I think that in the early stages of social impact bonds, governments will be reluctant to see them fail. We've seen this in different parts of the world. There are guarantees in some cases, but not in all. But I think the model has so much backing in places like the United Kingdom that failure to meet bond requirements would be approached in a fairly delicate way. My suspicion is that supports would be put in place to make the model a success.
It's difficult to evaluate what's happened so far. There's some evidence of some success, but in cases like Peterborough, the first one, the whole thing was wrapped up far too early for it to be called a success.
Don't [Inaudible--Editor] if you don't make the target. You don't get paid, nominally, but I think that is a very simplistic way of looking at an experiment. Governments would tend to make them work.