Direct and Indirect Impacts of Agricultural Land-Use Due to Increased Production of Grain-Based Ethanol in Kansas, USA: A County-Level Analysis

Applied economists have dedicated much literature to the design and specification of acreage allocation models for land use decisions. These models can play an important role in understanding how acreages might shift in the event of new policy or in a changing agricultural landscape, especially with the increasing demand for biofuel production. Furthermore, the increase in ethanol production over the past two decades has impacted markets for corn and grain sorghum in Kansas, USA, especially with the prospect of grain sorghum being classified as an advanced biofuel feedstock. The increased production of the ethanol within Kansas has had an impact on local markets and the intensity of agricultural production in these areas. The presence of an ethanol plant can impact the prices of crops in the local market; the outlets for marketing grain by farmers and elevators; and the intensity of crop production in the vicinity of the plant. The direct changes in land-use production from a stronger ethanol industry in the state will likely lead to indirect changes in the use of agricultural land, as well.

The purpose of this paper is to examine the direct and indirect impact on agricultural land-use from increased ethanol production in Kansas.

The study utilizes acreage allocation models designed in the agricultural economics literature and builds on these previous acreage allocation studies by correcting for spatial autocorrelation; spatial proximity to ethanol refineries using capacity as a function of distance to a refinery as an explanatory factor in the model; and by incorporating changes in agricultural land-use in neighboring counties to capture potential indirect land-use changes. Bioenergy is an important agricultural topic to the state of Kansas, and understanding how the presence of ethanol plants effects producers’ acreage decisions is critical as the bioenergy industry moves forward and new policies are developed.  This study uses a 1996-2009 Kansas county level dataset for analysis. Variables that will be examined include shifts in acreage allocated to crops in neighboring counties, expected input prices, production costs, livestock populations, government programs, site and soil characteristics, weather, and ethanol plant characteristics such as proximity to a county and size of the plant. Models will be estimated for dryland corn, wheat, sorghum and soybean. The methods in this paper will follow methods proposed by Wu and Brorsen (1995) which extended the HEAR model proposed by Kmenta (1996) to a set of seemingly unrelated equations. These methods will be expanded to take account of spatial dependence.

Results indicate that own- and cross-price effects of crops and the capacity/proximity to an ethanol refinery have had significant impacts on the crop acreage allocated to the production of corn and grain sorghum. In addition, using the spatial lags of shifts in crop production in neighboring counties allows for the derviation of spillover marginal effects of these on within-county acreage allocation to help capture indirect effects on crop land allocation decisions. Results from the paper will help to assess the agricultural land-use impacts from the expansion of the ethanol markets in the U.S. and the impacts of U.S. bioenergy policies.