Arguing for and against Social Policy Bonds

Over the years Social Policy Bonds have discussed in many forums. I tried to address some concerns raised in these forums in my book, but that is several years old, so I am now going into more detail. This post looks at some concerns that delegates raised at a meeting of the OECD during which I presented a paper on the Social Policy Bond idea applied to the environment. I will eventually collate these arguments and responses and upload them onto the Social Policy Bond website. 

--

If the polluters are few enough to overcome the free rider problem, then why are they not enough to collude in the bond market and thereby pay too cheap a price for the bonds issued?

The question is not about numbers of polluters. It’s about the number of bonds. Bidders for the bonds will compete to buy the bonds, then collude to achieve the goal that the bonds target.

The free rider problem arises when people own the bonds and do nothing to help achieve the goal. The more bonds in the hands of would-be free riders (passive bondholders), the lower the incentive for active bondholders to do anything. The value of the bonds on the market would keep falling. At some stage most bonds will end up in the hands of active investors. There might be few or many such investors, but they have every incentive to collude when they own bonds. They will, though, not necessarily collude to buy the bonds: the bonds will be sold on an open market, and, because anyone can buy them, collusion won’t be possible at that stage. So, there is always competition to buy the bonds, which maximises the returns to the issuing body. And there is always collusion between the shifting cast of bondholders, which maximises efficiency.

A more careful analysis is required about how governments would regulate information signals stemming from bondholders’ pollution abatement efforts and environmental measures on which bond payments are indexed.

The ‘environmental measures on which bond payments are indexed’ would be chosen to be, or to be inextricably and strongly correlated with the targeted goal. They would also be chosen to minimise the need for governments to regulate the associated information signals.

The idea needs a more careful analysis of the pros and cons of bond payment structures. Why do bonds simply pay a lump sum upon attainment of a fixed goal? Why not issue bonds that yield regular payments commensurate with pollution levels? If the bonds an all-or-nothing deal, the risk associated with the return could limit trading.

I agree. But the Social Policy Bond concept remains untried. Refinements can come later. Of course, separate ‘all-or-nothing’ bonds could be issued with targets of varying magnitude along the same scale.