Design and Objectives of Ftm - West Model

The FTM-West (“Fuel Treatment Market” model for U.S. West) is a dynamic partial market equilibrium model of regional softwood timber and wood product markets, designed to project future market impacts of expanded fuel treatment programs that remove trees to reduce fire hazard on forestlands in the U.S. West. The model solves sequentially the annual equilibria in wood markets over a historical period from 1997 to 2004, and projects annual equilibria from 2005 to 2020 using detailed assumptions about future thinning programs and future market trends. This paper explains how FTM-West was designed specifically to account for economic complexities that stem from unconventional size distributions of trees and logs removed in thinning operations as compared to conventional timber supply in the West. Tree size directly influences market value and harvest cost per unit volume of wood, while log size influences product yield, production capacity and processing costs at sawmills and plywood mills. The objective that has been achieved through development of FTM-West is to provide a tool to evaluate future market scenarios for large-scale fuel treatment programs with various thinning regimes that may have varying costs and may yield wood with divergent size class distributions. The model provides a capability to analyze and project how much harvestable wood the markets can absorb from thinning programs over time and the Peter Ince is a Research Forester and Henry Spelter is an Economist, both at the U.S.D.A. Forest Service, Forest Products Laboratory, Postal Box 2398, Madison, WI 53726-2398. FAX: 608-231-9592. regional timber price and timber harvest impacts of expanded thinning under varying assumptions about fuel treatment program subsidy or administrative costs, variations in thinning regime, or alternative projections of future product demands across the spectrum of products ranging from wood fuel to lumber, plywood and wood fiber products. INTRODUCTION Decades of fire suppression, reduced timber harvests on public lands since the 1980s, and a build up of standing timber inventories in fire-prone forested regions of the western United States have created conditions susceptible to catastrophic wildfires. Expanded programs of systematic stand density reduction through mechanical thinning on public lands may reduce the fuel buildup. Timber market consequences of such programs will depend on the scale of program and the type of treatment regime. This paper describes the design and objectives of an economic model that can project timber market impacts of expanded fuel treatment programs in the U.S. West. The “fuel treatment market” model for the U.S. West (FTM–West) employs the Price Endogenous Linear Programming System (PELPS). PELPS is a general economic modeling system developed originally at University of Wisconsin (Gilless and Buongiorno 1985, Calmels and others 1990, Zhang and others 1993) and more recently modified for applications at the Forest Products Laboratory (Lebow and others 2003). PELPS-based models employ the technique of spatial equilibrium modeling (Samuelson 1952), with periodic (e.g., annual) market equilibrium solutions obtained via economic optimization. Solutions are derived via maximization of consumer and producer surplus, subject to temporal production capacity constraints, transportation and production costs, and price-responsive raw material supply curves and product demand curves, all of which can be programmed realistically to shift over time and respond to endogenous shifts in market conditions. FTM–West employs the FPL version of PELPS (called FPL–PELPS); Lebow and others (2003) and earlier PELPS publications provide further mathematical details about the modeling system. PELPS has been used fairly widely for partial market equilibrium models of timber and forest products for many years (for example, Buongiorno and others 2003, Zhang and others 1996, ITTO 1993). STRUCTURE OF FTM-WEST Forest sector market models commonly include structural features of wood product markets such as a regional market structure with regional product demand curves, regional timber supply curves, interregional transportation costs, and regional production capacities and manufacturing costs. Those general structural features were included also in FTM–West. In addition, FTM-West was designed with other features to account for economic complexities that can arise with utilization of wood from fuel treatment programs, which may have a divergent distribution of volume by tree size-class as compared to conventional timber supply (wood from fuel treatments may have a larger fraction of volume in smaller trees than conventional timber supply for example). General design features Among general design features, FTM–West included demands for more than a dozen forest product commodities encompassing the full spectrum of forest products produced from softwood timber in the U.S. West, three product demand regions, eight production or supply regions, and estimated wood supplies from conventional timber supply sources and from future fuel treatment programs (assumed to be primarily softwoods). Table 1 summarizes the regional and commodity structure of the model. The model includes demand only for forest products produced from softwood timber in the U.S. West, a partial representation of total U.S. and global demands for forest products. Fairly simple demand curves were specified in the model based on an assumption that demands for all products are inelastic (price elasticity of demand ranged from –0.3 to –0.8 among the various products). Aggregate demand quantities for each product were equated to product output data for the U.S. West in the base year (1997) and proportioned to each of the three product demand regions using estimates of regional shipments from the West. Product output was based on data published by industry associations, such as WWPA for lumber, AF&PA for pulp and paper, and APA-The Engineered Wood Association for plywood. FTM–West was designed to derive annual market equilibria sequentially over a 24-year period, 1997 to 2020, which permitted testing and calibration of model solutions against overlapping historical data (to 2004). Demand curves were shifted each year based on historical shifts in production in the U.S. Table 1 – Regional and commodity structure of FTM–West model Supply/production regions Demand regions Demand commodities Coast PNW (OR, WA) U.S. West Softwood lumber & boards Eastern Washington U.S. East Softwood plywood Eastern Oregon Export market Poles & posts California Paper (five grades) Idaho Supply commodities Paperboard (three grades) Montana “Pines” Market pulp Wyoming–South Dakota “Non-Pines” Hardboard Four-Corners (UT, CO, AZ, NM) (trees, logs, chips) Fuelwood West (1997 to 2004), and the model was programmed with a set of assumed future growth rates in regional demand (2005 to 2020) for each forest product commodity. Demand growth rate assumptions matched recent Forest Service RPA Assessment projections (2005 Draft RPA timber assessment report). Similarly, simple supply curves were used to model conventional softwood timber supply in each of the eight supply regions, while exogenous estimates of wood supply from treatment programs (upper bounds on harvest quantity and harvest costs) were introduced as policy or program variables. The estimates of wood supply from fuel treatment programs were obtained from the Fuel Treatment Evaluator, FTE v. 3.0 (Skog and others 2005). Most conventional timber supply in the U.S. West is currently obtained from timber harvest on state and private forestlands, subjected mainly to evenaged timber management. Thus, inelastic supply curves were used for conventional timber supply (with an assumed price elasticity of 0.7). Conventional timber supply curves were programmed to shift over time in direct proportion (1:1 ratio) to net growth in softwood timber inventory volumes on state and private timberland within each supply region. Annual net growth in state and private timber inventories are computed in the model by deducting from standing timber inventories the harvest volumes from the preceding year and adding timber volume growth based on recent growth rates in each region (Smith et al, 2004). Thus, FTM–West incorporated techniques similar to those used in the Forest Service RPA Assessment to model conventional timber supply (that is, inelastic supply curves shifted over time in proportion to projected net growth in timber