  -0.768 (-1.972)  1.386 (2.267)  -0.836 (-3.478)  Aggregate imports of all species of softwood lumber from Canada 10.116 (8.333) Note:  -1.299 (-2.068)  1.927 (2.009)  -0.901 (-2.828)  1.431 (5.073)  P,,P ,P , and GNP, are the logarithms of the price of aggregate wood imports (logs, lumber and panels), the price of Japanese domestic logs, the monthly average wage index in Japan, and per capita GNP in Japan, respectively. Numbers in parentheses are t-values. D  E  capita gross national product (GNP) are common to all three regressions. Cochrane-Orcutt estimates are obtained to correct for serial correlation. When comparing these results with those of Tables 5.1 and 5.2, it can be noted that  Gaston  Chapter Five  Page 97  the regressions again offer good fits, with correlation coefficients which are of a similar magnitude. The signs are as expected in each case, and the parameters are mostly significant (with the notable exception of the price parameter of the aggregate import of all wood products). It is interesting to note that in each case, the own-price elasticity of the aggregate import is somewhat lower than those reported in Tables 5.1 and 5.2. This is not consistent with the results reported by Chou and Buongiorno (1983) for the U S imports of plywood, where their regressions employing C E S quantity and prices indices are slightly higher than when using arithmetic means. With the estimates of the aggregate import demand elasticity and the constant elasticity of substitution for the component parts within each system, it is now possible to calculate the individual own- and cross-price elasticities. These are reported in Table 5.6 for the first aggregation, being all wood product imports broken into types (softwood lumber, softwood logs, etc.). It can be noted from the table that all of the own-price elasticities (the diagonal from upper left to lower right) have the expected negative sign, with both softwood and hardwood logs reflecting slightly inelastic demand and softwood lumber, hardwood lumber and panels reflecting elastic demands. The cross-price elasticities also have the expected signs, and are less than unitary in all cases. Note that the cross-price elasticities indicate a greater willingness to substitute lumber and panels in response to increases in the price of logs than the other way around. Table 5.6 also breaks out the substitution and market expansion effects of the  Gaston  Chapter Five  Table 5.6  Page 98  Calculated own- and cross-price elasticities of demand for the Japanese imports of all wood products by type. SLG  HLG  SLM  HLM  PAN  SLG  Sub X  -0.888 -0.048  0.552 -0.046  0.213 -0.018  0.063 -0.005  0.061 -0.005  HLG  Sub X  0.568 -0.048  -0.904 -0.046  0.213 -0.018  0.063 -0.005  0.061 -0.005  SLM  Sub X  0.568 -0.048  0.552 -0.046  -1.243 -0.018  0.063 -0.005  0.061 -0.005  HLM  Sub X  0.568 -0.048  0.552 -0.046  0.213 -0.018  -1.393 -0.005  0.061 -0.005  PAN  Sub X  0.568 -0.048  0.552 -0.046  0.213 -0.018  0.063 -0.005  -1.395 -0.005  0.390  0.379  0.146  0.043  0.042  Shares  * S L G , H L G , S L M , H L M , P A N refer to softwood logs, hardwood logs, softwood lumber, hardwood lumber, and aggregate wood-based panels, respectively, * The elasticity of substitution (o) = 1.456 * The elasticity of the Japanese demand for aggregate imports (3) = 0.122 * Shares (S) are calculated by value at the mean. * Own-price elasticity of each product type is calculated as (1-S|)o + Sj (3 * Cross-price elasticity between product types is calculated as Sj o - Sj (3 * Sub = substitution effect; X = market expansion effect  calculated elasticities, as discussed in Chapter 2.  A s the own-price elasticity of the  aggregate demand of all wood product imports (B., in equation 4.25) is extremely low, however, the market expansion effect is always negligible. Table 5.7 shows the calculated own- and cross-price elasticities of demand when imports are defined as softwood lumber by country of origin. The own-price elasticities are represented by the sum of the two values in each cell (the addition of the substitution and market expansion effects), and are again of the expected sign. All values reflect inelastic demand, with little variation by source of the softwood lumber. The signs of the cross-price elasticities, on the other hand, are not what one might initially expect. In all cases, the sum of the two values (again, the substitution and market  Gaston  Page 99  Chapter Five  Table 5.7  Calculated own- and cross-price elasticities of demand for the Japanese imports of softwood lumber by country of origin. Sub X  SLM -0.318 -0.382  Sub X  SLM  C a n  SLM  C a n  SLM  U S  SLM  N Z  SLM  R U  SLM  0 T H  * * * *  SLM  N Z  SLM  R U  SLM  0 T H  0.023 -0.028  0.016 -0.020  0.071 -0.086  0.315 -0.382  -0.425 -0.252  0.023 -0.028  0.016 -0.020  0.071 -0.086  Sub X  0.315 -0.382  0.208 -0.252  -0.610 -0.028  0.016 -0.020  0.071 -0.086  Sub X  0.315 -0.382  0.208 -0.252  0.023 -0.028  -0.617 -0.020  0.071 -0.086  Sub X  0.315 -0.382  0.208 -0.252  0.023 -0.028  0.016 -0.020  -0.562 -0.086  0.498  0.328  0.036  0.026  0.112  Shares * SLM  US  0.208 -0.252  , S L M , S L M , S L M , and S L M refer to softwood lumber from Canada, the U S , New Zealand, the former Soviet Union, and "other" countries, respectively The elasticity of substitution (o) = 0.6331 The elasticity of the Japanese demand for imports of all softwood lumber ((3) = 0.768 Shares (SJ are calculated by value at the mean. Own-price elasticity of each product type is calculated as (1-S;)o + S; (3 C a n  U S  N Z  R U  0 T H  * Cross-price elasticity between product types is calculated as Sj a - S (3 s  * Sub - substitution effect; X - market expansion effect  expansion effects) is negative; without investigation of the component parts one comes to the conclusion that softwood lumber from different sources are complements rather than substitutes. This is shown by the positive substitution effects not to be the case, however. The negative net effect occurs because the market expansion effect counteracts the substitution effect in all cases, as the constant elasticity of substitution is smaller in value than the elasticity of demand of the aggregate import of softwood lumber from all sources. The result of this "cancelling" effect is that the cross-price elasticities are very low in all cases. Finally, Table 5.8 shows the calculated own- and cross-price elasticities of demand  Page 100  Chapter Five  Gaston  Table 5.8  Calculated own- and cross-price elasticities of demand for the Japanese imports of Canadian softwood lumber by species. SLM  SLM  C 2  SLM  C 5  SLM  C 6  SLM  C 7  SLM  C 9  SLM  C 0  C 5  SLM  C 6  SLM  C 7  SLM  C 9  SLM  C 0  Sub X  -0.424 -0.145  0.075 -0.203  0.229 -0.623  0.022 -0.061  0.015 -0.039  0.083 -0.227  Sub X  0.053 -0.145  -0.403 -0.203  0.229 -0.623  0.022 -0.061  0.015 -0.039  0.083 -0.227  Sub X  0.053 -0.145  0.075 -0.203  -0.248 -0.623  0.022 -0.061  0.015 -0.039  0.083 -0.227  Sub X  0.053 -0.145  0.075 -0.203  0.229 -0.623  -0.455 -0.061  0.015 -0.039  0.083 -0.227  Sub X  0.053 -0.145  0.075 -0.203  0.229 -0.623  0.022 -0.061  -0.463 -0.039  0.083 -0.227  Sub X  0.053 -0.145  0.075 -0.203  0.229 -0.623  0.022 -0.061  0.015 -0.039  -0.394 -0.227  0.112  0.157  0.480  0.047  0.030  0.175  Shares  S L M , S L M , S L M , S L M , S L M , and S L M refer to Canadian Sitka spruce lumber, yellow cedar, hemlock, Douglas fir, "other" species, and "planed", respectively. The elasticity of substitution (o) = 0.477 The elasticity of the Japanese demand for imports of softwood lumber from Canada (3) = 1.299 Shares (S) are calculated by value at the mean. Own-price elasticity of each product type is calculated as (1-S|)o + S| 3 C 2  * * * *  SLM  C 2  C 5  C 6  C 7  C 9  C 0  * Cross-price elasticity between product types is calculated as Sj o - Sj 3 * Sub = substitution effect; X - market expansion effect  when focussing on the Japanese imports of Canadian softwood lumber by species. The signs are again as expected, being negative for both the own- and cross-price elasticities, noting once again that the market expansion effect in the latter is more than offsetting the substitution effect.  5.3  Comparison of the Two-Stage and Direct Estimates of the Own-Price Elasticities of Demand for Wood Product Imports in Japan It can be noted that there exists considerable difference between the own-price  elasticities reported in Tables 5.6 through 5.8 as compared to the values reported in Table  Gaston  Chapter Five  Page 101  5.2. Beginning with the disaggregations by product type in Table 5.6, being the estimates from the two-stage analysis, the own-price elasticities are shown to range from -0.888 for softwood logs to -1.395 for panels. This compares to the direct estimates from Table 5.2, where the range is from -0.090 for softwood logs to -2.531 for panel products. While the overall grouping is largely maintained, with more processed products being more elastic, with the two-stage process this is only by degree; in fact the elasticity estimate for panel products is shown to be indistinguishably different from hardwood lumber. A s regards the cross-price elasticities generated from the two-stage model, Table 5.6 shows that there appears to be a greater willingness for the Japanese buyer to substitute more processed products for less processed products than there is for the other way around. For example, while a price increase in softwood logs leads to a significant substitution with panel products, a price increase in panel products leads to minor substitution with softwood logs. A possible explanation for this is that as the price of imported panel products goes up, Japanese log processors recover a higher proportion of these products relative to lumber (i.e., substitute domestically produced panel products in response to price increases in imported panel products). A s a more general observation on the cross-price elasticities shown in Table 5.6, note the expected result of identical increases in the quantity demanded of, for example, hardwood logs, softwood and hardwood lumber, and panels in response to a price change in softwood logs. This somewhat counterintuitive result is, of course, a direct consequence of imposing an equal elasticity of substitution across product types. Comparisons of the direct and two-stage estimates of the own-price elasticities of  Gaston  Page  Chapter Five  102  softwood lumber by source and Canadian softwood lumber by species show even greater inconsistencies. In both cases, it is seen that the range in elasticity values is significantly narrower with the two-stage estimation procedure, suggesting that Japanese demand response to price changes does not vary dramatically by either source of the product or species. This is not what is implied by the direst estimation results. Given the assumption of the constant elasticity of substitution which underlies the two-stage procedure, however, this result is again not surprising. After all, this assumption states that the Japanese wood buyer views the substitution of softwood lumber from New Zealand for softwood lumber from the U S , for example, exactly the same as the buyer views substituting Canadian lumber. Given that softwood lumber from New Zealand has historically been considered to be of only packaging quality, such an assumption may indeed be far too restrictive. A s a reminder, the appropriate tests on the equality restriction of the elasticity of substitution (Section 5.2) was rejected. The Japanese import data used in this study does not support the notion of a constant elasticity of substitution across all substitution possibilities investigated. Ironically, this means that the procedure suggested by Armington (1969), and later forwarded by Chou and Buongiorno (1983), largely negates what Armington set out to accomplish to begin with.  This was to show that the assumption of homogeneous  "commodity" demand is potentially over-restrictive.  It would appear from the results  presented here that what is being suggested by Armington for dealing with this restriction  Gaston  Page 103  Chapter Five  ends up being self-defeating . 21  This is undoubtably what prompted Hseu and Buongiorno (1992), who also found that their data rejected the equality restriction on the elasticity of substitution, to pursue their research. A s was discussed in some detail in Chapter 2, however, their findings can only be considered empirical; that is, it is not supported by theoretical foundations. A s a matter of interest, a similar procedure to that followed by Hseu and Buongiorno was applied to the Japanese import data used in the present study. This is of interest as the results closely parallel the findings of Hseu and Buongiorno, and the direct own-price elasticity estimates shown in Table 5.2. While these results are not presented for reasons already outlined, it does reinforce the need for further research which addresses the noncommodity nature of wood products.  5.4  Non-Wood Substitution in Japan A s the final set of results, this section presents direct elasticity of demand  estimates for non-wood materials, with corresponding estimates for the cross-price elasticity for aggregated wood imports in Japan. It was shown in Chapter 3 that Japan has made growing use of non-wood materials in its construction of homes. This trend is emphasized by Figure 5.2, showing the non-wood housing starts over the study period. While the substitution of non-wood materials for wood is clear from the figure, quantifying the trend is not without its problems. Inclusion of price indices for iron/steel,  The results obtained by Chou and Buongiorno (1983) for the US imports of hardwood plywood support this finding, with little variation found in the own-price elasticities by country of origin. 21  Page 104  Gaston Chapter Five  Figure 5.2  Number of Non-Wood Housing Starts in Japan (Japanese Ministry of Construction)  ceramics/stone/clay or plastic products, either together or separately, tended to be associated with insignificant parameter values, possibly due to multicollinearity of these data with the import and/or domestic wood price data. Further, as the collection of a detailed time series on the quantities and prices for non-wood materials was beyond the scope of this study, this problem could not be circumvented with the two-stage approach employed in this study. Table 5.5 shows the same regression reported in Table 5.1, but with the addition of a price index for iron/steel. The parameter on this non-wood "substitute" comes as  Page 105  Gaston Chapter Five  Table 5.9  Cochrane-Orcutt estimates of the Japanese demand for aggregated wood imports, with the inclusion of a non-wood regressor.  Constant  Pi  PD  PE  13.305 (11.500)  -0.0062 (-0.0313)  0.879 (2.364)  -0.621 (-2.590)  Note:  Psteel  -0.312 (-0.958)  GNP  R -Adj.  Durbin's h  0.923 (4.719)  0.815  0.172  2  . P,, P , P , P and GNP, are the logarithms of the price of aggregate wood imports (logs, lumber and panels), the price of Japanese domestic logs, the monthly average wage index in Japan, the price index of iron/steel in Japan, and per capita GNP in Japan, respectively. Numbers in parentheses are t-values. D  E  steel  somewhat of a surprise, as it has a negative sign (although the parameter is not significant). This suggests that iron/steel acts as a complement to wood in the Japanese output of products (as measured by the per capita GNP), rather than as a substitute. Aside from the possibility of multicollinearity among these regressors (i.e., that the iron/steel parameter is not meaningful due to very high variances), one would have to conclude that for the Japanese output which requires wood as an input (which is primarily housing construction in Japan), iron/steel is also desired in combination.  5.5  Summary This chapter has presented the results of two econometric analyses. The first was  the direct estimate of the demand elasticities for Japanese imports of various wood products inputs, including various levels of aggregation. The second was the results of the two-stage approach initially suggested by Armington in 1969, with three levels of aggregation (all wood imports by type, softwood lumber imports by source, and Canadian softwood lumber imports by species). It was shown that the own-price elasticity estimates mostly meet a priori  Gaston  Chapter Five  P a g e 106  expectations in terms of sign and relative magnitude. The cross-price elasticities derived from the two-stage methodology also met a priori expectations in terms of their signs. It was found that with the exception of the first case, where imports of all wood products are broken into major product categories, that the net cross-price effects tend to be very small, with any substitution effect being counteracted by a market expansion effect. However, it was also shown that considerable difference exist between the direct and the two-stage own-price elasticity estimates.  This is attributed to the required  restriction of a constant elasticity of substitution across all input pairs being investigated in the two-stage estimations. Finally, it was shown that individual commodities within each level of aggregation, whether by product type, source or species act as distinct economic units. This is made evident in the direct estimates of the price elasticities of demand through the large variation in the own-price demand elasticities, and confirmed by the rejection of the test on the equality of the elasticities of substitution in the two-stage model. In the following chapter, the econometric results presented here are discussed in the context of Japanese wood product import trends from 1965 to 1993. While the theoretical foundation of the two-stage approach was found to be sound, the needed restrictions cast considerable doubt as to its practical application.  A s a result, the  discussion will limit itself to the direct own-price elasticity estimates offered in Table 5.2, and a few general observations regarding cross-price elasticities from the two-stage results.  Gaston  Page 107  Chapter Six  Chapter 6 Discussion of the Results  Considerable detail has been offered in the previous chapters on the analysis of forest product import substitutions by Japan. This included a review of the North American literature on wood product substitutions (Chapter 2), a description of the Japanese forest industry and wood usage trends (Chapter 3), the theoretical foundations for quantifying Japanese wood import demand price elasticities (Chapter 4), and, finally, a description of the empirical results (Chapter 5).  In this chapter, the relevance of these results is  discussed. To facilitate this discussion, it will be helpful to visualize the Japanese import quantities and corresponding real price trends, starting with the most aggregated Japanese import basket, and moving to the least aggregated. While import quantity trends were summarized in Chapter 3, the simultaneous consideration of real price trends helpful in understanding substitution effects.  22  will be  To this end, Figures 6.1 through 6.16  graphically summarize the Japanese Tariff Association data used in the econometric analysis.  6.1  Japanese Wood Product Imports, Aggregated by Product Type Starting with Figure 6.1, showing the volume and corresponding real prices of the  Japanese imports by major product types, a number of general observations can be made.  As discussed in Chapter 4, real prices were generated by adjusting nominal prices to 1993 dollars, using the Japanese GNP deflator. 22  Gaston  Chapter Six  Figure 6.1  Page 108  Japanese imports of wood products by major product types; volume and real value. SLM, HLM, SLG, HLG, and VS/PLY/PART/FIB refer to softwood lumber, hardwood lumber, softwood logs, hardwood logs, and combined panel products, respectively.  Gaston  Page 109  Chapter Six  First and foremost, with all of the categories of wood products, in aggregate, with the possible exception of hardwood logs, Japanese importers were paying less (real Yen per m ) for their imports in 1993 than they were in 1965. When one considers the strength of 3  the Japanese Yen relative to the currencies of any of the major fibre suppliers discussed in this study, this becomes clear. For example, in 1965 the aggregate price of softwood lumber is shown in the figure to be roughly 55 thousand Yen per m , and roughly 40 3  thousand Yen per m  3  in 1993, real. In terms of Canadian dollars, these two values are  roughly $170 per m and $500 per m , respectively. This growth in the purchasing power 3  3  of the Yen is very important to the interpretation of the results of this study, and will be revisited when discussing implications for the B C forest industry in Chapter 7. A further observation regarding the prices of the aggregated fibre imports is that they are not identical, neither in magnitudes nor in their precise trends overtime. Clearly, hardwood lumber has seen the largest price premium over other fibre categories, and the largest price volatility. This is followed by veneer sheets and other panel products and then by softwood lumber.  Finally, the lowest real price per m is shared by softwood and 3  hardwood logs. It can also be seen from the figure that softwood lumber and softwood logs share a similar price pattern. This is not too surprising given that both are primarily used for housing construction in Japan (for both structural and appearance purposes). That softwood lumber trades at a premium to softwood logs is also not too surprising, given the higher waste and cost associated with the Japanese processing of log imports as compared to lumber imports. In other words, the derived demand for logs, after subtracting all other costs associated with building a house, for example, is logically lower than for  Gaston Chapter Six  Page 110  lumber. When comparing hardwood lumber with hardwood logs, the price patterns are not as similar as for softwood, and the price premium for lumber over logs is much wider. Unlike the softwood comparison, these two products do not tend to go into the same end product. Hardwood lumber, as evidenced by its price premium, is prized for its appearance qualities. While this appearance quality is primarily exploited for furniture, it is used in housing to a smaller degree, mainly as panelling and trim. Hardwood logs, on the other hand, are largely used by Japan's sizeable plywood industry. A s regards the corresponding volumes imported, it can be seen that these data confirm the trends discussed in Chapter 3. First, there was a strong increase in volume of softwood and hardwood log imports from 1965 to 1973, partially due to substitution of imports for domestic production, and partially due to rapid increases in Japanese housing starts. This was followed by a period of flux from 1974 to 1981, when Japan witnessed a decline in economic activity and a corresponding decline in the number of wooden housing starts. Finally, there were slightly increasing and declining volumes for softwood logs and hardwood logs, respectively, from 1982 on, corresponding to small increases in wooden housing starts and export controls on hardwood logs by two major producers, respectively. Softwood lumber imports show a gradual increase throughout the study period, as does hardwood lumber on a smaller scale. This shows the gradual trend of substitution of lumber for logs. Panel imports, on the other hand, became prominent only from the early 1980s on, contributing to Japan's import substitution away from logs. Relating this visual picture to the elasticity results presented in Chapter 5 yields few  Gaston  Chapter Six  Page 111  surprises. Referring back to the results of the direct estimations for own-price elasticities for selected imports (Table 5.2), it was shown that both softwood and hardwood logs reflect highly inelastic demand in aggregate (in fact, not significantly different from zero), compared to a moderately inelastic value for softwood and a slightly elastic value for hardwood lumber. This suggests that there are more substitutes for imported lumber than there are for imported logs. A s mentioned in the previous chapter, this makes sense as Japan can substitute both domestic lumber and domestic logs for imported lumber. Further, domestic lumber can be made from both domestic and imported logs. Imported logs, on the other hand, can only be directly substituted with domestic logs. The panel products are shown to have a quite elastic demand, which when one extends the substitution possibility logic just presented, also makes common sense. Bearing in mind the limitations of the two-stage results presented in the previous chapter due to the restrictive nature of the constant elasticity of substitution assumption across all product pairs, it is interesting to note that there appears to be a greater willingness for the Japanese buyer to substitute more processed products for less processed products than conversely (see Table 5.6). For example, while a price increase in softwood logs leads to a significant substitution with panel products, a price increase in panel products leads to minor substitution with softwood logs. A possible explanation for this is that as the price of imported panel products goes up, Japanese log processors recover a higher proportion of these products relative to lumber (i.e., substitute domestically produced panel products in response to price increases in imported panel products).  Gaston  6.2  Chapter Six  Page 112  Japanese Softwood Lumber Imports, Aggregated by Source Figure 6.2 offers a graphical representation of the next level of disaggregation of the  Japanese Tariff Association import data: total softwood lumber imports by county of origin. Section 6.3 follows with a discussion of softwood lumber by species. This is then repeated in subsequent sections for hardwood lumber, softwood and hardwood logs, and panel products by country of origin, and, where appropriate, by species. The greatest detail is offered for softwood lumber, due to its importance to the B C forest industry. Starting once again with general observations on the price levels and trends, it can be seen that there are three basic groups. Going from least to most expensive, there is the former Soviet Union and New Zealand/Chile group, the Canada/US group, and the "other" group.  Once again it can be noted that all groups have witnessed real price  declines over the period of the study. Further, there was a slight widening between the first two groups over time, and the "other" group saw the largest price drop over time, as well as the largest price volatility. A s regards this last point, softwood lumber imports falling into the "other" category originated originally from Asia, and later from the South Seas, Asia, and Scandinavia. It is unclear why there was such a large price premium for softwood lumber from these sources. One possible explanation is that Japan was purchasing a particularly high quality softwood.  Another possible explanation is that these lower  volumes represent marginal purchases above and beyond Japan's main supply of softwood lumber, and a resulting willingness to pay more. A s regards the difference observed between the North America group and the New Zealand/Chile/former Soviet Union group, the lower prices indicated for the latter are  Gaston  Chapter Six  Page 113  Gaston  Page 114  Chapter Six  intuitively explained by quality differences. When one considers that North American lumber over most of this time period largely represented product from old growth P N W and B C forests, it may be a source of some surprise, in fact, that there was not a larger price spread than that shown. The average landed price in Japan for softwood lumber from Canada was roughly C D N $490 per m  3  in 1993, compared to C D N $275 per m from New 3  Zealand/Chile. It must be stressed, however, that these are averages. While this average likely represents a relatively small range of prices in the case of New Zealand/Chile, representing primarily radiata pine and, historically, of only a low grade, the North American average is over a wide range of species and grades. A s the question of species will be dealt with below, discussion is reserved to that point.  To give a sense of the  variability at this point, however, the 1993 value of Japanese imports of yellow cedar exceeded $ C D N 950 per m , which is again a potentially misleading average across 3  grades. The price of the highest grade of yellow cedar lumber reported by a leading B C producer/exporter in 1995, " C " clear and better, was nearly $2,700 per m3, f.a.s. , 23  recognizing that this is still aggregated over all sizes within this grade. A s a final comment on the price comparisons shown, it comes as some surprise that from 1991 on there was a price premium for softwood lumber imported from the U S over that from Canada. Even when aggregated over grades, it would have been expected that the price premium should have belonged to Canada. While most of the Canadian lumber shipped to Japan is still obtained from old growth timber, the U S have shipped significantly  Note that comparison of this value with the landed values in Japan reported in this study would require the addition of loading and transportation costs. 23  Gaston  Page 115  Chapter Six  lower quantities of old growth lumber for quite some time. At least a partial explanation for this observation comes from the fact that planed lumber in the 1990s made up a larger component of Canada's market mix to Japan than that from the U S . In 1993, the planed share for Canada and the U S was 8.14% and 4.35% of all wood products imported by Japan (logs, lumber and panels), respectively. This is an interesting point, one that will be returned to later, especially given that the average share of planed lumber over 1965 to 1993 for the two countries is virtually identical . 24  In terms of the market share of softwood lumber by exporting country, it can be seen that in spite of year to year differences, the overall trend has stayed relatively constant, with the exception of Canada's gain in share at the expense of the U S after 1989. Canada has maintained its position as the largest softwood lumber supplier to Japan, followed by the U S and distantly followed by other sources. The direct own-price elasticity estimates from Chapter 5 (Table 5.2) do suggest that the level of substitutability does vary for softwood lumber by source. The demand for Canadian softwood lumber is slightly inelastic, the demand for U S and "other" softwood lumber moderately elastic, and the demand for New Zealand lumber is quite elastic. The estimated value for the former Soviet Union, being a positive value of 0.045, is an unexpected result that cannot be explained beyond the comments made in the previous chapter.  When referring to planed lumber in this thesis, reference is being made to the Japan Tariff Association's category termed "planed, grooved or tongued" before 1988, and "S-P-F" plus "planed and sanded, n.e.s." from 1988 on. It is assumed that this corresponds to North America's loosely defined "kiln dried, dimension lumber", as opposed to green lumber going to Japan, being primarily squares and baby squares. 24  Gaston  Chapter Six  Page 116  As mentioned in the previous chapter, the calculated cross-price elasticities from the two-stage estimations for softwood lumber by source (Table 5.7) indicated the unexpected result of a negative net value. This is attributed to the non-zero elasticity of demand for softwood lumber in aggregate, in this case yielding a market expansion effect that more that reverses the substitution effect. For example, a 10% increase in the price of softwood lumber from Canada leads to a 3.15% increase in the demand for softwood lumber from any other country (the same increase once again due to the assumption of a constant elasticity of substitution across all pairs of countries). Counteracting this, however is a negative market contraction effect of 3.82%, leading to a net cross-price effect of - 0.067. The results against price increases from any country are similar, with very small cross-price effects between countries. Given the constant elasticity of substitution across pairs of lumber sources, the small cross-price values do not come as a great surprise. It suggests that Japanese buyers are not apt to make significant adjustments to their import mix of softwood lumber by source due to relative price changes, but rather substitute with other product types and/or their own domestic production. Such substitution is clearly indicated in Tables 5.1 and 5.2. Even without such a strong assumption as that imposed in the two-stage model, low crossprice elasticities could be expected for similar reasons.  6.3  Japanese Softwood Lumber Imports from Canada, Aggregated by Species Turning to Figure 6.3, detail is offered on the Japanese import of softwood lumber  from Canada by species. Note that this level of detail reveals that not all species had  Gaston  Chapter Six  Page 117  Gaston  Page 118  Chapter Six  declining real price trends over the period 1965 to 1993. Both yellow cedar and Douglas-fir showed modest price increases, with the former showing the greatest price volatility. If one restricts the period of observation to 1974 onward, however, only Douglas-fir showed modest price increases over time. Yellow cedar clearly enjoyed the highest price premiums over time. Planed lumber and hemlock lumber occupied the other end of the spectrum, showing the lowest and least volatile prices over the study period, although the small volumes of planed lumber traded at high relative values early in the study period. Sitka spruce, Douglas-fir and "other" lumber prices are positioned somewhere in between. Even at the level of species detail, it is an important observation for this study that considerable price variation exists. The proposition that individual species of softwood lumber act as distinct economic units seems obvious from the figure, and was confirmed by rejection of the equality restriction on the elasticities of substitution. The two most prominent species, in terms of volume, are shown to be hemlock, which has shown fairly consistent levels throughout, and planed lumber, which has seen exponential growth since the mid-1970s. It is also interesting to note that these are the two lowest priced species. The volumes of all other species imported from C a n a d a  25  have gradually risen over  the study period and in recent years have represented very similar volumes. The increased market share of planed lumber in Japan over time has largely been at the expense of the  'The vast majority of Canadian exports from Japan are from BC. See Table 4.2.  Gaston  Chapter Six  Page 119  hemlock share. The direct estimates of own-price elasticities of demand vary considerably over species (Table 5.2). Yellow cedar shows the least elastic values, indicating a lack of willingness of Japanese buyers to reduce quantities purchased in response to price increases (note, too that this species is also the highest priced). The own-price elasticities for Sitka spruce, Douglas-fir and "other" are similar to each other, demonstrating unitary to slightly elastic demand. The most elastic demand is demonstrated by planed lumber, with an elasticity value of -3.591. Curiously, the elasticity parameter on hemlock is not significant. Before moving on to the Japanese imports of softwood lumber from other sources, it must once again be pointed out that the prices shown in Figure 6.3 represent an average over all grades of lumber within a species, and over ranges of size, etc., within a grade. While the range of own-price elasticities noted above largely appeals to a priori reasoning in terms of planed versus non-planed lumber, no further quantification of quality by species can be made from the data used in this study. Qualitatively, the fact that yellow cedar lumber demonstrates an inelastic demand, compared to Douglas-fir's slightly elastic demand, also appeals to a priori reasoning, as it is yellow cedar which is thought to be the "superior" product in terms of its value in appearance usage. However, without trade data by grade, this cannot be properly quantified. For example, the highest grade of Douglas-fir may be more highly valued than the lowest grade of yellow cedar; all that can be said is that yellow cedar on average demonstrates fewer substitution possibilities than Douglas-fir on average, and that the former has historically traded at a premium average value as a  Gaston  Chapter Six  Page 120  result. Further, while it is obvious that all of the species except hemlock traded at average price premiums over planed lumber, not so obvious is that these premiums are understated.  The reason for this understatement is largely technological.  Japanese  buyers have been willing to pay a premium price for lumber which contains clear material for the "appearance effect" used in their traditional post-and-beam construction (Chapter 3). However, this appearance effect can be simulated, thanks to the technology which makes possible the use of thin laminates and glueing. In other words, Japanese builders now need less clear raw material to produce an equivalent number of post and beams, yet they are still willing to pay a price premium for the input. In the absence of this technology, therefore, the price premiums would likely have been considerably higher than those shown in Figure 6.3 . 26  To add a sense of the range of quality that exists within any one species, a brief summary of the 1995 grade distribution of softwood lumber exports to Japan by a leading B C producer/exporter are briefly discussed. Unfortunately, trade data by grade are only available for this firm for 1994 on, disallowing direct comparison to the econometric analysis. Of the 1994 shipments of this coastal B C firm, softwood lumber shipments to Japan were primarily made up of fir and hemlock, with smaller quantities of Sitka spruce, red cedar, yellow cedar and S-P-F. Prices ranged from nearly $2,500 per m to less than $60 3  Although there was no evidence of structural change to support this, it is possible that the acceptance of veneered posts was too slow to register as a break at a particular point in time. 26  Gaston  Page 121  Chapter Six  per m , C D N . A s these prices are aggregated over all sizes, one would also expect a wide 3  range in prices within each of these grades. It is important to note that these data indicate that prices were much more grade specific than species specific.  Even with red cedar and S - P - F , which had the lowest  average prices, prices reached levels of roughly $1,000 and $700 per m , respectively, for 3  the clear grades. These grade data revealed that there was a relatively small volume of clear grade shipments, being roughly 10% by volume and 20% by value in 1994. Shipments of all species and grades (entirely clear) which exceeded $1000 per m represented less than 3  3% of the volume and a little over 8% of the value of this company's export mix. Shipments of lumber at prices exceeding $500 per m represented roughly 17% of the total volume 3  and 32% of the total value. These still mostly represent high grades down to #4 clear and #1 merchantable. Shipments exceeding $400 per m represented roughly 47% and 6 3 % 3  of the total volume and value, respectively, while shipments exceeding $300 per m  3  represented 74% and 85%, respectively. These are largely merchantable grades. The remainder of this company's coastal shipments, at less than 15% by value, represent structural, utility and economy grades. The average price over all grades is roughly $560 per m . 3  There are two important points which emerge from these data in the context of the analysis done in the present study.  First, the variation in price by grade exceeds the  variation in price by species. This suggests that the results of the present study, detailing only species, should be interpreted with caution, and strongly supports the need to extend  Gaston Chapter Six  Page 122  this analysis to the level of grade. Second, these data clearly point out that while the volumes of the very high quality lumber are relatively small, coastal B C shipments include only minor quantities of structural grade lumber. Given that the majority of Canadian lumber exports to Japan are from BC, and the majority of the lumber exports to Japan from the Interior of B C are S-P-F, this suggests that the vast majority of Canadian exports of all non-S-P-F species originate from old-growth forest of the B C Coast are of appearance quality and range from merchantable to clear grades.  6.4  Japanese Softwood Lumber Imports from Non-Canadian Sources, Aggregated by Species To complete the picture of softwood lumber, Figures 6.4 through 6.7 graphically  illustrate the quantity and real price trends for U S , former Soviet Union, NZ/Chile and "other" country imports, respectively. In Figure 6.4, showing softwood lumber imports from the U S , the picture is seen to be similar to that for Canada. In terms of volumes, the U S also saw dramatic increases in planed lumber shipments, although dropping from 1988 on. The U S , however, had higher shipments of Sitka spruce from 1965 to the late 1970s. The major differences in price were for red cedar (there were no significant quantities of yellow cedar lumber reported), which showed dramatic premiums until the mid 1980s, at which point only hemlock and planed lumber were lower priced. Also, unlike Canada, Douglas-fir lumber enjoyed a price premium over Sitka spruce lumber throughout much of the study period. Finally, from 1989 on, "other" lumber held a price premium over all other species.  Once again,  Gaston  Page 123  Chapter Six  2,000,000  Sitka  —•— Hemlock  —a—  Red Cedar  —«— Planed  —»— Other  Doug-fir  160  Figure 6.4  Japanese imports of US softwood lumber by species; volume and real value.  Gaston  Chapter Six  Page 124  this premium is most likely due to the negligible volumes traded, showing a higher willingness to pay on such marginal purchases. The own-price elasticities of demand (Table 5.2) are similar to those shown for Canada, insofar as the demand for cedar is inelastic as compared to a very elastic demand for planed lumber. The Japanese import picture from the former Soviet Union, shown in Figure 6.5, shows that there are only three relevant species, dominated by abies/picea species for most of the study period, followed by pine (except from 1977 to 1988) and larix. The overall volume has remained fairly consistent over time, but the volume by species has been quite volatile. Finally, it should be noted that the overall volume is relatively minor as compared to softwood lumber imports from North America. With the exception of 1979 and 1980, real prices showed a steady decline over the entire study period, with pine and abies/picea species showing a considerable premium over larix. This price decline is more prominent than for lumber from all other countries except New Zealand/Chile. A s mentioned in the aggregate discussion, lumber imported from the former Soviet Union shares New Zealand's position of having the lowest prices compared to other countries. The own-price elasticities by species are not reported in Table 5.2 as there was no improvement in the results over the aggregate regression reported earlier. In Figure 6.6, showing the New Zealand/Chile situation, it can be seen that pine (in this case radiata pine) has been the only significant species imported by Japan historically. The volume consistently increased over time and the real price showed more dramatic  Gaston  Page 125  Chapter Six  140,000  •  Pinus  _ — Abies/Picea  •  Larix  70 -,  Figure 6.5  Japanese imports of former USSR softwood lumber by species; volume and real value.  Gaston  Chapter Six  Page 126  Gaston  Chapter Six  Page 127  Decreases than from any other source of Japanese supply. A s shown in Table 5.2, the demand for New Zealand/Chile radiata pine is quite elastic. Finally, Figure 6.7 illustrates the softwood lumber trade from "other" countries. Note the similarity of the planed volume trend compared to North America, and the sporadic hemlock volume. In terms of price, it can be noted that the overall level is higher than for the countries already discussed, and that the price premium for yellow cedar is even more prominent. Also, the price trends tended to be flatter (ranging from minor to zero real declines over time), and in the case of cedar more positive. The own-price elasticities are different from the countries already discussed in two respects, being highly elastic for hemlock, and inelastic for planed lumber. The source of this planed lumber is almost exclusively Scandinavia and Western Europe (mostly the former). "Other" hemlock is almost exclusively obtained from Asia.  6.5  J a p a n e s e Softwood L o g Imports, Aggregated by S o u r c e Figure 6.8 illustrates the quantity and real price trends for softwood logs by country  of origin. Once again, real prices decline in all cases, with differences in price levels and trends by grouping. The nature of the grouping is identical to the case of softwood lumber, but this time the imports from "other" countries were even more volatile and tended to show smaller price premiums (or after 1989, price discounts). This volatility can most likely be attributed to the small volumes. "Other" logs were primarily imported from Asia in the early part of the study period, and by 1993 from Scandinavia, Asia and Eastern Europe, in that order.  Gaston  Chapter Six  Page 128  400,000  —m— Hemlock  —#*— W7Y C e d a r ^ — Planed  —s— Doug-fir  —e— Other  600  19*65' '  Figure 6.7  ' 19'69'  l_r  i9 73~ ' 1  r—  '19*77*  '  ' 19*81 '  '  ' 19'85'  '  '19'89'  '  ' 19*93  Japanese imports of "other" softwood lumber by species; volume and real value.  Gaston  Chapter Six  Page 129  Gaston  Chapter Six  Page 130  The Canadian dollar equivalents of the Japanese landed prices for softwood logs from North America and the former Soviet Union, respectively, are roughly $440 per m  3  and $150 per m , which shows a somewhat larger spread than that of softwood lumber. 3  The own-price elasticities for softwood logs vary considerably by origin (Table 5.2). The value for the U S indicates a quite inelastic demand, as is the case for the former Soviet Union (although the price parameter is insignificant). The demand is shown to be quite elastic for New Zealand/Chile, Canada and "other" softwood logs. The own-price elasticities of softwood logs imported from both the former Soviet Union and Canada come as a surprise for intuitive reasons, primarily from a quality perspective. While an inelastic value for the Soviet Union may be partially explained by reasons cited earlier (contractual arrangements preventing price from properly explaining quantity demanded), the reason for the elastic value for Canada is more difficult to explain. Given the export restrictions that Canada imposes on its log producers, which require that logs are proven to be in excess of domestic demand before exports are allowed, it is possible that quantities purchased by Japan correspond to periods of log surpluses and resulting depressed prices. In other words, if the shipment of logs to Japan is driven by supply, the market supply assumptions employed in this study would fail to capture this information. This is a potential limitation of this study that cannot be solved with existing data constraints.  6.6  J a p a n e s e Softwood L o g Imports, Aggregated by S p e c i e s Figures 6.9 and 6.10 offer detail on softwood log trade by species for the U S and  Gaston  Chapter Six  Page 131  Page 132  Gaston Chapter Six  the former Soviet Union. Aside from New Zealand/Chile, these two sources account for the vast majority of softwood log imports by J a p a n . 27  Beginning with the U S , there are two general observations which stand out. First, hemlock and Douglas-fir have dominated the log trade. Hemlock had the largest share in the beginning of the study period, while Douglas-fir took over from 1978 on. This is in part due to a preference change by the Japanese market, where only hemlock was initially considered a suitable substitute for their domestic species used for appearance applications. While this was primarily due to the whiteness of hemlock wood, over time Japan began accepting the slightly more yellow Douglas-fir . 28  The second general observation regarding Japanese imports of softwood logs from the U S is that with the exception of yellow cedar, which enjoyed a considerable price premium over the entire study period, all species had very similar price trends and relatively small price spreads. As compared to lumber, therefore, it appears that there is less uniqueness associated with softwood logs by species. This second point is at least partially supported by the quantitative analysis offered in Chapter 5. A s can be seen in Table 5.2, own-price elasticities narrowly ranged from 0.10 to -0.55 (inelastic demand). Figure 6.10 completes the picture for softwood logs, showing the trade summaries by species for the former Soviet Union. Other than a increase in the spread of volume by  As softwood logs from New Zealand/Chile and "other" countries are dominated by a single species, there is no reason to repeat the discussion offered earlier. 27  lt was also the case that the small sawmills, which dominated in the earlier part of the study period, had a distinct preference for hemlock over Douglas-fir. Evidently, this was not the case with the large sawmills, which grew in importance toward the latter part of the study period (Robertson and Waggener, 1995). 28  Gaston  Page 133  Chapter Six  3,500,000  ° 19'65 '  Figure 6.10  '  ' 19'69 '  '  ' 19"73 '  '  ' 19"77 '  '  ' 19'81 '  '  ' 19'85 '  '  ' 19'89 '  1  ' 1993  Japanese imports of former USSR softwood logs by species; volume and real value.  Gaston  Chapter Six  Page 134  species, with abies/picea species remaining dominant, note that there is little difference from the aggregate result presented in Section 6.5. The own-price elasticities by species are not presented in Table 5.2 due to the same lack of significance as for the aggregate.  6.7  J a p a n e s e Hardwood L u m b e r and L o g Imports, Aggregated by S o u r c e Figure 6.11 details Japanese imports of hardwood lumber by country of origin. It  is interesting to note that there was a wide variation in real prices, which gradually diminished overtime. In 1965 there was a dramatic price premium for hardwood lumber from the US, followed by hardwood lumber from "other" countries (primarily Asia, with less amounts from Europe, Central and South America, and Africa), and finally hardwood lumber from the South Seas. Also of interest, hardwood lumber imports from the South Seas are one of the very few products discussed in this study that actually witnessed real price growth in Japan over the study period (albeit minor). Table 5.2 showed a near unitary own-price demand elasticity for hardwood lumber, which is somewhat curious given the price premiums paid by Japan over softwood lumber. A partial explanation for this is found by noting Japan's large imports of hardwood logs (discussed below), which can generate an abundant source of domestic substitution for imported hardwood lumber by changing the lumber recovery. Although there is no lack of data on Japanese imports of hardwood lumber by species as well as source, trade is not detailed here due to the relatively small volume compared to other products and the desire to ultimately focus on implications for the B C forest industry.  Gaston  Page 135  Chapter Six  1,600,000  South Seas  — — US  —«— Other  350  Figure 6.11  Japanese imports of hardwood lumber by county of origin; volume and real value.  Gaston  Chapter Six  Page 136  Unlike hardwood lumber, hardwood logs have been dominated by imports from the South S e a s alone over the study period, as shown in Figure 6.12. Minor quantities at historic price premiums have been imported from "other" sources, which in 1965 was dominated by shipments from the US, followed by Asia, the former Soviet Union and Africa, in that order. By 1993, the "other" category was dominated by the former Soviet Union and Africa, followed by Asia and the U S .  6.8  J a p a n e s e Panel P r o d u c t Imports, Aggregated by S o u r c e The final set of illustrations in this chapter, shown in Figures 6.13 through 6.15,  detail the veneer and panel products trade. It is easy to see that all of these products have gained in importance over the study period, with the South Seas being the most dominant player.  Board products show the lowest volume, and are only significant after 1985. Both veneer sheets and plywood showed decreasing price volatility over time, while  board products had more sporadic pricing throughout (note that there were no fibreboard imports by Japan prior to 1979). A s indicated in Table 5.2, the Japanese demand for panel products in aggregate is quite elastic.  Gaston  Page 137  Chapter Six  30,000,000  19W  '19'69'  '  '19V3'  '  '19'81'  'ifrf  —•— South S e a s  '  '19'85'  '  'l9'89'  '  '1993  —«»— Other  120 -,  19'65 '  Figure 6.12  '  ' 19'69 '  '  ' 19V3 '  '  R  79 77 T  -1  '  ' 19'81 '  '  ' 19'85 '  1  ' 1989 '  '  ' 19'93  Japanese imports of hardwood logs by county of origin; volume and real value.  Gaston  Chapter Six  Page 138  Gaston  Chapter Six  Page 139  Gaston  Chapter Six  Page 140  Gaston  Page 141  Chapter Seven  Chapter 7 Contributions, Limitations and Implications for Further Research  By investigating the Japanese demand for wood imports disaggregated by product, region and species, this research has shown that wood inputs are imperfect substitutes in production. This important finding suggests that hiding wood characteristics through data aggregation potentially obscures important dimensions of both forest product trade and forest policy. This chapter summarizes the results which have led to this observation, including a discussion on the implications of these results for the B C forest industry, followed by the limitations of this study and recommendations for further research.  7.1  Research Contributions and Implications for the BC Forest Industry The primary contribution of this study is to offer some of the needed background  information to develop a more detailed understanding of international wood product trade. To date, there have been very few studies which have investigated the factor demand for wood beyond very broad product categories, such as "softwood logs", or "softwood lumber". It is hoped that this study has succeeded in demonstrating the need to move beyond such limiting product aggregations. A s a background study, no hard conclusions nor recommendations can be made. This task is left for future extensions and additions to the present research. However, there are a number of implications of this study for the B C forest industry which can be discussed, particularly those which relate to the marketing of B C solid wood products. This  Gaston  Chapter Seven  Page 142  discussion is offered below (Sections 7.1.2 and 7.1.3), following a brief summary of the research findings.  7.1.1  Summary of the Results This study began with four primary research hypotheses. The first was that B C  wood species, in the form of logs, lumber or further processed products, behave as distinct economic goods.  A s discussed in Chapters 5 and 6, and summarized below, this  hypothesis is clearly accepted for species of softwood lumber, including imports of softwood lumber from Canada in aggregate compared to imports from other sources. The second hypothesis was that the market share of individual B C products in Japan is dependent on relative prices with substitute products.  This hypothesis is  accepted insofar as it was determined that individual wood product imports are substitutes, albeit imperfect. At the same time, however, while some products were shown to have a high own-price elasticity of demand in Japan, the cross-price effects were shown to be very low. This suggests that Japanese buyers do not adjust their import mix as a result of relative price changes, rather that they adjust their overall level of wood product imports. Due to the noted limitations of the two-stage methodology employed, however, this result should be interpreted with caution. The third hypothesis was that Japan's wood product import mix is effected by its domestic log supply and non-wood alternatives. The first part of this hypothesis is clearly accepted; non-wood alternatives, however, were shown to act as complements, not substitutes.  Gaston  Chapter Seven  Page 143  The final hypothesis was that a structural change over the course of the study period affected the Japanese demand for logs, lumber or further processed products.  This  hypothesis could not be accepted for any of the direct estimates of wood product imports, either for various levels of aggregations or as individual products.  The main conclusions of the analysis offered in this study, discussed in detail in the previous two chapters, are:  1)  Individual wood products, whether aggregated by product type, country of origin, or species, behave as distinct economic units. This was quantitatively extended to include quality insofar as trade detail was available for planed versus non-planed lumber. This was qualitatively extended to quality through the knowledge and informal judgements about the nature of wood products typical of individual sources.  2)  In terms of product type, the own-price elasticity of demand was found to be the smallest for softwood logs, largest for panel products, with lumber's elasticity lying somewhere in between. Said another way, softwood logs displayed the fewest number of substitutes, and panel products displayed the greatest number of substitutes.  3)  In terms of the county of origin for softwood lumber, the own-price elasticity of demand was found to be the smallest for Canada (not including the former Soviet Union, which represented a small proportion of softwood lumber imports), largest for NZ/Chile, with softwood lumber from other sources lying somewhere in between.  4)  In terms of the species of softwood lumber form Canada, planed lumber (such as S-P-F) was shown to have by far the highest own-price elasticity, while yellow cedar showed the smallest.  5)  In the case of softwood logs, the lowest own-price elasticity was for U S logs, and the highest was for NZ/Chile logs. The own-price elasticity for the former Soviet Union has a positive sign, a non-intuitive result which may be explained by the presence of long-term contracts. Compared to lumber, there was little difference by species shown for the own-price elasticities for U S logs.  6)  The cross-price elasticities were found to be highly inelastic, demonstrating a  Gaston  Chapter Seven  Page 144  willingness to substitute one wood product for another to some degree (imperfect substitutes). Generally, it was found that the substitution effect was offset by a market expansion effect. Japanese buyers appear to be more willing to substitute more processed products in response to price increases for less processed products than the other way around. This may reflect changes in the product mix of domestically processed logs. 7)  Increases in the price of Japan's domestic logs cause the Japanese to substitute imported wood products. Given the significance of the Japanese domestic log supply over the course of the study period, symmetry would suggest that increases in the prices of imported wood products lead to significant substitution with domestic logs.  8)  In spite of the increasing purchasing power of the Yen over the period covered by this study, it was determined that Japanese buyers have remained price sensitive, yet have been willing to pay considerable price premiums for certain products.  9)  There was no evidence of structural change in the Japanese market for wood imports.  10)  In Japanese markets, iron/steel and wood inputs appear to be complementary, likely due to the importance of both non-wood housing starts and non-residential construction.  7.1.2  Implications of the Research for BC Wood Product Marketing An obvious application of the results of this study is in BC's choice of markets. While  B C has been a significant source of softwood lumber for the Japanese market over the entire period of this study, a little over a decade ago Canadian softwood lumber exports to Japan as a percent of all markets was only 8% by value, or less than 4% by volume (Canadian Forestry Service, 1984). At this time, lumber produced from old growth coastal timber largely found its way to the mix of dimension lumber destined for the U S housing market. Further, high grades of lumber from the B C Interior were not commonly separated from the S - P - F mix destined for this same U S market. This U S market has remained a  Gaston  Page 145  Chapter Seven  "commodity" market where it is difficult to differentiate the product to price advantage.  It  is for this reason that the coastal forest industry, fuelled by the recession of the early 1980s and falling timber supplies, realized the need to diversify its customer base.  This  realization materialized in the B C Interior industry shortly thereafter. Over the past decade, the forest industry has indeed become more market oriented. Scarcity induced price premiums for certain grades or attributes of lumber have led to better log sorting, cutting to customer demanded sizes, the adoption of kiln drying and international quality certification, etc. Recognizing the difficulties in employing a diversification strategy in the US, Canada (primarily BC) increased its share of softwood lumber exports to Japan to 21 % by value in 1992 (Natural Resources Canada, 1993), as well as increasing exports to Europe. The overall market mix in 1992 is shown in Table 7.1. The difference in the nature of the lumber product shipped to the U S versus Japan is supported by examining the average price per cubic metre, with shipments to Japan being worth more than twice per cubic metre than those to the U S . While this is not shown to be true in the case of softwood logs, any comparison here is likely to be obscured by Canada's log export restrictions.  Note that Canada (primarily BC) exported only 22% of its lumber, by volume,  to non-US destinations. Of this 22%, it can also be seen that roughly 60% of these exports went to Japan, with the balance destined mostly to Europe. It is this emphasis on the Japanese market that this section will now address. First of all, it should be made clear that Japan is not B C ' s only potential market for differentiated lumber products.  A s shown in Table 7.1, exports to most off-shore  Gaston  Page 146  Chapter Seven  Table 7.1  Destination of Canadian Softwood Lumber and Log Exports,  1992.  Softwood Lumber m  000$ US Africa  $ per m  3  4,195,276  65.76%  30,848,622  78.42%  136  89,766  0.23%  177  15,846  0.25%  Algeria  11,942  0.19%  51,975  0.13%  230  Europe  629,209  9.86%  2,549,763  6.48%  247  Belgium  69,548  1.09%  206,774  0.53%  336  France  26,517  0.42%  78,263  0.20%  339  Germany  67,736  1.06%  114,996  0.29%  589  Italy  77,537  1.22%  146,458  0.37%  529 189  UK  349,393  5.48%  1,847,858  4.70%  Asia  1,464,601  22.96%  5,598,317  14.23%  262  Japan  1,325,474  20.78%  4,732,838  12.03%  280  Australia  62,164  0.97%  201,218  0.51%  309  Other  12,533  0.20%  51,446  0.13%  244  World  6,379,629  100.00%  39,339,132  100.00%  162  3  Softwood Logs US  20,951  14.04%  151,000  13.64%  139  Asia  128,082  85.83%  956,000  86.36%  134  Japan  124,148  83.19%  915,000  82.66%  136  Other  203  0.14%  0  0.00%  World  149,236  186.10%  1,107,000  100.00%  Source: Natural Resources  Canada. "Selected Forestry  135  Statistics". 1993.  destinations in 1992 were at higher average prices than for sales to Japan (in the case of Germany, in fact, more than twice as high as the average export price to Japan). Further, growing Asian economies hold promise for adding to the list of potential markets for BC's wood products.  Given the advantages of trading in non-commodity markets, the  advantages of further diversification seems justified. When one introduces the uncertainty of future prices in Japan, to which the discussion now turns, the potential advantages  Gaston  Chapter Seven  Page 147  become even more clear. From a B C marketing point of view, perhaps the single most important element in trying to predict what the future holds in terms of wood product prices sold to Japan (vis-avis premiums over the other B C markets) is the Japanese/Canadian exchange rate . A s 29  has been noted, Japan has actually been paying less each year, on average, for even high quality products such as softwood logs from the P N W and softwood lumber from the B C Coast. The question that cannot help but come to mind is whether they would be willing to pay more? What is likely to happen if the Japanese Yen stabilizes or weakens in the future? There are three possible scenarios which could unfold given this eventuality. The first is that Japan will start paying higher real prices for at least the higher quality commodities which are, after all, becoming more and more scarce, technological advances notwithstanding.  From the Canadian or American point of view, there would be little  change in the Canadian or U S dollar price trends that have been seen all along (down for low quality, up for high quality). The second possibility is that Japan will resist higher Yen prices by substituting other commodities. Unless the Yen does not depreciate equally over all currencies, this could mean substituting appearance quality wood with structural quality wood and/or nonwood substitutes. A s was suggested in the previous chapter, this would also include  The Canadian-Japanese exchange rate is not the only rate BC exporters should be worried about. It has been estimated that a 1 cent increase in the Canadian dollar relative to the US dollar, maintained for one year, translates into a loss in revenues to the Canadian forest industry of $450 million (Price Waterhouse, 1995). 29  Gaston  Page 148  Chapter Seven  continued substitution of lumber and panel products against rising log prices. The growing acceptance of platform-frame construction would support this possibility. The third possibility is that Japan reverses its growing dependence on wood imports, and invests in its domestic forest resource. In fact, it has been proposed in this study that the only reason that Japan imports as much wood as it does is because the vast majority of its domestic supply lies outside the extensive margin. But this tight extensive margin is a direct result of "cheap imports" (in Yen terms), with which domestic sources have difficulty in competing.  Remembering the resource description offered in Chapter 3,  including the extensive post war plantings which would become merchantable in light of rising prices, it is possible that BC's biggest future competitor for the Japanese market will be Japan itself . 0  This observation should, however, be tempered with Japan's growing  non-timber valuation of its forest resource. The results of the present study suggest that all three possibilities could come into play. Given the very small own-price elasticities shown for many of the wood products imported by Japan, the scenario of higher real Yen prices is certainly possible for these wood products. The second and third possibility are best supported when taken together. The present study suggests that price rises for those wood products with high own-price elasticities lead to substitution with domestically produced lumber and panel products (from imported and domestic log supplies). Given that all three possibilities are supported,  There is some doubt, however, as to Japan's ability to harvest significant quantities of appearance grade timber, at least in the short- to medium-run. As there has been little financial incentive to do so, it has been suggested that Japan's forest plantations have not been managed to maximize value (personal communication, John Powles, Director for Asia, Council of Forest Industry). 30  Gaston  Chapter Seven  Page 149  depending on the wood product under consideration, this suggests that the existence of a price premium for some wood products over others is likely to persist. However, it must be cautioned that the elasticity values offered in this study were estimated over a period of declining real Yen prices, with an overall down trend. Is the incidence of low own-price elasticities for selected wood products likely to hold at higher real prices than Japan has historically paid? It is possible that the only reason the P N W and B C have enjoyed price premiums for high grade logs and lumber is because of the strength of the Japanese Y e n . In spite of the fact that B C has sold similar products for even higher prices to European buyers (Table 7.1), there is some validity to the argument that the Japanese have driven up the world price as a direct result of their strong currency, and would drive down this world price in the advent of a falling Y e n . Even the preliminary evidence offered in this study, however, caste considerable doubt on this hypothesis. It is more likely that Japan has made every effort to pay as little as necessary to secure these fibre supplies. This was confirmed by the range of own-price elasticities presented in this study. Had Japan been willing to "pay anything" as a result of its strong purchasing power (in the income sense, leading to "cheap" imports), own-price elasticities would have all been low, regardless of species or origin. To the contrary, wood products such as planed lumber from the U S and Canada, and logs from NZ/Chile display highly elastic demands. If the price goes up even a little (due to a falling Yen) the quantity demanded goes down considerably. In short, Japan has been shown to be price conscious in spite of its "wealth".  Gaston  7.1.3  Chapter Seven  Page 150  Implications of the Research for BC Forest Policy When describing the motivation for this research in the introductory chapter, it was  pointed out that over the next couple of decades, B C is going to witness a significant reduction in the volume of available timber. Given the selected forest rotations for B C ' s second growth stands and the existing silvicultural efforts, it was also pointed out that the quality of timber is also going to be lower. These facts contribute to the market implications discussed in Section 7.1.2 insofar as they describe the nature of the future product. Given that the B C Crown controls the vast majority of the forest land base, it is primarily forest policy, not industry market strategies, that will impact on the future ability to adopt a market orientation. It was argued in the previous section that there is no reason to expect that the market premium for certain wood products will disappear. While there may be no way to predict what attributes of the wood in the future will command such price premiums, one fact does remain: if the level of silviculture and/or length of forest rotations does not change from present practices, there will be no old growth quality timber available at some point in the future. In addition to this study's implications for a more diversified B C forest product market, then, the related implication is for a continued diversity in the product itself. Even if there is not a premium market for products produced from clear, slowly grown, large timber in the long-run, this timber can always be used for alternative purposes (including the potential of the forests for the production of non-timber values). Without provision for these products, however, the option will be lost. While there are a number of government initiatives which could help facilitate the  Gaston  Chapter Seven  Page 151  preservation of future product diversity, discussion of such initiatives is clearly beyond the scope of this thesis. There is a need for considerably more research in this area, which will be briefly summarized later in this chapter, following a discussion of the limitations of the present study.  7.2  Limitations The most obvious limitation comes from the investigation of the second hypothesis  of this study, being to quantify the degree and nature of import substitution as a result of a price increase in a specific product. Due to reasons of multicollinear data, determining such cross-price effects requires a unique methodology, one which Armington (1969) provided and which has largely been accepted in the literature.  However, it has been  shown here that this methodology is not without its own limitations, centred around overly restrictive assumptions, and that these translate directly into limitations for the present study. While Hseu and Buongiorno (1993) attempted to deal with the problem of the restrictive assumptions used by Armington, it has been shown that this was not done successfully. The second limitation of this study is the lack of adequate secondary data on the quality of wood products traded in the Pacific Rim market. This fact limited quality aspects of the discussion to planed versus non-planed lumber, with further comparisons by species or source being largely a matter of judgment. A s a related point, detail on highly processed products (beyond wood-based panels) was not included in the present analysis due to the small historical volumes reported. The volumes of such products, including engineered  Gaston  Chapter Seven  Page 152  wood, has been growing rapidly over the last few years of the data set, and will undoubtably be an important component in the future. Limitations resulting from the lack of data can be broken down into two components: the treatment of wood input supply, and the use of a single output production function. In the estimation of the Japanese wood factor demands, it was assumed that the market supply was completely elastic, with Japanese buyers behaving as a "price takers". In other words, Japan cannot affect the price of its wood purchases by varying the amount purchased. The "price taker" assumption, while common practice in demand studies of this nature, may not be justified. Japan is one of the world's largest wood product importers, and may indeed exercise some level of market power in its purchases. In fact, if Japan does indeed face an upward sloping supply function (global excess supply) for particular products, it would effect the elasticities of substitution discussed in the present study. This would included Japan's choice between imported and domestic products, and, where the degree of market power varies, in one imported product relative to another. Unfortunately, potential limitations created by the supply assumption can not be addressed without data which allows for the estimation of supply functions by geographic source, by product, and by species. The second methodological limitation of this study involves the underlying production function from which the derived demand equations for the factor demands were obtained. It was proposed that the Japanese demand for wood products is derived from the per capita G N P in Japan, regardless of the differing species and implied quality. This assumption was supported by the recognition that Japanese housing construction, being  Gaston  Page 153  Chapter Seven  the largest single end use for wood imports , requires a range of qualities. Appearance 31  grades are required for posts and beams, panelling, and so on; structural grades are required for the non-visible house components of post and beam construction, and for a growing percentage of platform frame and prefabricated housing construction; and "utility" grades are required for sub-flooring, filler for laminated posts, concrete moulds, and so on. It is possible, however, that this single output production function is overly simplistic to adequately deal with quality issues, particularly when quantifying substitutions (cross-price elasticities). For example, the buyer of structural or utility grades, which are "capital" or "producer" goods, will likely treat the demand for fibre in the true "derived" sense. That is, if the sale price of a house goes up, all else being equal, the house builder would be willing to pay that much more for the inputs. Appearance grades, however, may more closely resemble "consumer" goods, insofar as consumer income, tastes, education, tradition, etc., are all capable of shifting the quantity demanded independent of price. The buyers of this fibre would be exhibiting direct willingness to pay for specific characteristics. It was shown in Chapter 3 that the growing percentage of platform-frame and prefabricated housing relies on construction grade imports, and could largely explain the demand for Japan's imports of S-P-F and other planed, dimension lumber. In the context of the present study, it could be suggested that in response to higher softwood log prices, Japan has shown a willingness to substitute one form of housing construction for another. The limitations of this study, however, do not allow this to be quantified. Aside from the  Further, imported wood products are used outside of the construction sector, such as imported logs as pulpwood. 31  Gaston  Page 154  Chapter Seven  requirement of more detailed trade data , it must be recognized that all three construction 32  types use both appearance and structural grades of lumber. A further limitation to the production function employed in this study, is that wood inputs were expressed in their cubic metre equivalents, regardless of the product type. This fails to recognize the range of costs associated with the Japanese processing of logs versus lumber versus further processed inputs. Although this treatment can be justified by using a general output indicator such as G N P as opposed to housing starts, and by noting Japan's preference for re-manufacturing even processed imports, added detail on the Japanese cost structure could prove insightful. The choice of Japan as the demand focus could be considered a further limitation to this study. Although there were good reasons for making this choice, as discussed in Chapter 1, it may limit the overall applicability of the results. The first potential limitation is that real prices for almost all of Japan's wood product imports dropped over the course of the study period, corresponding to a fairly consistent increase in the purchasing power of the Japanese Y e n . This suggests that it was not possible to define more than a fairly narrow price range within the demand functions characterising Japanese wood purchases. In other words, it is possible that the estimated demand elasticities would have been significantly different in the absence of a strong Yen. A second, and related limitation, is that in light of Japan's growing purchasing power, this country may not be representative of other markets. Further, given that Japan  A s a reminder, there exists a wide range of quality within both appearance and structural classifications of lumber. 32  Gaston  Page 155  Chapter Seven  is rather unique in its post and beam construction preferences, the nature of the demand for appearance grade wood products outside of this country has not been identified.  7.3  Implications for Further R e s e a r c h Given what has been learned about the negative effect of Armington's needed  assumptions for his two-stage approach to product demand within a market, there is a strong need to continue to challenge its application. This is particularly important in light of the wide acceptance of his approach, and the potentially misleading published results. The difficulty in dealing with multicollinear data problems has more than likely contributed to the lack of research dealing with heterogeneous product trade to be found in the literature. This is obviously not a trivial problem, and its solution is well beyond the scope of the present research. Yet the importance of developing methodologies that better address the unique nature of products must be stressed. In addition to the methodological problems associated with modelling heterogenous products, it must be pointed out that the underlying limitation discussed in the previous section, being the lack of disaggregated wood product trade data, is likely to limit potential extensions of this study. A s a qualifying recommendation for further research, then, it is suggested that the place to start is in overcoming these data obstacles. It is hoped that a continued recognition of the importance of these data will expose primary sources and/or the future availability of better secondary sources. A s stated, most of the limitations of this study could be addressed in the absence of these data restrictions.  Given the noted limitation of not recognizing the different  Gaston  Chapter Seven  Page 156  demand characteristics of structural lumber, for example, as compared to appearance grade lumber, more detailed data would allow for a more sophisticated, multiple output production function. Due to its importance, this point deserves additional consideration in the context of its implications for further research. A s was apparent from the data on lumber grades from a major producer/exporter from the B C Coast (Section 6.3), there exists a considerable range in prices within the definition of "appearance" or "structural" lumber, or even within detailed grade classifications. Keeping in mind that these data represent exports of coastal production only, this range in price would be even wider if one were to include interior production. In other words, the necessity to describe "quality" as planed versus non-planed lumber in this study, for example, can be misleading. It must be recognized that planed and non-planed lumber alike can be of high or low quality, and that the implications for B C ' s marketing strategy must take these quality ranges into account. It is possible, for example, that the higher quality S - P - F lumber from the B C Interior is demanded for post and beam construction in Japan, while the lower quality green squares from the B C Coast are not. Aside from the quality of the resource, future studies should also include attention to the quality of the product. By this it is meant that there has been a growth in the production of, and the demand for, further processed products, such as panels and engineered products. While the present study has included an analysis of veneer and panel products, data were not available to document the demand for further value-added products such as engineered wood products, wood "systems", and other further manufactured products such as door and window frames. In fact, it is also an important  Gaston  Page 157  Chapter Seven  implication for further research that the present study was not able to differentiate between, for example, S - P - F lumber from the B C interior cut to metric sizes, compared to S - P - F lumber from the U S which, if destined for the Japanese market, likely requires a higher degree of remanufacture.  In short, "value-added" may evolve in small, rather than  dramatic, steps (Cohen, 1992). More generally, as the present study has focussed on the demand for wood products by category, species and source for a single buyer, a natural extension of this research would be to broaden the level of detail to include product grade within a species, and to investigate other international demanders. It is further recommended that future research include an analysis of the major wood producing countries' supply functions with as much attention to product detail as possible. Aside from improving extensions of the present analysis as noted, this would allow for a better understanding of future trade flow scenarios, and offer price forecasting abilities. Finally, as this study has utilized a static analysis, estimating short-run (one-year) demand responses, it would be desirable to utilize a dynamic approach to obtain longer-run demand elasticities. In summary, this thesis has shown that there has been a wide range in the strength of the Japanese demand for various wood products. To the extent possible, it was shown that the Japanese have been willing to pay for quality, and that this it is not likely to change in the future.  From the demand side, this research needs to be extended to further  quantify this demand for quality, and to extend the analysis beyond Japan. This includes, but is not limited to, developing better methodologies for investigating cross-price effects.  Gaston  Chapter Seven  Page 158  Even more importantly, this research should be complemented with research on the supply of wood products by category, species, and by grade, concentrating initially on B C . Only at this point will it be possible to utilize more powerful tools, such as spatial trade modelling, as an aid in price forecasting, identifying future market potentials, and as a tool in forest policy.  Page 159  Gaston Bibliography  B I B L I O G R A P H Y  Adams, D.M. 1977. Effects of National Forest Timber Harvest on Softwood Stumpage, Lumber, and Plywood Markets: An Econometric Analysis. Oregon State University, School of Forestry, Research Bulletin 15, Corvallis. 50 pp. Adams, D.M., R. Boyd and J. Angle. 1992. Evaluating the Stability of Softwood Lumber Demand Elasticity by End-Use Sector: A Stochastic Parameter Approach. Forest Science 38(4): 825-841. Allen, R.D.G. 1953. Mathematical England). 548 pp.  Analysis  for Economists.  Macmillan and Co., Limited (London,  Alston, J.M., C A . Carter, R. Green and D. Pick. 1990. Whither Armington Trade Models. American Agricultural Economics Association, May: 455-467. Armington, P.S. 1969. A Theory of Demand for Products Distinguished by Place of Production. International Monetary Fund Staff Paper 16: 159-177. Babula, R. 1978. An Armington Model of U.S. Cotton Exports. Journal of Agricultural Research 39: 12-22.  Economics  Brooks, D. 1993. Market Conditions for Tropical Timber Products, appendix in The  Economic Linkages Between the International Trade in Tropical Timber and the Sustainable Management of Tropical Forests, Barbier, era/., London Environmental Economics Centre,  London, U.K. Buongiorno, J. 1979. Income and Price Elasticities of Demand for Sawn Wood and Wood-based Panels: a Pooled Cross-section and Time-series Analysis. Canadian Journal of Forest Research 9(2): 141-147. Buongiorno, J., J.P. Chavas and J. Uusivuori. 1988. Exchange Rates, Canadian Lumber Imports, and United States Prices: A Time-Series Analysis. Canadian Journal of Forest Resources 18: 1587-1594. Canadian Global Almanac. 1995. MacMillan Canada, Toronto, Ontario. 792 pp. Cardellichio, P.A., C S . Binkley and V.K. Zausaev. 1989. Potential Expansion of Soviet Far East Log Exports to the Pacific Rim. Working Paper 21, Center for International Trade in Forest Products, University of Washington, Seattle. 23 pp. Chen, N.J., G.C.W. Ames and A.L. Hammett. 1988. Implications of a Tariff on Imported Canadian Softwood Lumber. Canadian Journal of Agricultural Economics 36: 69-81. Chou, J.J and J . Buongirno. 1982. United States Demand for Hardwood Plywood Imports: a  Gaston  Bibliography  Page 160  Distributed Lag Model. Agricultural Systems 8: 225-239. Chou, J.J. and J. Buongiorno. 1983. United States Demand for Hardwood Plywood Imports by Country of Origin. Forest Science 29(2): 225-237. Cohen, David H. 1992. Adding Value Incrementally: A Strategy to Enhance Solid Wood Exports to Japan. Forest Products Journal 42(2): 40-44. Constantino, Luis F. 1986. Modelling Wood Quality, Productivity, Demands and Supplies in the Sawmilling Industry: British Columbia Coast and the Pacific Northwest Westside. Unpublished Ph.D. Thesis, Department of Forest Resources Management, University of British Columbia. 289 pp. Constantino, Luis F. 1988. Analysis of the International and Domestic Demand for Indonesian Wood Products. Report to the Food and Agriculture Organization (mimeo.), Dept. of Rural Economy, University of Alberta, Edmonton. Constantino, Luis F. and David Haley. 1988. Wood Quality and the Input and Output Choices of Sawmilling Producers for the British Columbia Coast and the United States Pacific Northwest, West Side. Canadian Journal of Forest Research 18(2): 202-208. FAO Yearbook. Various Issues. FAO Forestry Series No. 27. Food and Agriculture Organization of the United Nations, Rome, Italy. Flora, Donald F. 1986. An Equilibrium Model of Pacific Rim Trade in Small Softwood Logs. Canadian Journal of Forest Research 16: 1000-1006. Flora, Donald F. 1991. Trade Issues in the Northern Pacific Rim Countries. Presented at the Society of American Foresters: Economics, Policy and Law Working Group, San Francisco, California. 7 pp. Flora, Donald F. 1992. A Mesoeconomic Perspective on Wood Quality. Paper presented at "Pruning Conifers in Northwestern North America: Opportunities, Techniques, Impacts", Olympia, Washington. 7 pp. Flora, Donald F. 1993. Forest Sustainability and the Pacific Rim. Presented to the Sustainable Development in the Pacific Rim International Exchange Conference, Lewis-Clark State College. 6 pp. Flora, Donald F., Andrea L. Anderson and Wendy J. McGinnis. 1991-a. Pacific Rim Log Trade: Determinants and Trends. USDA Research Paper PNW-RP-432, Forest Service, Pacific Northwest Research Station, Seattle, Washington. 72 pp. Flora, Donald F., Andrea L. Anderson and Wendy J. McGinnis. 1991-b. Future Pacific Rim Flows and Prices of Softwood Logs, Differentiated by Grade. USDA Research Paper PNW-RP433, Forest Service, Pacific Northwest Research Station, Seattle, Washington. 22 pp.  Gaston  Bibliography  Page 161  Flora, Donald F. and Wendy J. McGinnis. 1989. Alaska Midgrade Logs: Supply and Offshore Demand. Research Paper PNW-RP-411, Forest Service, Pacific Northwest Research Station, Seattle, Washington. 13 pp. Flora, Donald F. and Christine Lane. 1994. Timber Trade Forecasting and Policy Analysis with Non-Optimizing Equilibrium Models. Presented at conference, "Current Issues in International Trade", Oregon State University, Corvallis, Oregon. 6 pp. Flora, Donald F., Christine Lane and Richard Haynes. 1993. Wood Products Trade, Forest Replanning and Forest Habitat Conservation in the U.S. Northwest. Review Draft, USDA Forest Service, Pacific Northwest Research Station, Seattle and Portland. 8 pp. Flora, Donald F., Wendy J. McGinnis and Christine L. Lane. 1993. The Export Premium: Why Some Logs are Worth More Abroad. USDA Research Paper PNW-RP-462, Forest Service, Pacific Northwest Research Station, Seattle, Washington. 18 pp. Flora, Donald F., Ulla Woller and Michael Neergaard. 1990. Tradeoffs and Interdependence in the Alaska Cant and Log Markets. Research Paper PNW-RP-422, Forest Service, Pacific Northwest Research Station, Seattle, Washington. 11 pp. Gaston, C.W., D. Cohen and R. Prins. 1994. Environmentalism as a Driver for Wood Product Quality. Working Paper 204, Forest Economics and Policy Analysis Research Unit, University of British Columbia, Vancouver. 22 pp. Gellner, B., L. Constantino and M. Percy. 1991. Dynamic Adjustments in the United States and Canadian Construction Industries. Canadian Journal of Forest Research 21: 326-332. Grennes, T., P.R. Johnson and M. Thursby. 1978. The Economics of World Trade. Praeger Publishers (New York). 129 pp. Haley, D. and M.K. Luckert. 1990. Forest Tenures in Canada: A Framework for Policy Analysis. Forestry Canada Information Report E-X-43, Ottawa, Ontario Haley, D. and M.K. Luckert. 1995. Tenures as Economic Instruments for Achieving Objectives of Forest Policy in British Columbia. Executive Workshop on Economic Instruments for Protection of Forest Resources, Faculty of Law, University of Victoria, British Columbia. 36 pp. Haynes, Richard W. and Roger D. Fight. 1992. Price Projections for Selected Grades of DouglasFir, Coastal Hem-Fir, Inland Hem-Fir, and Ponderosa Pine Lumber. Research Paper PNWRP-447, Forest Service, Pacific Northwest Research Station, Portland, Oregon. 20 pp. Hseu, J. and J. Buongiorno. 1992. Price Elasticities of Substitution Between Species in the Demand for U.S. Softwood Lumber Imports from Canada. Canadian Journal of Forest Research 23: 591-597.  Gaston  Bibliography  P a g e 162  Iwai, Yohsiya. 1986. Timber Producing Districts in Japan and the demand for Housing Timber. Paper in IUFRO, Division 4, The Current State Of Japanese Forestry (V), The Japanese Forest Economic Society, Tokyo: 1-11. Jacques, R., M. Martin and R. Samson. 1982. Analysis of the Demand for Canadian Lumber. Canadian Forest Service, Environment Canada, Ottawa. 11 pp. Japan Tariff Association. Various Issues. Imports of Commodities  by Source.  Softwood  Tokyo, Japan.  Japan Wood Products Information and Research Centre. Various Issues. Wood Supply Demand Information Service, Tokyo, Japan and Seattle, Washington. Johnston, J. 1984. Econometric PP-  Methods.  and  Third Edition, McGraw-Hill Publishing (New York). 568  Judge, G.G., R.C. Hill, W.E. Griffiths, H. Lutkepohl and T. Lee. 1988. Introduction to the Theory and Practice of Econometrics. Second Edition, John Wiley & Sons (New York). 1024 pp. Kalt, J.P. 1994. Report for the First Administrative Review in Certain Softwood Lumber Products from Canada. Prepared for the International Trade Administration, United States Department of Commerce. 291 pp. Kato, Takashi. 1982. Comparison of Softwood Lumber Manufacturing and Selling Costs Between the Pacific Coast of North America and Japan. Paper in IUFRO, Division 4, The Current State Of Japanese Forestry (II), The Japanese Forest Economic Society, Tokyo: 30-39. Kennedy, P. 1992. A Guide to Econometrics. PP-  Third Printing, The MIT Press (Cambridge). 410  Lewandrowski, J. 1989. A Regional  Model of the U. S. Softwood Lumber Industry: Including the Role of Price Expectations, the Role of Finished Product Inventory, and the Impacts of Trade Restrictions on Canadian Softwood Products. Ph.D. Thesis, North Carolina State  University. 220 pp. McKillop, W.L.M, T.W. Stuart, and P.J. Geissler. 1980. Competition Between Wood Products and Substitute Structural Products: An Econometric Analysis. Forest Science 26(1): 134 - 148. Mochida, Haruyuki. 1989. Problems of Cost Reduction in Japanese Forestry. Paper in IUFRO, Division 4, The Current State Of Japanese Forestry (VI), The Japanese Forest Economic Society, Tokyo: 39-49. Moffett, Jeffrey L. 1993. A Comparison of Product Diffusion and Distributed Lag Models for Estimating Wood/Non-wood Substitution in the US Window Market. Working Paper 42, Center for International Trade in Forest Products, University of Washington, Seattle. 29 pp. Moffett, Jeffrey L. And Thomas R. Waggener. 1992. The Development of the Japanese Wood  Gaston  Page 163  Bibliography  Trade: Historical Perspective and Current Trends. Working Paper 38, Center for International Trade in Forest Products, University of Washington, Seattle. 125 pp. Mori, Yoshiaki. 1992. Timber Market in Japan: An Econometric Analysis. Memoirs of the of Agriculture 139: 179-191, Kyoto University, Japan. Nicholson, W. 1989. Microeconomic Theory: Basic Principles The Dryden Press (Chicago). 793 pp.  and Extensions.  College  Fourth Edition.  Otsuka, Fumiko. 1992. Japanese Market fir Dimensional Lumber: a Gravity Model Unpublished M.Sc. Thesis, University of British Columbia, Vancouver. 92 pp.  Approach.  Pearse, Peter H. 1993. Determination of Harvest Rates in the Transition to Sustained Yield. Mimeo., Department of Forest Resources Management, University of British Columbia, Vancouver. 19 pp. plus appendix. Penson, J. And R. Babula. 1988. Japanese Monetary Policies and U.S. Agricultural Exports. Journal of Agricultural Economics Research 40: 11-18. Perez-Garcia, J.M. 1993. Global Forestry Impacts of Reducing Softwood Supplies from North America. CINTRAFOR Working Paper, #43, University of Washington, Seattle, Washington. 35 pp. Pesonen, Miikka. 1993. Japanese Market for Scandinavian Wood Products. Department Forest Economics Reports, No. 1, University of Helsinki. 116 pp. plus appendix.  of  Phelps, Susan E. 1993. A Summary of Elasticities of Demand and Supply for North American Softwood Lumber. Research Note, Forestry Canada, Economic Studies Division, Policy and Economics Directorate, Ottawa. 13 pp. Price Waterhouse. 1995. Analysis of Recent British Columbia Government Forest Policy and Land Use Initiatives. Prepared for the Forest Alliance of British Columbia, Vancouver. 66 pp. plus appendix. Prins, Robert G. 1993a. The Economic and Environmental Impacts of Reduced Timber Harvests and Increased Softwood Lumber Prices. Working Paper, Forestry Canada, Economic Studies Division, Policy and Economics Directorate, Ottawa. 39 pp. Prins, Robert G. 1993b. Substitution Between Tropical and Temperate Wood Products: A Literature Review. Research Note, Forestry Canada, Economic Studies Division, Policy and Economics Directorate, Ottawa. 12 pp. Random Lengths. Various Issues. Yearbook, Oregon.  Random Lengths Publications Inc. Eugene,  Robertson, Guy and Thomas R. Waggener. 1995. The Japanese Market for Softwood Sawnwood and Changing Pacific Rim Wood Supply Conditions: Implications for US Pacific  Gaston  Bibliography  Page 164  Northwest Producers. Working Paper 52, Center for International Trade in Forest Products, University of Washington, Seattle.71 pp. plus appendix. Robinson, V.L. 1974. An Econometric Model of Softwood Lumber and Stumpage Markets, 19471967. Forest Science 20(2): 171-179. Rockel, M L and J Buongiorno. 1982. Derived Demand for Wood and Other Inputs in Residential Construction: a Cost Function Approach. Forest Science 28(2): 207-219. Savin, N.E. and K.J. White. 1977. The Durbin-Watson Test for Serial Correlation with Extreme Sample Sizes or Many Regressors. Econometrica 45: 1989-1996. Sedjo, Roger A., A. Clark Wiseman, David J. Brooks and Kenneth S. Lyon. 1994. Changing Timber Supply and the Japanese Market. Discussion Paper 94-25, Resources for the Future, Washington, D.C. 31 pp. Singh, B.K. and J.C. Nautiyal. 1986. An Econometric Analysis of Markets for Canadian Lumber. Wood and Fiber Science 18(3): 382-396. Smith, Ramsay. 1989. Use of Pacific Northwest Wood Products in Japan. Reprint Series 18, Center for International Trade in Forest Products, University of Washington, Seattle. 7 pp. Sohngen, B.L. and R.W. Haynes. 1994. The Great Price Spike of '93: An Analysis of Lumber and Stumpage Prices in the Pacific Northwest. Research Paper PNW-RP-476, USDA Forest Service. 20 pp. Spelter, H. 1985. A Product Diffusion Approach to Modelling Softwood Lumber Demand. Forest Science 31 (3): 685-700. Spelter, H. 1992. Technology-driven Substitution in the Forest Sector - the Variable Price Elasticity Model Revisited. In Forest Sector Analysis - Proceedings of IUFRO Centennial Conference, Berlin, Germany: 24 -29. Sutton, W.J.R. 1994. The World's need for Wood. In Proceedings No. 7319, The Globalization of Wood Supply: Supply. Processes, Products and Markets, Madison Wisconsin: 21-29. van Kooten, G.C. 1993. Land Resource Economics and Sustainable Development: Economic Policies and the Common Good. UBC Press (Vancouver). 450 pp. Varian, H.R. 1984. Microeconomic Analysis. Second Edition, WW Norton & Company (New York). 348 pp. Varian, H.R. 1993. Intermediate Microeconomics: A Modern Approach. Third Edition, WW Norton & Company (New York). 623 pp. plus appendix. Vincent, Jeffrey R., David J. Brooks and Alamgir K. Gandapur. 1991. Substitution Between  Gaston Bibliography  Page 165  Tropical and Temperate Sawlogs. Forest Science 37: 1484-1491. Waggener, T.R., G.F. Schreuder, and H.M Hoganson. 1978. Elasticities of Demand for Forest Products over Time. Office report on file at the Pacific Northwest Forest and Range Experiment Station, Portland, Oregon. 112 pp. Webb, A.J., E.E Figueroa, W.E. Wecker and A.J. McCalla. Impact of the Soviet Grain Embargo; a Comparison of Models. Journal of Policy Modelling 11:361 -89. Youn, Y.C. and S.C. Yum. 1992. A Study on the Demand and Supply of Timber in South Korea. Paper presented at the Symposium on Forest Sector, Trade and Environmental Impact Models: Theory and Applications, CINTRAFOR, Seattle, Washington. Yu, Xiaoming and Yoshiaki Mori. 1990. Timber Demand in Japan. Paper in International Trade in Forest Products Around the Pacific Rim, Y.C. Youn and G.F. Schreuder, editors, Institute of Forestry and Forest Products, Seoul National University: 206-216.  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items