Economics of Pesticide Reduction and Biological Control in Field Vegetables—A Cross Country Comparison

National pesticide reduction programmes vary considerably among EU member states. Countries can be divided into two groups: those that have adopted measures to achieve quantitative reductions in pesticide use, and was those that have taken national initiatives without reduction targets. Denmark and the Netherlands included reduction targets in their crop protection legislation in 1986 and 1990, respectively. This policy has resulted in induced innovation in these two countries through the development of reduced-dosage and pesticide-free pest control methods. In contrast, Germany has taken a very different route by working more on a stricter authorisation process. Low pesticide-use technologies are rarely applied in this country to date.

Chapter 2 sets out to select three vegetable crops for further investigation based on area planted, quantity produced and value of production per hectare in 1993–2003. Carrot and onion turn out to be important in all three countries (based on total quantity produced and area planted). The choice of leek is less obvious and seems to be more inspired by non-statistical reasons such as the availability of environmentally friendlier pest control methods. Chapter 2 discusses the differences in pesticide use legislation, and particularly the significant reduction in use figures that was achieved with the reduction programmes in Denmark and the Netherlands. The latter discussion is followed by a detailed description of the prevalent pest, weeds and diseases in the three selected crops and the differences in the common control methods between the three countries.

Chapter 3 presents an overview of the economic literature and decides on the appropriate method for the empirical analysis. The analysis is carried out from the point of view of the individual farmer. Optimal control by crop and pest control method depends on pest incidence, control effectiveness, potential maximum yield and the price by the quality class of the vegetable. To ensure a fair comparison of the control strategies, the analysis also needs to account for the stochastic natural and market conditions under which these are applied. Bio-economic modelling is selected as the appropriate tool to explore the potential for environmentally friendlier practices in German vegetable production. The author opts for stochastic partial budgeting to combine and compare insect and weed control strategies. Data come from field surveys and secondary data. Expert elicitation (Delphi procedure) is used to validate the partial budgeting models. The model output is the cumulative frequency distribution functions of the net returns of the alternative pest control strategies for the three crops. Next, stochastic dominance analysis is used to compare the cumulative frequency distribution functions and to identify efficient strategies considering risk aversion.

Chapter 4 discusses data collection. The survey respondents were located in areas with similar agro-ecological conditions in the three countries, which is assumed to eliminate country-specific influences by crop. A detailed account is given of the prophylactic and curative pest control efforts including the use intensity and effectiveness as observed by the growers. Yields and prices by quality class are compared and the pest situation (infestation levels and frequency) as perceived by the grower is described. The total sample of 134 growers is categorised by crop, country and organic, conventional and integrated production methods. This procedure results in small group sizes (1–14 growers) and no statistical tests are performed on the group means and their standard deviations. It is concluded that, in practice, integrated pest control can no longer be distinguished from conventional practice: what used to be called integrated has now become the conventional practice.

With the information gathered from both the field surveys in Denmark and the Netherlands and also the experts, Chapter 5 addresses the potential of reduced pesticide dosage regimes for Germany. The environmentally friendlier regimes are for insect and weed control only because in the survey no reduced-dosage regimes were observed for the control of fungal disease in the three crops. The measures for insect (cutworm and carrot fly) and weed control in washed carrots from Denmark are translated into 27 distinct combined control strategies for Germany. Based on survey data from the Netherlands, six combined control strategies are distinguished for the control of thrips and weeds in German leek production and another 18 combined strategies for the control of thrips, onion fly and weeds in dry onion production. Next, net returns are simulated for each individual strategy and crop. The farm economics of the few organic control methods available are analysed in a similar manner. It is concluded that reductions in pesticide use and simultaneous increases in net return are possible in German leek, onion and carrot production. The results also show that the reduced application and biological pest control do not necessarily lead to a higher risk from pest control.

Finally, Chapter 6 addresses constraints to adoption of the best management practices identified in Chapter 5. Whereas in the previous chapters any non-treatment effect is intentionally excluded, here they are the focus. For this purpose, the pesticide application frequency by crop by farm is regressed by ordinary least squares (OLS) against other variables from the survey (conventional and integrated growers only).

The bio-economic modelling part of the book is rigorous and very well documented. The author has collected and processed a large amount of data—the book includes no less than 115 tables. I found it somewhat surprising that despite the data intensity the book never mentions the actual reduction in pesticide use possible by switching to the more profitable and more environmentally friendlier strategies. With some effort, this reduction (by volume and in terms of application frequency) can be recovered from the background material included in the book.

I was less convinced by the econometric exercise in Chapter 6. With the OLS procedure, the data on the application frequency is treated as continuous whereas in fact this is a discrete variable, which takes only strictly positive integer variables. The estimation should take into account the non-negative, non-zero integer nature of the treatment count data (Hubbell, 1997). Suitable candidates are the zero-truncated Poisson model or the zero-truncated negative binomial model and these can be estimated with procedures included in recent versions of STATA and LIMDEP.

The book is the published version of a doctoral thesis from the Institute for Horticultural Economics at the University of Hanover under the supervision of Professor Hermann Waibel, well known for his work on pesticide economics. The bio-economic partial budgeting approach discussed in the book has been successfully applied to pesticide-use issues elsewhere (Pemsl et al., 2004; Pemsl and Waibel, 2007).

References