Use current global spendings as a baseline and adjust from time to time. If you have more money available than researchers to pay you can decrease spendings, if

Use current global spendings as a baseline and adjust from time to time. If you have more money available than researchers to pay you can decrease spendings, if you desire faster progress throw more money at it. Besides the added complexity of a global scale undertaking I see no reasons that it is fundamentally harder to do then say planning the budget for road construction and maintenance or other national infrastructure spendings. EDIT: In the case of drugs we also have a somewhat exceptional situation, we have a pretty good idea of what the demand is and what the benefits of satisfying a specific demand will be.