UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Global softwood lumber trade : a spatial equilibrium model Delcourt, Gregg Vincent 1995

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

Item Metadata

Download

Media
831-ubc_1995-0439.pdf [ 4.8MB ]
Metadata
JSON: 831-1.0086943.json
JSON-LD: 831-1.0086943-ld.json
RDF/XML (Pretty): 831-1.0086943-rdf.xml
RDF/JSON: 831-1.0086943-rdf.json
Turtle: 831-1.0086943-turtle.txt
N-Triples: 831-1.0086943-rdf-ntriples.txt
Original Record: 831-1.0086943-source.json
Full Text
831-1.0086943-fulltext.txt
Citation
831-1.0086943.ris

Full Text

GLOBAL SOFTWOOD LUMBER TRADE: A SPATIAL EQUILIBRIUM MODEL by GREGG VINCENT DELCOURT B.Sc. (Agr.), University of British Columbia, 1993  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in  v  THE FACULTY OF GRADUATE STUDIES (Department of Agricultural Economics)  We accept this thesis as confonriing to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA July 1995 © Gregg Vincent Delcourt, 1995  In presenting this thesis in partial fulfilment  of the requirements for an advanced  degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department  or  by his or  her  representatives.  It  is understood that copying or  publication of this thesis for financial gain shall not be allowed without my written permission.  Department The University of British Columbia Vancouver, Canada  DE-6 (2/88)  11  ABSTRACT This study uses a partial spatial equilibrium model to analyze changes in global softwood lumber trade. Based on work by Adams and Haynes (1980) and Cardellichio et al. (1990), trade flows from 15 different regions are analyzed using elasticity estimates and production, trade and price data from the year 1987.  Unlike the earlier models, this one  focuses on British Columbia (BC) and how proposed changes in the Annual Allowable Cut (AAC) will impact trade flows and prices. The model establishes a base case of expected future conditions from which a variety of different scenarios are tested. Prices, production, consumption and trade flows are calculated over a 38 year time period, 1987-2025. The results of the model indicate that changes in BC production cause an increase in the world average price until production responds to this shortage.  As BC experiences a decline in  production, exports to the United States are replaced by the increased production in the US South.  However, BC continues to increase its exports to Japan and Interior Canada as  demand for softwood lumber grows over time.  Increases in transportation costs affect  production and tradeflowsby forcing many importers to increase production and supply their domestic market. BC exports to Japan increase in response to this redistribution of trade. Welfare results indicate that, in the long-term, BC is better-off with a reduduction in AAC. A decline in production in BC causes world prices to rise, therby causing producer surplus to increase in the BC market. This comes at relatively small cost to the BC domestic consumer. These results suggest that provincial forest companies are not jointly maximizing returnsfromlumber production.  iii  TABLE OF CONTENTS ABSTRACT  ii  TABLE OF CONTENTS  iii  LIST OF TABLES  v  LIST OF FIGURES  vi  ACKNOWLEDGMENTS  vii  CHAPTER ONE  1  INTRODUCTION  1  1.1 INTRODUCTION 1.2 BACKGROUND INFORMATION 1.2.1 Demand Regions 1.2.2 Supply Regions 1.3 PURPOSE OF THE STUDY 1.4 OUTLINE OF STUDY  1 2 6 8 9 10  CHAPTER 2  11  LITERATURE REVIEW  11  2.1 LITERATURE REVIEW 2.2 TRADE MODELS 2.2.1 Two-Region, Nonspatial Models 2.2.2 Multi-region, Nonspatial Price Equilibrium Models 2.2.3 Spatial Equilibrium Models 2.2.4 Trade flow and market share models 2.2.5 Transportation models 2.3 FOREST SECTOR MODELS  11 12 12 12 13 14 15 15  CHAPTER 3  25  THEORETICAL MODEL  25  3.1 CONCEPTUAL MODEL 3.2 BACKGROUND OF THE SPATIAL EQUILIBRIUM MODEL 3.2.1 The Two Region Case 3.2.3 Multi-region Case....  25 26 27 29  CHAPTER 4  32  EMPIRCAL MODEL  32  4.1 THE LUMBER TRADE MODEL  32  iv 4.2 OVERVIEW OF MODEL 4.3 THE MODEL 4.3.1 The Regions 4.3.2 Data ... 4.4 MODEL SPECIFICATION 4.4.1 Consumption Function 4.4.2 Production Function 4.4.3 Trade Functions 4.5.4 Transportation Costs 4.5 GENERAL ASSUMPTIONS OF THE SE MODEL 4.5.1 Homogeneity 4.5.2 Competitive Market Assumption 4.5.3 Preferences 4.5.4 Inertia 4.6 MODEL ASSUMPTIONS 4.7 MODEL VALIDATION  32 33 34 -35 36 37 39 41 42 44 44 .46 46 47 48 50  CHAPTER 5  53  BASE CASE RESULTS AND SIMULATIONS  53  5.1 BASE CASE SIMULATION 5.3 REGIONAL ASSUMPTIONS 5.3.1 British Columbia..... 5.3.3 Pacific Northwest 5.3.4 US South 5.3.5 Chile and New Zealand 5.4 WELFARE MEASURES 5.5 THE BASE CASE FORECAST 5.5.1 Export Regions 5.5.2 Import Regions... 5.5.3 Prices 5.6 ALTERNATIVE SCENARIOS 5.6.1 Decrease in BC Production Welfare Changes 5.6.2 Increased Production in the Former USSR Welfare Changes  53 56 57 58 59 60 61 62 62 64 65 66 67 69 70 71  CHAPTER 6  72  SUMMARY AND CONCLUSIONS  72  6.1 SUMMARY AND CONCLUSIONS 6.2 MODEL LIMITATIONS 6.3 SUGGESTIONS FOR FURTHER STUDY  72 74 75  REFERENCES  110  V  LIST OF TABLES  Table 4.1:  Exporting and Importing Regions  77  Table 4.2:  Total Quantity Produced  78  Table 4.3:  Lumber Supply and Demand Elasticities  Table 4.4:  Transportation Costs  80  Table 4.5:  Value Differential  80  Table 4.6:  Regional Assumptions of Future Supply Conditions - Base Case  81  Table 5.1:  Simulated Production, Consumption, Exports and Prices  82  Table 5.2:  Simulated Production, Consumption, Imports and Prices  87  Table 5.3:  Direction of Trade - Base Case  91  Table 5.4:  Projected Real Softwood Lumber Prices  93  Table 5.5:  Projected Softwood Lumber Price Index  93  Table 5.6:  Direction of Trade - Reduced AAC in BC (Scenario a)  94  Table 5.7:  Direction of Trade- Reduced AAC in BC (Scenario b)  96  Table 5.8:  Direction of Trade - Increased Production in the Former USSR  98  Table 5.9:  Change in Producer and Consumer Surplus - Scenario: Change in AA Ca.... 100  Table 5.10:  Change in Producer and Consumer Surplus - Scenario: Change in AACb.... 101  Table 5.11:  Change in Producer and Consumer Surplus - Scenario: Increased Production in USSR  79  102  vi  LIST OF FIGURES Figure 1.1:  Comparative Forest Productivity  103  Figure 1.2:  BC Lumber Shipments  104  Figure 2.1:  Two Region Trade Model  105  Figure 5.1:  Projected Exports - 1987-2025  106  Figure 5.2:  Projected Imports - 1987-2025  ...107  Figure 5.3:  World Softwood Lumber Exports - Market Share  108  Figure 5.4:  World Softwood Lumber Imports - Market Share  109  Vll  ACKNOWLEDGMENTS  I would like to thank my advisor, Dr. Casey van Kooten, for his assistance, guidance and encouragement throughout my academic career. He has been a great help and a great 11:30am work-out partner during my stay at UBC. Dr. Mary Bohman, a member of my thesis committee, has always been a great help and excellent teacher since thefirstclass I took as an undergraduate. Her direction was highly useful in the writing of this thesis. As well, I would like to thank Dr. W.R. Wilson who gave me the opportunity to develop this project. His input was greatly appreciated. I would also like to thank my colleagues at Agricultural Economics and FEPA for their help and input. A special thanks to Casey for thefinancialassistance he has provided me during my years at UBC. I would also like to thank my family for their encouragement and support. Finally, I offer my deepest thanks to Michelle Johansen for her love, reassurance and understanding throughout my work; I will always be grateful.  1  CHAPTER ONE INTRODUCTION 1.1  INTRODUCTION  BC is experiencing a reduction in the supply of merchantable timber. Pressure from the public and environmental groups has forced the provincial government to legislate new harvesting regulations including smaller clear-cuts and more intensive reforestation activities. Projections of harvest levels indicate there will likely be a fall-down period in coming years. This has facilitated the proposal of a reduction in the current annual allowable cut (AAC) by 20 to 25 percentfromits current levels (Ministry of Forest 1994; Smyth 1994). Such a proposition could cause serious problems for the BC forest industry with its current capitalintensive structure. Unemployment, mill closures and increased production costs are only a few of the foreseeable problems that such an event could cause. The commodity-driven BC forest industry should consider investing more money in areas where it can compete more effectively.  High quality old-growth timber is BC's competitive advantage and should be  marketed in a manner that yields the highest net revenue per cubic metre. World softwood lumber supply is increasing, however there are wide variations in regional production. Emerging regions and other alternative sources are threatening BC's market share. The US South, New Zealand and Chile are producing softwood lumber on rotation periods many times shorter than in BC. The former USSR is another producer that has enormous potential to supply softwood lumber.  2 This study examines the effects of forest policies and increased global supply on international trade flows and, in particular, the BC softwood lumber industry. A partial equilibrium trade model is used to predict how different regions react to different supply and demand conditions.  The model predicts changes in softwood lumber trade, production,  consumption and prices for 7 demand regions and 8 supply regions over a 38 year period spanning from 1987 to 2025.  The base case scenario, which is derived from explicit  assumptions about future supply and demand conditions, yields projections regarding future market conditions. From the base case, alternative circumstances are tested and the results are discussed. The paper examines the following scenarios: 1. a reduction in provincial AAC with an increase in AAC over time, 2. a reduction in provincial AAC with no increase in AAC over time, and 3. an increase in suppliesfromalternative sources.  1.2 BACKGROUND INFORMATION BC continues to dominate the international softwood lumber market. In 1991 BC accounted for almost one-third of all softwood lumber exports and one-tenth of global production (COFI 1993). However declining domestic supply,risingwood costs and deflated international prices have created uncertainty about the future direction of the BC forest industry.  Provincial legislation is requiring a reduction in AAC at the same time that  stumpage costs are increasing. BCfirmsare objecting to higher stumpage costs saying that they are unable to compete in times of depressed prices (Hamilton 1995). The forest industry  3 is being forced to alter its focus as these issues, as well as government, special interest groups and international pressure create an uncertain environment for the marketing of BC wood products. As timber demand rises, traditional timber production and consumption patterns are modified. Production changes include the utilization of marginal species and smaller trees, investments in plantation forests (including on agricultural land), and biotechnological investments in enhancing tree growth (intensive margin); and the use of marginal forest areas and removal of remaining trees from agricultural land (extensive margin). Plantation forests have become common-place in non-traditional wood producing regions such as parts of South America.  Other plantation intensive regions include Australia, New Zealand and the US  South. In times of inadequate supply, New Zealand radiata pine has become a substitute for ponderosa pine and other alternative US species (Apthorp 1994). As these plantations mature and become areas of reliable timber supply, consumers of BC lumber may look to alternative supplies. The US has been investing in plantation forest for over half a century. Much of the US South is second and third growth timber. The US reforested over one million hectares (ha) in 1987 and continues to plant at a rate of approximately a million ha every year. Other countries such as New Zealand, Chile, Australia and Brazil are investing in plantations with high yield/short rotation species and are expected to make an impact on world supply in the near future. Another area with abundant resources is the former USSR. Since BC lumber has a quality advantage over these regions, direct comparisons must be made carefully.  4 Due to its large inventory and low rates of regeneration, BC forests have had little investment in replanting and intensive reforestation Until the last couple of decades. In fact, silviculture investment still makes little economical sense in most regions (Benson 1988; Thompson et al. 1990). In the immediate future, BC will continue to compete effectively as an international supplier although future supply conditions are in question. As indicated in Figure 1.1, BC growth rates are one-fifth of those in the southern hemisphere, making forest management plans even more important over the long-term. BC needs to pursue a forest management plan whereby BC's resources are used in a manner that maximizes provincial welfare, not harvested timber volume. As BC supplies decrease, the cost and access of timber becomes an additional constraint for BC producers. Continual harvest of old-growth alters specie composition and end-use products. The 60-100 year rotation length of BC forests creates a great deal of uncertainty regarding silvicultural investment and harvesting patterns. Since future demand for lumber is uncertain, investment in alternative species are a consideration for the BC forest industry. Currently timber is harvested for lumber, and the residual is used for pulp and paper. An option exists, for some regions, to grow timber used exclusively for pulp and paper. Hardwoods, such as alder and poplars, with much shorter rotation lengths are favourable for this allocation of fibre. Although future timber demands are uncertain, BC does recognize that its comparative advantage over the emerging regions lies in its supply of higher quality appearance grade timber.  Flora, Anderson and McGinnis (1991) conclude that commodity grades will  experience flat or declining real prices due to increased supply in plantations and higher rates  5 of utilization, but that 'performance' grade lumber will experience a steady and strong rise in prices. There has already been a steady increase in 'performance' grade lumber prices over the past 15 years, and industry has responded with new technology and higher recovery rates, while consumers have responded by accepting lower grades and veneer overlays, and by substituting non-wood products. An example of technology and substitutability of appearance grade lumber is found in wooden doors. Until the 1970s these products were made of solid material using an estimated 81 foot board measure (fbm) of wood per door. When the same door is made of a construction grade core with a veneer overlay there is a reduction of over 90% in clear material used (MacMillan Bloedel 1994). Evidently rising prices and declining supplies are affecting technology, production, tastes and the consumption of lumber. Technological change in lumber products may work to BC's disadvantage.  As  technology develops, fiber could substitute for sawtimber and cause a decline in demand for BC's construction material. Historic trade patterns will prevail over time, but it is the eroding of market share that is of concern in BC lumber trade. Estimates of future timber supplies can be made through existing and future planting and growing stock (however there is still a high degree of uncertainty), whereas demand fluctuates due to a variety of market, social and political conditions. The following is a summary of demand and supply regions that are of particular interest in global lumber trade.  6 1.2.1 Demand Regions The BC forest industry is one that invests little in value-added goods and specializes in large volumes of commodity forest products. Exports of lumber outside Canada account for well over 75 percent of total volume produced. The US imports about 58 percent of total production (COFI 1993). Other consumers of BC lumber are Canada (20 percent), Japan (14 percent), and the UK (3 percent). Figure 1.2 illustrates how exports have fluctuated over time. Thefigurealso shows how BC production is closely tied to US imports. Domestic production in BC's two primary export markets, the US and Japan, differ considerably. The US is the world's largest producer and consumer of softwood lumber, whereas Japan imports most of its timber leaving its domestic forest lands relatively untouched. Both of these markets have strong currencies compared to the Canadian dollar, thereby making BC wood products more attractive. The two demand regions are briefly discussed below. The US is the world's number one producer of softwood lumber accounting for approximately 25 percent of world production, of which less than 10 percent is shipped outside of the country. Until the late 1980s the US West dominated US lumber production, however the areas preserved for the spotted owl has caused a dramatic decrease in timber harvest in this area and has allowed the US South to increase its market share of US lumber production by over 10 percent. US South production is expected to continue to grow to account for over 50 percent of US lumber market share by the year 2000 (Widman 1994). Although the US is the largest producer, it is also the largest importer and consumer. In 1992, over 58 percent (19.6 million cubic metres) of BC lumber was sent to the US (COFI  7 1993) . This dependency on the US market could cause BC exporters to scramble to find alternative markets as plantations from the US South become more competitive (Smyth 1994) . Japan is BC's other primary international export destination. There is a demand for large dimension appearance grade timber in Japan. BC old-growth possesses favourable qualities which are well suited for use in the Japanese construction industry for decorative purposes. Structural lumber is being considered more closely as the Kobe earthquake of 1995 proved the viability of the 2-by-4frameconstruction style. Houses of Vancouver Village in Kobe were completely undamaged compared to houses built using the traditional post and beam structure (Column One 1995). Exports to Japan have more than doubled over the past 10 years and trade is favorable between the two regions due to location. BC's market share in Japan's sawtimber trade has increased considerably, but cost advantages in Chile, Brazil and New Zealand could cause a change in the demand for 'performance' grade lumber. Japan's ability to provide for its domestic demand should also be considered. As Japan's domestic stock of timber matures, it will become more self-sufficient in timber resources. Over 40 percent of Japan's forest base has been planted over the past 100 years. After W.W.II there was a large depletion of forest resources in order to rebuild bombed cities. In the years that followed, timber shortages ensued and afforestation projects did not solve short-term shortages. In 1963 Japan started importingfromthe US and USSR. Import prices were more attractive than harvesting less accessible areas and by 1970 imports accounted for over half of Japanese wood consumption. Japan still relies extensively on imported softwood  8 lumber, but when prices rise and forest areas mature, Japan may look towards its domestic resources.  1.2.2 Supply Regions  Global supply of softwood lumber is increasing over time due to increased harvests and developments in technology. However, as resources become less abundant and demand conditions change, countries that were net exporters could become importers. For example, India and China are satisfying their current domestic demands, but this could change depending on a variety of factors including an increase in living standards and changes in tastes. Other countries, such as Japan, have been net importers, but, as their domestic supply matures, demands for overseas' timber could lessen. Other factors that cause trade patterns to change include environmental concerns and newly emerging suppliers. Environmental concerns encompass the spotted owl controversy in the Pacific Northwest and old-growth preservation in BC. Increased supplies could come from regions such as the former USSR, and, as mentioned above, New Zealand and Chile.  The ability of the former USSR  developing into a large player in lumber trade is under a great amount of uncertainty due to the political situation in the area. This issue is addressed later in the study. The supply regions chosen in this paper, making up about 85 percent of softwood lumber, include Canada, the US, Scandinavia, Chile and New Zealand.  A more  comprehensive list of exporting nations is desirable; however, due to data and model restrictions a partial list is used. Only softwood lumber is investigated since hardwood lumber  9  does not yet appear to be a substitute. By concentrating on softwood lumber trade, the model can be used to predict the supply response of the BC industry.  1.3 PURPOSE OF THE STUDY  In this study, a trade model that estimates the effects of BC policy on long-term global lumber trade is developed.  The model identifies the changes in regional production,  consumption, prices and trade when global welfare is maximized and a steady-state equilibrium is established.  The transition from current market conditions to equilibrium  conditions depends on clear assumptions and explicit modeling techniques. The provincial government's introduction of new policies in forest management is of particular interest because it affects investment decisions, land use and costs of production. By modeling lumber trade, responses to price and resource constraints can be used to estimate future demands. The forecasts act as indicators to the forest industry of long-run equilibrium conditions in global markets. These estimates aid in the development of policy and in the making of rational investment decisions—private and public. By modeling different policy scenarios, projections of future trade flows are generated and different scenarios are compared. decisions.  The results are of interest for policy recommendations and for investment  10  1.4 OUTLINE OF STUDY  This study is divided into six chapters including this introductory chapter. Chapter 2 is a literature review of the various classes of trade models and their applications in forest trade. Recent research and forest trade models are then examined followed by a brief discussion of their limitations.  Chapter 3 details the theoretical background of the spatial equilibrium  model. A two-region case is first explained diagrammatically followed by the mathematical theory necessary to solve a multi-region case. The limitations of the model are also discussed within this chapter. Chapter 4 presents the model and clearly details the construction of the model. The constraints and assumptions necessary in model development follow. Chapter 5 presents the base-case model and relates the results to projections found in other studies. Different scenarios are then forecasted and compared with the results of the base projections. These scenarios include BC forest policy change in the size of the AAC, the effects of alternative producers supplying export markets and the effects of an energy crisis that causes transport costs to increase.  Chapter 6 summarizes the findings and details the  recommendations suggested by the model.  11  CHAPTER 2 LITERATURE REVIEW 2.1 LITERATURE REVIEW  Research in spatial models has progressed a great deal since the early 1960s as the importance of international trade in all sectors has demanded more in-depth investigation. Inter-regional trade is increasing as restrictions such as tariff and non-tariff barriers, embargoes and technology are overcome. As a result,free-tradeagreements and regional trading blocks are being formed, thus creating a demand for research in spatially optimal trade conditions. Policies, economic development and different business strategies require analysis using international trade models. The development of spatial models has come primarily from agriculture and general economic research; the forest sector did not venture into this area until the early 1970s. Spatial models can be classified as: 1) Two-region, nonspatial models; 2) Multi-region, nonspatial price equilibrium models; 3) Spatial equilibrium models; 4) Trade flow and market share models; and 5) Transportation models. Each model is explained below, including a discussion of its advantages, disadvantages and limitations.  12  2.2 TRADE MODELS  2.2.1 Two-Region, Nonspatial Models  The two-region, nonspatial model divides the market into two geographical regions: the target region and the rest of the world (ROW). The target region is either a net importer or exporter and has either an excess demand or supply curve. If it is a small player in the world market, it is deemed a "small country" and has no influence on world price. If it has a significant amount of market share, it is deemed a "large country," price is endogenous, and trade prices and quantities are affected. This model is the most aggregate of all trade models and has been used very little in the forest sector. Its primary advantage is its ability to focus on a specific country or region. It is useful when only crude approximations are needed for a more detailed analysis of a domestic market. The disadvantage of this model is its high level of aggregation. The ROW category groups all bilateral trade together, thereby making supply and demand elasticities unreliable. Also, the decision of whether the region is a "small country" or "large country" is arbitrary and can lead to problems in model estimation.  2.2.2 Multi-region, Nonspatial Price Equilibrium Models  The multi-region, nonspatial price equilibrium model calculates quantities traded by each region, but it does not calculate trade flows among regions. Excess demand and excess  13 supply are used to calculate a global equilibrium price that has been adjusted to include the transport costs between each region. Since this model cannot identify bilateral trade flows, it cannot be used to predict specific bilateral trade restrictions.  It does, however, effectively model tariff or quota  restrictions imposed on all regions by a single region. The primary disadvantage of this model is the assumption of a single global equilibrium price.  2.2.3 Spatial Equilibrium Models  The spatial equilibrium (SE) model calculates prices, quantities and bilateral trade flows endogenously in the model. location of trading partners.  Transportation costs are used to quantify the spatial  When these costs are minimized, optimal trade conditions  prevail. A partial SE model is one that only considers one commodity while assuming all other factors constant. The SE model is a desirable modeling technique when long-term analysis is required. Over time a market moves toward competitive equilibrium and trade routes maximizing individual welfare are developed. Thus, trade relationships that do not exist may very well develop over the long-term. The SE model is able to predict these new trade patterns. Adams and Haynes (1987) discuss the advantages of using the SE model and note that "...numerous geographic regions are readily accommodated...with little additional solution cost or increase in model complexity. This stands in distinct contrast to any other general modeling approaches."  Also, the SE model provides theflexibilityto perform changing  14 policy simulations by simply altering the objective function or constraints to imitate the proposed policy. The limitations of the SE model focus on the assumptions implicit in the model. These assumptions include perfect competition, homogeneous products and per unit transportation costs. A detailed analysis of this model is found in Chapter 3 as it is the modeltype used in this study.  2.2.4 Trade flow and market share models  Trade flow and market share models explain bilateral trade flows between specific regions for specific relationships.  (This contrasts with the SE model where trade flows  between all regions are determined.) These models attempt to explain differences in import demand characteristics depending on the region and the product.  For each region and  commodity, a market share relationship exists; it is a function of relative import prices and substitutes, and other derived demand shifters such as output price, economic indicators and end-use activity measures (Adams and Haynes 1987). Cardellichio and Veltkamp (1981) use this form of model to explain imports and inter-regional shipments in the softwood lumber and plywood markets.  A market share relation is developed for each region to explain each  supplier's preference to supply less distant markets. Another form of this model uses explicit supply and demand equations for each bilateral trade flow. For each region, a solution of equilibrium prices andflowsis determined. Data requirements are extensive for these models, but tradeflowsare generally more accurate than in the SE model. Short-term projections are more accurate as the distributed lag of price  15 and volume achieves the appropriate inertia properties. The major limitation of these models is their inability to model trade flows that did not exist historically. Also, the models become very large and data requirements are extensive when many producers and consumers are modeled.  2.2.5 Transportation models  The transportation model minimizes transportation costs of bilateral trade flows given the unit costs of shipment between each port (Koopmans 1948). This model is similar to the SE model where trade flows are transportation cost minimizing; however, the transportation model is less general than the SE model. Supply, demand and prices are set outside the model and bilateral trade flows are approximated.  2.3 FOREST SECTOR MODELS  There have been major developments and extensive undertakings in forest models recently. The SE model is used most frequently and viewed advantageous for modeling trade and long-term policy effects (Adams and Haynes 1987).  Studies have become more  complicated and the number of regions and products have become substantial in number (Cardellichio et al. 1989; Boyd, Doroodian and Abdul-Latif 1993; Kallio, Dykstra and Binkley 1987). The methodology and results of these studies are applied in the current study.  16 The first noteworthy forest market model was the Timber Assessment Market Model (TAMM)by Adams and Haynes (1980). TAMM is a SE market model. The methodology of T A M M is still the basis of numerous models, including most of the models discussed below and the current one. TAMM models nine supply regions and six demand regions in North America for two major forest sectors (final products and stumpage)  The final products  (including lumber) are modeled with their individual supply and demand curves linked to the stumpage supply. Due to difficulties experienced in estimating the simple linear demand curve, the demand functions for lumber are derived using the national demand elasticity and regional demand quantities and prices. Regional demand elasticities are derived by relating 1  the national US lumber demand elasticity, -0.35, to the national/regional price ratio. This is based on the assumption that the national/regional elasticities ratio is equal to the price ratio. The basic form of the demand equations is:  where D is the quantity demanded of lumber in region / in year t; it  Pu is the delivered price of lumber in region i in year t; and Yo, Yi are the intercept and slope parameters, respectively.  Adams and Haynes (1980) refer to their regional elasticity derivation as the hybrid approach. The above demand equation is used in the current study, with some modifications.  The expected size and signs of the coefficients in the demand functions are not intuitive. This phenomena was also found by Berck (1979) who had to use different price variables. 1  17 The product supply functions in TAMM are represented by a lagged supply term, average product price at the mill, "stump to car" transport price and an over run factor. Lumber supply elasticities with respect to price range from 0.21 in the Western Pacific Northwest to 0.79 in the Southern US.  The estimate for the supply elasticity of Canada is  0.47, with an export supply elasticity estimated at 0.89. The output of TAMM consists of prices and quantities, harvest volumes and stumpage prices, and the distribution of shipments. The model is calibrated to 1978 values and achieves equilibrium for future time periods by solving for each year of the forecast. The model was validated by plotting actual values against predicted values for the ten-year period 1966 to 1976. The predicted values replicated actual values very closely, especially for lumber. US imports from Canada were overstated due to demand elasticities changing over time (Adams and Haynes 1980). Numerous simulations were then performed. The simulation of interest for the current study estimated the changes in exports when Canadian production costs increased over time. This scenario comes from the assumption that second growth timber will take longer to mature, thus creating a fall-down in supply and force firms to log the extensive margin. TAMM estimates that an increase in production costs of $6/MBF in 2010, $11/MBF in 2020 and S17/MBF in 2030 will cause a 14 percent decrease in Canadian timber production and 31% decrease in exports to the US. A similar analysis is performed in the current study, although the decrease in supply is due to provincial government policy. A SE model developed by Boyd and Krutilla (1987) is conceptually related to TAMM with some improvements. Boyd and Krutilla (1987) analyze lumber trade between 34 supply  18 regions and 39 demand regions in Canada and the United States.  They pay particular  attention to transportation costs, exchange rates and tariffs in the trade. Demand elasticities are derived indirectlyfromlocal construction activity and lumber prices. It is assumed that the construction component induces demand elasticities to vary across regions. The supply of US domestic lumber is assumed to be fairly inelastic since US government timber sales are unresponsive to price. They comment that since excess supply cannot be more inelastic than the domestic market (see section 3.2), Canada's exports are able to vary with price. They use TAMM's estimate of Canada's export supply elasticity andfindthat Canadian exporters can lose up to 7 percent of their pre-tariff welfare if the US imposed trade restrictions (a 10 percent ad valorem tariff ). If suppliers are more responsive to price changes (supply is more elastic), a tariff will be more effective in reducing exports. Changes in the foreign exchange market had little impact on the demand for Canadian lumber. Sedjo (1983) modeled the economic returns of different forest regions with an emphasis on the potential of plantation forests. This study is of interest because it establishes long-term supply potentials of plantations.  Sedjo (1983) found that plantations of South  America and the US South generate higher net present values (NPV) than temperate-climate wood-producing areas. Temperate regions experience negative NPV for scenarios that test high discount rates, high production costs and high transportation costs.  A comparative  advantage is found in areas of high yield/short rotation plantations. The implications of these findings suggest that BC's forest industry could come under pressure once its absolute advantage (in volume of timber) is minirnized.  19 The International Institute for Applied Systems Analysis Forest Sector Project developed the Global Trade Model (HASA GTM) (Kallio, Dykstra and Binkley 1987). The HASA GTM uses a SE model to represent the global forest sector and international trade in forest products. The model estimates production, consumption, trade and prices of 16 forest products. It also projects values for the year 2030 given a variety of scenarios of structural change. These scenarios include changes in global economic growth, changes in the strength of currency exchange rates, trade liberalization, increased supply from the USSR, and environmental effectsfromacid rain and global warming. The basic model is a partial market equilibrium economic model with linear constraints and a non-linear objective function. The model links the components of each region's forest sectors to address the complexity of the forest industry. Timber supply, processing, demand and trade are all linked to their respective markets. There are 10 iterations of the model in a 50 year forecast where equilibrium for any time period is dependent on the previous time period. The model does not make future time periods endogenous (i.e., it is not a dynamic model). The results of the IIAS A GTM indicate trends in the forest sector that are important to the current study. The model predicts that Canada's share of softwood lumber trade will increase with a majority being shipped to Japan and the Rest of the World. The model also finds that there is not a significant increase in real prices of softwood lumber. However, when the model allows for increased supplyfromthe USSR, the results indicate that the main losers are Canada, the US, Southeast Asia, Brazil and Chile. As in any empirical model, there are numerous assumptions and generalities that create limitations in the estimating properties of the model. The fundamental limitation in this model  20 is its inability to address inter-temporal optimization; decisions in time t+1 have no effect at decisions in time t. This causes a bias in the model that is also a problem in most trade models. Another problem of the HASA GTM is the reliability of its database (Cardellichio and Adams 1987).  Data are inconsistent and unreliable for some cases and replication is  problematic. Although the model has the above limitations, it does provide a framework for analysis of the forest sector trade and different trade scenarios. Sedjo and Lyons (1991) developed an optimal control model, the Timber Supply Model (TSM), that addresses the changing age and volume of forests and the changing state of the forest. Inter-temporal investment can be modeled to allow for optimal rotation length and optimal old-growth depletion. The TSM uses a partial equilibrium approach by assuming that the forest sector is a price taker with respect to macro economic parameters such as interest rates, prices of intermediate goods and factor input prices. The model calculates each region's supply and global demand; it does not predict bilateral trade flows. It is of interest for the current study because it investigates the draw down of inventories of old-growth stands and models the transition to second-growth and plantation timber. The TSM estimates that the global demand for timber will increase at 0.6 percent to 0.9 percent a year to the year 2035. These results are similar to the HASA GTM that estimates a growth rate of 1.2 percent per year to the year 2030. Sedjo and Lyons (1991) also predict that real prices of industrial roundwood will remain relatively constant over the estimated time horizon. A revised TSM (Sedjo et al. 1994) incorporates the expected decline in harvest in BC and the US West. The results show that average prices only increase about 5 percent over the original TSM base case results. Increased productionfromthe US South, Scandinavia and  21 emerging regions offset the reduction in supply in the Pacific Northwest. When demand is assumed much stronger, prices increase 30 percent over the base case and regions such as Eastern Canada become an important alternative source. These results are consistent with Messmer and Booth (1993) who found that Ontario harvest levels are sensitive to relatively minor changes in price. Boyd, Doroodian and Adul-Latif (1993) attempt to quantify the consequences of reducing and eliminating lumber (pine andfir)tariff restrictions between Canada and the US, and the US and Mexico, in response to the North American Free Trade Agreement. They analyze trade flows, prices and welfare effects of all three countries using a SE model. Since transportation costs act like a tariff in the model, they simulate the tariff rate by raising transportation costs.  Own-price elasticities are estimates from Adams, Boyd and Angles  (1992) and cross-price elasticities between fir and pine are derived indirectly by relating construction activity to the input amounts of pine,firand capital. In order to keep shipments of fir out of pine producing regions they impose exceedingly high transportation costs (US$220/m ). 3  (This method is used in the current study to restrict trade between some  regions.)  The results suggest that the removal of tariffs on lumber cause consumption to increase and lumber to be allocated more efficiently. British Columbia is expected to gain over US$6 million dollars per year in afreetrade scenario. The total change in welfare from the removal of tariffs is estimated as a US$35 million per year gain. Boyd, Doroodian and Adul-Latif (1993) conclude that the increase is relatively small compared to total industry revenues (about 2 percent), due in part to the lack of demand for lumber in Mexico. The  22 insignificant size of the gainsfromtrade liberalization is a common finding of studies modeling trade liberalization.  The most significant and most comprehensive forest trade model is the CINTRAFOR Global Trade Model (CGTM) developed by the Center for International Trade in Forest Products (CINTRAFOR) at the University of Washington (Cardellichio et al 1989). This is a SE model that builds on HAS A GTM. The CGTM modifies the IJASA GTM by covering less products (10 instead of 16) and substantially more regions (40 rather than 18). The CGTM reduced the number of products by combining all pulp and paper into one category and dropping fuelwood. It creates new categories to differentiate coniferous and non-coniferous products. As in the GTM the model is comprised of four components: timber supply, product supply, product demand and trade.  The demand and supply curves are determined  endogenously for all but 16 regions where they are determined exogenously by the user. These exogenous regions are ones that have poor data, specialized products or little trade, and include the former USSR, Eastern Europe, Africa and all of South America except for Chile. The user must provide output and consumption levels for these regions. The CGTM, as with the GTM, maximizes an objective function to determine optimal global welfare.  The demand function uses a non-linear form and, for some regions, uses  elasticitiesfromother studies. The supply curves are determined in the same manner as the GTM. Also similar to the GTM, trade is banned between regions with negligible trade and regions that are not expected to trade in the future. The CGTM estimates transportation costs by simply taking the gross difference between import and export prices. This causes transport costs and other transfer costs (tariff and non-tariff barriers) to be included as one value.  23 Projections for coniferous softwood markets suggest that there will be a large increase in consumption in Western Europe, China and the US West. Production increases are projected to occur in Eastern Canada, the US South and the US North. Policy simulations with the CGTM are made through marginal increases rather than simulating extreme absolute changes since marginal policy changes are viewed more realistic. Perez-Garcia (1993) uses the CGTM to study the global impacts of a reduction in softwood from North America. The model reports considerable increases in log prices for all regions over the next 50 years and suggests price increases of 20 percent in the Pacific Northwest, 60 percent in the US South, 90 percent in Interior BC and 20 percent in Chile. In the study, Perez-Garcia (1993) estimates two scenarios: a reduction of 33 million cubic meters from Western Canada and the US West, and an increase in log exports from Siberia (with the reduction in North American supply). Under the first scenario, the results indicate further price increases and substantial welfare losses to lumber consumers. Canadian consumers lose US$141 million and US consumers lose US$970 million. Lumber producers gain US$512 million in Canada and US$754 million in the US. Globally, lumber consumers lose about US$2.5 billion dollars annually. The second scenario does not change welfare conditions much for Canada and the US, but it does reduce global losses to only US$1.5 billion. The study states that high cost forest producers are the overall winners. The preservation of one hectare by a low cost producer is offset by the harvest of 1.12 to 1.61 hectares by high cost producers. The ratio is even higher if Siberia enters into the export market. These results have global environmental implications.  24 The current study draws on the methodology, data and results of the papers reported above. Particular attention is given to the methodology of TAMM and CGTM. It is these models that the current study builds on by modifying scenarios, updating data and altering supply and demand specifications.  The results of the above models are compared to the  current study in the analysis that follows.  25  CHAPTER 3 THEORETICAL MODEL 3.1 CONCEPTUAL MODEL  The trade model presented in this study is a spatial equilibrium model that optimizes welfare by estimating quantities traded, international prices and trade flows among regions. Under the assumptions of the SE model, countries with high costs of production or inadequate supplies look to world markets to satisfy domestic demand at lower prices. Likewise regions supply world markets at higher prices than they can otherwise acquire domestically. Regions continue to trade as long as prices, net of transportation and other transfer costs, differ from domestic prices. welfare subject to certain constraints.  An iterative process is used to maximize the  A spatial price equilibrium is established when the  demand price is equal to the sum of the supply price and transportation costs for all regions. Consumers rearrange their consumption bundle in favour of cheaper suppliers until this equilibrium is reached. A SE model is used so bilateral tradeflowscan be observed. As more countries enter the world market, interactions become more complex and optimal trade flows become less obvious for each region. The SE model assumes cost minimizing (profit maximizing) behavior by the consumer (producer).  Therefore, by solving for a global maximum, regional welfare is maximized.  Enke (1951) gives a precise summary of the one commodity model as follows:  26 "There are three [or more] regions trading a homogeneous good. Each region constitutes a single and distinct market. The regions of each possible pair of regions are separated—but not isolated—by a transportation cost per physical unit which is independent of volume. There are no legal restrictions to limit the actions of profit-seeking traders in each region. For each region the functions which relate local price are known, and consequently the magnitude of the difference which will be exported or imported at each local price is also known. Given these trade functions and transportation costs, we wish to ascertain: 1) the net price in each region, 2) the quantity of exports or imports for each region, 3) which regions export, import, or do neither, 4) the aggregate trade in the commodity, [and] 5) the volume and direction of trade between each possible pair of regions."  3.2 BACKGROUND OF THE SPATIAL EQUILIBRIUM MODEL  The partial SE, international trade model for one commodity wasfirstdeveloped by Samuelson (1952). Samuelson formalized Cournot's (1838) price relations problem between two spatially separate markets and related it to Enke's (1951) paper, which generalized the problem of inter-spatial markets. Enke used electric circuits to illustrate the solution, whereas Samuelson approached the problem using mathematics. Prior to Samuelson no one had  27 proposed a method to quantify the complex relationships in spatial analysis.  Samuelson  (1952) stated that the model is solved "by trial and error or by a systematic procedure of varying shipments in the direction of increasing social pay-off" until a unique maximum exists. In the days of Samuelson's work, complex problems were difficult to estimate due to the complicated iterative procedures; today, computers perform these calculations in a fraction of the time.  3.2.1 The Two Region Case  To illustrate the concept of price equilibrium, the two region case is used. Consider Figure 2.1. It is obviousfromthis graph that trade will occur since equilibrium prices differ in the two regions. Without trade, price in Region A is P and the price in Region B is P . If a A  B  region's domestic price is greater or lower than the other region's price, net of transportation costs, it will engage in trade until an equilibrium is established. If a region cannot clear its market domestically, it becomes an exporter and its excess supply (ES) function is derived by laterally subtracting the demand curve from the supply curve at every price greater than the no-trade equilibrium in the domestic market. Likewise, the excess demand (ED) curve is derived from the importer's domestic market by laterally subtracting the supply curvefromthe demand curve at every price less than equilibrium in the domestic market. Notice that the trade functions (ES and ED) are more elastic than their respective functions in the domestic markets.  28 To derive Region A's excess demand elasticity, domestic prices, quantities and the slopes of the domestic supply and demand curves are required. Given that the quantity of excess demand is equal to the quantity demanded less the quantity supplied,  =  Q -Q , D  S  the elasticity of excess demand, €ED , is derived as follows:  cP  (3.1)  dP cP P _dQ P  (3.2)  D  dP QED 8Q  &ED ~  D  QED  &  P Q  SQs P Qs  D  SP QED Q  D  £ ED ~  & QED  &  Qo 'ED  (3.3)  QED QS  (3.4) QED  where P is the domestic price, £D is the elasticity of demand and £, is the elasticity of supply. (The elasticity of the excess supply function is derived in a similar manner.) Note that if £  s  andQy are greater than zero, the excess demand curve will be more elastic (more price responsive) than its domestic demand curve. Intuitively, this can be explained by analyzing the demand for domestically produced goods. As trade lowers prices in the importing region, demand is diverted away from the domestic market to the international market where prices are lower. A comparable analysis can be used to relate the supply function in the exporting region to its excess supply curve. Under conditions of zero transportation costs, a price equilibrium occurs where the excess-supply and excess-demand curves intersect.  Region A imports quantity Q* from  29 Region B and world price is equal to P*. This is the solution of a nonspatial price equilibrium without transportation costs. The inclusion of transportation costs of wz per unit increases import prices, decreases export prices and reduces the amount traded. In this case Q** (< Q*) is traded and transport costs are equal to area xywz in Figure 2.1. When the area between the excess demand and excess supply curves is at a maximum, net of transportation costs, a spatial price equilibrium is calibrated. Samuelson (1952) termed this area (Axz+Byw) as the net social payoff (NSP).  He explains that this area is "artificial" in magnitude since the  "Invisible Hand has led us to maximization, [and] we need not necessarily attach any social welfare significance to the result." The NSP is also equivalent to the sum of aed in Region A and gkj in Region B. Constraints are imposed on the model in order to define a feasible solution space. The constraints form bounds on the solution to restrict the values of the flow parameters. The minimum constraints needed to calibrate the model include total exports equal total imports, total exports of a region are greater than or equal to the sum of all importsfromthat region, and the non-negativity of prices and quantities.  3.2.3 Multi-region Case  The two region case is adequate for explaining the structure of the model, but it does not explain multi-lateral trade movements.  By minimizing transportation costs between  regions, a direction of trade matrix can be determined. The dual of this problem, maximizing  30 NSP, yields the same results.  A spatial price equilibrium is found when all regions have  maximized their individual welfare. The multi-region case considers n regions that supply and demand a given commodity. Each region is considered an independent market where quantity hj is supplied at price p with t  a inverse supply function, Sj{hj),j=l,...,n. Quantity demanded, d , at price % is explained with t  the inverse of the demand function, Di(d), i=l,...,n. It is assumed that S/hj) is a continuous monotone increasing function in h h >0, and that D (d) is a continuous decreasing monotone h  function in d , d >0. t  s  t  Quantities of bilateral trade, ty, are priced at c,,  t  per unit—the  transportation costs from region j to region /. Equilibrium occurs when the following condition holds for all /' and j:  SM) + c -DXd,) =0if / , > 0,  (3.5)  >0if t = 0.  (3.6)  tj  v  The NSP is derived by maximizing the sum of the area under the excess demand curves less the sum of the area under the excess supply curves and transportation costs (ED and ES curves are derivedfromequations 3.7 through 3.11 below).  Optimal trade flows are  determined by simply solving the following optimization problem:  (3.7) o  jo  31 subject to:  (1)  Pj,*»yj,q  (2)  2>-z>=o,  (3)  -c7  t  1+z  (3.8)  ^0 Vi,j,  (3.9) (3.10)  >,,>0,  (4) yj-Zvj^ '  (3.11)  0  where q is the quantity of imports of region /' and y is the quantity of exports of region / The t  t  constraints form bounds on the optimal solution space. Constraint (3.8) ensures prices and quantities are positive. Constraint (3.9) ensures all markets clear. Constraint (3.10) ensures that what is supplied to region / is at least equal to what is consumed in region /'. Constraint (3.11) ensures that the supply of region j is at least as big as what region j exports. The Kuhn-Tucker conditions are equivalent to the above constraints. Since the objective function is the sum of two concave functions and a linear function, it is a concave function and the Kuhn-Tucker conditions are necessary and sufficient for equilibrium values q yj and t (Florian and Los h  tj  1982). Inverse functions are used because the constraints to the problem are in terms of quantity. This framework provides the basis of the spatial equilibrium problem to be solved in this study.  32  CHAPTER 4 EMPIRCAL M O D E L 4.1 THE LUMBER TRADE M O D E L  The object of this study is to develop a trade model that replicates current lumber trade and use it to estimate future conditions given specific policy decisions.  The one-  commodity SE model is used in this study because of its ability to estimate trade flows and its explicit consideration of regional price differences through transportation costs.  Only  softwood lumber is modeled (and not hardwood lumber) since it is of particular interest to the BC forest industry. The model isfirstcalibrated to replicate trade flows from 1987 and then used to project trade until the year 2025. The model performs 6 iterations over a 38 year forecast period where equilibrium for each period is dependent on the supply and demand conditions of the previous period. Implicit assumptions are made on the model regarding future market conditions for each time period. Initial projections are referred to as the "base case" and are used to compare against alternative policy scenarios that are performed on the model.  4.2 OVERVIEW OF M O D E L  A partial equilibrium model (versus a general equilibrium model) is used in this study under the assumption that changes in the softwood lumber industry do not disturb nonforestry sectors of the economy. This implies that the softwood lumber industry acts as a price taker with respect to changes in interest rates and factor inputs. Although the SE model may not  33  produce predictions as accurate as other short-term models, it does provide information on long-term equilibrium and the competitive advantage of each region (Cardellichio et al. 1989; Adams and Haynes 1987). The model generates optimal trade patterns and predicts changes in the international demand for softwood lumber. A total of 15 regions are considered in the current study. Each region is active in the international softwood lumber market as either an importer or an exporter. Using reported supply and demand elasticities, domestic markets are modeled and trade functions are derived for the international market.  A region will be represented by either an excess-demand  function (if it is an importer) or an excess-supply function (if it is an exporter) in global trade. Transportation costs between regions are estimated and used to calibrate the spatial nature of international trade. By maximizing the area between the trade functions net of transportation costs and subject to a number of constraints, optimum trade flows are determined.  4.3 THE MODEL  The model is set-up and estimated using the Microsoft Excel software package. Through an iterative approach, each region's optimal trade flow is calculated by optimizing global welfare. The Excel software package is used because of its availability and simplicity. It allows the users to try different trade scenarios, perform sensitivity analysis or impose additional constraints with little difficulty. Also, since Excel can be programmed using the Visual Basic programming language, the model can be created as a stand alone executable file to be used with the Excel spreadsheet program.  34 4.3.1 The Regions  The model estimates 15 regions (detailed in table 4.1) that account for approximately 93 percent of international trade in softwood lumber. Of these regions, 8 supply lumber to the international market and 7 are demanders of international lumber. In 1992 BC and Interior Canada accounted for over three-quarters of Canada's softwood lumber exports and one-third of world exports (National Forestry Database 1993). The US West and US South primarily supply the domestic US market, but also export some softwood lumber overseas. The Scandinavian countries of Sweden and Finland form the other primary exporting region; their export share is about 16 percent of total world exports. Chile and New Zealand are emerging supply regions with growing market shares and are of interest when estimating future trade flows.  The remaining exporting regions are grouped together as Rest of World Exports  (ROW) and own only about 7 percent of the market share. The demand for softwood lumber is dominated by the US where almost two-fifths of all foreign exports are destined; the US imports almost all of thisfromCanada (FAO 1992). The US North is the largest demand region within the US, accounting for about one-third of the importsfromall regions. Central Canada imports primarilyfromBC with some imports from different regions in the US. Western Europe imports about 32 percent and Japan accounts for about 8 percent. The remainder of the importers is included in Rest of World Imports (ROW) and competes for about 10 percent of the market.  35  4.3.2 Data  The model uses price and quantity data taken from a variety of sources due to inadequate availability of some regional production and trade data. CINTRAFOR's GTM (Cardellichio et al. 1989) offers a comprehensive database of 1987 quantities and prices for most regions in the model, in particular the different US regions. This database, although somewhat dated, is used because it offers a consistent source of information for most regions included in this study. Also, by using 1987 values, validation of forecasts can be made by comparing actual and predicted values. Table 4.2 displays 1987 production data. Prices are reported in constant 1980 US dollars and quantities are reported in millions of cubic metres. All prices are prices paid in the domestic market and converted to US dollars using 1987 exchange rates. BC production and trade data are industry statisticsfromCOFI (1993). This source is used because it reports BC shipments to Canadian provinces.  The matrix of  distances between regions measures the kilometres between the major port in each region (Defense Mapping Agency 1976). Freight costs are reported in 1980 US dollars per cubic metre per 100 kilometre. There is an abundance of literature reporting domestic supply and demand elasticities for Canada and especially the US (see Gaston, Cohen and Prins 1994). Elasticities for other regions are less accessible and somewhat deceiving without the price/quantity ratio of the estimation. This model uses elasticity estimates by Cardellichio et al. (1989) for all regions. Table 4.3 reports the own-price supply and demand elasticities along with 1987 production and trade data for each region. These elasticities are comparable to the elasticities used by the Timber Assessment Market Model of Adams and Haynes (1980).  36  4.4 MODEL SPECIFICATION  In an ideal trade model, information on all market interactions would be included in each region's supply and demand equations. The development of a complete forest model is somewhat elusive due primarily to the lack of data. Also, by making the model more complex, the results may or may not improve over extended projection periods. This study relates prices to quantities using linear domestic supply and demand curves to estimate the international trade functions. Linear functions are used for a number of reasons.  First, they are integrable and  robust in determining an equilibrium. Second, elasticities are non-constant.  As prices rise,  price becomes more elastic and resembles market conditions more closely.  Third, demand  curves with constant elasticities less than 1 1 . 0 1 indicate a negative marginal revenue and therefore cannot be profit maximizing; it is impossible for marginal revenue to intersect marginal cost since marginal revenue is continuously less than zero and marginal cost is continuously greater than zero. The use of linear functions for demand and supply curves eliminate these problems associated with using non-linear functions. The supply and demand curves allow for changes in exogenous variables through an intercept shifter.  By shifting the constant, a uniform increase or decrease in quantity is  assumed for all price/quantity pairs. A change in slope, or a rotation about the equilibrium, may represent real world conditions more accurately, but pose a problem in solving for equilibrium values. The scenarios presented in this model can be modeled using a intercept  37  shifter since the changes in supply are assumed to cause uniform changes in price for all quantity/price ratios.  4.4.1 Consumption Function  For each region, the demand function explains the relationship between domestic prices and the level of consumption. The demand function is derived using prices, quantities and elasticities from existing literature. Elasticity estimatesfromCardellichio et al. (1989) are used to evaluate domestic demand curves in this study since this is also the source of price and quantity data. Given that lumber is consumed primarily by the housing market, elements outside the forest sector can influence lumber demand. Interest rates, income growth, population and technology can shift demand over time.  Due to incomplete data, these factors are not  explicitly included in the demand function; however, they are included as an intercept shifter. It would be desirable to have slope shifters, however due to the sensitivity of the model and time constraints, these factors are grouped together in the intercept. Given that a partial equilibrium is desired, the primary requirement for each region is that the demand function must depend on price. The final demand function for region /', i=l,..,18, and time period t is specified as follows:  (4.1) where:  38  Q is lumber consumption in millions of cubic metres, P is the real US price of lumber per cubic metre, and a, P are the intercept and slope parameters, respectively.  Changes in demandfromnon-price effects are measured over time by making a endogenous. Each time period is dependent on the changes of previous time periods. Let oti be defined as follows: (4.2)  where: rtt is the global per annum increase in demand for softwood lumber, t is the year of the current period, and t-1 is the year of the previous period.  The inclusion of this effect allows the model to be used in estimating future trade flows by simulating changes in the nonforest sector of each region. It is obvious that since n,=0 in the base case, eta equals a^ui). It follows that as global demand increases, consumers will demand more at a given price, thereby shifting the intercept of the demand curve. The demand equation must be in its inverted form to be used in the algorithm of the objective function. Therefore the estimated domestic demand equation used in the model is defined as: a  + (",)('-('-!) . 1 {Pi)  (4.3)  39  It is assumed that the slope of the demand equation remains constant over time.  4.4.2 Production Function  Each domestic lumber supply function is denned as the relationship between the quantity produced and the price. Changes in the supply of lumber in domestic markets are felt in the international market. Domestic elasticities, prices and quantities are used to estimate the supply curve. As in demand elasticity estimates, there is a wide range in the values of supply elasticities. Long-run and short-run elasticities vary due to the ability to earn positive profits in the short-run. Long-run elasticities are used because they resemble a competitive market more accurately. The supply of lumber is dependent on numerous factors other than price. Technology will cause an increase in utilization rates in lumber as well as develop other sectors which may divert input materials awayfromlumber. Investment in enhanced silviculture and plantations cause an increase in future timber supplies. Supply is also dependent on government policy, which is currently evident in BC (Ministry of Forests 1994). These factors have not been explicitly estimated due to lack of data. Instead, they have been acknowledged in the supply function as intercept shifters for future periods. As with the demand equation, slope shifters may enhance results, however this should be addressed in a sensitivity analysis during further research. Thefinalsupply function for region /, i=l,...,18, for time period t is specified as: Yt^Cu+dtf  (4.4)  40  where: Y is lumber supplied in mmm  3  .P is the real US price of lumber per m 5  3  c, d are the intercept and slope parameters, respectively.  Parameter c is made endogenous to measure the effect that changes in non-price factors have over time. Let c, be defined as follows:  «»= *  4(M)  +  (  a  (  4  -  5  )  where: giis the rate of change in global supply due to technological developments, hi is the rate of change in regional supply due to increasing rates of harvests, t is the year of the current period, and t-1 is the year of the previous period.  Forecasts of fixture lumber supply can be made by allowing these to change with expected changes in technology and harvest rates. Supply changes are easier to predict in the short-term since harvests from standing inventories can be projected over time. In the first period, c equals c .i) since g and h equal zero. The supply equation must be inverted so it it  l(t  t  t  is a function of quantity. Therefore the estimated domestic supply equation used in the model is defined as:  41  s  p  =  c +(fi-frX'-('-l) , 1 y d d tf  x  (  4  6  )  (  The domestic supply curve will shift according to regional supply conditions and policy decisions.  4.4.3 Trade Functions  Each region will be either a net importer, a net exporter or neither, depending on the domestic equilibrium price. The trade functions are generated by using each region's 1987 price and quantity data and estimating their respective excess demand or excess supply elasticity. If a region is a net exporter, the slope of the excess supply curve is derived from the elasticity of excess supply, the quantity traded and the domestic price. The intercept of the excess supply curve is equal to the equilibrium price in the domestic market. The excess demand curve is derived using this same method except substituting the excess supply elasticity with its excess demand elasticity. These equations are then used in the objective function to calculate optimal trade relations. The slope of the trade functions remains constant over time since the slope of the domestic supply and demand curves do not change. Changes in domestic conditions are measured in the international market through changes in the market clearing price in each region. Thus, the intercepts of the excess supply and demand curves are altered. International trade conditions are dependent on domestic conditions and the relative size of the domestic  markets. It is obvious that supply changes in BC will have a more dramatic affect on price than a small volume exporter such as Chile.  4.5.4 Transportation Costs  Transportation costs, used to quantify the relative distances between trading partners, play an important role in determining flows and direction of trade.  Transport costs of  commodities with low value-to-weight ratios, such as lumber, are even more important, especially when looking at a region's ability to compete with distant markets. BC's close proximity to the giant US market gives BC forest companies a geographic advantage over competing foreign producers. Foreign marginal producers cannot afford to trade due to their proximity to importing regions. Transportation costs can be difficult to estimate due to highly variable freight rates. Costs vary between regions and over different distances, particularly with respect to port handling fees.  Long-term contracts, energy costs, port facilities, backhaul availability,  commodity and length of haul are all determinants of shipping costs and account for the highly unstable transport costs. In the SE model, transport costs are assumed to be independent of volume and a function only of distance. Sedjo (1983) uses the following function to estimate freight costs as a function of distance: FR = 1.6* (16 + (4 *D)) where: FR is the freight rate per thousand board feet (MBF); and  (4.7)  43 D is the distance in thousands of nautical miles.  Using the above linear relationship is problematic in representing the differences between import and export prices. Since the SE model assumes that transport costs are equal to the price difference between two regions, a strategy used by Cardellichio et al. (1989) is adopted and used to model transportation and other transfer costs in this study. The price equilibrium is defined as follows:  P^Pj+Tji  + Cjt,  (4.8)  where: P is the average product price, 7" is the transportation cost, C is the value adjustment or quality differential, and /* and j are the exporting region and importing regions, respectively.  The difference in price between regions, 7}, + C , is referred to as the transfer costs. Q, can be }i  either positive or negative, depending on tariff and non-tariff barriers as well as the average quality of the lumber shipped compared to the average quality of the domestic market supply. Tariff rates on softwood lumber varyfromregion to region and, in the model, are included in the Cp term. By using this method, transportation costs are simplified and concessions can be made regarding quality differences and trade barriers between regions.  44 This study uses transportation costs derived from the functional form presented by Sedjo (1983) to equal 7},. When modeling changes in transport costs, Q, is held constant and only Tjj is changed. Transportation costs and the value adjustments are detailed in table 4.4 and 4.5, respectively. Adjustments costs less than zero simply imply that the average value of the export is greater than the average value of lumber produced in the importing region. This does not affect the solution procedure.  4.5 GENERAL ASSUMPTIONS OF THE SE MODEL  The assumptions of the SE model, as stated by Enke (1951) (see Chapter 3), require further attention.  The SE model assumes a perfectly competitive market where world  consumption is equal to world production at any time period. Homogeneous goods, per unit transportation costs andfreetrade are simple rules that form the benchmark of the "normal" competitive model. Some problems arise, however, when this model is applied to real market conditions. Additional assumptions need to be imposed on the model in an attempt to create more representative conditions.  4.5.1 Homogeneity  The SE model assumes that each commodity is viewed by the consumer as a homogeneous good. This implies that all lumber is of the same quality and available to all consumers at a single price.  This is an unrealistic assumption as there are often large  45 differences in prices of lumber depending on the grade, size of lumber, method of processing and the species of log. Larger logs require less handling per cubic metre of output, produce more output per cubic metre and usually possess the desired characteristics of performance grade lumber (and therefore return higher profits). Quality differences arise across species as well as within species. For example, performance grade lumber fetches higher prices than structural lumber. In the construction of the post and beam style Japanese housing, structural wood is visible and therefore clear lumber is desired. Also, culturally, the Japanese view knots and defects as structural weaknesses (Sedjo 1983, p 25). Dealing with a commodity such as lumber is a difficult task when creating a forest model. Unlike pulp and paper, thousands of different products and qualities comprise the sawtimber commodity. When estimating supply and demand functions, each function must specify a particular region, a particular time period and a particular final product. Most models assume that sawtimber is a homogeneous product that is perfectly substitutable with timberfromother regions and among species. Published data aggregate many heterogeneous lumber products together when evaluating lumber volumes making no attempt to define quality differences. To address the problem of homogeneity, Sedjo (1983) assumed that a representative basket of sawtimber is produced containing equal proportions of differing qualities for each region. He makes a distinction between coniferous and nonconiferous products by arbitrarily setting hardwood sawtimber prices 10 percent below those of softwood sawtimber. He also discounts plantation softwood by 10 percent to account for the higher percentage of lower quality lumber producedfromthis input. The CGTM makes no attempt to address the lumber  46  quality problem and simply regards it as a limitations of the spatial equilibrium model (Cardellichio et al. 1989). The model does, however, make a distinction between hardwoods and softwoods, and tests for their substitutability. This study does not make reference to different qualities in the base case.  4.5.2 Competitive Market Assumption  In determining a spatial equilibrium it is assumed that a competitive market exists across all regions. Neo-classical long-run conditions hold so that profits equal zero, supply equals demand and average total cost equals price. The SE model maximizes the objective function to yield an efficient allocation of goods within given constraints (Pareto optimality). Efficient allocations are chosen by afirmwhere profits are maximized for a given price. The model used in this study assumes competitive equilibrium and includes parameters in the demand and supply functions to account for exogenous circumstances.  4.5.3 Preferences  The heterogeneity of lumber makes it difficult to model changes in tastes. Consumption of high quality lumber may remain inflated even when lower quality lumber is available at a lower price. Flora, Anderson and McGuiness (1991) found that the offshore demand facing the US had a price elasticity of -1.95 for construction grade logs compared to a price elasticity of -0.80 for performance grade logs. This indicates that higher quality grades  47 are more resilient to changes in prices and more likely to show inertia in trade. It is therefore believed that Canadian softwood lumber will continue to experience high demand even at higher prices (Wallace 1987). Since tastes are driven in part by price, over time tastes will change and demand for higher priced goods will decline. The model is unable to pick up this price-quality interaction directly, and is therefore assumed that each region's exports and imports consist of a representative sample of similar quality lumber.  4.5.4 Inertia  It is a assumed that a trading agent will maximize profit by choosing the lowest cost and therefore the most profitable trade route available. This is not always the case, however, due to trade inertia. Trade inertia is the extent to which historical patterns of trade prevail over time (Kornai 1987).  Once one country has established a trading relationship with  another, it is more likely to continue to trade with that country than to spend the time and money in developing another trade relationship. Many factors influence a nation to carry on a trading relationship even when relative price and cost differentials indicate otherwise. Trade between regions has been driven by numerous factors including culture, geographic location, political structure and accessibility.  Other factors include availability of information,  preferences, costs of changing to other markets and long-term contracts. Prohibitive trade barriers also affect the inertia of trade as do tariffs, non-tariff barriers, embargoes, cartels and lack of information. As international trade becomes more efficient, inertia will play a remote  48 role. For homogeneous goods, inertia is less of a concern than with products that are more diverse in nature. Cardellichio et al. (1989) chose not to place bounds on inertia due to the artificial effect that constraints have on the model. Constraints on inertia do not allow the model to estimate an equilibrium; rather, it forces the model to equate to some preconceived level of trade. The UASA GTM accounts for inertia by placing upper and lower bounds on bilateral trade. By imposing these restrictions, the speed of adjustment is controlled. This study follows the methodology of the CGTM and places no bounds on inertia.  4.6 MODEL ASSUMPTIONS  Since the SE model is unable to predict bi-directional trade flows, it is necessary to make assumptions regarding the production and consumption characteristics of some regions. It is believed that valuable results can be obtained regarding regions that are both importers and exporters. The reason why a region imports lumber even though it already has an excess supply is uncertain, however it could be a function of quality, short-term inventory deficits or location. These factors cannot be modeled in a SE model. In an attempt to gain insight regarding these areas that both import and export, these regions are divided into separate trade areas and designated as either an importer or an exporter. The US South, US West and Canada (excluding BC) are divided into smaller trading blocks. The US West is broken into two regions, the Interior and the Coast. The Interior region includes the US states of Montana, Idaho, Colorado, Nevada, Wyoming, Utah,  49 Arizona, New Mexico, and South Dakota. This region produced approximately 11.6 million cubic metres in 1987 and 7.9 million cubic metres in 1992. The Coast region is made up of Washington and Oregon and produced 31.9 million cubic metres in 1987 and 24.1 million cubic metres in 1992 (Warren 1994). Due to the size of the Interior region and comparatively low production, it is assumed that this region is an importer (importing all of the 3.5 million cubic metres destined to the Pacific Northwest (PNW)). The Coast region is assumed to be the export region because of its size and port facilities. Cardellichio et al. (1989) reported that 16.6 million cubic metres of softwood lumber were shippedfromthe PNW in 1987. The US South is another region that has been divided into an import and an export region for the purpose of this study. The import region is the Atlantic region and the export region is the South-Central region. The US South Atlantic is assumed an importer due to its population base. Both regions produce similar amounts of softwood lumber and have similar production capacities, however the Atlantic region has less extensive inventories (Haynes, Adams and Mills 1995). The division of the South into these two groups is undesirable since the actual quantities that are imported and exported by each region are unclear. The US South reported imports of 9.9 million cubic metres and exports of 4.5 million cubic metres in 1987 (Cardellichio et al. 1989). The creation of these two regions enables inter-state trade to be analyzed. Interior Canada and Eastern Canada are the final regions that need clarification regarding regional assumptions.  Interior Canada (the import region) is comprised of the  Prairies and Ontario, and Eastern Canada (the export region) includes the Maritimes and Quebec. Although Interior Canada exports softwood lumber (primarily to the US), it is  50 assumed an importer since it is Canada's largest consumer. Interior Canada produced 10.3 million cubic metres in 1987 and 9.4 million cubic metres in 1992. The model assumes that 2  Interior Canada imports all shipments destined to Canada. Eastern Canada is assumed an exporter due to its quantity of production. The province of Quebec is Canada's second largest producer, next to BC. It produces over 18 percent of total Canadian production and exports over half this amount to regions outside Canada. The remainder is either consumed within the province or shipped to Interior Canada. In this model it is assumed that Eastern Canada exports all of Canada's softwood lumber, net of BC's share. Although the above assumptions are not ideal, it enables the model to predict regional trade flows. Assumptions about expected future supply conditions are detailed in table 4.6 and are discussed in Chapter 5. Less aggregate data are needed to create a set trade regions that are either importers or exporters.  4.7 MODEL VALIDATION  A simulation model must be tested to see if it is a valid paradigm of the scenario being represented. Of primary concern is the models suitability in testing scenarios for which the model is developed.  The current model is intended to measure, over time, the long-term  effects of policy change by the BC provincial government. Without prior knowledge of the intended use of the model, validation is meaningless. Therefore to test the model's ability to predict policy changes, the model is put through a variety of tests.  Consumption data is unclear since it is only reported as "apparent consumption" (i.e., consumption equals production less exports plus imports).  2  51 First, the model is validated through simulation. In this thesis, the base year is calibrated to 1987 values using existing data. The 'base case' scenario is evaluated and trade flows are simulated in the absence of any policy intervention or exogenous supply factors. Testing the model's ability to accurately project future tradeflowsis an excellent indicator of its predicting capabilities. When actual raw data are compared with predicted values the model should replicate production, consumption and direction of trade with some degree of accuracy. The current model predicts 1987 production values almost exactly (table 4.2). When 1992 values are compared, short-term marketfluctuationscause projected values to divergefromactual values. Second, future simulations of the model are performed and compared with research by industry, government and academia to determine their validity.  It is the goal of the  programmer to develop a model replicating future tradeflowsand it should be in agreement with expectations of future conditions. The current model performs well when compared to other studies (FAO 1991; Cardellichio etal. 1990). A third method of model validation is to perform historical simulations. To perform historical simulations spanning the same time period as performed for the future is difficult and probably of little value. There has been a great deal of structural change in the forest industry over the past 40 years. Since these changes have not been made endogenous, the model is unable to account for these technical developments. Also, the model has been calibrated using the consumer and producer preferences of 1987. Current behaviour is probably more indicative of the future rather than historical market behaviour (Adams and Haynes 1980). Therefore, a model's ability to predict historical trade relationships may be of little value in determining future behaviour. Furthermore, inaccuracies may occur due to short-term estimations. SE models are able to predict  52  long-term trade flow, but short-term fluctuations are extremely difficult, and often undesirable, to model. Due to the dependence of lumber demand on the housing market, the cyclical nature of the economy causes lumber demand to follow the peaks and valleys of the housing market. This causes instabilities and inconsistencies in short-term projections making these estimates questionable at best. The current trade model has not been tested for historical tradeflowsdue to the above reasoning. Thefinaltest of model validation is the model's response to policy changes. The basic test is to see if trade flows, prices and quantities react in an intuitive manner when alternative scenarios are proposed. For example, if a US tariff caused BC lumber import prices to rise 10 percent, it would be expected that the model should show a decrease in BC lumber exports to the US. Further tests on policy include comparing the absolute and relative values of simulation results with actual, projected and intuitively reasonable values. Comparisons may be made with other models, studies orfromindustry estimates.  53  CHAPTER 5 BASE CASE RESULTS AND SIMULATIONS 5.1 BASE CASE SIMULATION This section presents the results of the trade model. The model is calibrated to 1987 price and quantity values, with forecasts generated for the years 1992, 1997, 2002, 2007, 2012 and 2025. The results of these forecasts are referred to as the base case. The base case scenario projects future quantities, prices and bilateral trade flows using expected trends andfixedpolicy conditions at the present time. Projections are based on a basic set of assumptions believed to be most probable in future market conditions (Sedjo and Lyons 1990; FAO 1991; Haynes, Adams and Mills 1995). The base case scenario is not deemed the most likely scenario; rather, it is an indication of how future trade could unfold given present conditions. The base case forms the basis from which all scenarios are compared and is used as a guideline to measure the impacts of a variety of changing market conditions and policy changes. By analysing how different scenarios cause trade to diverge from the base case, insight can be gained regarding the long-term behaviour of global lumber trade. The results may be interpreted and used to aid in policy decisions for industry development. The base case is developedfrom1987 equilibrium conditionsfromwhich assumptions are made about current market conditions. In order to calibrate the modelfromactual trade values in 1987, some constraints were imposed on 1987 trade flows. Although it is optimal to use as few constraints as possible, due to imperfect world markets, it is often necessary to impose trade flow restrictions between some region as well as force trade between other regions. Although this does not yield a true competitive equilibrium solution, it will replicate real world conditions more  54 closely. Real world conditions may or may not encourage trade due to tariff and non-tariff barrier, differing product qualities and long-term contracts. Over time, however it is theorised that equilibrium conditions will prevail. From the 1987 equilibrium conditions, projections are then made for futuretimeperiods. Projections after 1987 are allowed to calibrate to equilibrium values with only the minimum constraints used in a SE model (as discussed in Chapter 3). Future projections are made in 5 year intervals until the year 2012 and then a final projection is made for the year 2025. Assumptions are made about future supply and demand conditions throughout the projection period. These assumptions are discussed below.  5.2 ASSUMPTIONS OF THE BASE CASE  A set of basic conditions are assumed for all regions in the model. These assumptions are based on previous studies (Sedjo and Lyons 1990; FAO 1991; Haynes, Adams and Mills 1995), current conditions and historical trends. The assumptions made in the development of the model are as follows: 1. World demand for lumber increases at an initial rate of 1.4 percent per year until 2002 and then 1.5 percent per year afterward (FAO 1991); this increase is due to population 3  growth and therisein global economic conditions.  Sedjo and Lyons (1990) and the BC Ministry of Forests (1994, p. 217) forecast softwood demand to increase at 1 percent per year.  3  55 2. Global lumber production increases at an initial rate of 0.5 percent per year (Sedjo and Lyons 1990); this increase is due to advancements in biotechnology, harvest, milling and distribution. 3. Exchange rates remain at 1987 values. 4. Relative transportation costs are constant and, therefore, are unchanged.  Using the above assumptions, the base case is generated for the different trade regions in the model. Assumptions 1 and 2 are calculated infiveyear intervals, using the production, consumption and price data of the previous period to determine intercept values for the current period. This process ensures non-linear changes in supply and demand (e.g., demand increases at an increasing rate due to the geometric growth in population).  There is no  increase in demand during the first period since historical data indicate a decline in demand during this period. It is assumed that there is a continual upward trend in the demand for softwood lumber throughout the remainder of the projection period.  The only regional  assumption on demand is one regarding the regions contained in the ROW. The ROW is assumed to demand less softwood lumber than the world average after the year 1992. Many regions within this group have wide access to domestic hardwood lumber supplies and it assumed that, as these supplies develop, regions within ROW will substitute softwood lumber with the hardwood variety.  4  World hardwood consumption is predicted to grow faster than softwood consumption. Consumption is primarilyfromeach producer's domestic market (Waggener, Schreuder and Eastin 1990). 4  56 Assumption (2) is used to account for increased efficiency in production over time. This assumption does not include changes in regional supply due to plantations or reforestation, but does include increases in supply due to the exploitation of existing inventories. Assumptions regardingregionalsupply are discussed below. Constant 1987 exchange rates are assumed throughout the model due to uncertainty in future money markets. It is unlikely that predictions of future money markets would yield better results than the status quo. Buongiomo, Chavas and Uusivuori (1988) determine that it is prices and not exchange rates that are responsible for long-run changes in imports of lumber. Transportation costs are also assumed constant. Under free trade conditions there is the free movement of knowledge, capital and labour, thereby allowing each region equal access to the same technology. Any relative cost advantages will be equated over time.  5.3 REGIONAL ASSUMPTIONS  Knowledge of future supply conditions in each region makes it necessary to include assumptions for certain regions in the development of the base case. Overall global lumber production will increase in the future; however, regional output varies depending on the region. Only assumptions on production are made for individual regions (except for ROW imports) since future production levels can be inferredfromstanding inventory. These assumptions are simplistic and their accuracy debatable, but the recognition of these conditions is important  57 and their inclusion is necessary. A sensitivity analysis of cases where high and low values of each assumption is advisable, but time did not permit such tests for the base case. The base case follows the supply assumptions detailed in table 4.6.  Each supply  change is in response to current inventory assessments and current harvesting technology. This allows assumptions to be made regarding the availability of future quantities of mature supply during any particular period. Most supply assumptions come from Sedjo and Lyons (1990) and Haynes, Adams and Mills (1995) who look at timber inventories in their respective studies. Assumptions regarding individual regions are discussed below.  5.3.1 British Columbia  BC continues to dominate global softwood lumber trade, accounting for almost one-third of total volume traded. The dependence on the US as the world's primary consumer creates a dependence on the US economy and its inherent business cycles. In 1993, BC exported almost three-quarters of its international exports to its southern neighbours, and these trends appear to continue into the future. Increasing supplies in the South, and other foreign supplies, fulfill the void left by the stagnant BC production. Political uncertainty and environmental concerns may force US importers to choose domestic suppliers where imports are guaranteed. Assumptions imposed on the British Columbia export region include a gradual reduction in supply due to the provincial government's reduction of harvestable timber area (Ministry of Forests 1994). Although supplies are becoming more remote and less accessible, a one-time reduction is not assumed in the base case since there are still abundant supply areas  58 to harvest. The model is forecasted assuming that the current AAC (71.6 million cubic metres) will continue into the future. In the base case, BC is assumed to continue lumber production at levels approximately equal to its current output (37 million cubic metres). Although such an assumption is probably unrealistic, it gives a basis to compare changes in trade caused by future declines in BC supply. Present day conditions are used as the base case to compare the effects of the predicted "fall-down" in lumber production.  5.3.2 Eastern Canada  Since Eastern Canada is not yet harvesting at its AAC, some production growth is allowed in the base ease, but after the year 2002 production increases are modest. Although much of Eastern Canada is forested (about two-fifths of Canada's inventory), the location of the timber is often in remote areas requiring long distance transport to processing facilities. The inaccessibility of certain regions make harvesting and reforestation difficult and costly. The model assumes that supply remains relatively constant over time.  5.3.3 Pacific Northwest  The US West regions are assumed to undergo radical reductions in harvestable timber in the short-term. Declining stocks on private lands and the elimination of federal forest lands have caused a worsening of timber supply conditions. Until the late 1980s, timber from government forestland supplied sawtimber for domestic sawmills in the order of about 13.9  59 million cubic metres per year. This ended in 1991 when a federal court injunction shut down most of the national forest program in Washington and Oregon to investigate the environmental impacts on the northern spotted owl. US President Bill Clinton proposed plan, referred to as Option 9, is forcing harvestsfromfederal lands to be cut to one-quarter the 1985-1989 average (Smyth 1993). In the current model, it is assumed that there is a partial reduction in productionfrom1992 to 1997 and a further 20 percent reduction in the next time period. The 20 percent reduction is in response to Option 9 and the declining production of private land. This trend is predicted to stabilize due to growing inventories on private and state land, but not until 2010-2020 (Haynes, Adams and Mills 1995).  5.3.4 US South  The US South is expected to continue its intensive management and reforestation practices. Already the region has been able to increase production to modern day records, up 5 192 thousand cubic metres from 1990 to 1993.  The region is anticipated to increase  production by over 50 percent by the year 2000 (Haynes, Adams and Mills 1995). This trend is expected to continue in the Atlantic region until 2010 and until 2025 in the Central region, or until inventories and harvests start to decline on private lands. Increased demand and falling US supplies are already causing smaller timber to be harvested in the US South. Higher chip prices make the harvest of smaller timber more profitable (Smyth 1993). Also, 5  as lumber prices rise, a shift in timberfrompulp and paper to lumber could occur.  The move downfrom20 centimetre chip-N-saw logs to 15 centimetre diameter in the US South has been required by more severe competition for timber, and higher stumpage and lumber prices (Smyth 1993). 5  60 The model acknowledges the abundant inventories of the US South by imposing 4 percent annual increases for the two US South regions. After the year 2007, the US South Atlantic only experiences growthfromtechnological developments.  5.3.5 Chile and New Zealand  The forest industries of Chile and New Zealand have highly developed plantations capable of producing high volume/short rotation timber. These resources are expected to continue to increase in the future. Although they are viewed inferior quality in current day standards, future demand conditions are changing and, therefore, the model does not make concessions for the quality differences of these products. Future harvests of Chile's plantation forests are estimated to increase by over 100 percent by the year 2005 (Cortes 1988). These projections are estimated from Chile's well stocked plantations. Inventory data indicate that 82 percent of the plantation forests are between 5 and 20 years old, on a 24 year rotation. The current model assumes a 4 percent annual increase in productionfromthe year 1987 to 2002 and then a 3 percent increase thereafter. Similar assumption are made regarding New Zealand production. A majority of their 1.3 million ha plantation forest are less than twenty years old, assuming 25 year rotations. Exports are expected to continue to increase as well (Neilson and Smith 1993). It should be recognized that, although these regions are important in supplying the Pacific Rim, they are both only marginal suppliers in terms of volume.  61  5.4 WELFARE MEASURES  To represent the overall well-being of each region, consumer and producer surplus values are calculated. These measurements calculate the area under the demand and supply curve for each importing and exporting region, respectively. It is used as an assessment of how changes in prices and quantities effect the over-all welfare of a region. For example, it is not obvious if a region's welfare increases if exports increase but price decreases. Increases in supply and demand will have diverse effects on the different regions depending on the slope of the respective supply and demand curves of each region. Consumer surplus (CS) is calculated by taking the area under the demand curve less the total cost of the amount consumed. Likewise the producer surplus (PS) is calculated by subtracting the area under the supply curve from total revenue. Infigure2.1, before trade occurs, the CS of the importing region, Region A, is equal to area faP . It is obvious from 4  thisfigurethat when the region begins trading and prices remain different across regions, CS will increase. Once an optimal trade relationship is established, CS is equal to area fdP , an 1  increase of P^adF. The PS of Region A decreasesfromOaP to OeP*. The over-all welfare 4  gainfromtrade is equal to area aed. Under different scenarios, each region will be better-off than in a non-trade case, however it is of interest to this study to see how changes in supply and demand conditions effect regional welfare. Global welfare is calculated as the sum of regional CS and PS for all regions. Comparisons of welfare measurements across studies are not analysed. Differences in the model structure and its basic set of assumptions make comparisons an irrelevant measurement.  62  5.5 THE BASE CASE FORECAST  Figures 5.1 and 5.2 show the results of the base case forecasts of imports and exports. Total trade increases by almost one-third over the entire projection period. BC exports remain relatively flat whereas the US South Central experiences immense growth in exports. Chile and New Zealand also experience large increases. The US North imports increase substantially because of increased demand that is much larger than the increase in domestic production. The US South Atlantic decline in imports is due to the increased harvest of its privately owned forests. Production, consumption, trade and prices for the export and import regions are detailed in tables 5.1 and 5.2, respectively. Table 5.3 reports the direction of trade results for the base case.  5.5.1 Export Regions  The model predicts that, although BC production remains constant, exports do not decline substantially. The base case predicts that BC lumber destined to the US North, currently BC's primary destination, is replaced by exportsfromthe US South Central and Eastern Canada. Lower transportation costs and a secure supply source could be the reason for a movement awayfromBC lumber. BC diverts this quantity to Japan, the US West Interior and Central Canada. As the US South inventories mature, the US market becomes less important to BC as an importer as exports decline to less than 14.0 million cubic metres. By the year 2025, BC exports over half of its total  63 exports to Japan and Central Canada. Exports to the US South Atlantic have been replaced by the US South Central. The US West experiences the largest decrease in exports because of the spotted owl controversy.  Although the decline is substantial in the 1990s, the base case predicts that  inventories and harvestsfromplantations will increase production to 1987 levels by the year 2025. Also, by 2025 Japan becomes the US West Coast's most important export destination. Continued demand for high quality softwood lumber is expected in Japan. Eastern Canada increases production by almost a third over the 38 year span of the model. Technological advances and increased prices are responsible for this increase. Most of the increase continues to go to the US North region, replacing BC and the US West as suppliers. Total exports increase by approximately 3.0 million cubic metres. Production levels in the US South Central are predicted to increase to levels of 35.812 million cubic metres by the 2025, most of which is either consumed in its domestic market or 6  exported to the US North. The model predicts that the US South Atlantic waits until after 2012 to start importingfromthe US South Central. The US South Atlantic shuts BC out of this market in favour of the US South Central. Scandinavia's increase in production is absorbed by Western Europe and also by the US North. Production levels increase by over 8 million cubic metres to 24.8 million cubic metres by the year 2012. (FAO (1991) predicts that production will be 25.4 million cubic metres by 2010.) It appears that Scandinavia attempts to develop alternative export markets in the US because of increased competitionfromthe ROW region.  Haynes, Adams and Mills (1995) projected that current inventories indicate that by the year 2030 production will reach 38.0 million cubic metres. 6  64 Chile and New Zealand increase production by substantial amounts, although they do not make an impact on the world market. Most of the productionfromboth regions is bound for their domestic markets or the ROW. Australia and Southeast Asia are the primary importers of the ROW group. Chile and New Zealand do enter the European market, but this seems unlikely given the distance between markets. The ROW exporters play a more significant role over time. As demand increases, more supplies are exported to Western Europe. As the former USSR develops its infrastructure, it will play a more prominent role in lumber trade. Currently the east primarily exports logs to Southeast Asia and most of the lumber comesfromthe west.  5.5.2 Import Regions  The most substantial change in imports occurred in the US North where, as discussed above, the US South Central replaced a market currently dominated by BC and the US West Coast. The US North also increased its importsfromthe ROW. The remaining regions did not show any radical changes in trading partners. Japan increased importsfromboth BC and the US West Coast over the 38 year span. It appears that there is a preference for US West Coast lumber over BC lumber; however, restricted production causes importsfromthe US West to fall. Alternative suppliers—Chile, New Zealand and ROW—begin to enter the Japanese market, and appear to be replacing BC and US West lumber to a small degree. Imports from BC decline in 2025 in response to  65 increased production in the US West and alternative sources, although BC still holds over 50 percent of the market share. Western Europe increases importsfromthe ROW in response to increased domestic demand. Scandinavia remains its primary exporter, although its market share is declining due to cheaper importsfromthe ROW. The US South Central increases its exports to Western Europe over time. ExportsfromBC fall to zero in 1997 in response to a realignment of competitive position and non-tariff barriers, including regulations banning green lumber importsfromNorth America.  7  The US West Interior increases its imports to offset the decline in domestic production. BC continues to be supply a majority of the lumber demanded in this region. The US West Coast ships small quantities to the Interior despite higher prices in Japan. By 2025, total importsfromBC approximately equal total production in the US West Interior.  5.5.3 Prices  Projectionsfromthe base case indicate an increase in real prices over time. Real price changes for the base case are detailed in table 5.4. The projected softwood lumber price index is reported in table 5.5. Over time, the real price of BC softwood lumber increases at an average rate of approximately 1.0 percent a year. Most of this price increase occurs in the  The European Community, in 1993, placed a ban on green lumber from North America based on concern about the pinewood nematode found in some BC softwood species. Lumber is required to be kiln dried prior to export to the European Community. 7  66 2012-2025 period (at 1.3 percent per year) when higher domestic demand restricts exports. The changes in price are consistent with Haynes (1990).  5.6 ALTERNATIVE SCENARIOS  The model's ability to represent alternative scenarios relies on each region's demand and supply curve specifications. Each alternative scenario is calculated by shifting the supply curve of a representative region and, therefore, then calculated using the underlying assumptions of the base case. The sensitivity of the model to changes in the intercept rely on the slope of the domestic demand curve in the respective market. Although shifting the constant is not an ideal method to model changes in individual markets, changes in slope cause the model to calculate large fluctuations in trade and make results unrealistic. From the base case projections, a variety of alternative conditions are predicted for the future.  The base case offers one future scenario, with results strongly influenced by the  economic conditions of the base year and the assumptions implicit in the model. The model's ability to analyze alternative scenarios allows the examination of a variety of different future conditions affecting domestic supply and demand conditions. This section looks at some alternative futures. The alternative conditions analyzed, along with a brief description of their importance, are as follows.  67 1. Decrease in B C Production. A 25 percent reduction in BC AAC is modelled by restricting production to 75 percent of the base case levels. The following two scenarios are examined regarding AAC reduction: a) a one-time reduction in AAC where, over time, there is a slow increase in supplies as the Forest Renewal Plan increases future yields (i.e., there is an allowable cut effect); and b) a one-time reduction in AAC with no increase in supply over time.  2. Increase in Supplies from Alternative Sources. The former USSR is of particular interest due to its immense resource base.  A one-time increase in production is modelled where  productionfromthe ROW increased by 20 percent.  Tables 5.1 and 5.2 summarise the production, consumption, trade and prices for the each alternative scenarios. All results are detailed by region and scenario. Direction of trade results are displayed in tables 5.6-5.8. The "winners" and "losers" of each scenario are determined through changes in consumer and producer surplus values. These measurements are found in tables 5.95.11.  5.6.1 Decrease in B C Production  It is well documented that a fall-down in timber supply is probable in BC due to the imbalance in age classes. A reduction in BC's AAC may aid in reducing the impact of the  68 projected fall-down (Ministry of Forests 1994; Smyth 1994). Current harvesting levels cannot sustain the forest industry over the long term. In response, the provincial government is developing a new forest management strategy and assessing each region's timber supplies. This report is to be completed in December 1996; it is predicted that a 25 percent reduction in the current AAC will be needed. Other reductions in BC harvest occurfromthe conversion of forestlands to parks and 8  slow regeneration on previously harvested areas. Scenario 1(a) illustrates the market power that BC has on world lumber trade. The onetime reduction in 1997 causes global lumber prices to increase since other producers cannot adjust quickly to the drop in global supply. Price increases are felt by all import regions in response to a redistribution of lumber trade. BC diverts supplies away from the US North and continues to supply the Canadian, Japanese and US West markets. This deficit in the US North isfilledby the US South Central region. As production in BC increases to 1987 output, prices fall below the base case levels. This response is due to the increase in global production under this scenario. Scenario 1(b) projects the reduction in AAC with no recovery in output. Total exports never recover to the base case output as they did in thefirstscenario. Prices remain inflated in all regions and domestic production increases in regions that consume BC lumber. Exports to Japan decline over time and are replaced by the recovering US West Coast production (table 5.7). BC continues to export to Japan, however the absolute quantities are declining over time. Exports from BC to the US West and the Interior Canada remain relatively constant.  The Commission on Resources and Environment (CORE) recently completed a detailed land-use plan for BC. A wide variety of interest groups were invited to present their land use demands. Among other decisions, the Vancouver Land Use Plan added 480,000 hectares to the park system (CORE 1994).  69 Figure 5.3 and 5.4 illustrate the market share in the alternative scenarios and the base case for exports and imports, respectively. With a decrease in AAC, it is evident that BC market share declines. Most notable is the increase in market share of the US South.  Welfare Changes  Although production decreases in BC, the province experiences a net increase in welfare in the short-run (table 5.9).  In scenario 1(a) higher prices cause consumer surplus to decline,  however producers are better-off through inflated revenues from higher global prices and a reallocation of exports. In the long-run, BC producers are worse off. Increased global demand and higher prices stimulate competition and technological advancements in lumber production. When BC increases production due to the ACE, world prices decline. Japan's producers feel the effect of lower BC prices through a reduction in demandfromtheir domestic market. Imports from BC increase above the base case levels. Lower BC prices also affect the US West producers. Overtime,consumers in Central Canada, BC, Japan and the US West gainfromthe reduction in AAC. Increased importsfromBC cause therisein CS. The biggest loserfromthe AAC reduction is the consumer in the US South Atlantic and the US North. Both regions experience a net loss in consumer welfare of approximately US$176.7 and US$128.4 million, respectively. This is primarily a result of higher prices in their domestic markets and in Eastern Canada. Scenario 1(b) (table 5.10) indicates that most producers will be better-off under a scenario of reduced BC production. So, even though BC experiences a drastic decline in production, the  70 net welfare of the softwood sector increases.  The reduction in BC's AAC has a similar effect on  other exporting regions. Higher prices and a redistribution of exports effectively increase producer surplus in all regions, except for Chile and New Zealand in 2025. Lumber producers in importing countries also gainfromthe reduction in BC production through higher demand for domestic lumber (due to higher import prices). Over-all, however, regional welfare in the importing regions decline; higher prices effectively reduce consumer surplus.  Japan experiences the greatest  reduction in consumer surplus (US$534 million) because of its dependence on BC lumber.  5.6.2 Increased Production in the Former USSR  Overtimenew supply regions develop and compete with existing export countries. It is reported that the former USSR has a forest area of over 800 million ha of which 52 percent of the world's coniferous forests are located (Neilson 1994). Growth rates are very low (0.5-1.5 cubic metres per ha per year) and quality is often inadequate for harvest. However, the absolute size of the resource creates an alternative source to many regions. In 1983 Japanese log imports totalled 6.4 million cubic metres, but has since fallen to about half that level due to political, social and physical access to the resource. This simply emphasises the instability and unpredictability of this region. Inadequate infrastructure and political turmoil have slowed growth in this region. In the base case, the former USSR is grouped within the ROW exporters. ROW production is assumed to increase by over 50 percent (40 million cubic metres)from1992-2025. World prices are projected to fall and global production and exports increase. Western Europe decreases its domestic production in favour of cheaper importsfromthe ROW. BC reduces its  71 total production in response to lower prices and diverts trade awayfromthe US West Interior in favour of Japan, Interior Canada and the US North. Figure 5.3 andfigure5.4 illustrates the changes in market share when exports increase from the former USSR Again, BC market share declines, however BC does capture more of Japan's market for softwood lumber.  Welfare Changes  By the year 2025, an increase in outputfromthe former USSR causes world welfare to increase (see table 5.11). The most significant beneficiary is Western Europe. CS increases by over US$1.3 billion at the expense of a reduction of US$500 million in PS. All other consumers (except for the Rest of the World importers and small losses in Canada) gain from lower prices brought on by the increased production. Producers are net losers due to lower prices. Scandinavia is the worst off since its exports are replaced by the former USSR. The BC market is relatively unaffected by the increase in production, and the redistribution of trade does, in fact, positively affect BC producers.  72  CHAPTER 6 SUMMARY AND CONCLUSIONS 6.1 SUMMARY AND CONCLUSIONS  The purpose of this study is to examine the effects of potential scenarios on global lumber trade. By using a spatial equilibrium model, a long-term equilibrium is established that projects an efficient distribution of lumber. Trade flows between regions are calculated for a base period using a set of assumptions regarding future supply and demand conditions. Assumptions are made using the results of previous studies and current inventory assessments. The model is dependent on price and quantity data, and domestic elasticity estimates. Transportation costs are assumed to equal the difference in prices between any two trading nations. Production, consumption, trade and prices statistics are calculated using this model. The results of the base case indicate that global production increases to 472.8 million cubic metres in 2012 and to 541.8 million cubic metres in 2025. FAO (1991) project that the total lumber demand for softwood and hardwood lumber will be 742.0 million cubic metres in 2010. (Currently hardwoods make-up about 30 percent of all lumber.) Production results of the current study and the CGTM (Cardellichio et al. 1989) are reasonably close for the year 2002 and 2000.  9  Direct comparisons, however, are difficult due to a different set of  assumptions regarding supply and demand conditions in the base case. From the base case projections, alternative scenarios are calculated and then analyzed. The model reacted intuitively to changes in market conditions. The results of the above  9  The CGTM only reports results up to the year 2000.  73 scenarios are significant when considering policy judgments and future investment decision. The results of the current study indicate that BC is indeed a "large country" in terms of softwood lumber trade. This is evident in the projected price increases when a reduction in BC's AAC is modeled. Another conclusion that can be drawnfromthe study is one regarding BC's export destinations. BC will increase exports to Japan and decrease exports to the US. Perez-Garcia (1993) reaches the same conclusion when he modeled a reduction in production. As the US South increases production, BC increases exports to Japan and Interior Canada. Increased production in Chile and New Zealand is found to have little effect on traditional suppliers. Increased production in the US South, distance from market and lower quality are reasons why they are unable to capture more market share. The results from the model suggest that even when supply is low, the US South Atlantic is more likely to increase production than to importfromChile and New Zealand. The results from the welfare analysis reveal importantfindings.The study calculates an increase in welfare for BC with a sustained decrease in the AAC.  This is especially  important since 94 percent of BC forests are publicly owned. Forest companies can increase PS by reducing harvests.  The results indicate that the provincial government can charge  higher stumpage prices while still allowing forest companies to earn comparable PS. It is of interest to note that when BC increases its production after a reduction in AAC (Scenario 1(a)), the over-all welfare of the softwood lumber sector in BC is lower. This suggests that at current production levels higher stumpage prices can be demandedfromBC's producers to force lower rates of harvest. In fact, in the short term, a reduction in AAC causes a large increase in PS at little expense to its domestic consumers.  74 Another importantfindingof the study is the limited effects of the former USSR on BC; producers in BC actually experience an increase in revenues over time.  The former  USSR, on the other hand, has a decline in PS due to lower prices in the region; CS, however, increases drastically and the Rest of the World has a net gain in welfare. These results suggest that the former USSR must be careful in the extraction of its domestic resources. Since the former USSR can play a significant role as a supplier, market saturation is an important consideration in regards to the timing and quantity of additional production. By looking at the changes on welfare, it is evident that a decline in production does not have a negative effect on the province's welfare. BC lumber trade influences world prices enough to cause net revenues for BC forest companies to increase when faced with lower harvests; forest companies' concerns about AAC reduction may very well be ill-founded. The reduction in AAC, however, will cause a reduction in jobs unless labour intensive forest practices are enforced.  6.2 MODEL LIMITATIONS The spatial equilibrium model has inherent limitations that must be considered when results are being analyzed. These limitations include assumptions of homogeneous goods and perfect competition. The variability in transportation costs is another limitation that must be considered. Also, the SE model is unable to calibrate an equilibrium solution when one region is both an importer and an exporter. Some of these limitations can be overcome by less aggregate data and a further breakdown of regions, but such statistics are difficult or even impossible to obtain.  75 Elasticity estimates are another questionable area. Often point elasticities are reported without their respective price/quantity ratios.  This creates some questions regarding the  validity of the estimate. The current model uses elasticity estimates and price/quantity data from the same study, the CGTM (Cardellichio et al. 1989). To some extent, therefore, the results rely on the accuracy of the CGTM estimates. However, significant changes have been made to make the current model more robust. In the current study numerous assumptions are made regarding the regional breakdown in lumber trade. For a few regions (the US South and the US West) these assumptions are made because of lack of additional data. To solve this problem, the model must have either a less complex trade matrix or a more complete data set. Even though the model possesses the above limitations, the forecasts and simulations provide a starting point for future policy and trade considerations. The model tracked 1987 trade flows with a high degree of precision, thereby creating a basis for long-term equilibrium conditions. As addressed above, the results are not intended to provide an exact forecast of the future, rather it is designed to assist in the development of future trade relations.  6.3 SUGGESTIONS FOR FURTHER STUDY  The model presented in this study offers a number of different opportunities for policy analysis. First, however, a more detailed data set is required to model inter-regional trade. Furthermore, the domestic supply and demand curves should be estimated for each region rather than relying on elasticity estimatesfromprevious studies. This, however, could prove  76  to be extremely difficult and the model may not improve its predicting ability. Additional variables should be included in the supply and demand specifications so the model more accurately represents the softwood lumber market. For further analysis of lumber trade, projections should be performed by changing the slope of the domestic supply and demand curves in response to changes in market conditions. Intercept shifters may or may not accurately model these changes.  However sensitivity  analysis should be performed when performing such a test due to the sensitivity of the model to small changes in the slope.  77  Table 4.1 Exporting Regions  Importing Regions  British Columbia Eastern Canada US West Coast US South Central Scandinavia Chile New Zealand Rest of the World Exports (ROW)  Central Canada US West Interior US South Atlantic US North W. Europe Japan Rest of the World Imports (ROW)  b  0  0  h  8  d  1  8  1  Notes: a) Alberta, Saskatchewan, Manitoba and Ontario. b) Quebec and Atlantic provinces. c) Washington and Oregon. d) Montana, Idaho, Colorado, Nevada, Wyoming, Utah, Arizona, New Mexico and South Dakot e) Kentuky, Alabama, Tennessee, Arkansas, Louisiana, Oklahoma and Texas. f) Virginia, North Carolina, South Carolina, Florida and Georgia. g) The Notheast and North Central US States. h) Finland and Sweden. i) All Western Block countries (excluding Finland and Sweden).  Table 4.2 Total Quantity Produced millions of cubic metres  Export Regions British Columbia Eastern Canada US West Coast  Actual 1987 37.6* 14.955 * 31.24*  Predicted 1992 33.373 11.893  a  a  1987 36.912 14.817  1992 34.499 14.471  24.242  b  29.660  28.084  15.340  Scandinavia  10.3 18.7 *  c  18.815  d  10.263 18.605  10.887 19.277  Chile New Zealand  2.311 * 1.844*  2.582 2.465  d  2.674 2.177  2.358 2.029  76.759 *  n.a.  77.599  79.184  US South Central  ROW  Import Regions Interior Canada  Actual 1987  US West Interior US South Atlantic US North Western Europe Japan ROW  9.443 *  Predicted 1992 9  9.311  9.078  10.092 10.119  9.488 11.022  9.440 13.452  c  3.1*  3.540  c  2.957  2.869  33.9*  32.920  d  33.728  35.195  24.423  d  26.127 94.090  26.055 96.037  26.204 * 93.143 *  ** assumed from Cardellichio et al. (1989) ' COFI 1992 Warren 1994 b  4  1992  10.31 * 10.3 **  Cardellichio et al. (1989)  e  a  339  1987 b  n.a. not available *  d  1990 values (Haynes, Adams and Mills 1995) FAO 1994  n.a.  Table 4.3 Lumber Supply and Demand Elasticities Region  British Columbia Central Canada Eastern Canada US West US South US North Scandinavia Western Europe Chile New Zealand Japan Rest of the World  Own-price Elasticity of Supply  Own-price Elasticity of Demand  1.0 1.0 1.0 1.0 1.0 1.4 1.0 1.0 2.8 2.2 0.9 1.0  -0.3 -0.3 -0.3 -0.3 -0.3 -0.3 -0.3 -0.3 -0.3 -0.45 -0.67 -0.3  Source: Cardellichio et al. (1989)  80  Table 4.4 Transportation Costs in 1980 US$/cubic metre Export/Import Int. Canada $18.44 British Columbia Eastern Canada $10.85 US West Coast $18.44 US South Central $13.56 Scandinavia $20.88 Chile $26.58 New Zealand $38.78 ROW $38.78  US West Int. US South Atl. US North $11.93 $19.53 $18.98 $18.44 $13.56 $12.20 $10.85 $19.53 $18.98 $19.53 $10.85 $13.56 $35.80 $24.41 $23.86 $25.49 $20.88 $22.78 $28.47 $35.53 $36.61 $40.14 $45.56 $46.64  W. Europe $35.53 $19.53 $35.25 $22.51 $12.47 $31.73 $45.29 $20.34  Japan $22.51 $40.41 $22.51 $37.15 $38.51 $37.42 $24.41 $27.12  ROW $36.88 $38.78 $40.14 $45.56 $29.83 $34.17 $27.66 $10.85  W. Europe $47.47 $44.47 $9.75 $21.49 $11.53 $53.27 -$20.29 $9.66  Japan $109.49 $72.59 $71.49 $55.85 $34.49 $96.58 $49.59 $51.88  ROW $36.12 $15.22 -$5.14 -$11.56 -$15.83 $40.83 -$12.66 $9.15  Source Sedjo and Lyons 1983  Table 4.5 Value Differential in 1980 US$/cubic metre Export/Import Int. Canada British Columbia $2.56 Eastern Canada -$8.85 US West Coast -$35.44 US South Central -$31.56 Scandinavia -$58.88 Chile -$3.58 New Zealand -$75.78 ROW -$70.78  US West Int. US South Atl. US North $31.07 $28.47 $39.02 $5.56 $15.44 $26.80 -$5.85 -$9.53 $1.02 -$15.53 -$1.85 $5.44 -$51.80 -$35.41 -$24.86 $19.51 $29.12 $37.22 -$43.47 -$45.53 -$36.61 -$50.14 -$50.56 -$41.64  (Value differential =Import Price - Export Price - Transportation Costs)  Table 4.6 Regional Assumptions of Future Supply Conditions - Base Case (percent change per annum)  BC Eastern Canada US West US Sout Atlantic US South Central Chile New Zealand  1992-1997  1997-2002  2002-2007  2007-2025  -0.75 +0.5 -0.75 +2.0 +2.0 +4.0 +4.0  -0.75 +0.5 +4.0 +4.0 +4.0 +4.0 +4.0  -1 no growth no growth +5.0 +5.0 +4.0 +4.0  -1 no growth no growth no growth +1.0 +4.0 +4.0  Source: Haynes, Adams and Mills (1995)  82  Table 5.1 Simulated Production, Consumption, Exports and Prices (Quantities Reported in millions of cubic metres; prices reported in 1980 US$/cubic metre) British Columbia  Year Production 1987 1992 1997 2002 2007 2012 2025  Base Case  Prices 1987 1992 1997 2002 2007 2012 2025  Reduced AAC in BC Increased ScenarioScenario Production in a) b) Former USSR  36.912 34.499 38.626 37.492 36.360 37.660 37.036  36.912 34.499 25.751 25.489 28.896 31.231 36.905  36.912 34.499 25.751 25.401 23.896 26.609 27.600  36.912 34.499 37.569 37.686 35.260 37.548 37.421  4.430 4.499 4.789 5.062 5.383 5.760 6.647  4.430 4.499 4.681 5.040 5.366 5.813 6.676  4.430 4.499 4.681 4.953 5.310 5.626 6.506  4.430 4.499 4.826 5.061 5.426 5.772 6.644  32.482 30.000 33.837 32.430 30.977 31.900 30.389  32.482 30.000 21.070 20.448 23.530 25.418 30.229  32.482 30.000 21.070 20.448 18.585 20.983 21.094  32.482 30.000 32.742 32.625 29.833 31.776 30.777  $95 $90 $102 $106 $111 $113 $130  $95 $90 $110 $108 $111 $108 $128  $95 $90 $110 $114 $115 $121 $137  $95 $90 $99 $107 $108 $112 $131  Consumption 1987 1992 1997 2002 2007 2012 2025  Exports 1987 1992 1997 2002 2007 2012 2025  Eastern Canada  Year  Base Case  Production 1987 1992 1997 2002 2007 2012 2025  14.817 14.471 16.237 16.901 17.660 17.601 19.597  14.817 14.471 17.466 17.842 17.479 18.959 19.630  14.817 14.471 17.466 17.842 18.028 18.948 20.572  14.817 14.471 15.860 16.922 17.101 17.531 19.903  4.143 4.221 4.488 4.729 4.999 5.381 6.210  4.143 4.221 4.352 4.615 5.001 5.213 6.156  4.143 4.221 4.352 4.615 4.941 5.210 6.046  4.143 4.221 4.529 4.729 5.064 5.396 6.187  10.674 10.250 11.749 12.172 12.660 12.220 13.387  10.674 10.250 13.114 13.227 12.477 13.746 13.474  10;674 10.250 13.114 13.227 13.087 13.738 14.526  10.674 10.250 11.331 12.193 12.036 12.135 13.716  $115 $109 $120 $126 $131 $131 $146  $115 $109 $130 $133 $130 $141 $147  $115 $109 $130 $133 $134 $141 $154  $115 $109 $117 $126 $127 $130 $149  Consumption 1987 1992 1997 2002 2007 2012 2025  Exports 1987 1992 1997 2002 2007 2012 2025  Prices 1987 1992 1997 2002 2007 2012 2025  Reduced AAC in BC Increased ScenarioScenario Production in b) a) Former USSR  83  Table 5.1 (continued) Simulated Production, Consumption, Exports and Prices (Quantities Reported in millions of cubic metres; prices reported in 1980 US$/cubic metre) US West Coast  Year Production 1987 1992 1997 2002 2007 2012 2025  Base Case  US South Central  Reduced AAC in BC Increased Scenario Scenario Production in Former USSR ai b)  Year  Base Case  29.660 28.084 25.003 22.631 24.149 24.001 29.642  29.660 28.084 26.149 23.839 23.624 25.542 28.985  29.660 28.084 26.149 23.839 24.425 25.515 31.640  29.660 28.084 24.135 22.632 23.001 24.071 29.758  Production 1987 1992 1997 2002 2007 2012 2025  Consumption 1987 13.446 1992 13.609 1997 14.583 2002 15.503 2007 16.462 2012 17.717 2025 20.625  13.446 13.609 14.429 15.311 16.488 17.468 20.635  13.446 13.609 14.429 15.311 16.381 17.464 20.270  13.446 13.609 14.699 15.526 16.640 17.745 20.653  Consumption 1987 1992 1997 2002 2007 2012 2025  Exports 1987 1992 1997 2002 2007 2012 2025  16.214 14.475 10.420 7.128 7.686 6.284 9.017  16.214 14.475 11.720 8.527 7.135 8.074 8.350  16.214 14.475 11.720 8.527 8.044 8.051 11.370  16.214 14.475 9.435 7.107 6.361 6.326 9.105  Exports 1987 1992 1997 2002 2007 2012 2025  $133 $128 $141 $144 $151 $151 $169  $133 $128 $146 $150 $149 $158 $166  $133 $128 $146 $150 $153 $158 $178  $133 $128 $137 $144 $146 $150 $169  Prices 1987 1992 1997 2002 2007 2012 2025  Prices 1987 1992 1997 2002 2007 2012 2025  Reduced AAC in BC Increased Scenario Scenario Production in Former USSR a) b)  10.263 10.887 14.395 18.652 23.246 26.196 36.030  10.263 10.887 14.661 19.174 23.378 27.195 36.563  10.263 10.887 14.661 19.174 23.736 27.205 36.931  10.263 10.887 14.215 18.543 22.854 26.112 35.749  5.806 5.909 6.321 6.713 7.149 7.678 8.952  5.806 5.909 6.276 6.634 7.150 7.535 8.915  5.806 5.909 6.276 6.634 7.090 7.537 8.857  5.806 5.909 6.351 6.725 7.206 7.679 8.982  4.457 4.978 8.074 11.939 16.097 18.518 27.078  4.457 4.978 8.385 12.541 16.227 19.660 27.647  4.457 4.978 8.385 12.541 16.646 19.668 28.075  4.457 4.978 7.864 11.818 15.648 18.433 26.768  $136 $127 $141 $145 $151 $151 $169  $136 $127 $145 $151 $150 $162 $169  $136 $127 $145 $151 $155 $161 $173  $136 $127 $139 $145 $146 $152 $167  84  Table 5.1 (continued) Simulated Production, Consumption, Exports and Prices (Quantities Reported in millions of cubic metres; prices reported in 1980 US$/cubic metre) Scandinavia  Year  Base Case  Production 1987 1992 1997 2002 2007 2012 2025  Consumption 1987 1992 1997 2002 2007 2012 2025  Exports 1987 1992 1997 2002 2007 2012 2025  Prices 1987 1992 1997 2002 2007 2012 2025  Chile  Reduced AAC in BC Increased Scenario ScenarioProduction in Former USSR a) b)  Year  18.605 19.277 20.412 21.599 22.626 23.473 26.949  18.605 19.277 20.412 21.599 22.641 23.691 27.034  18.605 19.277 19.443 20.179 21.618 22.090 25.150  Production 1987 1992 1997 2002 2007 2012 2025  6.410 6.410 6.387 6.387 6.981 6.958 7.420 -7.376 7.907 7.879 8.424 8.441 9.900 9.887  6.410 6.387 6.958 7.376 7.877 8.419 9.875  6.410 6.387 7.058 7.526 7.995 8.601 10.112  18.605 19.277 20.191 21.177 22.384 23.687 26.850  12.196 12.891 13.210 13.757 14.477 15.263 16.950  12.196 12.891 13.454 14.224 14.748 15.032 17.062  12.196 12.891 13.454 14.224 14.763 15.272 17.159  12.196 12.891 12.385 12.653 13.623 13.488 15.038  $155 $157 $161 $165 $170 $177 $190  $155 $157 $163 $168 $172 $175 $191  $155 $157 $163 $168 $172 $176 $191  $155 $157 $154 $157 $164 $164 $177  Base Case  Reduced AAC in BC Increased Scenario Scenario Production in Former USSR b) a)  2.674 2.358 3.218 4.312 5.253 6.607 10.072  2.674 2.358 3.446 4.370 5.581 6.735 10.235  2.674 2.358 3.446 4.370 5.655 6.754 10.223  2.674 2.358 2.908 4.199 4.760 6.225 9.598  Consumption 1987 1992 1997 2002 2007 2012 2025  1.277 1.304 1.415 1.491 1.592 1.685 1.986  1.277 1.304 1.401 1.490 1.575 1.682 1.984  1.277 1.304 1.401 1.490 1.570 1.681 1.985  1.277 1.304 1.434 1.495 1.618 1.701 2.000  Exports 1987 1992 1997 2002 2007 2012 2025  1.397 1.053 1.803 2.821 3.661 4.922 8.087  1.397 1.053 2.046 2.880 4.007 5.052 8.251  1.397 1.053 2.046 2.880 4.084 5.073 8.237  1.397 1.053 1.475 2.704 3.142 4.524 7.598  Prices 1987 1992 1997 2002 2007 2012 2025  $101 $94 $99 $104 $107 $113 $120  $101 $94 $102 $104 $111 $114 $120  $101 $94 $102 $104 $112 $114 $120  $101 $94 $94 $104 $101 $110 $118  85  Table 5.1 (continued) Simulated Production, Consumption, Exports and Prices (Quantities Reported in millions of cubic metres; prices reported in 1980 US$/cubic metre) New Zealand  Year  Base Case  Rest of the World  Reduced AAC in BC Increased Scenario Scenario Production in Former USSR a)  Year  Reduced AAC in BC Increased Base Scenario Scenario Production in Case a) Former USSR W  77.599 77.599 77.599 79.184 79.184 79.184 82.425 84.146 84.146 87.249 87.828 87.828 91.365 93.217 93.267 96.940 97.272 97.315 111.339 111.344 111.115  Production 1987 1992 1997 2002 2007 2012 2025  2.177 2.029 2.870 3.580 4.307 5.224 7.812  2.177 2.029 2.897 3.600 4.417 5.300 7.875  2.177 2.029 2.897 3.600 4.449 5.320 7.884  2.177 2.029 2.699 3.562 4.164 5.063 7.661  Production 1987 1992 1997 2002 2007 2012 2025  Consumption 1987 1992 1997 2002 2007 2012 2025  1.687 1.729 1.838 1.956 2.086 2.216 2.613  1:687 1.729 1.834 1.954 2.069 2.206 2.609  1.687 1.729 1.834 1.954 2.064 2.203 2.609  1.687 1.729 1.867 1.955 2,105 2.235 2.622  Consumption 1987 64.169 1992 64.238 1997 70.345 2002 74.574 2007 79.680 2012 84.828 2025 99.332  64.169 64.238 69.912 74.409 79.186 84.691 99.255  64.169 64.238 69.912 74.409 79.173 84.679 99.311  64.169 64.238 70.880 75.644 80.377 86.148 101.450  Exports 1987 1992 1997 2002 2007 2012 2025  0.490 0.300 1.032 1.624 2.221 3.008 5.199  0.490 0.300 1.063 1.646 2.347 3.093 5.266  0.490 0.300 1.063 1.646 2.385 3.117 5.275  0.490 0.300 0.833 1.607 2.058 2.828 5.039  Exports 1987 1992 1997 2002 2007 2012 2025  13.430 14.945 12.081 12.676 11.685 12.112 12.007  13.430 14.945 14.234 13.419 14.031 12.581 12.089  13.430 14.945 14.234 13.419 14.094 12.635 11.804  13.430 14.945 25.254 23.680 25.165 23.058 20.802  Prices 1987 1992 1997 2002 2007 2012 2025  $158 $149 $162 $164 $168 $173 $180  $158 $149 $163 $165 $171 $175 $180  $158 $149 $163 $165 $172 $175 $180  $158 $149 $156 $165 $164 $170 $180  Prices 1987 1992 1997 2002 2007 2012 2025  151.641 151.099 153.565 158.966 162.745 169.176 185.001  151.641 151.099 156.928 160.012 166.252 169.621 184.766  151.641 151.099 156.928 160.012 166.351 169.703 184.310  151.641 151.099 149.406 150.943 158.242 160.245 171.869  77.599 79.184 96.134 99.324 105.542 109.205 122.252  Table 5.1 (continued)  Simulated Production, Consumption, Exports and Prices (Quantities Reported in millions of cubic metres; prices reported in 1980 US$/cubic metre) Global Total  Year  Base Case  Reduced AAC in BC Scenario Scenario b) a)  Increased Production in Former USSR  Production 1987 1992 1997 2002 2007 2012 2025  192.707 190.790 202.967 211.995 224.723 237.915 278.378  192 707 190 790 194 929 203 740 219 217 235 706 278 486  192.707 190.790 194.929 203.653 216.096 231.356 272.999  192.707 190.790 212.963 223.048 234.299 247.844 287.492  101.368 101.896 110.760 117.448 125.260 133.689 156.264  101 368 101 896 109 843 116 829 124 715 133 050 156 117  101.368 101.896 109.843 116.741 124.407 132.819 155.460  101.368 101.896 111.645 118.661 126.433 135.277 158.649  91.339 91 339 88.894 88 894 92.207 85 086 94.547 86 911 99.464 94 503 104.227 102 656 122.113 122 368  91.339 88.894 85.086 86.911 91.690 98.537 117.539  91.339 88.894 101.318 104.387 107.866 112.567 128.843  134.059 132.049 138.630 143.809 148.931 152.829 169.158  134.059 132.049 145.649 149.700 154.527 158.673 173.130  134.059 132.049 135.873 140.156 145.481 148.275 162.918  Consumption 1987 1992 1997 2002 2007 2012 2025  Exports 1987 1992 1997 2002 2007 2012 2025  Prices 1987 1992 1997 2002 2007 2012 2025  134 059 132 049 145 649 148 882 151 878 156 047 168 580  87  Table 5.2 Simulated Production, Consumption, Imports and Prices Interior Canada  Year Production 1987 1992 1997 2002 2007 2012 2025  Base Case  US West Interior  Reduced AAC in BC Increased Scenario ScenarioProduction in ai hi Former USSR  Year  Base Case  Reduced AAC in BC Increased Scenario ScenarioProduction in Former USSR a) b)  9.311 9.078 10.354 10.645 10.918 11.227 12.799  9.311 9.078 10.963 11.185 11.065 11.163 12.616  9.311 9.078 10.963 11.185 11.374 12.113 13.271  9.311 9.078 9.972 10.641 10.722 11.487 12.799  Production 1987 1992 1997 2002 2007 2012 2025  10.092 9.488 8.558 7.787 8.080 8.410 10.369  10.092 9.488 9.114 8.399 8.076 7.567 9.967  10.092 9.488 9.114 8.399 8.311 8.948 10.779  10.092 9.488 8.291 7.827 7.945 8.399 10.228  Consumption 1987 15.508 1992 15.732 1997 16.790 2002 17.823 2007 19.026 2012 20.421 23.632 2025  15.508 15.732 16.491 17.537 18.911 20.404 23.669  15.508 15.732 16.491 17.537 18.759 19.930 23.247  15.508 15.732 16.978 17.838 19.136 20.314 23.632  Consumption 1987 13.883 1992 14.078 1997 15.049 2002 15.982 2007 17.063 2012 18.283 2025 21.281  13.883 14.078 14.826 15.693 16.999 18.552 21.403  13.883 14.078 14.826 15.693 16.905 17.992 20.998  13.883 14.078 15.156 15.987 17.139 18.314 21.369  Imports 1987 1992 1997 2002 2007 2012 2025  6.196 6.653 6.436 7.178 8.108 9.194 10.834  6.196 6.653 5.528 6.352 7.846 9.241 11.053  6.196 6.653 5.528 6.352 7.385 7.817 9.976  6.196 6.653 7.006 7.197 8.413 8.828 10.832  Imports 1987 1992 1997 2002 2007 2012 2025  3.791 4.590 6.491 8.195 8.983 9.874 10.911  3.791 4.590 5.712 7.293 8.923 10.986 11.436  3.791 4.590 5.712 7.293 8.593 9.044 10.219  3.791 4.590 6.865 8.160 9.194 9.915 11.141  $116 $111 $124 $127 $131 $132 $151  $116 $111 $131 $134 $133 $131 $149  $116 $111 $131 $134 $137 $143 $157  $116 $111 $119 $127 $128 $135 $151  $137 $131 $145 $149 $153 $155 $175  $137 $131 $153 $159 $154 $145 $171  $137 $131 $153 $159 $158 $164 $181  $137 $131 $142 $150 $151 $155 $172  Prices 1987 1992 1997 2002 2007 2012 2025  Prices 1987 1992 1997 2002 2007 2012 2025  Table 5.2 (continued) Simulated Production, Consumption, Imports and Prices (Quantities Reported in millions of cubic metres; prices reported in 1980 US$/cubic metre)  US South Atlantic  Year Production 1987 1992 1997 2002 2007 2012 2025  Base Case  US North  Reduced AAC in BC Increased ScenarioScenario Production in Former USSR a) b)  Year  Base Case  Reduced AAC in BC Increased ScenarioScenario Production in Former USSR a) b)  10.119 11.022 14.410 18.620 21.297 22.973 25.669  10.119 11.022 14.794 19.890 21.591 23.772 26.158  10.119 11.022 14.794 19.890 21.768 23.821 26.294  10.119 11.022 14.146 18.579 21.077 22.882 25.597  Production 1987 1992 1997 2002 2007 2012 2025  2.957 2.869 3.290 3.481 3.716 3.864 4.527  2.957 2.869 3.589 3.679 3.706 4.042 4.552  2.957 2.869 3.589 3.679 3.824 4.094 4.698  2.957 2.869 3.148 3.440 3.598 3.829 4.527  Consumption 1987 20.307 1992 20.490 1997 22.004 23.399 2002 2007 24.949 26.773 2012 2025 31.286  20.307 20.490 21.778 22.699 24.863 26.397 31.049  20.307 20.490 21.778 22.699 24.759 26.368 30.966  20.307 20.490 22.160 23.392 25.042 26.788 31.290  Consumption 1987 28.487 28.810 1992 1997 30.982 2002 32.931 2007 35.101 2012 37.621 2025 44.122  28.487 28.810 30.376 32.501 35.071 37.208 43.962  28.487 28.810 30.376 32.501 34.831 37.090 43.636  28.487 28.810 31.269 33.027 35.358 37.723 44.168  Imports 1987 1992 1997 2002 2007 2012 2025  10.188 9.468 7.594 4.779 3.652 3.800 5.618  10.188 9.468 6.984 2.809 3.272 2.624 4.891  10.188 9.468 6.984 2.809 2.991 2.547 4.671  10.188 9.468 8.014 4.813 3.966 3.906 5.693  Imports 1987 1992 1997 2002 2007 2012 2025  25.529 25.942 27.692 29.451 31.385 33.757 39.595  25.529 25.942 26.787 28.822 31.365 33.167 39.410  25.529 25.942 26.787 28.822 31.007 32.996 38.938  25.529 25.942 28.121 29.587 31.761 33.894 39.641  $142 $138 $151 $154 $159 $160 $177  $142 $138 $156 $171 $160 $168 $179  $142 $138 $156 $171 $162 $168 $181  $142 $138 $147 $155 $157 $160 $178  $153 $148 $160 $164 $170 $172 $187  $153 $148 $171 $171 $169 $178 $187  $153 $148 $171 $171 $173 $180 $192  $153 $148 $155 $163 $166 $171 $188  Prices 1987 1992 1997 2002 2007 2012 2025  Prices 1987 1992 1997 2002 2007 2012 2025  89  Table 5.2 (continued) Simulated Production, Consumption, Imports and Prices (Quantities Reported in millions of cubic metres; prices reported in 1980 USS/cubic metre) Western Europe  Year Production 1987 1992 1997 2002 2007 2012 2025  Base Case  Japan  Reduced AAC in BC Increased ScenarioScenario Production in b) Former USSR a)  Year  Base Case  Reduced AAC in BC Increased Scenario Scenario Production in Former USSR ai b)  33.728 35.195 36.754 38.154 40.059 42.303 47.799  33.728 35.195 36.859 38.728 40.554 42.284 47.786  33.728 35.195 36.859 38.728 40.498 42.169 47.626  33.728 35.195 35.420 36.820 38.999 39.713 45.135  Production 1987 1992 1997 2002 2007 2012 2025  26.127 26.055 28.038 29.141 30.325 31.217 35.205  26.127 26.055 28.815 29.900 30.387 31.729 34.893  26.127 26.055 28.815 29.900 30.858 32.158 36.067  26.127 26.055 27.626 29.129 29.925 31.147 35.065  Consumption 1987 53.315 1992 52.492 1997 56.890 2002 60.269 2007 63.598 2012 66.810 2025 76.397  53.315 52.492 56.758 59.544 62.935 66.758 76.325  53.315 52.492 56.758 59.544 63.005 66.906 76.549  53.315 52.492 58.561 62.017 65.091 70.299 80.447  Consumption 1987 33.068 1992 33.724 1997 35.847 2002 37.981 2007 40.401 2012 43.306 2025 49.914  33.068 33.724 35.120 37.236 40.273 42.747 50.050  33.068 33.724 35.120 37.236 39.831 42.323 48.870  33.068 33.724 36.234 38.009 40.796 43.411 50.101  Imports 1987 1992 1997 2002 2007 2012 2025  19.586 17.297 20.137 22.114 23.539 24.507 28.599  19.586 17.297 19.898 20.817 22.382 24.474 28.540  19.586 17.297 19.898 20.817 22.507 24.737 28.923  19.586 17.297 23.141 25.197 26.092 30.586 35.312  Imports 1987 1992 1997 2002 2007 2012 2025  6.941 7.670 7.809 8.840 10.076 12.089 14.710  6.941 7.670 6.305 7.335 9.886 11.018 15.157  6.941 7.670 6.305 7.335 8.974 10.165 12.803  6.941 7.670 8.608 8.881 10.871 12.264 15.036  $179 $183 $186 $189 $194 $200 $215  $179 $183 $187 $192 $196 $200 $215  $179 $183 $187 $192 $196 $200 $214  $179 $183 $179 $182 $189 $187 $202  $228 $221 $234 $238 $243 $244 $263  $228 $221 $242 $246 $243 $249 $259  $228 $221 $242 $246 $248 $253 $270  $228 $221 $230 $238 $239 $244 $262  Prices 1987 1992 1997 2002 2007 2012 2025  Prices 1987 1992 1997 2002 2007 2012 2025  90 Table 5.2 (continued) Simulated Production, Consumption, Imports and Prices (Quantities Reported in millions of cubic metres; prices reported in 1980 US$/cubic metre) Rest of the World Reduced AAC in BC Increased Scenario ScenarioProduction in Former USSR a) b)  Year  Base Case  Production 1987 1992 1997 2002 2007 2012 2025  94.090 96.037 100.063 105.219 109.612 115.946 127.016  94.090 96.037 101.659 105.586 111.723 115.828 126.975  94.090 96.037 101.659 105.586 112.160 115.764 126.881  Consumption 1987 113.197 1992 113.310 1997 116.115 2002 119.209 2007 123.333 2012 126.952 2025 138.863  113.197 113.310 115.532 119.069 122.553 126.975 138.857  19.107 17.273 16.049 13.990 13.721 11.006 11.847  $172 $171 $174 $179 $182 $189 $195  Imports 1987 1992 1997 2002 2007 2012 2025  Prices 1987 1992 1997 2002 2007 2012 2025  Global Total Reduced AAC in BC Increased Scenario ScenarioProduction in Former USSR b) a)  Year  Base Case  94.090 96.037 97.491 100.421 106.821 114.396 127.557  Production 1987 1992 1997 2002 2007 2012 2025  186.425 189.743 201.470 213.046 224.006 235.940 263.382  186.425 189.743 205.794 217.367 227.102 236.384 262.947  186.425 189.743 205.794 217.367 228.793 239.066 265.617  186.425 189.743 196.095 206.855 219.085 231.854 260.909  113.197 113.310 115.532 119.069 122.393 126.996 138.890  113.197 113.310 117.056 120.973 124.389 127.571 138.745  Consumption 1987 277.764 1992 278.636 1997 293.677 2002 307.593 2007 323.470 2012 340.167 2025 385.496  277.764 278.636 290.879 304.278 321.604 339.041 385.315  277.764 278.636 290.879 304.278 320.483 337.603 383.156  277.764 278.636 297.414 311.242 326.952 344.420 389.752  19.107 17.273 13.872 13.483 10.830 11.146 11.882  19.107 17.273 13.872 13.483 10.233 11.232 12.008  19.107 17.273 19.565 20.552 17.569 13.175 11.188  Imports 1987 1992 1997 2002 2007 2012 2025  91.339 88.894 92.207 94.547 99.464 104.227 122.113  91.339 88.894 85.086 86.911 94.503 102.656 122.368  91.339 88.894 85.086 86.911 91.690 98.537 117.539  91.339 88.894 101.318 104.387 107.866 112.567 128.843  $172 $171 $177 $180 $186 $188 $195  $172 $171 $177 $180 $187 $188 $195  $172 $171 $169 $170 $177 $186 $197  Prices 1987 1992 1997 2002 2007 2012 2025  174.472 172.988 179.024 182.793 186.536 191.207 203.029  174.472 172.988 182.779 186.724 188.939 192.173 202.291  174.472 172.988 182.779 186.724 190.472 193.497 204.614  174.472 172.988 174.201 177.435 182.320 187.685 201.283  Table 5.3 Direction of Trade - Base Case (in millions of cubic metres) 1987 Export/Import Int Canada US West Int US South AtL US North W.Europe 9.88 3.39 British Columbia 4.33 2.81 6.99 0.34 0.07 1.09 6.55 1.24 Eastern Canada 2.09 7.33 0.00 US West Coast 1.30 0.72 0.19 1.73 1.06 0.23 0.02 US South Central Scandinavia 0.00 0.00 0.00 0.05 12.11 0.00 0.81 Chile 0.00 0.00 0.00 0.00 0.00 0.00 0.00 New Zealand 0.00 ROW 0.00 0.00 0.00 0.00 0.98 Total Imports 6.20 3.79 10.19 25.53 19.59  Japan 3.50 0.00 3.27 0.00 0.00 0.17 0.00 0.00 6.94  ROWTotal Export 1.59 32.48 1.39 10.67 1.50 16.21 1.23 4.46 12.20 0.04 0.42 1.40 0.49 0.49 12.45 13.43 91.34 19.11  1992 Export/Import Int Canada US West Int US South AU. British Colombia S.148 3.544 2.485 0.069 0.975 Eastern Canada 0.339 0.926 0.767 5.987 US West Coast 0.190 0.021 US South Central 0.239 0.000 0.000 0.000 Scandinavia Chile 0.000 0.019 0.000 0.000 0.000 0.000 New Zealand 0.000 0.002 0.000 ROW Total Imports 6.653 4.590 9.468  US North W.Europe 14.821 0.330 7.686 0.000 0.719 0.000 2.715 0.940 0.000 12.890 0.000 0J54 0.199 0.000 2.784 0.000 17.297 25.942  Japan 2.487 0.001 5.028 0.001 0.000 0.153 0.000 0.000 7.670  ROWTotal Export 1.186 30.000 1.180 10.250 1.048 14.475 0.871 4.978 12.891 0.000 0.728 1.053 0.300 0.101 12.160 14.945 17.273 88.894  1997 Export/Import Int Canada US West Int US South AtL British Columbia 5.130 5.253 2.451 0.346 0.061 1.217 Eastern Canada US West Coast 0.510 0.975 3.892 0.448 0.195 0.022 US South Central Scandinavia 0.000 0.000 0.000 0.000 0.003 0.001 Chile New Zealand 0.000 0.000 0.000 0.000 ROW 0.002 0.012 Total Imports 6.436 6.491 7.594  US North W.Europe 14.497 0.000 8.346 0.000 0.733 0.000 4.087 1.786 0.000 13.210 0.000 0.074 0.000 0.079 0.028 4.987 20.137 27.692  Japan 4.886 0.000 2.905 0.000 0.000 0.001 0.000 0.017 7.809  ROWTotal Export 1.620 33.837 1.779 11.749 1.404 10.420 1.536 8.074 0.000 13.210 1.724 1.803 0.952 1.032 7.034 12.081 16.049 92.207  2002 Export/Import Int Canada US West Int US South AtL US North W.Europe 5.46 6.19 2.18 10.53 0.00 British Columbia 0.07 Eastern Canada 0.35 1.64 9.77 0.00 US West Coast 0.90 1.46 0.91 0.69 0.00 US South Central 0.27 0.29 0.02 8.43 2.45 Scandinavia 0.00 0.00 0.00 0.01 13.74 0.02 Chile 0.02 0.00 0.00 0.01 New Zealand 0.16 0.16 0.00 0.00 0.32 0.03 0.00 0.02 0.01 5.59 ROW Total Imports 7.18 8.20 4.78 22.11 29.45  Japan 7.79 0.00 0.97 0.00 0.00 0.02 0.04 0.02 8.84  ROW Total Export 0.27 32.43 0.34 12.17 2.19 7.13 0.47 11.94 0.00 13.76 2.76 2.82 0.94 1.62 7.01 12.68 13.99 94.55  92  Table 5.3 (continued) Direction of Trade - Base Case (in millions of cubic metres) 2007 Export/Import Int Canada US West Int US South AtL US North W.Europe BC 6.505 7.505 1.600 7.229 0.000 Quebec and East Coast 0.352 0.066 1.061 10.038 0.000 US West Coast 0.501 0.987 0.493 2.191 0.000 US South Central 0.371 11.489 0.213 0.034 2.867 Scandinavia 0.000 0.000 0.001 0.001 14.473 0.167 Chile 0.152 0.185 0.127 0.162 New Zealand 0.058 0.058 0.076 0.051 0.546 ROW 0.153 0.001 0.201 0.257 5.491 Total Imports 8.108 8.983 3.652 31.385 23.539  Japan 7.079 0.000 2.588 0.000 0.001 0.186 0.012 0.210 10.076  ROW Total Exports 1.058 30.977 1.143 12.660 0.927 7.686 1.122 16.097 0.000 14.477 3.661 2.682 1.418 2.221 5.371 11.685 13.721 99.464  2012 Export/Import British Columbia Eastern Canada US West Coast US South Central Scandinavia Chile New Zealand ROW Total Imports  Int Canada US West Int US South AtL 7.633 8.245 2.659 0.346 0.071 0.962 0.756 1.171 0.021 0.456 0.301 0.023 0.000 0.000 0.000 0.000 0.056 0.000 0.003 0.028 0.075 0.000 0.002 0.060 9.194 9.874 3.800  US North W.Europe 5.345 0.000 10.765 0.000 0.637 0.000 16.789 0.830 0.000 15.263 0.112 0.281 0.003 0.247 0.106 7.886 33.757 24.507  Japan 8.011 0.000 3.666 0.000 0.000 0.098 0.168 0.146 12.089  ROW Total Exports 0.005 31.900 0.076 12.220 0.033 6.284 0.119 18.518 0.000 15.263 4.376 4.922 2.485 3.008 3.912 12.112 11.006 104.227  2025 Export/Import British Columbia Eastern Canada US West Coast US South Central Scandinavia Chile New Zealand ROW Total Imports  Int Canada US West Int US South AtL 8.660 9.516 0.000 0.084 0.059 0.083 0.843 0.596 0.000 1.206 0.646 5.408 0.003 0.000 0.025 0.018 0.062 0.034 0.003 0.031 0.037 0.017 0.002 0.031 10.834 10.911 5.618  US North W.Europe 4.464 0.000 13.105 0.000 1.157 0.000 16.972 2.680 1.818 15.093 0.222 0.454 0.003 0.421 1.853 9.951 39.595 28.599  Japan 7.744 0.000 6.421 0.000 0.011 0.118 0.262 0.154 14.710  ROW Total Exports 0.005 30.389 0.055 13.387 0.000 9.017 0.166 27.078 0.000 16.950 7.179 8.087 4.442 5.199 0.000 12.007 11.847 122.113  Table 5.4 Projected Real Softwood Lumber Prices (1980 US$ per cubic metre) Actual  Export Regions British Columbia Eastern Canada US West Coast US South Central Scandinavia Chile New Zealand ROW  1987 97 116 135 136 156 95 155 150 Actual  Import Regions  Interior Canada US West Interior US South Atlantic US North Western Europe Japan ROW  1987  118 140 145 155 180 229 170  Predicted  1987 95.23 114.93 133.47 135.51 155.21 100.52 157.58 151.64  1992 90.14 109.49 127.98 127.49 157.09 93.96 149.13 151.10  1997 101.90 120.38 140.97 141.47 160.70 98.82 162.02 153.57  2002 106.45 125.53 144.36 145.40 164.71 104.29 164.17 158.97  2007 110.78 131.41 151.19 150.65 170.36 107.03 167.53 162.75  2002  2007  2012 112.73 130.96 150.52 151.24 176.56 113.44 173.01 169.18  2025 130.07 146.44 168.89 168.67 190.10 120.21 180.08 185.00  Predicted  1987  1992  116.35 110.65 137.18 130.60 142.45 138.06 153.42 147.53 179.09 182.58 228.29 221.50 171.73 171.16  1997  123.76 145.37 150.86 160.45 186.18 234.44 174.13  127.39 130.80 149.43 153.42 154.34 159.25 164.43 169.90 188.74 193.79 238.33 242.76 178.97 182.19  2012  2025  131.60 155.43 160.37 171.94 200.39 244.06 188.75  151.23 174.63 177.30 187.14 214.97 263.08 195.20  2012  2025  1.18 1.14 1.13 1.12 1.14 1.13 1.10 1.12  1.37 1.27 1.27 1.24 1.22 1.20 1.14 1.22  2012 1.13 1.13 1.13 1.12 1.12 1.07 1.10  2025 1.30 1.27 1.24 1.22 1.20 1.15 1.14  Table 5.5 Projected Softwood Lumber Price Index (1987=100) Export Regions  British Columbia Eastern Canada US West Coast US South Central Scandinavia Chile New Zealand ROW Import Regions  Interior Canada US West Interior US South Atlantic US North Western Europe Japan ROW  1987  1992  1997  1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00  0.95 0.95 0.96 0.94 1.01 0.93 0.95 1.00  2002  1.07 1.05 1.06 •1.04 1,04 0.98 1.03 1.01  2007  1.12 1.09 1.08 1.07 1.06 1.04 1.04 1.05  1.16 1.14 1.13 1.11 1.10 1.06 1.06 1.07  1987 1.00 1.00 1.00 1.00 1.00 1.00 1.00  1992 0.95 0.95 0.97 0.96 1.02 0.97 1.00  1997 1.06 1.06 1.06 1.05 1.04 1.03 1.01  2002 1.09 1.09 1.08 1.07 1.05 1.04 1.04  2007 1.12 1.12 1.12 1.11 1.08 1.06 1.06  94 Table 5.6 Direction of Trade - Reduced AAC in BC (Scenario a) (in millions of cubic metres) 1987 Export/Import British Columbia Eastern Canada US West Coast US South Central Scandinavia Chile New Zealand ROW Total Imports  Int Canada 4.33 0.34 130 0.23 0.00 0.00 0.00 0.00 620  US West Int US South Ad. 2.81 6.99 0.07 1.09 0.72 2.09 0.19 0.02 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 3.79 10.19  US North W. Europe 9.88 339 6.55 124 733 0.00 1.73 1.06 0.05 12.11 0.00 0.81 0.00 0.00 0.00 0.98 25.53 19.59  Japan 3.50 0.00 3.27 0.00 0.00 0.17 0.00 0.00 6.94  ROW Total Exports 1.59 32.48 139 10.67 1.50 1621 123 4.46 0.04 1270 0.42 1.40 0.49 0.49 12.45 13.43 19.11 9134  Export/Import British Columbia Eastern Canada US West Coast US South Central Scandinavia Chile New Zealand ROW Total Imports  Int Canada 5.148 0339 0.926 0739 0.000 0.000 0.000 0.000 6.653  US West Int US South Ad. 3.544 2.485 0.069 0.975 0.767 5.987 0.190 0.021 0.000 0.000 0.019 0.000 0.000 0.000 0.002 0.000 4.590 9.468  US North W.Europe 14.821 0.330 7.686 0.000 0.719 0.000 2.715 0.940 0.000 12.890 0.000 0.154 0.000 0.199 0.000 2.784 25.942 17297  Japan 2.487 0.001 5.028 0.001 0.000 0.153 0.000 0.000 7.670  ROW Total Exports 1.186 30.000 1.180 10550 1.048 14.475 0.871 4.978 0.000 12.891 0.728 1.053 0.101 0.300 12.160 14.945 17273 88.894  1997 Export/Import British Columbia Eastern Canada USWestCoast US South Central Scandinavia Chile NewZealand ROW Total Imports  Int Canada 4.154 0.337 0.568 0.467 0.000 0.000 0.000 0.000 5.528  US West Int US South Ad. 4215 1.316 0.061 1.010 1202 4.387 0.199 0.022 0.000 0235 0.003 0.001 0.029 0.001 0.002 0.012 5.712 6.984  US North W.Europe 8368 0.000 11.025 0.000 0.857 0.000 5.953 0.920 0290 12.928 0212 0.074 0.051 0.078 0.030 5.898 26.787 19.898  Japan 2.455 0.000 3.756 0.000 0.000 0.001 0.075 0.018 6.305  ROW Total Exports 0.563 21.070 0.680 13.114 0.950 11.720 0.824 8385 0.000 13.454 1.754 2.046 0.829 1.063 8273 14234 13.872 85.086  Int Canada 429 035 122 027 0.00 0.02 0.18 0.03 635  US West Int US South Ad. 4.44 0.73 0.07 1.01 2.30 1.02 030 0.02 0.00 0.00 0.02 0.00 0.17 0.00 0.00 0.02 729 2.81  US North W.Europe 4.85 0.00 11.51 0.00 0.87 0.00 11.57 0.00 0.01 1421 0.00 0.01 0.00 032 0.01 627 28.82 20.82  Japan 5.90 0.00 1.36 0.00 0.00 0.02 0.04 0.02 7.34  ROW Total Exports 023 20.45 029 1323 1.77 8.53 0.38 12.54 0.00 1422 2.81 2.88 0.93 1.65 7.07 13.42 13.48 86.91  1992  2002 Export/Import British Columbia Eastern Canada USWestCoast US South Central Scandinavia Chile NewZealand ROW Total Imports  95 Table 5.6 (continued)  Direction of Trade - Reduced AAC in BC (Scenario a) (in millions of cubic metres)  2007 Export/Import British Columbia Eastern Canada US West Coast US South Central Scandinavia Chile New Zealand ROW Total Imports  Int Canada US West Int 5.717 7.592 0.313 0.065 0.473 0.841 0.329 0500 0.609 0.000 0.181 0.164 0.060 0.061 0.165 0.001 7.846 8.923  US South Ad. US North 1.341 0.842 0.880 11.146 0.497 3.331 0.034 15.533 0.001 0.001 0510 0.143 0.081 0.054 0529 0.315 3572 31.365  W.Europe 0.000 0.000 0.000 0.019 14.135 0.165 0.579 7.484 22.382  Japan 7.687 0.000 1.753 0.000 0.001 0504 0.013 0528 9.886  ROW Total Exports 0.352 23.530 0.074 12.477 0542 7.135 0.112 16527 0.000 14.748 2.940 4.007 1.501 2.347 5.609 14.031 10.830 94.503  2012 Export/Import British Columbia Eastern Canada US West Coast US South Central Scandinavia Chile New Zealand ROW Total Imports  Int Canada 7.358 0.358 0.833 0.476 0.094 0.063 0.003 0.057 9.241  US West Int 9.147 0.071 1371 0310 0.000 0.056 0.029 0.002 10.986  US South Ad. US North 1.504 2.042 0.922 12.320 0.021 0.656 0.023 17.924 0.012 0.002 0.008 0.114 0.075 0.003 0.060 0.107 2.624 33.167  W. Europe 0.000 0.000 0.000 0.809 14.845 0586 0550 8584 24.474  Japan 5.362 0.000 5.161 0.000 0.078 0.098 0.170 0.149 11.018  ROW Total Exports 0.005 25.418 0.076 13.746 0.033 8.074 0.118 19.660 0.000 15.032 4.427 5.052 2.565 3.093 3.922 12.581 11.146 102.656  2025 Export/Import British Columbia Eastern Canada US West Coast US South Central Scandinavia Chile New Zealand ROW Total Imports  Int Canada US West Int 9.480 10546 0.083 0.058 0.599 0.476 0.849 0.559 0.003 0.000 0.018 0.063 0.003 0.031 0.017 0.002 11.053 11.436  US South Ad. US North 0.315 1.895 0.083 13.195 0.000 0.996 4.367 19536 0.025 1.902 0.034 0539 0.037 0.003 0.031 1.943 4.891 39.410  W. Europe 0.000 0.000 0.000 2.479 15.122 0.522 0.479 9.938 28.540  Japan 8588 0.000 6578 0.000 0.011 0.125 0598 0.158 15.157  ROW Total Exports 0.005 30529 0.054 13.474 0.000 8.350 0.157 27.647 0.000 17.062 7550 8551 4.415 5566 0.000 12.089 11.882 122.368  96 Table 5.7 Direction of Trade - Reduced AAC in BC (Scenario b) (in millions of cubic metres) Int Canada US West IntUS South All US North W.Europe Japan 4.33 281 6.99 9.88 3.39 3.50 0.34 0.07 1.09 6.55 1.24 0.00 130 0.72 209 7.33 0.00 3.27 0.23 0.19 0.02 1.73 1.06 0.00 0.00 0.00 0.00 0.05 121 0.00 0.00 0.00 0.00 0.00 0.81 0.17 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.98 0.00 6.20 3.79 10.19 25.53 19.59 6.94  ROW Total Exports 1.59 3Z48 1.39 10.67 1.50 16.21 1.23 446 0.04 1220 0.42 1.49 0.49 0.49 1245 13.43 19.1 9134  1987 Export/Import British Columbia Eastern Canada USWestCoast US South Central Scandinavia Chile NewZealand ROW Total unpens 1991 Export/Import British Columbia Eastern Canada USWestCoast US South Central Scandinavia Chile NewZealand ROW Total Imparts 1997 Export/Import British Columbia Eastern Canada USWestCoast US South Central Scandinavia Chile NewZealand ROW Total Imports  Int Canada US West IntUS South Ait 5.148 3.544 2485 0.339 0.069 0.975 0.926 0.767 5.987 0.239 0.190 0.021 0.000 0.000 0.000 0.000 0.019 0.000 0.000 0.000 0.000 0.000 0.002 0.000 6.653 4.590 9.468  US North W.Europe 14.821 0.330 7.686 0.000 0.719 0.000 2715 0.940 0.000 12890 0.000 0.154 0.000 0.199 0.000 2784 25.942 17.297  Japan 2487 0.001 5.028 0.001 0.000 0.153 0.000 0.000 7.670  ROW Total Exports 1.186 30.000 1.180 10.250 1.048 14.475 0.871 4978 0.000 12891 0.728 1.053 0.101 0.300 12160 14945 17.273 88.894  InL Canada US West IntUS South AtL 4.154 4.215 1.316 0.337 0.061 1.010 0.568 1.202 4.387 0.467 0.199 0.022 0.000 0.000 0.235 0.000 0.003 0.001 0.000 0.029 0.001 0.000 0.002 0.012 5.528 5.712 6.984  US North W.Europe 8.368 0.000 11.025 0.000 0.857 0.000 5.953 0.920 0.290 12928 0.212 0.074 0.051 0.078 0.030 5.898 26.787 19.898  Japan 2455 0.000 3.756 0.000 0.000 0.001 0.075 0.018 6.305  ROW Total Exports 0.563 21.070 0.680 13.11 0.950 11.720 0.824 8385 0.000 13.454 1.754 2046 0.829 1.063 8.273 14.234 13.872 85.086  2002 Export/Import British Columbia Eastern Canada USWestCoast US South Central Scandinavia Chile NewZealand ROW Total Imports  Int Canada US West IntUS South AO. 429 4.44 0.73 0.35 0.07 1.0 1.22 230 1.02 0.27 0.30 0.02 0.00 0.00 0.00 0.02 0.02 0.00 0.18 0.17 0.00 0.03 0.00 0.02 6.35 7.29 281  US North 485 11.5 0.87 11.57 0.01 0.00 0.00 0.01 28.82  Japan 5.90 0.00 1.36 0.00 0.00 0.02 0.04 0.02 7.34  ROW Total Exports 0.23 20.45 0.29 13.23 1.77 8.53 0.38 1254 0.00 14.22 281 288 0.93 1.65 7.07 13.42 13.48 86.91  W.Europe 0.00 0.00 0.00 0.00 14.21 0.01 0.32 6.27 20.82  97 Table 5.7 (continued) Direction of Trade - Reduced AAC in BC (Scenario b) (in millions of cubic metres)  s  2007 Export/Import British Columbia Eastern Canada US West Coast US South Central Scandinavia Chile New Zealand ROW Total Imports  Int Canada 4.848 0.365 0.530 0.388 0.826 0.194 0.062 0.175 7.385  US West Int US South AtL 6.943 1.001 0.067 0.957 1.126 0.473 0.220 0.034 0.000 0.001 0.175 0.211 0.061 0.081 0.001 0.232 8.593 2991  US North 0.658 11.625 2324 15.891 0.001 0.141 0.054 0.313 31.007  W. Europe 0.000 0.000 0.000 0.000 13.933 0.165 0.586 7.821 22507  Japan 5.136 0.000 3.356 0.000 0.001 0.219 0.013 0.249 8.974  ROW 0.000 0.074 0.234 0.114 0.000 2978 1.529 5.304 10.233  Total Exports 18.585 13.087 8.044 16646 14.763 4.084 2385 14.094 91.690  2012 Export/Import British Columbia Eastern Canada US West Coast US South Central Scandinavia Chile New Zealand ROW Total Imparts  Int Canada 5.933 0.358 0.833 0.476 0.094 0.063 0.003 0.057 7.817  US West Int US South AtL 7.201 1.424 0.071 0.925 1.374 0.021 0310 0.023 0.000 0.012 0.056 0.008 0.029 0.075 0.002 0.060 9.044 2547  US North 1.886 12308 0.654 17.924 0.000 0.113 0.003 0.107 32995  W.Europe 0.000 0.000 0.000 0.816 15.087 0.286 0.250 8.298 24.737  Japan 4533 0.000 5.136 0.000 0.078 0.098 0.170 0.149 10.165  ROW 0.005 0.076 0.033 0.118 0.000 4.448 2588 3.963 11.232  Total Exports 20.983 13.738 8.051 19.668 15.272 5.073 3.117 12635 98.537  2025 Export/Import British Columbia Eastern Canada US West Coast US South Central Scandinavia Chile New Zealand ROW Total Imports  Int Canada 7.508 0.085 0.922 1.421 0.003 0.018 0.003 0.017 9.976  US West Int US South AtL 8.716 0.000 0.060 0.083 0.637 0.000 0.711 4.462 0.000 0.025 0.063 0.034 0.031 0.037 0.002 0.031 10.219 4.671  US North 1320 14.243 1.11 18.477 1.765 0.227 0.003 1.789 38.938  W.Europe 0.000 0.000 0.000 2841 15.355 0.477 0.441 9.810 28.923  Japan 3.545 0.000 8.698 0.000 0.011 0.120 0.275 0.155 12803  ROW 0.005 0.055 0.000 0.164 0.000 7.299 4485 0.000 12008  Total Exports 21.094 14.526 11370 28.075 17.159 8.237 5.275 11.804 117.539  Table 5.8 Direction of Trade - Increased Production in the Former USSR (in millions of cubic metres) 1987  Export/Import Int. CanadaUS West IntUS South Atl. US North W.Europe British Columbia 4.33 2.81 6.99 9.88 3.39 Eastern Canada 0.34 0.07 1.09 6.55 1.24 US West Coast 1.30 0.72 2.09 7.33 0.00 US South Central 0.23 0.19 0.02 1.73 1.06 Scandinavia 0.00 0.00 0.00 0.05 12.11 Chile 0.00 0.00 0.00 0.00 0.81 New Zealand 0.00 0.00 0.00 0.00 0.00 ROW 0.00 0.00 0.00 0.00 0.98 Total Imports 6.20 3.79 10.19 25.53 19.59  Japan 3.50 0.00 3.27 0.00 0.00 0.17 0.00 0.00 6.94  ROW Total Exports 1.59 32.48 1.39 10.67 1.50 16.21 1.23 4.46 0.04 12.20 0.42 1.40 0.49 0.49 12.45 13.43 19.11 91.34  Japan 2.487 0.001 5.028 0.001 0.000 0.153 0.000 0.000 7.670  ROW Total Exports 1.186 30.000 1.180 10.250 1.048 14.475 0.871 4.978 0.000 12.891 0.728 1.053 0.101 0.300 12.160 14.945 17.273 88.894  Japan 5.464 0.000 2.849 0.000 0.234 0.001 0.041 0.018 8.608  ROWTotalExports 0.600 32.742 0.607 11.331 0.577 9.435 0.805 7.864 0.047 12.385 1.304 1.475 0.658 0.833 14.966 25.254 19.565 101.318  Japan 7.83 0.00 0.98 0.00 0.00 0.02 0.04 0.02 8.88  ROW Total Exports 0.27 32.62 0.34 12.19 2.16 7.11 11.82 0.47 0.00 12.65 2.64 2.70 0.93 1.61 13.75 23.68 20.55 104.39  1992  Export/Import Int. CanadaUS West IntUS South Atl. US North W.Europe British Columbia 5.148 3.544 2.485 14.821 0.330 Eastern Canada 0.339 0.069 0.975 7.686 0.000 US West Coast 0.926 0.767 5.987 0.719 0.000 US South Central 0.239 0.190 0.021 2.715 0.940 Scandinavia 0.000 0.000 0.000 0.000 12.890 Chile 0.000 0.019 0.000 0.000 0.154 New Zealand 0.000 0.000 0.000 0.000 0.199 ROW 0.000 0.002 0.000 0.000 2.784 Total Imports 6.653 4.590 9.468 25.942 17.297 1997  Exportflmport Int. CanadaUS West Int.US South Atl. US North W.Europe British Columbia 5.678 5.640 2.612 12.749 0.000 Eastern Canada 0.351 0.061 1.284 9.029 0.000 US West Coast 0.508 0.959 3.829 0.712 0.000 US South Central 0.468 0.199 0.022 5.266 1.104 Scandinavia 0.000 0.000 0.255 0.201 11.647 Chile 0.000 0.003 0.001 0.092 0.074 New Zealand 0.000 0.000 0.000 0.044 0.090 ROW 0.000 0.003 0.012 0.028 10.226 Total Imports 7.006 6.865 8.014 28.121 23.141 2002  Export/Import Int. British Columbia Eastern Canada US West Coast US South Central Scandinavia Chile New Zealand ROW Total Imports  CanadaUS West IntUS South Atl. US North W.Europe 5.47 6.16 2.20 10.69 0.00 0.35 0.07 1.64 9.79 0.00 0.90 1.47 0.92 0.69 0.00 0.27 0.29 0.02 8.39 2.37 0.00 0.00 0.00 0.01 12.64 0.02 0.02 0.00 0.00 0.01 0.16 0.16 0.00 0.00 0.32 0.03 0.00 0.02 0.01 9.85 7.20 8.16 4.81 29.59 25.20  99 Table 5.8 (continued) Direction of Trade - Increased Production in the Former USSR (in millions of cubic metres) 2007 Export/Import BC Quebec and East Coast USWestCoast US South Central Scandinavia Chile NewZealand ROW Total Imports  Int Canada USWestlnt US South Ad. US North W.Europe 6.637 7.771 1.754 5216 0.000 0.357 0.066 1.169 9.737 0.000 0.485 0.925 0.491 1.679 0.000 0.380 0216 0.034 . 14.101 0.186 0.165 0.000 0.001 0.000 13.456 0.171 0.157 0.190 0.124 0.162 0.059 0.058 0.077 0.509 0.045 0.162 0.001 0250 0.858 11.780 8.413 9.194 3.966 31.761 26.092  Japan 7.812 0.000 2230 0.000 0.001 0.107 0.009 0.712 10.871  ROW Total Exports 0.645 29.833 0.707 12.036 6361 0.551 0.731 15.648 0.000 13.623 2231 3.142 1.301 2.058 25.165 11.403 17.569 107.866  Int Canada USWestlnt US South Ad. US North W.Europe 7.126 8284 2.735 5.563 0.000 0.344 0.071 0.965 10.680 0.000 0.752 1.175 0.021 0.641 0.000 0.453 0300 0.023 16.703 0.835 0.026 0.000 0.028 0.000 13.377 0.000 0.056 0.000 0.111 0277 0.003 0.028 0.074 0.003 0244 0.124 0.002 0.060 0.193 15.852 8.828 9.915 3.906 33.894 30.586  Japan 8.063 0.000 3.705 0.000 0.025 0.097 0.169 0204 12264  ROW Total Exports 0.005 31.776 0.076 12.135 6326 0.033 0.119 18.433 0.031 13.488 4.524 3.982 2.306 2.828 6.622 23.058 13.175 112.567  Int Canada USWestlnt US South Ad. US North W.Europe 8.656 9.743 0.000 4.400 0.000 0.084 0.059 0.083 13.434 0.000 0.842 0.5% 0.000 1.151 0.000 1.208 0.648 5.483 16.541 2.722 0.003 0.000 0.025 1.767 13233 0.018 0.062 0.034 0223 0.458 0.003 0.031 0.037 0.003 0.425 0.017 0.031 0.002 2.123 18.473 10.832 11.141 5.693 39.641 35.312  Japan 7.972 0.000 6.516 0.000 0.011 0.118 0264 0.156 15.036  ROW Total Exports 0.005 30.777 0.055 13.716 0.000 9.105 0.165 26.768 0.000 15.038 6.686 7.598 4277 5.039 0.000 20.802 11.188 128.843  2012 Ejqxift/Import British Columbia Eastern Canada USWestCoast US South Central Scandinavia Chile NewZealand ROW Total Imports  2025 Export/Import British Columbia Eastern Canada USWestCoast US South Central Scandinavia Chile NewZealand ROW Total Imports  100 Table 5.9 Change in Producer and Consumer  Scenario: Change in AACa in millions 1980 US$  tw Export Regions  British Columbia Eastern Canada US West Coast US South Central Scandinavia Chile New Zealand ROW Total  Central Canada US West Interior US South Atlantic US North W. Europe Japan ROW Total  2025  CS CS PS -8.17 423.99 13.99 121.57 -23.54 -37.22 7.31 -99.21 259.66 -41.25 66.70 -25.83 -26.69 69.40 -10.49 -0.48 1.69 -0.99 2.57 -2.00 -0.99 -58.84 -95.47 85.39 1030.97 -100.39 -309.48 2SS2L8Z ms9.11 PS cs 720.14 i m « 602,9$ 3035.56 1979.31 1317.51 $9537 1669 05 198.72 379.39 2S8.40 447,62 15989.51 588X59  1987 Import Regions  Change in Welfare 2002  2002  CS 3062 61 541.69 -128.86 -155.19 3259.08 692.21 -386.29 4933 40 720.67 -255.95 73S1 47 42331 -183.85 6022.21 3020,20 5662 85 268331 -290.19 -83.14 31974.58 62296.20 1616035-1483.47 CS  Note: CS refers to Consumer Surplus and PS refers to Producer  PS -152.74 4.79 -122.10 0.16 17.41 -1.74 0.14 -22.19 -276.27  2025  PS CS PS 70.70 22.09 -27.86 136.65 88.06 -49.03 26.88 189.12 -176.69 38.49 -128.42 -2.75 -23.30 108.56 -8.82 179.08 70.17 -100.58 -3.88 -26.25 58.54 781.13 -151.96 -188.40  101 Table 5.10 Change in Producer and Consumer Surplus  Scenario: Change in AACb in millions 1980 US$  Base Year Export Regions  190 CS  British Columbia Eastern Canada USWestCoast US South Central Scandinavia Chile NewZealand ROW Total  720,14 602.98 StiSSJft 1317,51 1669.05 198.72 28140 1598931 2382L88  Import Regions  1987 CS  Central Canada US West Interior US South Atlantic US North W.Europe Japan ROW Total  Change in Welfare PS  1757A9 851*47 1979.31 £95.37 1443,8)3 mM 447.62 $883.59 I343$M  PS  3062,61 541,69 3259,08 692,21 4933.40 720.67 73*1.47 423.31 6022.21 3020.20 5662.85 2683,31 31974.58 S073.9S  miHM tMMM  2002 CS  PS  2025 CS  PS  -40.24 748.43 -68.36 287.05 -37.22 121.57 -70.36 146.48 -99.21 259.66 -243.32 430.11 -41.25 66.70 -66.43 54.93 -26.69 69.40 -19.88 30.83 -0.48 1.69 -0.23 -5.77 -0.99 2.57 -1.93 -1.04 -95.47 85.39 -16.02 -65.25 -341.55 1355.41 -486.54 877.33 2002 CS  -128.86 -155.19 -386.29 -255.95 -183.85 -290.19 -83.14  PS  2025 CS  70.70 136.65 189.12 38.49 108.56 179.08 58.54 -N83.47 781.13  PS  -229.89 71.27 -202.01 95.10 -238.91 48.48 -387.48 25.65 49.02 -41.20 -534.12 199.90 18.70 -45.84 -1524.69 353.35  102 Table 5.11 Change in Producer and Consumer Surplus  Scenario: Increased Production in Former USSR in millions 1980 US$ Base Tear 1987 Export Regions  British Columbia Eastern Canada US West Coast US South Central Scandinavia Chile New Zealand ROW Total  Change in Welfare 2002  1987 Import Regions  Central Canada US West Interior US South Atlantic US North W. Europe Japan ROW Total  2025  PS CS PS CS PS CS 720.14 1757.49 -0.52 7.08 -1.50 49.97 602.98 85J.47 0.17 2.74 -9.97 45.10 3035.56 1979,31 11.76 -21.94 19.67 -6.63 1317.51 695.37 6.36 -8.57 20.65 -21.88 64.31 -157.27 1669.05 1443,88 172.28 -281.66 1.44 198.72 379.39 -7.68 6.90 -14.79 288.40 447,62 -0.56 5.45 4.62 -1.88 15989.51 588159 624.44 -636.15 1651.57 -1199.02 -1430.80 2382188 13438.11 707.39-816.351864.21 2002  2025  3062.61 541.69 3259.08 692.21 4933.40 720.67 7381.47 423.31 6022.21 3020.20 5662,85 2683.31 31974 58 8078.95  CS 6.77 2.58 -4.17 57.51 452.83 11.27 1057.37  W296.m umjs  1584.16 -1070.591463.73-383.73  CS  PS  Note: CS refers to Consumer Surplus and PS refers to Producer  PS -0.46 0.67 4.82 -7.26 -241.15 -0.40 -826.81  CS -0.53 63.56 2.58 37.09 1345.74 96.65 -81.37  PS 0.79 -37.54 5.09 3.32 -500.95 -30.06 175.63  103  Figure 1.1 Comparative Forest Productivity MAI (m /ha/year) 3  m'/ha/yr  Source: FRDA(1992)  Figure 1. 2 British Columbia Lumber Shipments 1983-92 (million board feet) 16000 14000 12000 10000 8000 6000 4000 2000  o  o _J  1983  o1_  1984  -1  1985  1986  1987  1988  1989  I  1990  I  1991  I  1992  year -Total Production —•—Canada -A—USA — - U K —d—Japan  Source: COFI (1993)  Figure 5.1  Projected Exports 1987-2025 35.000  T  0.000 I 1987  I 1992  1 1997  1 2002  1 2007  1 2012  1 2017  1— 2022  year —•— British Columbia —a—Eastern Canada —A—US West Coast M  Scandinavia  •  Chile  —I—NewZealand  X US South Central ROW  107  Figure 5. 2  Projected Imports 1987-2025 40.000  T  35.000 4-  30.000  25.000 T E u  3 20.000 i ID  e o E  15.000 +  10.000  5.000  0.000 1987  H  1992  1  1997  1  1  2002  2007  1  2012  h  2017  -+-  2022  year - Central Canada  -US West riterior  - US South Atlantic  -W. Europe  - Japan  •ROW  -US North  108 Figure 5. 3  World Softwood Lumber Exports - Market Share  100%  III I I I  BROW •New Zealand 0 Chile • Scandinavia • US South Central  •  • US West Coast • Eastern Canada D British Columbia  +-  -+-  2002 Scenario  or  "3  2  % 3  2025  0C.  109  Figure 5. 4 World Softwood Lumber Imports - Market Share  IDROW  BJapan •Western Europe • US North • US South Atlantic BUS West Interior £3 Interior Canada  1087  2002  Scenario  2025  110  REFERENCES  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). Adams, D.M. and R.W. Haynes. 1980. "The 1980 Softwood Timber Assessment Market Model: Structure, Projections, and Land Policy Simulations." Forest Science, Monograph 22, Supplement to Forest Science 26 (3). Adams, D.M. and R.W. Haynes. 1987. "Interregional Modeling." In M. Kallio, D.P. Dykstra and C. Binkley (Editors) The Global Forest Sector: An Analytical Perspective. John Wiley and Sons, New York. Chapter 16. Apthorp, B. 1994. "Improving US-bound lumber." New Zealand Forest Industries, September: 37. Benson, C. A. 1988. "A Need for Extensive Forest Management," The Forestry Chronicle. October:421-430. Berck, P. 1979. "The economics of timber: a renewable resource in the long run." Bell Journal of Economics 10:447-462. Boyd, R., K. Doroodian and S. Abdul-Latif. 1993. "The Effects of Tariff Removals on the North American Lumber Trade," Canadian Journal ofAgricultural Economics, 41(3):311-28. Boyd, R. and K. Krutilla. 1987. "The welfare impacts of US trade restrictions against the Canadian softwood lumber industry: A spatial equilibrium," Canadian Journal of Economics 20: 17-34. Buongiorno, J., J. Chavas and J. Uusivuori. 1988. "Exchange rates, Canadian lumber imports and United States prices: a time series analysis." Canadian Journal ofForest Resources, 18: 1587-1594. Cardellichio, Peter A., Yeo Chang Youn, Darius M. Adams, Rin Won Joo and John T. Chmelk. 1989. "A Preliminary Analysis of Timber and Timber Products Production, Consumption, Trade, and Prices in the Pacific Rim Until 2000", CINTRAFOR Working Paper, #22, University of Washington, Seattle Washington. Cardellichio, P.A. and D.M. Adams. 1987. "US experience with the Global Trade Model." In P. Cardellichio, D. Adams and R. Haynes (Editors) Forest Sector and Trade  Ill Models: Theory and Applications. Proceedings of an International Symposium, November 3-4, 1987. CINTRAFOR, University of Washington, Seattle Washington. Cardellichio, P. A. and J. Veltkamp. 198 i. "Demand for Pacific Northwest timber and timber products." Study Module H.-A, Forest Policy Project, Washington State University. COFI. 1993. "British Columbia Forest Industry Statistical Table." Council ofForest Industries of British Columbia, April COFI. 1992. "British Columbia Forest Industry Statistical Table." Council ofForest Industries ofBritish Columbia, April. Column One. 1995. "Shake out." The Vancouver Sun, May 6:E1 Commission on Resource and Environment. 1994. Vancouver Island Land Use Plan, 3 volumes. Victoria: CORE, July.  Cortes, H. 1988. "Chile: Its current and future role as a major exporter of forest products." In J. Gruenfeld (Editor) Marketing Logs, Timber, and Chips of ht Pacific Rim—Some Emphasis on the Middle East. Seattle, Washington. Cournot, A. A. 1838. Mathematical Principles of the Theory of Wealth. Chapter X. Defense Mapping Agency. 1976. Distance Between Ports, Pub. 1551, Hydrographic Center, Defense Mapping Agency, US Government, Washington, DC Enke, S. 1951. "Equilibrium Among Spatially Separated Markets: Solution by Electric Analogue," Econometrica, 19:40-47. FAO (Food and Agricultural Organization). 1994. Timber Bulletin. Food and Agriculture Organization of the United Nations, United Nations Economic Commission for Europe, New York. FAO. 1992. FAO Forest Products Yearbook. FAO of the United Nations Publications, Rome.  FAO. 1991. Forest Products: World Outlook Projections: Projections of Consumption an Production of Wood-Based Products to 2010, FAO Forestry Paper 84, Rome, Italy. FAO. 1992. FAO Forest Products Yearbook. FAO of the United Nations Publications, Rome. Flora, Donald F., Andrea L. Anderson and Wendy J. McGinnis. 1991. "Future Pacific Rim Flows and Prices of Softwood Logs, Differentiated by Grade." United States Department of Agriculture, Forest Service Research Paper PNW-RP-433 January 1993.  112  Florian, M. and M. Los. 1982. "A New Look at Static Spatial Price Equilibrium Models", Regional Science and Urban Economics, 12: 579-597. FRDA 1992. The Wood Industry in B.C. The Next 20 years. Summit Conference, H. A. Simons, Vancouver. Gaston, C , D. Cohen and R. Prins. 1994 "Environmentalism as a Driver for Wood Product Quality", FEPA Working Paper #204, University of British Columbia, Vancouver, BC. Hamilton, Gordon. 1995. "Stumpage killing BC mills, Barber says." The Vancouver Sun, May 6A16. Haynes, R.W., D.M. Adams and J.R. Mills. 1995. "The RPA Timber Assessment Update." USDA Forest Service General Technical Report, RM-GTR-259, Fort Collins, Colorado. Haynes, R.W. 1990. "An Analysis of the Timber Situation in the United States: 1989-2040." USDA Forest Service General Technical Report, RM-199, Fort Collins, Colorado. Japanese Timber Union's Association. 1990. Seizai Koujo no Genkyo to Bunseki (Current Condition and An Analysis of Sawmills). Fourteenth Edition, Tokyo. Kallio, M., D. Dykstra and C. Binkley. 1987. The Global Sector: An Analytical Perspective. John Wiley and Sons, New York. Koopmans, T. 1948. "Optimum Utilization of the Transportation System." Econometrica, 17:136-46. Kornai, G. 1987 "Historical Analysis of International Trade in Forest Products." In M. Kallio, D.P. Dykstra and C. Binkley (Editors) The Global Forest Sector: An Analytical Perspective. John Wiley and Sons, New York. Chapter 18. MacMillan Bloedel. 1994. "Future World Sawlog Prices: A review of the Underlying Factors and Alternative Approaches." MacMillan Bloedel Working Paper, Corporate Forestry. Messmer, M. and D. Booth. 1993. "An Economic Timber Supply Analysis for Ontario." Policy and Economics Directorate Working Paper, Forestry Canada. Ministry of Forests. 1994. Forest, Range and Recreation Resource Analysis. Ministry of Forests, Public Affairs Branch, Victoria, BC.  113 National Forestry Database. 1992. Compendium of Canadian Forestry Statistics. Canadian Council of Forest Ministers, Ottawa. National Forestry Database. 1992. Compendium of Canadian Forestry Statistics. Canadian Council of Forest Ministers, Ottawa. Natural Resources Canada. 1993. Selected Forestry Statistics Canada 1992. Policy and Economics Directorate, Natural Resources Canada, Canadian Forest Service. Neilson,D.A. 1994. Vast Timber Resources But...." Asian Timber. January : 16-18. Neilson, D.A. and M.R. Smith. 1993. "World prototype for sustainable timber resource." Asian Timber, December: 16-19. Perez-Garcia, J. 1993. "Global Forestry Impacts of Reducing Softwood Lumber Supplies from North America" CINTRAFOR Working Paper, #43, University of Washington, Seattle, Washington. Samuelson, P. 1952. "Spatial Price Equilibrium and Linear Programming." The American Economic Review, 42: 283-303. Sedjo, R. A. 1983. The Comparative Economics ofPlantation Forestry. Resources of the Future, Washington, DC. Sedjo, R.A. and K.S. Lyon. 1991. The Long-Term Adequacy of World Timber Supply. Resources for the Future, Washington, DC. Sedjo, R.A., A.C. Wiseman, D.J. Brooks and K.S. Lyons. 1994. "Global Forest Products Trade: The Consequences of Domestic Forest Land-Use Policy." Resources for the Future Discussion Paper, Resources for the Future, Washington, DC. Smyth, D. 1994. "Timber supply crisis really arrives in'94." Wood Technology, Jan./Feb:23-24. Smyth, D. 1993. "Pent-up US housing demand and the timber supply crisis collide in 1994." IWA-Canada Working Paper, Research Department, Vancouver, BC. Thompson, W.A., P.H. Pearse, G.C. van Kooten and I. Vertinsky. 1990. "Rehabilitating the Backlog of Unstocked Forest Lands in British Columbia: A Preliminary Simulation Analysis of Alternative Strategies," FEPA Working Paper #147, University of British Columbia, Vancouver, BC.  114 Waggener, T.R., G.F. Schreuder and IX. Eastin. 1990. "Production and Trade in Tropical Hardwoods: An Asian-Pacific Case Study." CINTRAFOR Working Paper, #29, University of Washington, Seattle, Washington. Wallace, P. 1987. Southern pine dealt setback by costlier Canadian Lumber." Wall Street Journal, 27 January, 1987:37. Warren, D.D. 1994. "Production, Prices, Employment and Trade in Northwest Forest Industries, First Quarter 1994." United States Department ofAgriculture, Forest Service Research Paper PNW-RP-204, November. Widman. 1994. "TheUS South." Widman's World WoodReview, April: 1-5.  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

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"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0086943/manifest

Comment

Related Items