Assessing and attributing the benefits from varietal improvement research in Brazil

This study has revealed the importance of taking greater care in the attribution of benefits and costs of research in a context in which the attribution problems are made more transparent through the availability of information on the genetic history of crop varieties-information on which institution released a particular variety and its parents. Nevertheless, implementation of the methods used in this study requires a great deal of information on the experimental and commercial performance and adoption rates of individual varieties, and such information is often not readily available. In many cases the results from experimental trials are not kept in an appropriate form, if they are kept at all for the longer time periods required for this kind of work, and information on adoption is often sketchy at best. Even with good information on genetic histories, performance, and adoption patterns, we are obliged to use arbitrary but nonetheless transparent procedures to apportion credit among institutions. Other types of (nonvarietal) technologies may pose different, and in some senses even greater, challenges both in terms of conceptualizing how to address them and in obtaining data; especially, perhaps, privately produced technologies. However, if our results are any guide it will be important to give greater attention to attribution issues in studies of research benefits of all types.