On Tuesday, Rick Cohen appeared on THE KOJO NNAMDI SHOW on WAMU in Washington, D.C., to discuss social impact bonds with Jeffrey Liebman, who is a professor of public policy at Harvard’s Kennedy School of government and director of Harvard’s Social Impact Bond Lab. TUNE IN HERE TO LISTEN TO THE EXCHANGE!

At the Council on Foundations’ annual meeting, during a plenary on the state of philanthropy, White House social innovation and civic participation director Jonathan Greenblatt touted the potential of “social impact bonds,” or, in the Obama administration’s nomenclature, “pay for success.” It’s hard to imagine a concept that has taken off quite like social impact bonds—which aren’t actually bonds, but more like equity investments in social problems with a government payout of costs plus an investment return if the programs meet predetermined outcomes.

Given that there are only four social impact bond projects underway in the U.S., the hoopla for SIBs is something else. SIBs might be an interesting tool for attracting private capital for the multi-year capitalization of a variety of social programs, but so far they have been rather narrowly focused on efforts to reduce prison recidivism (New York City), increase employment—and reduce recidivism—for former incarcerated persons (New York State), reduce youth recidivism (Massachusetts), and promote early childhood education (Utah).

None of the U.S. examples, nor the prison recidivism project from the UK, have even reached their first payment points, but that hasn’t tempered advocates from imagining all kinds of SIB uses. Just yesterday, for example, two researchers opined that social impact bonds could solve the nation’s urban blight crisis, by privately financing through social impact bonds “promising intervention(s)” that would “improve blighted spaces and yields [sic] savings” in “reduced police, fire, and public welfare costs.”

Two of the authors of that SIB blight idea, John K. Roman and Kelly A. Walsh, are among the four authors of a new study from the Urban Institute, titled Five Steps to Pay for Success, that focuses on projects aimed at improvements in the juvenile and criminal justice systems. Like other reports that have been issued on SIBs, particularly instructionals from the Rockefeller Foundation (A New Tool for Scaling Impact and an exuberant SIB infographic), the Urban Report is enthusiastic about the SIB potential and detailed in its description of how SIBs work.

The report is quite brief in its treatment of the potential downsides of SIBs and PFS, noting Kyle McKay’s analysis prepared for the Maryland state legislature as criticizing operational aspects of social impact bonds as one potential area of concern and what they call “philosophical” critiques regarding the implications of introducing private for-profit capital into the social services arena as “an abrogation of government’s responsibilities to address social problems.” Roman, et al., dismiss the criticisms with the observation that the SIB/PFS field is relatively recent, thus “it remains to be seen if these critiques will be borne out as the field develops.” The same could of course be said about social impact bonds themselves, which are yet to be proven beyond the experiments they are, much less than replicated with demonstrable government savings.

On Washington, D.C.’s National Public Radio station, WAMU, the Kojo Nnamdi Show focused on social impact bonds this week, prompted by the announcement last month that the Washington, D.C. government was going to examine the potential of a SIB focused on teen pregnancy. Our role on the show, it seems, was to temper what might be the tendencies toward irrational exuberance on the parts of some promoters of SIBs, who sometimes slip into language suggesting that the market discipline purportedly inserted into social programming by private capital is much more broadly applicable to a range of social problems than the experience so far suggests.

Having done substantial public-private financing in prior senior positions with national financial intermediaries, we have some enthusiasm-tempering considerations that SIB/PFS advocates and critics might reflect upon:

  1. If these are proven models of social intervention, why structure them as single-project SIBs rather than broader policy changes? The SIB promoters all say that they pick exceptionally strong programs and providers carrying out interventions already documented and proven to work. If that is the case, why wouldn’t government simply turn the intervention into a larger, broader program rather than limiting it to one site? That’s a bit of what happened in the UK with the Peterborough Prison project, as the government decided to turn the project into a national program rather than waiting several years for Peterborough project results.
  2. How much of a return on investment do socially motivated SIB/PFS investors really need? The highly publicized Rikers Island recidivism project is getting a $9.6 million investment from Goldman Sachs with the idea that if the project succeeds, Goldman will leave with a 22 percent return on investment, or $2.1 million. However, $7.2 million of Goldman’s $9.6 million investment is guaranteed by Bloomberg Philanthropies, meaning that Goldman Sachs is only actually risking $2.4 million. The result, in terms of actual risk, is that Goldman Sachs is risking $2.4 million to potentially earn $2.1 million, an 87.5 percent return. Does social innovation with private capital mean having to offer lucrative returns of 22 percent, much less 87 percent, when governments can sell tax-exempt bonds for significantly less cost to the taxpayer?
  3. Do insured investments, such as Goldman Sachs’ with Bloomberg backing, really test the market for SIBs? Clearly, even with a strong program like the recidivism project at Rikers, the massive Goldman Sachs was leery of a major investment and protected itself with Bloomberg’s money. Imagine if Goldman Sachs were to lose its $2.4 million (since Bloomberg pays the rest). Would Goldman write the amount off as an above-the-line business loss? Might it try to get a charitable tax deduction for its expenditure in support of the nonprofit venture, given Goldman Sachs’ big announcement a few years ago of a major pledge to charitable giving as a result of its mammoth profits and executive bonuses? Might other potential SIB/PFS investors, such as the frequently mentioned Bank of America, look for Community Reinvestment Act credit, an idea pitched in the Urban report?
  4. Have promoters of SIB/PFS investment structures forgotten alternative, well-known mechanisms for private investment, such as low-income tax credits and New Market Tax Credits, which have long demonstrated the willingness of private capital to invest in social programs—at much lower rates of return than those anticipated by Goldman Sachs? To dismiss those tax credit investments as somehow less risky than SIBs because they are collateralized in part by real estate misses the entire high risk and reward history of public-private partnerships in urban development.
  5. Shouldn’t SIB/PFS promoters advocate for adequate overhead and full cost reimbursement in government grants and contracts for nonprofits in general rather than just in structured SIBs? The advocates point out that SIBs fund both the program and overhead costs of the nonprofit sponsors, frequently underfunded or unfunded in government programs. The answer to that would be to push for standardized, adequate overhead rates as a matter of policy in all federal government contracting, as the National Council of Nonprofits did in getting the Office of Management and Budget to establish a change in the rules to give nonprofits the option of at least a minimum 10 percent overhead or indirect rate on government contracts with the ability to negotiate higher rates.
  6. Doesn’t the SIB focus on established providers work to exclude those nonprofits without the upfront working capital to plan and design projects? For most nonprofits, even with shortfalls in overhead and program reimbursements, the biggest gap is in their ability to access front-end risk funding to plan and design projects and programs. If SIB/PFS projects were to address that funding hurdle, they would be helping a much broader array of nonprofits. Some financial intermediaries have tried to address these needs through predevelopment financing and “recoverable grants,” so the concept is hardly unknown or untested.
  7. Similarly, doesn’t the focus on established providers with proven, evidence-based programs mean that SIB/PFS structures are funding what works and not really risking capital to explore potential new ideas—and new nonprofit entrants? That’s the higher risk possibility, funding nonprofits with new ideas that haven’t been so well established that government could simply adopt the ideas as more broadly applicable programs. If the programs are well established and proven, government could and should fund more broadly, like Peterborough, but those nonprofits that are trying things that are new and truly experimental, that’s where private risk capital could really advance social change.
  8. Is it possible that a potential SIB/PFS downside is that private capital might overly influence the decision-making and priorities of government through the SIB/PFS model? As one advocate testified in a Congressional hearing, a SIB “improves decision-making by bring market discipline to government decisions about which programs to expand, as investors will only put their dollars behind programs with a strong evidence base.” If government overly focuses on programs that will attract private investors, the results might work to the investors’ benefit, but not necessarily to the benefit of appropriately identifying and prioritizing social initiatives that don’t generate private capital interest. Should private investors determine “which programs to expand,” or should public debate and discussion in a democratic process about human needs be the determining factors?

The discussion on the Kojo Nnamdi Show raised these and other issues in terms of the benefits that social impact bonds might bring to social programming and the cautions that governmental agencies, nonprofits, and investors themselves might want to heed going forward. A willingness to experiment in the social policy arena is a great thing to have both in government and in the private sector. Because there is private money in the mix, such as Goldman’s or Bank of America’s, critics should not jump to conclusions that the results will be automatically toxic. By the same token, however, SIB/PFS advocates might want to remember that the “market” doesn’t make everything right, doesn’t automatically make better choices or decisions than government, and doesn’t guarantee better outcomes.