In the fields of economics, ecology and social affairs, small differences in where you start can have a huge impact on where you end up. This is the path dependency that led to driving on the left hand side of the road (in the UK), or the near-universal use of the (supposedly) inefficient QWERTY keyboard. Since tiny causes can have large effects on complex systems then 'even knowing 99% of what you need to know leaves you vulnerable to large errors. And 100% knowledge is impossible.' (Source).
All this is one reason for considering Social Policy Bonds for, especially, those social and environmental goals that have many possible causes and are characterised by time lags and seeming intractability. The peaceful resolution of conflicts, for instance, or the promotion of biodiversity, or the avoidance of natural or man-made catastrophe.
As society becomes increasingly complex, you would think that policy instruments that reward positive outcomes but do not prejudge how to achieve them, such as Social Policy Bonds, might be considered more widely. I did try to interest the Santa Fe Institute in Social Policy Bonds. It conducts research into complexity. However, I didn't receive a response to my approach (made nearly two years ago).