Incentives, groups and research innovation

Very interesting story on a recent article in the The Economist, where a prize for innovation resulted in far better results than the funding of individual research proposals. I believe this article underscores two important principles:

1. It takes an ecological/selection model of innovation rather than the traditional deterministic model of betting on one team or individual. In other words, it seeds broadly by making the funds available to whoever can deliver a good product, and then let competition determine who has the best fit. This strategy increases the breadth of the research portfolio, allowing research policy makers to pick the best performing options.
2. It focuses on teams rather than individuals. By giving the prize to whoever can come up with the best product, it is natural that teams with self-selected, hand-picked members be formed. This is in contrast with the traditional method of choosing one central principal investigator.

Of importance, I believe this is one method rather than the method, meaning that it probably works better in some situations and not so well in others. From a Research on Research perspective, it will be interesting to start listing which methods can be used, and then which method combinations result in the best outcomes under different scenarios.