Actually, that kind of speaks to my next question to some degree, I think, because I think that's actually one of the criticisms there are of the current model of crime prevention programs. There's a lack of really having any kind of hard metrics to measure the results. Would you see, though, the idea of a social impact bond having a better way to have more definitive results that we could actually measure? I would think that would have to be part of the social impact bond type of model. Obviously, in order to pay for successful outcomes, there must be a way that you can actually measure that there are outcomes that are meeting the results you're seeking. I know you've sort of identified that just briefly in your previous response, but would you see that as being one of the positives here?
