The priority-setting approach described by Herdt generates a ranking of alternative research programs.14 It does not, however, describe an optimal portfolio of investments. For a research manager, however, the optimal portfolio is of keen interest. Other priority-setting methodologies can generate recommendations for portfolios. For example, Evenson, Dey, and Hossain recommend research portfolios for different regions and agroecosystems.10
The portfolios described here and in similar studies, however, are based on equalizing net benefits per dollar of funding across research programs. But note that research programs involve uncertain outcomes. Some programs have low probabilities of success, but high payoffs. Others have high probabilities of success, but relatively modest payoffs. Research planners can usefully approach their task as a portfolio selection problem. Like stock market investors who choose a portfolio of investments to trade off risk against return, research planners should seek a balanced portfolio of safe and risky investments. For example, maintenance breeding for disease and pest resistance is generally low risk but has a low return. The search for drought tolerance, by contrast, is high risk but has a high return.
In the economic literature, it is often argued that public actors can ignore the riskiness of investment projects because individual projects are small relative to the size of the state sector and relative to the economy as a whole. For a research manager, however, the riskiness of a project is highly relevant. Few research programs can afford to gamble everything on a low probability, high payoff project. The Rockefeller Foundation's investment in rice biotechnology was probably appropriate for a private foundation with limited accountability to legislators or the general public. A public research institution, by contrast, may need to invest in a less ambitious portfolio of projects, diversified across commodities and research techniques.
This problem seems not to have been explored as extensively as might be warranted. For most institutions, priority setting and portfolio selection need to be based on some explicit treatment of risk.
Was this article helpful?