Pay for Success and the Savings Trap

Saving money isn't the point. The conversation should be about government effectiveness and positive outcomes.

savings-trap
(Shutterstock)
"So, how does this really save our jurisdiction money?" I get this question about "pay for success" projects all the time, and I dread it -- not because I can't answer it but because it's pernicious. It sounds practical and technocratic, but it's neither. It conceals an intricate web of misunderstanding about why and how to fund, manage and evaluate public programs.

What's more, this is largely a self-inflicted wound. From its inception nearly a decade ago, many of us involved with pay for success (PFS) have described it as a way to deliver "cashable savings." It's not. Framing it that way is both an over-simplification and a fundamental misreading of how the public sector works.

PFS projects are a way for governments to fund social programs on the basis of their performance. Parties agree on what outcomes constitute success and how to measure them, and then put a price tag on how valuable each outcome is. But most nonprofits that provide program services can't afford to wait to get paid until outcomes are achieved and can't bear the financial risks of a negative result. That's why working capital for PFS programs is often provided by "impact investors," those who seek to generate social and environmental impact alongside a financial return.

Cost savings aren't the point -- not for PFS projects and not for governments. The point is to create safe, just and prosperous communities at a reasonable cost to taxpayers. The idea that, for example, helping the homeless find stable housing has to save money is ludicrous; we support the homeless because the kind of society we want doesn't let the most vulnerable among us live, and too often die, on the streets. In plenty of areas this is obvious: K-12 education is incredibly expensive, doesn't show any fiscal return for decades and pays off over a lifetime -- and is one of our most important public institutions.

Of course, some programs do drive down costs, and good cost-benefit analyses are valuable tools for making decisions. But we should be humble about our ability to predict the future. Sophisticated models can't tell us whether a new drug will stop a disease in its tracks, or whether treatment costs will plummet, or whether policies will change. Each of these dynamics changes the "savings" picture dramatically.

What is really misleading about the idea of cashable savings, though, is that "cashing" them is ultimately about effective governing. If, for example, a PFS program reduces days people spend in jail by 25 percent, it will mean fewer incremental costs, such as for inmates' food and clothing. But most of the real value will come from personnel. There will be less work for staff members -- corrections officers, janitors -- to do. A county can decide to "cash" those "savings" by downsizing jail staff, or it can decide to put these staffers to work on other tasks. Deciding to extract savings from the budget is a political question separate from programmatic success. It's the job of public leaders to decide what to do with the value created by good programs: extract and reallocate it, or reinvest it.

A better way to talk about programmatic success is in terms of effectiveness. Instead of focusing on the cost of services, we need to start focusing on the cost of getting good outcomes. Take the cost of high school. If our state spends $10,000 per student per year and graduates half of the class, we're paying $80,000 per graduate. If by spending another $1,250 per student per year we can boost graduation rates to 90 percent, we won't "save" any money, but we'll get a lot more bang for taxpayer's buck by paying only $50,000 per graduate.

This is what pay for success is about: not cashable savings but government effectiveness.

We need to start having a more sophisticated conversation about the value of programs and stop hiding behind the false simplicity of cashable savings. Governments need to redefine the terms of engagement to more thoughtfully make tough decisions about where to spend taxpayers' money. That's the only way we'll be able to accomplish something refreshingly bipartisan: getting more value out of our government spending.

Director of advisory services for Social Finance
Special Projects
Sponsored Stories
Sponsored
Creating meaningful citizen experiences in a post-COVID world requires embracing digital initiatives like secure and ethical data sharing, artificial intelligence and more.
Sponsored
GHD identified four themes critical for municipalities to address to reach net-zero by 2050. Will you be ready?
Sponsored
As more state and local jurisdictions have placed a priority on creating sustainable and resilient communities, many have set strong targets to reduce the energy use and greenhouse gases (GHGs) associated with commercial and residential buildings.
Sponsored
As more people get vaccinated and states begin to roll back some of the restrictions put in place due to the COVID-19 pandemic — schools, agencies and workplaces are working on a plan on how to safely return to normal.
Sponsored
The solutions will be a permanent part of government even after the pandemic is over.
Sponsored
See simple ways agencies can improve the citizen engagement experience and make online work environments safer without busting the budget.
Sponsored
Whether your agency is already a well-oiled DevOps machine, or whether you’re just in the beginning stages of adopting a new software development methodology, one thing is certain: The security of your product is a top-of-mind concern.
Sponsored
The World Economic Forum predicts that by 2022, over half of the workforce will require significant reskilling or upskilling to do their jobs—and this data was published prior to the pandemic.
Sponsored
Part math problem and part unrealized social impact, recycling is at a tipping point. While there are critical system improvements to be made, in the end, success depends on millions of small decisions and actions by people.