I think it's important to recognize that, again, we're not talking about replacing. In many cases we're talking about additional programming.
The risk is not to the organization. The organization itself is guaranteed the funding. It's the return to the investors. The whole idea is that you have to guarantee funding to that organization. Let's say the federal government says they're only prepared to pay based on outcomes. The work still has to be done by the organization—the charity, the non-profit—and that's where we go to the investors and tell them we need to find the money to support this organization while it does this work, this experiment. So the investors' money is at risk. The organization—the charity, the non-profit—gets their money no matter what. The government then essentially repays the investors only based on results. There has to be an agreement going in as to how you measure, and often you use proxies to do that measurement.
I hope I answered your question.