Last update: March 10th, 1998
| 34. | Name: | Jos Verstegen | |
| Organiz.: | Agricultural Economics Research Institut | ||
| Co-Authors: | Jos A.A.M. Verstegen Agr | ||
| Title: | Assessing the Economic Impact of Management Information Systems in Agriculture Using Panel Analysis, Farm Classification Methods and Experimental Economics | ||
| Quantifying economic benefits of management information systems (MIS) is difficult because of the specific product of MIS, i.e., processing of data into useful management information. MIS costs are relatively easy to determine. They include costs for interest and depreciation related to the investments in hardware and software (updates), and operating costs, such as labor costs, MIS training costs, and help desk costs. However, MIS benefits are more difficult to evaluate, because of the wide range of decisions and activities that can be affected by MIS and the crucial role of heterogeneous MIS users in creating MIS benefits. Economic value of MIS arises from: (1) higher benefits because MIS users choose better decision alternatives than nonusers, (2) higher benefits because MIS users make decisions more timely than nonusers, and (3) loss-avoidance, for instance, because MIS users can control a larger firm with the same decision quality as before firm expansion whereas nonusers cannot. In agriculture, MIS-users tend to be better educated, operate large farms, are younger, and typically have more contacts with colleagues using computers, or have children interested in computers. A panel study was set up to sort out the effect of "better management" from the actual benefits of MIS in pig farming. Data were collected on animal recording practice, year of MIS adoption, and annual herd performances from 1982 to 1991. Panel analysis in a mixed-effects model using ordinary least-squares procedures allowed for a separation in farm-specific and (common) trend effects. Adjusted for other effects in the statistical model, average pig farms adopting MIS increased their yearly income per sow by $20 per year, indicating a return on investment of between 220% and 348%. Tests for autocorrelation, influential observations, and nonequivalent control time-series indicated that this outcome is robust. However, the statistical model also revealed a highly significant interaction between farm and MIS effect, indicating that differences between farmers in the effect of MIS use exist. The heterogeneity in MIS impact was further analysed comparing two conceptually different farm classification methods within the same research population: the sociological "style of farming" approach and the farm-economic "management level" approach. The sociological style of farming concept is defined as the specific structuring of farm aspects and is based on a complex of opinions of what farming should be like, shared by a group of farmers. It entails self-classification of farmers. The management level classification is based on an index method for scoring management. Management experts evaluate farmers on six management factors, i.e., education and training, modernity of facilities, farm policy, tactical planning, operational planning, and social aspects. For analyzing heterogeneity of farmers on MIS impact, the "management level" approach was considered superior to the "style of farming" approach, because management levels could explain part of the variation in MIS profitability between farms whereas styles of farming could not. Management levels of pig farmers were positively correlated with MIS profitability (r =0.35, p = 0.02). Although farmers with high management levels tend to be better informed than farmers with low management levels, they derive more added value from MIS. An economics experiment with pig farmers was conducted to yield insight into whether laboratory economics experiments can be used as an alternative to surveys for determining the profitability of MIS in sow farming. Many of the pig farmers also participated in the above-mentioned survey study. Instead of linking MIS use to farm results directly, the effect of different information levels on decision making was investigated under controlled laboratory conditions. An investment project selection problem was constructed to be an abstract experimental economics analogue of the sow replacement problem. In an MIS group, MIS estimates were derived by within-subjects comparisons of decision quality with and without MIS features. MIS features consisted of processed animal data and sorted animal overviews. A baseline group was included to control for learning and exhaustion effects during an experimental session. The result was that decision quality of subjects in the MIS group significantly improved when offered MIS features. Although an overall effect of MIS was found in both the survey and the experiment, survey and experimental MIS estimates were not significantly correlated. Possible explanations for these uncorrelated estimates are differences between the laboratory and the natural environment in levels of communication and decision making routine. Directions for future experimental research are discussed in the paper. | |||