Pay For Success Bonds

President Barack Obama's proposed FY 2012 budget included a provision that provides for up to $100 million of Pay for Performance bonds, the US incarnation of Britain's Social Impact Bonds.  The bonds will be issued by local governments (or other entities) to fund programs designed to attack America's thornier social problems.  It allows for long term capital to be committed where  government resources can only be appropriated one year at a time.  The bonds are paid off under contract from the governmental entity based on the success of the program funded, yielding up to 13.5% if the UK model is followed.

Observations:

1. These bonds provide all of the funding for the enterprise.  Thus investors are assuming equity risk is the investment.
2. The tax provisions for these bonds are not addressed in any of the materials that I have read.  The tax treatment could have a huge impact on the pool of capital available for these instruments.
3. The bonds will be subject to appropriation risk in addition to enterprise risk.  Unless there is some escrow of funds, there can be no assurance that funds will be available to pay off the bonds regardless of the success of the project.
4. Bonds will be highly illiquid.
5. This "investment"  amounts to venture capital in the non-profit sector.  A typical venture capitalist would not consider a 13.5% cap on return as attractive.  Highly favorable tax treatment would be required to make the investment economically attractive.
6. On the other hand, favorable tax treatment is irrelevant to non-profits who would would be natural candidates for providing this kind of funding.  These institutions are already providing funding, through grants, to those programs deemed most likely to succeed.  Therefore, these bonds could have the perverse effect of funding marginal programs which are likely to have a low success rate.

It seems to me that the major benefit of this vehicle is to bridge the timing factor of government appropriations, and assist a program in raising capital to put into place infrastructure to deliver services to the community.  It occurs to me, though, that if the services are useful and truly valuable, creating a revenue stream should be rather straightforward, and capitalizing the revenue stream could be handled by the private markets.  As described, these bonds provide seed capital for start up businesses, with very low potential profits.  I see the potential market for these bonds as very small.