A Dynamic Programming Analysis of Optimal Farmland Investment Decisions: An Application to Central Illinois High-Quality Farmland

169 p.Thesis (Ph.D.)--University of Illinois at Urbana-Champaign, 1987.The purpose of this research is to examine farmland investment decisions using a dynamic programming framework. Consideration is given to the dynamic, stochastic nature of farmland returns, the linkages between farmland returns and farmland prices, and the effects of the above dynamic factors on a farm's financial structure. The dynamic programming model's objective is to maximize the expected value of after-tax wealth. Its decision variable is the number of acres to purchase or sell at the beginning of each year.Analysis of optimal decisions for a Central Illinois high-quality farmland setting indicates that the optimal amount of farmland to purchase (sell) increases (decreases) with higher farmland returns, lower current farmland prices, higher farmland price movements (the difference between the current and lagged farmland price), and lower debt-to-farm-asset ratios. In addition, conditional probability methods are used to evaluate farm sizes and debt-to-farm-asset ratios that result if optimal decisions are made. Results suggest that low debt-to-farm-asset ratios should be maintained between 1985 and 1990.Optimal decisions are compared from dynamic programming models with differing tax codes, farmland availability assumptions (the ability to purchase or sell farmland at the current farmland price), interest rates, and consumption withdrawals. Results suggest that these numerical parameters are relatively less important than return and price levels and return and price movements. In additions, a model that restricts the ability to sell farmland is solved in order to examine costs associated with a reluctance to sell farmland. Results indicate that not selling farmland can lead to lower wealth accumulations.Optimal decisions from the dynamic programming model are compared to decisions from a traditional capital budgeting model. The defined capital budgeting model uses only the expected values of random variables when determining farmland investment decisions and does not consider future farmland investment decisions when analyzing current farmland investment decisions. Because the dynamic programming model considers these factors, decisions from the capital budgeting model are sub-optimal relative to the dynamic programming model's decisions.U of I OnlyRestricted to the U of I community idenfinitely during batch ingest of legacy ETD