If You Want Government Programs to Work, Think Like a Designer

While it’s relatively new in the digital world, design thinking was actually pioneered years ago in government. For example, the Clinton-Gore Reinventing Government initiative sponsored a pilot called “Oregon Option,” where nine federal programs for child health services were combined in Oregon to increase the number of children immunized against childhood diseases. The funding flexibility was contingent on increased accountability for delivering results. The pilot worked, but with a lack of support and a change in the state’s governor (who was the champion), it faded.

A couple of years ago, the concept was resurrected. Congress authorized a pilot program, Partnership Performance Pilots for Disconnected Youth (P3), to integrate services around a target population in exchange for results—in this case, at-risk youth. Nine pilots were launched last year and the government just announced a competition for 20 more this year.

How does this approach work?  Patrick Lester, in a new report for the IBM Center, explores how seven federal departments and agencies are collaborating with states, localities, Indian tribes, and nonprofits to create a basket of services and strategies to address the complex needs of disconnected youth. He concludes that it is probably too soon to tell how effective they have been since they were just finalized in April, but he outlines progress to date and next steps.

According to Lester: “At its heart, the P3 program is a test of several interrelated strategies.” It is a test of the effects of increased flexibility and a test of agencies’ and other partners’ abilities to tie together multiple programs and providers with a common focus on targeted populations with complex needs.

According to the Department of Education, the program’s managing partner, P3 is intended to “test the hypothesis that additional flexibility for states, localities, and tribes, in the form of blending funds and waivers of certain programmatic requirements,” will result in improved outcomes for program participants. In addition, notes Lester: “P3 is a test of the idea that compliance-based regulatory accountability can be reduced in favor of increased performance-based accountability.”

In this case, the target population is low income disconnected youth, aged 14-24, who are either homeless, in foster care, involved in the juvenile justice system, unemployed, or not enrolled in school or are at risk of dropping out.

A formal notice inviting pilot applicants was published in November 2014, offering up to $7.1 million for up to 10 programs. Nine were awarded in September 2015. The recipients had to negotiate a legally binding performance agreement (not unlike the process for Social Impact Bonds), which describe the approved waivers of federal regulatory requirements, the blending of federal funds from different agencies, the performance metrics to be used, and the mechanisms that would be used to enforce the agreement. Most performance agreements were signed by late March 2016.

Approved programs range in size from 80 participants to 8,000 participants, and the target populations includes rural youth, tribal youth, teen mothers, and youth living in public housing.

Lester says that with the first round just underway, “it is too soon to know with certainty” how well these goals are being met. However, he offers some preliminary observations about several key aspects of the program regarding its use of collaborative approaches, administrative flexibility, and performance metrics.

More Coordinated Service Delivery

Lester observes: “Although a single nonprofit or public agency can provide comprehensive or tailored services, the needs of disconnected youth often require coordination among multiple public agencies and service providers.” Since disconnected youth are served by multiple organizations with different priorities—such as increasing high school graduation rates, getting participants into the workforce, or preventing teen pregnancy—it is important to create partnerships among these organizations to establish common agendas and reinforcing goals for these youth. The nine pilots are doing this.

One approach to creating mutually reinforcing activities is to use logic models and theories of change, notes Lester. These are generally described in grant applications, as well as later in local memoranda of understanding and the federal performance agreements. As Michael Twyman, executive director of an Indianapolis-based P3 pilot program said: “That is a smarter way for us to work. It allows us to bring more holistic services to individuals.”

Continuous communication is also important to effective collaboration, writes Lester: “Although responsibilities are often reflected in formal legal arrangements . . . the actual interactions among local partners are often more dynamic, requiring ongoing communication and adjustment that rely heavily on relationships and trust.”

Modest Use of Administrative Flexibility

The statute authorizing the P3 program allows participating federal agencies to waive any statutory, regulatory, or other administrative requirements within certain constraints (they may not waive civil rights provisions, for example). The requests for waivers “are intended to be bottom-up, driven from the local level.”

According to Lester’s analysis, most waiver requests to date involve “changes to eligibility requirements, allowable fund use, and performance reporting. Local projects may also propose blending federal funds [from different agencies and programs] into a single stream with a single set of administrative requirements.” He concludes: “So far, most of the approved waivers appear to be modest, both in numbers and scope.”

The statutory waiver authority expires for the first round of pilot projects in 2018. Waivers granted in the second and third round of pilots will expire in 2019 and 2020, respectively.

Performance Metrics Are Evolving

Lester writes: “After reaching agreement on broad goals, [the next step] is to translate these goals into concrete, measurable indicators that allow for tracking progress and making adjustments over time . . . each project’s performance agreement includes projectwide metrics that track educational, employment, and other key outcomes.”

Local project leaders propose the metrics, and final metrics are then negotiated with the relevant federal agencies. Thus far, metrics are “rooted in varying points of the project’s logic model,” Lester said. This allows tracking against milestones as well as “diagnosing and addressing potential problems when they arise.”

In addition, most of the pilot sites rely on “existing data systems to provide the outcome data needed to track their performance,” according to Lester. The grant applications “built in the expectation that the pilots would bring such pre-existing capacity . . . including executing data-sharing agreements, managing and linking data, maintain data quality, and protecting privacy.” For example, the Chicago Young Parents Program relies primarily on its pre-existing Head Start data system with additional data provided by its nonprofit partners.

What Happens Next? 

In late April, the administration announced the first of two sets of competition for additional pilots. On June 20-21, the nine existing P3 pilots will participate in a joint Community of Practice meeting in Washington to share their lessons learned.

Lester recommends that federal officials and Congress support more aggressive use of P3’s waiver authority, via the use of more proactive technical assistance to pilot recipients. He also recommends extending the length of time allowed for P3 pilots and waivers (the first round expires Sept. 30, 2018) because “the projects must be given enough time to reach their full potential,” such as five years.

Will this new push for the use of customer-focused design thinking take hold? Technology makes it easier to collaborate and collect data than it was two decades ago, but the real key to success will be whether the efforts make a difference—and if governments are willing to streamline the path to get there.