July 20, 2016

From Input to Impact: Pay-for-Success Drives Government Performance (Part II)

Row of hanging lightbulbs

In this second and final installment in our miniseries on pay-for-success models, Magdalena Seol interviews George Overholser, CEO and Co-founder of Third Sector Capital Partners. The first installment in this series discussed and evaluated the different forms of  pay-for-success. This interview has been edited for clarity and length.

In 2007, George Overholser delivered a speech in Florida, at America Forward’s Gathering of Leaders, in which he proposed a $1 billion social investment fund to advance proven intervention programs with a combination of social returns to beneficiaries and fiscal returns to government. This was the first write-up of his idea of “money back guarantees.” Mark Warner, who was governor of Virginia then, told him, “George, don’t even try it. It will never work.” Soon after, social impact bonds were developed in the U.K. independently. There are currently 11 such projects going on in the US, the first one being launched by New York City in 2012 to reduce juvenile recidivism.

Q. What is so special about pay-for-success?

There are two fundamental underlying assumptions. First, the vast majority of social spending in America is done by government and philanthropy is comparatively tiny in just about any issue you can think of — poverty, education, recidivism, homelessness, wellness, etc. In my own philanthropy, I quickly came to realize that you get nothing done unless you change the government. Second, the information revolution is finally reaching government, giving us a low-cost way to see if people’s lives are improved by the programs that governments purchase. What’s been revealed is that in nine out of 10 instances, there is no discernable impact seen between the persons who have been served by programs compared to those who have not; you simply can’t see the impact. However, one time out of 10, there really is an impact. Pay for success is about reallocating government dollars away from services that don’t work and towards those that actually do bring about improvements to people’s lives. PFS is not about the finance or capital market. It is about “reallocating government dollars.”

Q. It’s been almost 10 years since the speech in Florida. Have there been any changes in the PFS world?

Third Sector Capital launched 5 years ago in 2011. At that time, the discussion in the UK was all about tapping into capital markets. The thing that’s changed is moving from social impact bond as a centerpiece to pay for success contracting as the centerpiece. Pay for success is “contracting.” Social impact bond is “financing.” Financing is helpful but it’s not like actually paying for services in the end; it just provides temporary loans. A shift in focus from financing towards paying for what works — that has been a change.

A more recent trend is the Social Impact Guarantee (SIG). It seems subtle, but instead of making loans, SIG is a form of insurance. It’s an easier way to get all the things done. Government pays and if the program fails to have impact, the SIG pays the money back. In this way, government is behaving in a way that it is more familiar, and it’s the private side that does all the financial work.

By far the biggest change is a shift from a world where none of the government spending streams were “PFS-enabled” to a world where existing spending streams are becoming PFS-enabled. Most spending streams are still attached to laws that do not permit PFS-style contracting. You can’t do multiyear contingent contracts — it says in the law that this money is to reimburse people for following some recipe and here is the recipe. Well, we want to change the recipe and we don’t want a reimbursement but a performance payment. And so for the first five years, whenever we’d work on a new project, we had to create a new source of funding. It usually took one to three years just to launch the projects, and five years to have the laws changed. Now, PFS-enabled government budgets are already there — they are now being internalized into existing government budgets. We have billions of dollars of the federal spending streams that have become PFS-enabled. This is a very important change that is happening now. This change will accelerate the growth of PFS. The Work Force Innovation Opportunity Act, for example, now has up to 15 percent of spending marked as PFS-enabled.

Q. How do you deal with the issue of change in government administrations?

Clearly, the place where Third Sector Capital Partners starts is the government, but if you take away any part, the whole thing wouldn’t work. We mostly work with the state- and county-level governments. Given that the government administrations change, it is illegal to do a multiyear contract — you can’t obligate the future administration to be forced into a contract. One of the benefits of this whole approach is involving philanthropists and banks — the projects can withstand the administration changes. It resolves the continuity issue, which is a big problem in government programs. Private investors are not willing to make a loan unless there is a longer timeframe. Part of our expertise is to keep the project for five years. One way to do that is to have a penalty on the contract. A big part of the contract is a “shut-down provision.” If the government fails to appropriate funds for next year, then it’s no longer a pay for success contract, it’s just a pay.

Q. What are the benefits of PFS to other parties — private funders and nonprofits?

A growing realization is that PFS is a spectacular philanthropic product. Philanthropists like this new approach because the alternative is to try paying by themselves for the social programs. But they are too small and they worry about creating dependency relationships with the programs they support. With PFS, they pay, and if it works, the government replenishes the money. Instead of one check and done, PFS makes once and forever possible. With one check, you create something forever.

Both the SIB and SIG protect the nonprofits. In the deals nonprofits make with the government, they’re saying that we’re willing to risk not being paid, but in exchange, you’re going to make it happen so that we don’t have to follow your recipe. In both models, there is some third-party financing to protect the nonprofit if the nonprofit doesn’t get paid. One of the service providers that we’re currently working with is such a big provider that they don’t need a third party to pay — they’re self-insurant.

Q. You mentioned the increased data availability in the beginning. Can you say more?

Administrative data is a by-product of running a government. For example, there is something called the National Student Clearing House. Governments want to make loans to students. In order to do so, they need to make sure that the students are actually in college. So, every college participating in this loan program has to send the government their student database so that the government can confirm they are enrolled. This means that the government has a database that has every college-goer in the country. The data is updated every semester, and is available for anyone who says I can raise the college-going rates of the high school graduates. You have privacy laws that you need to work with, but the data is available. These are government data, and there are databases for foster care systems, prisons, homelessness, health care, etc. Every one of these databases is available to measure the outcomes. More data is available to figure out which projects or programs would work. In a case where there is no data, we wouldn’t get into the project.

Q. What do you expect to see in the PFS world?

There are many different ways to have performance-based contracts that have to do with outcomes. Let’s say here’s a contract that serves 100; if you do well, you will be paid to serve 200 next year; if you do poorly, it will be 50 next year. This is not strictly a pay for success contract but it still has a performance feedback loop into it. So, one way to think of PFS is, “Do you have an outcome and performance feedback loop that is connected to the money?” There are many impact evaluations out there and they might be interesting for reading. But unless they are connected to the money they don’t change anything. The real way to think of PFS is to connect the money to the impact evaluation. I think this is where it’s heading.


The views expressed in the Government Innovators Network blog are those of the individual author(s) and do not necessarily reflect those of the Ash Center for Democratic Governance and Innovation, the John F. Kennedy School of Government, or of Harvard University.

Related Topics

Related Topics