Style and performance of agricultural market advisory services.

Marketing decisions are an important part of farm business management. Farmers are interested in enhancing farm income and reducing income variability when marketing crops. There are many tools to assist farmers in such marketing decisions. Several surveys, including Patrick, Musser, and Eckman (1998) and Schroeder et al. (1998), report that farmers specifically view one of these tools, professional market advisory services, as an important source of marketing information and advice. For a subscription fee, advisory services provide market information and specific recommendations on marketing transactions (Isengildina et al. 2006).

The general finding from early studies of market advisory services is that farmers could outperform the market by a few cents in corn and soybeans by following the recommendations of services (e.g., Gehrt and Good 1993; Martines-Filho 1996). However, these studies are based on a small number of services evaluated in a given crop year and/or a small number of time-series observations for each service. The Agricultural Market Advisory Services (AgMAS) project was initiated at the University of Illinois in 1994 with the goal of providing a more complete and rigorous evaluation of advisory service performance in crop marketing. The performance of at least 23 advisory programs has been tracked for each crop year between 1995 and 2004. In the most recent evaluations, Irwin, Good, and Martines-Filho (2006) and Irwin et al. (2006) find positive performance for advisory services when compared to the average price offered by the market, with price differences around 3 cents/bu. for corn and 15 cents/bu. for soybeans.

The main goal of AgMAS research to date has been to analyze the performance of advisory services as a group with less attention given to differences in performance across services. AgMAS results show wide cross-sectional differences in the performance of services. For example, the difference between maximum and minimum net advisory prices in 2001 was $0.87/bu. for corn and $0.93/bu. for soybeans, which represents a substantial proportion of average market prices for that crop year, $2.00/bu. for corn and $5.34/bu. for soybeans. The wide range in performance suggests there are substantial economic rewards to successful prediction of advisory service performance.

Previous studies find notable differences in the marketing transactions recommended by advisory services (e.g., Colino et al. 2006a, 2006b). Some services recommend only four or five cash transactions during the marketing year, others recommend numerous "selective hedging" transactions in futures or options, and some occasionally recommend speculative long positions in futures or options. Pennings et al. (2004) suggest that the recommendations reflect the "marketing philosophy" or "marketing style" of an advisory service. The differences in style and performance of advisory services lead naturally to the question of whether style characteristics are related to net returns from following the recommendations. A number of studies (e.g., Sharpe 1992; Brown and Goetzmann 2003) examine the relationship between style characteristics and performance of financial investments, such as mutual funds and hedge funds, but not for agricultural market advisory services.

The purpose of this article is to develop measures of marketing style for advisory services and estimate the relationship between style characteristics and pricing performance. Specifically, this study evaluates advisory services tracked by the AgMAS project for the 1995 through 2004 corn and soybean crops. Three style characteristics are considered for advisory services: intensity of futures and options use, degree of activeness in marketing, and seasonality of sales. Since previous research (Pennings et al. 2004) shows that farmers prefer advisory services that match their own marketing philosophy, a detailed descriptive analysis of each style characteristic is presented first. Next, the relationship between style characteristics and pricing performance is estimated using panel data regression models. The final section summarizes the results and offers conclusions.

Style Characteristics

Style analysis of financial investments typically is based on the relationship between returns of the investment in question and various "passive" asset return indexes (Sharpe 1992). This approach is difficult to apply to the performance of market advisory services because identification of the relevant asset return indexes is problematical. A more useful approach is drawn from research on the crop marketing behavior of farmers and corporate hedging programs.

Several studies survey farmers about their crop marketing decisions. Survey questions focus mainly on the intensity of use of different marketing instruments and seasonality of sales. For instance, in Sartwelle et al. (2000), producers are asked about the percentage of crops sold under cash, forward, futures, and options contracts. Goodwin and Schroeder (1994) investigate whether farmers use forward or futures transactions to price crops during the preharvest season. McNew and Musser (2002) employ a different approach to study farmers' marketing behavior. In their study, a game is used to observe hypothetical selling decisions of grain marketing clubs; specifically, farmers' decisions regarding preharvest crop sales in futures and options markets.

Recent research by Cunningham, Brorsen, and Anderson (2004) investigates the marketing style of wheat farmers based on actual transactions as collected from grain elevators. They consider the length of storage period, number of transactions per year, and days between sales to describe marketing behavior. Emphasis is placed on the degree of "activeness" in marketing as measured by the variability of the storage period and the number of transactions. The idea behind these measures is that seasonal timing and number of sales will vary more from year-to-year when a decision-maker believes he/she can forecast future price movements. On the other hand, farmers who do not forecast price movements have a similar pattern in crop sales across years.

Williams (2001) argues that degree of activeness is a particularly relevant characteristic for farmers evaluating market advisory services. More specifically, a program can be considered conservative if sales are made in small increments spread over the marketing season. Also, conservative advisors generally recommend a sell-and-hold strategy rather than being in-and out-of-the-market several times during the marketing period. Behaviors related to a more active marketing style include selling large proportions of the crop in one transaction, frequently reversing positions, and trading production several times a year.

Corporate hedging programs represent a parallel problem to the one analyzed in the current article. Research in this area investigates the proportion of a firms' output that is hedged by futures and options transactions and the advantages of hedging in terms of adding value to the firm (e.g., Nance, Smith, and Smithson 1993). Recently, attention has been given to speculative motives in corporate hedging programs (e.g., Faulkender 2005). These studies classify a hedging program's objective as market timing or risk reduction based on hedge ratio variability. Risk management programs with highly variable hedge ratios are considered active, or selective, hedging programs. Active hedging programs are related to managers having "views" on future price movements and incorporating speculative elements in hedging programs based on their market views.

The research reviewed above suggests that three types of style factors are most relevant for a farmer selecting an advisory service and, therefore, will be used here. The first style characteristic is the intensity of derivatives use, measured by the proportion of the crop traded in futures and options markets. The second style factor is the degree of activeness, which is based on four different measures. The third characteristic is the seasonality of sales, measured by the percentage of crop marketed in each of five periods of the crop year.

Data

The sample for this study is drawn from advisory programs evaluated by the AgMAS project for the 1995 through 2004 corn and soybean crops. The term "advisory program" is used because several advisory services have more than one distinct marketing program. When it was first launched, AgMAS monitored and evaluated a sample of 25 advisory programs, including the most popular among Midwest farmers. Additions and deletions to this original sample occurred for a variety of reasons, resulting in 23-27 programs each crop year, with a total of 41 programs considered in at least one crop year. For this study, a subgroup of 34 programs in corn and 33 in soybeans is selected. These programs are included in at least two of the ten crop years between 1995 and 2004. A minimum of two observations per program is required to estimate the statistical model used in this study. A list of the included advisory programs and years they were evaluated by AgMAS is presented in table 1. (1)