• Opinion
  • November 23, 2018
  • 10 minutes
  • 3

Government outsourcing is failing — here’s how to get it back on track

Opinion: After several big failures, we need to look again at commissioning

This opinion piece was written by Mara Airoldi, director of the Government Outcomes (GO) Lab, a partnership between the Blavatnik School of Government and the UK Cabinet Office to establish a centre of excellence for innovative public sector commissioning. This piece was originally published in the Oxford Government Review, and it also appears in our innovative public partnerships newsfeed.


“Carillion collapse to cost UK taxpayer £148m”

“Privatisation of probation threatening to become ‘contracting disaster’”

“G4S stripped of control over ‘failed’ Birmingham jail”

Provocative statements have been hitting headlines across the UK, accelerated by the collapse of Carillion in January 2018. This construction giant buckled under £1.5billion debt (US$1.9billion), and, as it holds so many UK government contracts, from building hospitals to managing schools, there was fear that the disruption caused by the failure might spread to other outsourced public services.

Such fears were compounded when the government announced the early cancellation of private probation contracts — through which private contractors supported the rehabilitation of low to medium risk offenders — in July 2018. Following a highly critical report by the House of Commons’ Justice Select Committee, the government was forced to take decisive action to avert a Carillion-style collapse.

With such high-profile failures in the outsourcing of public services, questions about the relationship between the state and the private sector are high on policymakers’ agenda. And, with a hefty bill for the taxpayer, they’re towards the top of the public agenda too.

At present, the debate about outsourcing is both highly politicised and highly polarised. With many taking a binary stance on whether to outsource or not, it has also become unproductive.

An alternative to outsourcing on trust?

There are many policy tools that have been considered as alternatives to simply trusting that outsourced services will deliver. One in particular is “Payment by Results” (PbR). With PbR, independent providers are only paid by government when the desired outcomes have been achieved.

At first glance the logic is appealing: if we only pay when we get the results we want, we can guarantee high standards of public service and value for the taxpayer.

If only.

In reality there is little evidence for the effectiveness of PbR. One issue is that a focus on specified and measured outcomes can lead to “gaming”, where providers opportunistically deliver outcomes that benefit themselves not the client. It could also see them avoid or neglect providing a service to people who may jeopardise those targets.

The debate about outsourcing is both highly politicised and highly polarised

This is a criticism levelled at the Work Program, which sought to get 785,000 long-term-unemployed people into work but was accused of side-lining the most difficult cases. The same program highlighted another problem: deficiencies in contract set-up. The government ended up paying even the worst-performing contracts, since it was more expensive to terminate than to pay out.

PbR is not, then, a simple story with a problem and resolution. The idea that it will give value for money and deliver effective public services is not a guarantee. Nor is it an open-and-closed failure story either. Perhaps we could think of it as a novel with complex characters and plot twists; perhaps a disappointing ending, but one that is worthy of attention.

Research helps identify which elements matter. These include both technical aspects, such as program design and how the contract is arranged, and “softer” aspects such as how relationships between stakeholders are maintained. We shouldn’t throw the PbR book into the fire without giving it a second read. This is where the GO Lab joins the conversation.

Exploring what works and learning the lessons

The Government Outcomes (GO) Lab is a centre for academic research and practice based at the Blavatnik School of Government. It was launched two years ago as a partnership between the school and the UK government, both of whom recognised the need for robust research and debate around payment by results.

Agnostic on PbR, we are intrigued by its promise, but well aware of the challenges. The aim is to find what does work and understand how the lessons learned can be applied more widely.

In recent years, there has been a “new wave” of PbR known as “outcomes-based commissioning” (OBC), which aspires to retain the benefits of PbR while avoiding the pitfalls. We are currently focusing on social impact bonds (SIBs), which are differentiated by the involvement of social investors who have a joint financial and social motivation. A SIB brings together three key partners: a commissioner, a service provider, and an independent investor — who may be mainstream, socially motivated and/or charitable.

The UK has been a pioneer in this area, launching the world’s first SIB in 2010 to support ex-offenders in Peterborough. Since then, there have been over 40 SIBs across the country in a range of policy areas, from supporting homeless people to find housing in Manchester to helping children stay out of residential care in Essex and young adults to find employment across London.

Collaboration, prevention, innovation

Our latest report, published in July 2018, adds to a growing body of evidence about SIBs’ effectiveness.

It shows that SIBs may help to overcome three perennial public sector challenges. They may support collaboration, allowing local authorities and service providers to work together and “wrap around” citizens to meet their needs. They may encourage earlier intervention to prevent a crisis, saving money in the longer term. They may also bolster innovation, as risk is transferred to the investor.

For questions about outsourcing, an ideological debate is not going to cut it

We are in the early stages of our research, but what is already clear is that it is no good trying to answer a yes/no question about whether SIBs work. Multiple elements matter in SIBs and we want to understand whether or how they can work together.

It is also important to look at whether SIBs work better than other forms of contracting. A project may have been effective, but was it because of the SIB, or would it have worked anyway?

Asking the right questions

Asking the question “do SIBs work?” is unproductive. At the GO Lab, we are interested in what SIBs and PbR can contribute to the future of policymaking. We want to build on the successes, log these in the policy toolkit and encourage open discussion, so we can learn from failures rather than shy away from them.

Zoom back out to the debate around outsourcing and the same logic applies. Rather than debate whether we should outsource public services or do it all in-house, we need to ask a different set of questions.

We must look again at the story to see who the complex characters were. What were the conflicts and were any of them resolved? What were the plot twists and could the story have turned out differently if it was written another way? Even if we didn’t enjoy the ending, were there redeeming features? Maybe it wasn’t a five-star novel, but was it worth one or two?

For questions about outsourcing, an ideological debate is not going to cut it. There is a much more nuanced discussion to be had based on robust empirical evidence and open discussion. To outsource or not to outsource is not the question. — Mara Airoldi

(Picture credit: Unsplash/Helloquence)

Discussion

Leave a Reply

to leave a comment.
Master the skills you need for the public service.

Discover inspiring resources, tools and policies.