(Illustration by nazarkru)
In November 2012, an article in The Telegraph reported that a flagship UK government welfare-to-work scheme was “worse than doing nothing.” In May 2016, the Independent splashed another headline to make the government wince: “Watchdog Criticises Government’s Privatisation of Probation Service.” And in December 2016, BBC News reported on an attempt to “turn around the lives” of families facing difficulty: “Troubled Families Turnaround Claim Misleading, Say MPs.”
What do these three ugly headlines have in common? The answer is quite straightforward: The three government programs they refer to all made use of a Payment by Results arrangement—a form of outcomes-based contracting largely synonymous with Pay for Success or results-based financing. Instead of paying local service-delivery agencies and private providers upfront, the government agreed to pay at least partly in arrears, based not on specified services but on the outcomes program participants achieved.
Though the front-page splashes have subsided, the UK government’s enthusiasm for Payment by Results has not. The recidivism program mentioned above closed in 2020, but the other two subsist, albeit rebranded and much evolved.
Closely related programs also continue to flourish. Take the “impact bond” (or “social outcomes contract”), which adds a third party—a social investor that provides risk-bearing capital which it will lose if results are not forthcoming. A dataset compiled by the International Network for Data on Impact and Government Outcomes (INDIGO), which our research group leads, shows that impact bonds continue to launch at pace worldwide, totaling 206 so far.
Given the apparent political and reputational risks for governments in using these mechanisms, what explains the continued interest? Are they a lost cause, or—with modification—could they help build a robust social economy that better supports people and the planet?
Working the Space Between Policy and People
Governments have long experimented with ways to influence what happens in the space between a policy goal (such as reducing unemployment or recidivism, or improving state care of children) and the lived experience of the people a policy is targeting. As the researchers Rik van Berkel and Vando Borghi suggest in a 2007 article, a web of people and organizations linking policy makers and frontline workers make many crucial decisions, including: What values will guide service provision? How much discretion will frontline workers have? What role will service users play? How will the system hold service providers to account? How will it steer the behavior of providers, frontline workers, and service users?
The traditional, post-war model of government was rule-based, bureaucratic, and hierarchical; it decided and controlled these considerations from the top-down. As the 20th century progressed, these approaches gradually fell out of favor in developed countries. Instead, governments adopted plans and performance targets from corporate governance, and New Public Management was born. Since the 1980s, these governments have increasingly used private companies and NGOs to deliver public services, and as a result, the space between policy and delivery is now congested with a wide range of organizations.
This deliberate move by governments to break their own monopoly over public service delivery has often led to a situation where, instead of being the sole provider, government is the sole customer (on service users’ behalf). In many cases, this has led to a race to the bottom, as providers try to out-compete each other on price. This not only generates damaging headlines, but also prevents providers from becoming profitable or even sustainable, and often yields poor-quality services. The aforementioned recidivism program, for example, pushed big corporate providers to the brink, side-lined and alienated charities with decades of experience engaging former offenders, and did not produce its intended social outcome of reducing re-offending. On top of this, it cost taxpayers nearly 500 million pounds (about $709 million) to end the program early.
This quasi-market for public services has often resembled a medieval court, where an endless stream of jealous courtiers continuously elbow each other out of the way to win the affections of an all-powerful monarch, who just doesn’t have enough favors to go around. Perhaps it would be preferable to instead model a beehive—a buzzing and productive environment where communication and co-operation prevail. Though still more conceptual than real, network governance—the coordinated provision of social services via multiple, interdependent providers and agencies, using relationships built on trust—may be possible. Instead of the so-called “targets and terror” of New Public Management, where those in power discipline providers for perceived poor performance, network governance favors co-operation, co-dependence, and shared responsibility and decision-making between organizations.
The challenge is how to make network governance real. Here, partnerships emerge as an integral tool. Organizations sometimes invoke “partnership” as a shrewd rhetorical device. The word adds positive overtones to processes that might otherwise be unwelcome (such as privatization) and can make relationships appear less transactional than they really are (such as within supply chains). But real partnerships look noticeably different from this: The participating organizations integrate their actions based on a coherent strategy designed to achieve a common set of objectives. Like any type of collaboration, real partnerships carry power imbalances, but they are distinct in that the more powerful organizations use their clout to advance the collaborative endeavour, rather than as an instrument of control. As Peter Shergold, former head of the Australian civil service and now chancellor of Western Sydney University, has put it, this means public servants “forsake the simplicity of control for the complexity of influence.”
Recognizing a Real Partnership When You See It
True partnerships usually share three defining features: durability, risk-sharing, and the explicit articulation of a shared purpose. These elements do not happen automatically, and we cannot simply wish them into existence.
Payment by Results contracts notionally demonstrate all three:
- They tend to be longer-term (multi-year) contracting arrangements.
- The parties share risk. Since government pays providers only when they meet the agreed indicators of success, providers take on implementation and performance risks. Government meanwhile retains political and reputational risk, as noted earlier.
- The parties articulate a shared definition of success via contractual outcome metrics.
In reality, though, Payment by Results contracts are vulnerable to neglectful practices, whereby providers focus their energies on serving the people who are most likely to achieve the specified outcomes, at the expense of those with more complex needs. (Such practices are sufficiently prevalent to have their own lexicon: Creaming,” “parking,” and “churn” all describe tactics for gaming a system driven by perverse incentives). International academic studies like those by Dan Finn in 2009 and Ludo Struyven and Geert Steurs in 2005 show that contract design can either facilitate or buttress against these practices, but it is an enormously challenging and largely trial-and-error process.
Eleanor Carter (co-author of this essay) and Adam Whitworth at the University of Sheffield have described some of the contract design features that aim to stamp out negative practices. One approach is to segment the target population, offering higher payments for higher need people. Another is to split payments into milestones that reward progress toward outcomes, not just the outcomes themselves, recognizing the longer journey some people might have to make. Others include giving the recipients of the service more power: To choose which services they receive, based on star rating signaling quality; to make complaints which are followed up; or to stop using a service altogether.
But while technical fixes like these aim to prevent providers from hacking the system, no Payment by Results contract can be perfect. Providers have more knowledge of what is happening on the ground than governments do, and can therefore find ways to navigate terms and conditions to their advantage if they are determined to. Equally, governments can use the emergence of unexpected circumstances (like a pandemic) to take advantage of providers, by forcing them to comply with terms that are no longer economically viable.
Co-Designing Outcomes to Build a Foundation of Trust
Recent Nobel prize winner Oliver Hart believes the antidote to these problems is establishing stronger, more trusting relationships between partners. Relying on trust to underpin Payment by Results contracts risks sounding naïve. But while the public sector’s culture of regulatory compliance does not lend itself to counting on something so idiosyncratic and uncountable, the idea has found more traction in the private sector. Hart’s research, alongside David Frydlinger and Kate Vitasek’s in the commercial world, for example, has advanced the practice of “formal relational contracting” as a potential route to avoiding self-serving practices in contractual engagements. They suggest that parties adopt guiding principles such as loyalty and equity, and include them within the contract.
It would seem that if private sector partners can build trust, surely Payment by Results partners can. But how? Historically, governments have been the ones to decide on and define Payment by Results outcomes. But paired with the highly competitive procurement processes we mentioned earlier that encourage race-to-the-bottom pricing, these contracts have often become a tool of coercion for governments at providers’ expense. In turn, providers resort to neglecting parts of the target population to make the sums add up.
But it is possible to use outcomes in a different way: as a collaborative opportunity for governments and providers to develop a deep, shared understanding of the social challenge at hand, and then agree on a contract that gives providers the freedom to experiment in solving it. The lessons of relational contracting in the private sector suggest that the additional effort required to co-design a contract could underpin a more trusting partnership between the parties over the long-term, which in turn could underpin more effective delivery. Rather than dictating terms, governments would invest time up-front to agree on a set out of outcomes with providers, negotiate how they would be measured and paid for, and discuss what might happen in different success scenarios. This would help build relationships that would serve the partnership throughout delivery; instead of the parties circling one another with suspicion, they would be better poised to adopt a mindset of continuous improvement, establish a culture of continuous learning, and realize a shared vision of success that would ultimately lead to better outcomes for people using the service.
The degree to which this sort of relational practice in the private sector might translate to the public and social sectors is a matter of fierce debate, not least because public contracts are (rightly) subject to public scrutiny in a way that private ones are not. Nonetheless, a handful of promising examples exist among the social impact bonds launched as part of the UK’s outcomes fund, which our research group is evaluating.
Moving Contracting Practice Forward
Payment by Results contracts based on a process like this are radically different from those that have come before, but there is at least one major inhibitor to the model’s widespread adoption. All Payment by Results contracts—both the competitive, punitive type and the co-designed, collaborative type—tend to incur high transaction costs due to their novelty, relationship complexity, and often lengthy negotiation periods. It is currently almost impossible to put a monetary figure on the full costs of developing and managing a co-designed, collaborative contract; parties rarely explicitly report them. But the greater the effort to agree on mutually favorable terms up front—and thus build trust between partners—the higher the transaction costs will likely rise.
Still, for all their apparent limitations, we are beginning to see evidence that a new type of Payment by Results-style contract could provide the fillip public officials need to gain confidence working in a more relational, trust-based way. Given this, we strongly encourage practitioners and researchers to weigh not only the costs associated with developing such contracts, but also the transaction benefits. Processes that may seem lengthy or expensive at the start can contribute greatly to trust-building and shared understanding. That may well pay off through the more effective performance of the contract, and ultimately, lead to better outcomes for people using services.
Support SSIR’s coverage of cross-sector solutions to global challenges.
Help us further the reach of innovative ideas. Donate today.
Read more stories by Eleanor Carter & Nigel Ball.