How broad?

Though this discussion is unlikely to have practical application in the next few years (or decades), I will pose the question anyway: should a single Social Policy Bond issue target one social ill, or more than one? Say that we are aiming to reduce climate change and its impacts. Doing so, we could issue Climate Stability Bonds that target a range a physical, biological, social and financial indicators. Because targeting climate change makes sense only at a global level, and because of the wide range of indicator variables necessarily targeted, I am inclined to think that, yes, it would be better to issue Climate Stability Bonds independently of measures to target other disasters. By this I mean that Disaster Prevention Bonds, which could encompass a range of natural or man-made disasters, could or should be issued separately from Climate Stability Bonds. 

Then again, should political violence (wars and civil wars) be targeted by Disaster Prevention Bonds, or should they be the subject of a more narrowly targeted bond issue, such as Conflict Reduction Bonds?

The answers depend partly, as I've suggested, on the range of indicators to be targeted. It would be simpler to keep track of a smaller number of variables, which makes monitoring and verification easier and would focus the minds of bondholders and their agents more effectively. Much would also depend on who would be paying for the bonds' redemption: a global problem like climate change is more likely attract funds from a wider range of governments than a more general bond issue that folds climate change in with other disasters to which some countries are less prone. Since taxpayers would supply the public-sector funding, popular support is relevant here too. 

Against this argument, is that it could be more efficient to target the broadest possible goal, so that bondholders would have incentives, depending on events, say, to shift resources away from targeting climate change and towards conflict reduction, if they believe that doing so would minimise human suffering at least cost. They would be more likely to do this if there were no mismatch between the time horizon of the bond issues; for example, if Climate Stability Bonds required an array of indicators to fall within approved ranges for thirty years, and Conflict Reduction Bonds likewise stipulated a thirty-year period of sustained peace before they could be redeemed. My thinking, though, is that any such efficiency gains would be outweighed by the lack of focus that a very broad targeting implies, and the enhanced difficulty of monitoring a very wide range of targeted indicators. 

For these reasons, and for the funding reason, even Conflict Reduction Bonds might be too broad to attract much interest, which is why I have also written about Middle East Peace Bonds, though any region could of course be the subject of bonds aimed at conflict reduction. 

To summarise, I would think that at a national level, broader is generally better when considering the targets of a Social Policy Bond issue. At the global level, I think that the broadest possible bond issue might be suboptimal. However, I'm open to discussion.