Assessing the risk and return of optimal portfolios of U.S. timberland and farmland

Using synthetic returns for timberland in the U.S. South and NCREIF data for farm crops from 2000:Q1 to 2016:Q4, we build efficient frontiers under the mean-conditional valueat-risk (CVaR) framework. Recognizing the availability of the investable universe of natural resource assets at any given time, we incorporate constraints and evaluate their impacts in two hypothetical scenarios. The optimal tangency portfolios have risks of 0.16% and 0.55%, and returns of 1.42% and 1.38% on a quarterly basis. We use Monte Carlo simulation to estimate the VaR and CVaR of the optimal portfolios for a 10-year horizon and find that risk increases with investment size.

Duke Scholars

Author Bin Mei Environmental Sciences and Policy