Incorporating Fire Risk into Optimal Decisions in Forestry Using a No-Arbitrage Approach

This paper investigates the impact of including the risk of fire in an optimal tree harvesting model at the stand level, assuming timber prices follow a mean-reverting stochastic process. The relevant partial differential equation is derived under different assumptions about hedging the risk of fire. The assumption that fire risk is fully diversifiable is contrasted with the assumption that it can be hedged with another asset. It is conjectured that the risk-neutral probability of fire exceeds the historical probability of fire, which will affect forest land valuation. An empirical example is presented for two different silvicultural regimes