Softwood Lumber Products in the United States: Substitutes, Complements, or Unrelated? Rao V. Nagubadi, Daowei Zhang, Jeffrey P. Prestemon, and David N. Wear ABSTRACT. This study addresses an important dimension concerning the softwood lum- ber trade dispute between United States and Canada-substitutability among imported and domestically produced species. We employ the restricted translog subcost function ap- proach to study this issue based on the monthly data of US softwood products consumption and prices between Jan. 1989 and July 2001. The results show that the spruce-pine-fir lumber species group, mainly imported from Canada, is largely unrelated to domestically produced treated southern yellow pine, Douglas-fir, and other species groups, but is a substitute to untreated southern yellow pine and engineered wood products. Furthermore, untreated southern yellow pine is facing more severe competition from structural panels rather than from the imported Canadian spruce-pine-fir group. FOR.SCI. 50(4):416-426. Key Words: Trade dispute, US-Canada, southern pine, spruce-pine-fir, engineered wood products, translog subcost function. MONG THE TRADE CONFLICTS between Canada and ing that Canadian softwood lumber imports hurt them. A the United States, the softwood lumber dispute has key argument used by these producers was that US and A been the most important in terms of trade volumes Canadian softwood lumber are substitutes. Their statistical and values, complexity, procedure, politicization, and dura- evidence was that the price correlation between spruce- tion (Zhang 1997, Gagne 1999). Softwood lumber is one of pine-fir (an aggregate product class consisting of Picea spp., the largest commodities produced by the forest industry in Pinus spp., Abies spp.) (SPF, the main softwood lumber both countries. In 2001, Canada exported 43.75 million species group imported from Canada) and southern yellow cubic meters of softwood lumber to the United States, pine (mainly, Pinus echinata Mill., P. taedu L., P. elliotii accounting for 34% of total US softwood lumber consump- Engelm.) (SYP, the largest single species groups produced tion. This constituted 85% of all softwood lumber exports in the United States) was around 0.8 based on historical data from Canada and 93% of all softwood lumber imports into (Ragosta et al. 2000). More sophisticated empirical analyses the United States. With the expiration of the Softwood of lumber demand (e.g., Buongiorno et al. 1988, Lewand- Lumber Agreement (SLA) that reigned for 5 years from rowski et a]. 1994) also lend credence to this argument. In 1996 to 2001, the softwood lumber trade confrontation a latest development, on May 2, 2002, the US International between the United States and Canada has entered a new Trade Commission voted unanimously that the US lumber phase. producers were threatened with injury from lumber im- In Apr. 2001, some US lumber producers filed com- ported from Canada, clearing the way for new duties. The plaints to the US Department of Commerce (USDC), claim- duties, averaging 27.22% ad valorem, are large compared to Rao Nagubadi, Postdoctoral Fellow, Daowei Zhang, Professor, School of Forestry & Wildlife Sciences, Auburn Univer- sity, AL 36849-5418 -Phone: (334) 844-1067; Fax: (334) 844-108.4;[email protected]. Jeffrey Prestemon, Research Forester, David Wear, Project Leader, Economics of Forest Protection and Management, Southern Research Station, USDA Forest Service, PO Box 12254, Research Triangle Park, NC 27709. Acknowledgments: Comments from two anonymous reviewers and an Associate Editor of this journal were greatly helpful. The financial support from USDA Forest Service (Cooperative Agreement SRS 01-CA-11330143-419)is grate- fully acknowledged. Manuscript received April 21, 2003, accepted February 12, 2004 Copyright O 2004 by the Society of American Foresters 416 Forest Science SO(4) 2004 Reprinted from Forest Science, Vol. 50, No. 4, August 2004. Not for further reproduction. duties on other imported countries and zero tariffs on most (Adams et al. 1986) and that Canadian lumber imports do other forest products from Canada. not affect US lumber prices (Buongiorno et al. 1988). Other US consumer groups and the Canadian lumber industry, studies (e.g., Wear and Lee 1993, Zhang 2001) address the in contrast, claim that Canadian lumber imports and soft- welfare consequences of various t~aderestrictions on Cana- wood lumber produced in the US are not substitutes and that dian lumber. Studies examining species substitution focus Canadian imports meet US consumer demand and do not on tropical and temperate sawlog imports in Japan (Vincent injure US lumber producers (ACAH 2002, NAHB 2002). et al. 1991), tropical log imports from different world re- The National Association of Home Builders (NAHB) ar- gions in Japan (Vincent et al. 1990), and domestic and gues that the different types of lumber for house framing are imported industrial woods in the 36 most important wood- not interchangeable and that one wood product cannot be importing countries (Uusivuori and Kuuluvainen 2001). substituted for another without significantly harming US Hseu and Buongiorno (1993) find significant elasticity of consumers by forcing them to spend more on a house. That demand for Canadian lumber imports with respect to US is, because SPF from Canada is better suited for wall domestic lumber price, suggesting the possibility of substi- framing and SYP is best used for beams and joists, they tution. However, Hseu and Buongiorno (1 993) concentrate may, in fact, even be complements. Hence, raising tariffs on on substitution among softwood lumber species imported imported SPF would adversely affect demand for both SYP from Canada, not total US consumption, and their data were and SPF. prior to 1989. Lewandrowski et al. (1994) examine the Southern pine dominates outdoor wood construction ap- substitutability among imported lumber species and domes- plications such as decking and fencing and it is used exten- tic lumber species in the United States. That study, also sively in large dimension (timber framing) applications. based on data from the 1970s and 1980s, does not distin- While SPF is the largest consumed lumber product in the guish between treated and untreated SYP and therefore does United States, it is also the least domestically produced and not address the period after the emergence of large volumes largest imported lumber product (Figure 1). On the other of treated SYP (treated SYP production increased 15-fold hand, SYP is largely domestically produced. Canadian SPF between 1977 and 1987, to a level where it essentially is the preferred wood for wall framing due to its workabil- remains today). Neither study could have incorporated in- ity, strength, stability, and low density. Indeed, both species formation related to higher prices for naturally durable groups are used in the construction industry, often in the western species such as redwood (Sequoia sempervirons [D. same building. Don.] Endl.) (Olson et al. 1988, Berck and Bentley 1997) or The purpose of this article is to look at the issue of the evolution in building technologies that involve greater substitutability and complementarity between various soft- use of structural panels. This study, in contrast, is based on wood lumber products in US consumer demand with em- later data and includes all important domestic and imported pirical analysis based on economic theory. Insights gener- species groups as well as structural panels and other engi- ated from this study could help inform debates in the current neered wood products. US-Canada softwood lumber trade dispute and might have The rest of the article is organized as follows. In the next implications on the competitiveness of the US lumber in- section, the theory on which our analysis is based and the dustry and forest-based economic development in the dif- empirical specification are described. In the third and fourth ferent regions of the United States. sections, data used in this analysis and empirical results are Earlier studies conclude that the exchange rate influences presented. The final section draws some conclusions and the share of Canadian lumber in the US lumber market policy implications. Consumption ed Production Imports .. .. -. -- -- --- SP-F SYP -U SYP-R DF WSP Other Figure 1. Softwood products consumption, production, and imports in the United States: 2001. Fnr.rst Science SOj4) 2004 417 Methods To capture a full picture of the softwood products mar- ket, we include wood structural panels (WSP) that include softwood plywood, oriented strand board (OSB), and wa- ferhoard in our analysis. Following Fuss (1977) and Uusi- vuori and Kuuluvainen (2001), we adopt a two-stage ap- proach in our analysis. First, the aggregate production func- tion of softwood-utilizing industries (housing construction and remodeling and repair industry) has several major input categories, capital (K), labor (L), energy (E), softwood products (S), and other materials (M);each may consist of several components. Considering that the softwood- where CS is the total cost of softwood products in the utilizing industries can choose among a number of softwood softwood-utilizing industries; pS8 are softwood product prices, i = 1-6 (1, SPF; 2, SYP-untreated; 3, SYP-treated; species or products, St, i = 1, 2, . , N, and assuming a 4, Douglas-fir; WSP; and 6, other species); T, is from homothetic weakly separable production technology, the 5, Sept. 1994 to Mar. 1996 and Apr. to July 2001 (the free underlying production function can be written as trade period), 0 otherwise; T, is 1 from Apr. 1996 to Mar. 2001 (the SLA period), 0 otherwise; Q is the quantity of softwood products consumed by the softwood-utilizing in- dustry; and a and P are coefficients. Period 1 (T,) is from where Y denotes the gross output of the softwood-utilizing Jan. 1989 to Aug. 1994, during which the Memorandum of industries, S, the total consumption of softwood products, is Understanding (MOU), the interim duty, and the counter- an aggregate function for softwood products.