Determinants of Non-Timber Values in Wisconsin Northern Hardwoods

A working definition of non-timber value is the difference between the revenues attainable by implementing an infinite horizon timber revenue maximizing cutting rule, and the value of the observed harvest. This non-timber value was estimated for the stands of the Forest Inventory Analysis data in the maple-beech-birch type in Wisconsin. Non-timber values averaged 23 $/ha&r, and were higher than timber revenues, even on industry lands. Hedonic regression was then used to estimate the marginal effects on non-timber value of stand attributes and socioeconomic variables. The results gave a vector of non-timber values for trees of different size and species. The marginal contributions of trees to non-timber values were significantly higher in National Forests. INTRODUCTION Revealed and stated preference methods of evaluating non-market goods have been researched extensively in the past twenty years (Freeman 1993). In particular, the role of non-timber values has long been recognized in the theory of optimal harvesting decisions (Hartman 1976, Bowes and Krutilla 1985). Compelling empirical evidence exists supporting the relevance of these values to woodland owners. Birch (1994) found that amongst the reasons for woodland ownership, “recreation” and “aesthetic enjoyment”, ranked consistently higher than timber production. Much of the published work has been aimed at deriving non-timber values for benefit-cost analysis. Revealed preference methods, especially based on travel costs, have been used in a number of studies (Michaelson, 1975; Brook&ii and Coursey, 1987). Dennis (1989) related harvesting behavior to forest characteristics in a theoretical model that can also accommodate non-&ber values. Recently, Lee (1997) derived non-timber values for owners of even-aged forests in the southern United States, with hedonic discrete choice models. Non-timber values have also been derived by contingent vahiation. Walsh et al. 1990, and Loomis et al. 1994 estimated the public willingness to pay to protect forests from insects and fires. These studies corroborate the hypothesis that non-timber benefits are substantial, and may exceed timber revenues (Lockwood et al. 1992). The objective of this paper is to report on the magnitude of non-timber value, and its determinants, for maple-birch forests in Wisconsin. THEORY Forest owners are assumed to prefer some combinations of forest stand states and timber revenues to others. For uneven-aged stands, the state can be represented concisely by the number of trees of different size and species per unit of land, i.e. the tree distribution. For the purpose of deriving an infinite horizon optimal timber-harvesting rule in the context of uneven-aged management, Lin and Buongiomo (1998) defme N possible stand states. For every state i there is an optimal decision k’, solution of N recursive equations (Ross, 1983): Vt+’ = mp[r(i,k)+ d g p(j 1 k)Vi 1 _ /1\ K j=l 1’1 i= 1 ,..., N;t = l,,.., m where r is the present value of the timber income over t years, r(i,k) is the immediate timb return from cutting a stand f&m initial state i t&&and state k, &I&) is the probability of the stand moving f?om state k to statei, and d is the discount factor. Each state i corresponds to a tree distribution. A decision means cutting the stand tiom state i to state R. The best decision is unique and depends only on the state. ’ Riccardo Scarpa is a Research Ass&ant and Joseph Buongiomo ([email protected]) is a Professor in the Department of Forest Ecology and Management, University of Wisconsin, Madison. Karen Lee is a Research Forester at the Southern Forest Experiment Station, Research Triangle Park, North Carolina. The USDA Forest Service, Southern Forest Experiment Station, supported the research leading to this paper. Additional support came from M&tire-Stennis grant D946, the School of Natural Resources, University of Wisconsin, Madison, and the USDA Forest Service North Central Forest Experiment Station through collaboration with Michael Vasievich.