Three essays on agricultural insurance and farm real estate investment

Abstract

This dissertation consists of three essays discussing topics on agricultural insurance and farm real estate investment. The first essay focuses on diversifying systemic risk in crop insurance portfolios. Portfolio risk in crop insurance due to the systemic nature of crop yield losses has inhibited the development of private crop insurance markets. Government subsidy or reinsurance has therefore been used to support crop insurance programs. We investigate the possibility of converting systemic crop yield risk into “poolable†risk. Specifically, we examine whether it is possible to remove the co-movement as well as tail dependence of crop yield variables by enlarging the risk pool across different crops and countries. Hierarchical Kendall copula models are used to allow for potential non-linear correlations of the high-dimensional risk factors. A Bayesian estimation approach is applied to account for estimation risk in the copula parameters. The results indicate that the systemic risk in crop insurance can be eliminated by combining crop insurance policies across crops and countries.The second essay attempts to provide an explanation for the high return-low risk paradox in farmland investment. We investigate both the nominal and real returns of a farmland portfolio from a forward-looking perspective. Land values and cash rents are slow to adjust and therefore the return from owning land is likely to be time-varying and serially correlated. Time-series and copula modeling techniques are used to construct the optimal portfolio and to evaluate the risk-return profile. The results indicate that it takes a number of years for the expected return to reach the long-term equilibrium. From a forward-looking perspective, the attractive average return level observed historically can only be attained over a long investment period. The risk involved in the long investment period, however, is also considerably higher than the historical sample volatility. This is due to autocorrelation in the return series. These findings help explain the “high return and low risk†puzzle observed in historical farmland returns.The third essay examines the predictive power of capital market risk factors for farmland returns. Farmland value slightly increased in 2017 even though farm income was lower. This development suggests the rate of return required by investors for farmland asset has been reduced. A similar phenomenon has been observed in the equity market which also suggests reduced equity risk premium. One possible explanation for the decreasing required rate of return is an increased money supply. Previous research suggests that the money supply affects several macroeconomic risk factors through different transmission channels, which in turn influence investor behaviors and asset returns. This article examines the predictive power of these risk factors for farmland asset returns. Both linear and neural network models are used and the forecast accuracy is compared across different models. The results indicate that farmland return prediction is significantly improved by adding capital market excess return as an explanatory variable. Adding additional risk factors, however, does not improve the prediction with the sample used in this study.