There have been a few cases in the media where research has been done into the outcomes on the ground of public spending.
(I am afraid I only have these results verbally, but they accord with less well researched results I have seen elsewhere.)
The study was into a family in Swindon that over a period of ten years was in various sorts of low grade trouble. A number of public services contributed to the support this family received, at an average cost for each year of £235,000. The family was in no better shape at the end of this investment than they were at the start. The proportion of the expenditure that was construed to have been spent on the family to their immediate benefit was 3%.
The question of what the POSIWID of public services like social care or the police or the benefit system might be is crucial in the current climate. As a starter for 10 I would suggest that in the large it has been important to say that something is being done. In the small, it has been important to say that best practice has been followed. I think it is moot whether we will move at all to transparent decision making on the basis of evidence of what makes a difference.