Financial Classification of Farm Businesses using Fuzzy Systems

Readers may make verbatim copies of this document for non-commercial purposes by any means, provided that this copyright notice appears on all such copies. Abstract This paper attempts to improve the ability of farm managers and lenders to forecast expected financial performance of farm businesses using a neuro-fuzzy inference system. Ex-ante farn financial performance is examined to identify the farms that are likely to become financially stressed before they actually become so. The financial risks associated with farming have significantly increased, due to increasing requirements from continuous technology advances, but also because of tighter profit margins and depressed commodity prices. Blank (2000) reports that from 1999 to 2000, the USDA's index of prices paid by all farmers for inputs increased about 19% while the index of prices received for outputs dropped 7%. The size required to take advantage of non-scalable technologies, as well as extended average cash flow periods of 18 to 24 months, make farmers highly dependent on borrowed capital, and ultimately lenders. These lenders have traditionally operated on a collateral basis. However, the nature of farm capital investments is changing, shifting from investment in hard assets such as land to investment in technology or education, with no or unknown market values. This change in the nature of capital investments, combined with uncertainties associated with farm return on equity may hamper farmers' abilities to attract external equity or debt capital. The USDA (1999) reports that real return on farm assets financed by debt has been on average –3.8% in recent years, pointing out the continuing inability of farmers and lenders to accurately forecast future financial performance of farm businesses. Such inability has led to two major farm financial crises over the last century, including one in the 1980s during which many farm operators disappeared (Bierlen and Featherstone, 1998). Even though the 1980s crisis is over, better techniques to monitor and predict farm financial conditions would allow farmers to make better managerial decisions, assist lenders in reducing the amount of non-performing loans, and ultimately help reduce the cost to borrow. This paper is an attempt to improve the ability of farm managers and lenders to forecast expected financial performance of farm businesses using a neuro-fuzzy inference system (NFIS). 1 Many of the studies that examine financial stress and performance do not analyze farm financial stress but focus on ex-post identifying financial failure points out, little can be done at the …