Pay for Success and the Human Value Chain

After a bit of a break, I am finally getting back to the Trans-Atlantic dialogue I had going between myself and UK blogger Privatising Schools. This post is the first in a Q&A series about social impact bonds and pay for success finance.

Privatising Schools: Question Seven

A key focus of your research is the connection between education technology and finance. This aspect of your work answers a lot of questions for me. I’d struggled to understand why the privatisation movement was being pushed not only by companies with a direct interest in selling ‘education services’ – the edu-businesses, the big outsourcing firms, the tech industry – but also by the financial sector.

Why are hedge fund and private equity firm managers sitting on the boards of US charter school chains, or of academy chains here in England? Why is our so-called ‘academies minister’ a former private equity boss? Why have these people suddenly developed a philanthropic interest in education?

You have shown how financial interests, centred in New York and London but operating on a global scale – via the Global Impact Investing Network, for example – are seeking to ‘financialise’ elements of the old welfare state, using complex new instruments like the social impact bond (or SIB). The aim is to create a new investment market within public services – healthcare, education, social services – which have been hollowed out by years of austerity.

As you know, England is the home of the social impact bond. The very first SIB was launched at HMP Peterborough, a privately-managed prison, in 2010. (As it happens, one of the inventors of the SIB, Toby Eccles, is another former Ark employee.) SIBs were then exported to the US, in the form of the ‘Pay for Success’ model. The first SIB in the US, I think I’m right in saying, was launched by Goldman Sachs in 2012, and was used to finance an anti-recidivism programme similar to the Peterborough scheme.

Three years later, Obama’s Every Student Succeeds Act plugged Pay for Success directly into the US public education system, by tying federal funding directly to this new financial instrument.

Could you say more about this? How do social impact bonds work? Why is technology so integral to this new investment vehicle?

My Response:

The premise of a social impact bond is that investors pay for public service delivery with the understanding that if the interventions they’ve invested in “work,” the government pays them back, plus a pre-determined, built-in profit. “Successful” interventions purportedly save the government money by reducing the need for special education, addiction treatment, incarceration, and chronic health management services. The idea is that people identified as being “at risk” of becoming a “burden” on society are “treated” via interventions that are supposed to reduce their likelihood of accessing services in the future.

Of course this structure can be gamed, as evidenced by the Salt Lake City, Utah pre-kindergarten SIB that supposedly reduced the need for special education services for participating children by 99 percent. Most well funded preschool programs, which the Goldman-Sachs-backed project was not, yield at best a 50 percent reduction. So, either the screening tool used to identify children for the program was flawed, or children who may have needed special education services were denied access. This New York Times article, “Success Metrics Questioned in School Program Funded by Goldman” provides additional background.

Social Impact Bonds, or Pay for Success as it is becoming better known, is simply a new method of privatization. Impose austerity so government departments, including education, cannot function; proclaim the system “broken;” and use the created dysfunction as an excuse to outsource services. Worse, by using manipulated data to direct anticipated public expenditures into financiers’ coffers in the present, it hobbles future government operations.

This “Third Way” approach to financing “the public good” has found bi-partisan support in the United States. Liberal interests are happy that services continue to be provided, and conservatives are pleased government is only paying for “what works,” based on data. This summer Democrats, Republicans, investors, consultants, and non-profit service providers all gathered in the Senate Building to celebrate the passage of the Social Impact Partnership Pay for Results Act (SIPPRA), which is set to inject $100 million in federal funding to this nascent “Pay for Success” market. Other beneficiaries of “outcomes-based” contracting include telecommunications companies, cloud-based computing companies, and law firms with expertise in Blockchain smart contracts.

In 2008, the Rockefeller Foundation provided funding to B-Lab to develop the standardized impact metrics that would undergird the system. This framework, now managed as the IRIS initiative, maintains a catalog of over 2,500 metrics aligned to the UN’s Sustainability Goals. Many have specific associated cost savings that are used to calculate profit margins for investors. Predictive analytics establish baselines for individuals and populations, estimate the likelihood they will be “impacted” in a “positive way” (as shown by data), and assess what savings might be generated from “fixing” them via interventions. All of this entails significant levels of surveillance.

Under this model, service delivery becomes increasingly digitized since large quantities of data are required to prove the “success” of programs. Evaluators aren’t equipped to review qualitative data, but instead rely on dashboards that aggregate data streams. Services are platformed to ease data extraction. This dynamic is driving the growth of the ed-tech, tele-health, and tele-therapy industries. In school settings, data dashboards include online learning and assessment data, biometric and health data, behavioral data, attendance data, and school climate data.

The “what works” approach to government and Big Data go hand in hand. The lives of those needing publicly funded services, including students and the poor, are being turned into metrics to feed the impact data pipeline. In the United States, New York University’s Governance Lab has teamed up with UK-based New Philanthropy Capital to coordinate a network of data-centers to push this transition to so-called “evidence-based” policy making.

The NYU affiliated labs are funded by the same entities pushing social impact investing: MacArthur Foundation, Omidyar Network, Pritzker Children’s Initiative, Social Innovation Fund, and the Stanford Center on Philanthropy and Civil Society. I created a relationship map of the labs here. They include: Actionable Intelligence for Social Policy at the University of Pennsylvania, the Rhode Island Innovative Policy Lab at Brown University, Urban Labs at the University of Chicago, the California Policy Lab of UC Berkeley and UCLA, Washington State Institute for Public Policy, and the Ministry of Justice Data Lab.

Many question whether this approach can be made truly profitable for investors. That may be, in part, why despite there being increased hype around social impact bonds, only about a dozen have been launched in the US so far. I was able to locate what I believe to be an important insight into the true nature of the planned profit taking. On page 7 of Ready Nation’s report “Early Childhood Pay for Success Social Impact Finance: A PKSE Bond Example to Increase School Readiness and Reduce Special Education Costs” Robert Dugger, a hedge fund manager who works closely with the University of Chicago Economist Jim Heckman, notes the goal of a 2010 working group was to “facilitate the creation of ‘invest in kids’ bonds that can underwritten individually or aggregated into asset-based securities, which can be invested in by individuals and institutions worldwide.” The key here is asset-backed securities.

Kauffman ReadyNation SIB 2.jpg

The majority of the financiers’ profit could then be generated not from the modest “success” awards built into the outcomes-based contracts, but rather from trading against the debt associated with the provision of outsourced services. They are building a mechanism for legalized gambling on humanity writ large. They seek to create derivatives markets informed by real-time flows of data from the digitized lives of people using public services, including students. Developments in Internet of Things-enabled tracking of behaviors through fit bits, educational games, and compliance monitoring apps as well as the increasing sophistication of Artificial Intelligence-informed high frequency trading will make this possible.

SocialSuite How It Works

They haven’t quite yet found the means to scale it, but they continue to experiment in earnest. A new “impact security” recently developed by NPX is being piloted around prison labor in San Quentin under the name “The Last Mile” with backing from Omidyar Network now. Several entities, including Alice in the UK and SocialSuite in Australia, are structuring impact data platforms around Blockchain technologies. Governments, including the state of Illinois, are examining ways to put public benefits on Blockchain to track outcomes and impact. Eventually all the pieces will click. The next “Big Short” will likely be not on homes but on the data and predictive analytics that shape our lives and future opportunities.

One thought on “Pay for Success and the Human Value Chain”

  1. I just finished watching Homecoming, a new series on Amazon. It is a suspenseful show, and I just realized by reading this that it’s all about people who run pay for success and people who get caught up in a pay for success scheme. I won’t give away the ending.

    Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s