Stumpage Price Distributions and Correlations and Their Impact On Timberland Investment Modeling: An Analysis Using Maine Stumpage Prices (1961–2017)

Abstract

This paper analyzes more than 55 years of Maine stumpage prices and finds that the normal distribution does not best characterize many of the primary species and products located in the state. Nontrivial correlations are also identified among many of the species and product stumpage prices. The implications of these findings are discussed for the value of species and age diversification and the use of financial simulations to assess projected return distributions from timberland investments. Specifically, we look at two hypothetical timberland investment scenarios with varying amounts of species and age diversification and demonstrate the differences in the projected return distributions timber analysts would obtain with simple (normal distribution and independence) versus data-based (best-fit distributions and correlations) assumptions. The study findings imply that the use of simple assumptions (normally distributed and independent prices) in simulation analyses of timberland investments may lead to under- or overestimated projected return distributions, and that data-based assumptions (best fit distributions and correlations) are preferable for these types of analyses. Additionally, the correlations reported suggest species and age diversification may reduce financial risk of Maine timberland investments. Finally, the study findings indicate that the foundational assumptions of option-based techniques to value and assess timberland investments may be violated. The violations of the assumptions could potentially bias these analyses and may require new analytical approaches to accommodate alternative statistical distribution assumptions.