The answer, of course, is that the environment, within which the Centers work, has changed. Rapid advances in molecular biology and information and communication technology have opened up new opportunities for the use of science to alleviate poverty and manage natural resources, while at the same time creating new opportunities for collaboration and merging of certain genomics activities. The expanded role of the private sector in agricultural research, the expanded use of exclusive rights for research inputs and outputs, and new and emerging challenges facing agriculture and poor people in developing countries combined with successful solution of many of the problems individual Centers were aimed at solving, all contribute to the need for programmatic and structural changes in the CGIAR.

In the early nineties, as the number of Centers supported by the CGIAR grew, voices arguing for more inter-Center activities and a stronger and more coordinated effort grew stronger. The result was a series of new organizational structures such as inter-Center programs, system-wide programs, merging of Centers, and most recently, the creation of Challenge Programs. Some of these new structures were driven by programmatic needs, emerging priorities and opportunities for increased efficiency and effectiveness. Others were unfortunately driven by the desires for structural change for its own sake. The former generally succeeded, the latter failed.

One of the critical questions facing the new Science Council is how the CGIAR can capture the benefits from a more cohesive system program focused on key emerging issues for which international public goods-type agricultural research is needed without losing the potential benefits from decentralized and flexible decision-making and research. In other words, how do we help amplify the impact of the CGIAR beyond the sum of the impact of the 240 projects and programs and how do we maintain a critical mass of research to solve the most important problems.

I believe the answer is to be found, not initially in structural changes, but in a more cohesive system program driven by a small number (say 10-15) of well-defined system priorities, which aims to maximize the impact per dollar spent by the CGIAR, while, at the same time, keeping a small portion of funding for exploratory research that may result in new system priorities. If the CGIAR stakeholders can agree to move in this direction, the next step would be to agree on the priorities. This might be difficult, because priorities imply not only inclusion but also exclusion of certain research activities that may be of particular interest to one or more stakeholder. A review of the priorities guiding the 240 on-going projects, system-wide programs, and challenge programs might be a place to start. Results from the soon to be completed priority and strategy work by the Science Council and priority-setting work by the Global Forum on Agricultural Research (GFAR) could provide very important input. In fact, efforts to move towards a small number of system priorities should be an integral part of the Science Council's on-going work on priorities and strategies.

Each priority area would aim to help solve a critical current or emerging problem facing poor people in developing countries. An overall goal and timeframe would be specified along with objectives, timelines, and annual milestones for each major activity within each program. Center performance would be measured on the basis of achieving relevant milestones, and on the complementarity and synergy achieved as its research is effectively blended into larger system priority programs. The execution of the agreed-upon work program would remain with the Centers and their collaborators and inter-Center collaboration and structural changes would be determined by programmatic needs. Some programs may best be done by one Center in collaboration with NARS and advanced institutions while others may require input from many or all of the Centers.

Would such an approach work for the CGIAR? I believe it could, but only if donors are willing to focus most of their financial support on those system priorities and pay full overhead and only if Centers are allowed sufficient flexibility to carry out a small amount of exploratory research and related activities, that are not part of the system priorities but are essential to keep a program dynamic and moving towards emerging opportunities and needs. Research is best carried out in an environment of learning, exploration, flexibility, and decentralized decision-making. A problem-solving applied research institution such as the CGIAR must keep its eyes on the prize, but it must also maintain an environment conducive to innovation.

Per Pinstrup-Andersen