Using ‘nudge’ techniques to lower housing benefit fraud and error
Working with five UK councils, Capita tested the impact of behavioural science ‘nudge’ techniques to reduce fraud and error across housing benefit.
Increasing the number of housing benefits claimants that report a change in their circumstances is an important factor in reducing fraud and error in housing benefit payments. So important that the Fraud and Error Reduction Incentive Scheme (FERIS) was introduced in 2014, offering financial rewards to local authorities who do reduce fraud and error in their housing benefit cases. Unsurprising considering it costs the UK £millions every year, taking money away from vital services.
Testing the effectiveness of nudges
Capita supported five partnering councils – the London Borough of Harrow, the London Borough of Southwark, Wycombe District Council, Vale of White Horse District Council and South Oxfordshire District Council – to apply to FERIS for funding.
Following the funding award, we designed a collaborative experiment with the councils to design a study which tested the effectiveness of small changes to housing benefit correspondence to residents.
The experiment introduced four tested ‘nudges’, or treatments. Each nudge was given to a group of 513 randomly selected recipients spread across all five councils. The experimental groups were:
- Treatment one: separate communication about reporting changes in benefit status
- Treatment two: separate communication but redrafted using simpler and more straightforward language about the legal responsibilities of recipients
- Treatment three: separate communication but redrafted according to ‘social norm’ principles, designed to persuade recipients
- Treatment four: combination of both nudges used on group two and group three
- Control: no communications received over research period.
Excluding the control group, the study resulted in 266 changes reported to the local authorities from 2,124 residents.
The most effective treatment was treatment one, which resulted in 15% of residents reporting changes in circumstance. Treatments two and four resulted in 13% of residents reporting changes in circumstance, while treatment three saw 9% of residents report changes. In the control group, which did not receive any additional communications, 7% of residents reported a change in circumstance.
Although the most effective treatment in terms of impact on the number of responses was treatment one, treatment two produced a significantly higher financial return to the councils (£19,000, rather than £8,000 for treatment one) over the course of research, as it generated many more reports of overpayments than underpayments. The recommendation in Capita’s report is that the councils introduce a dedicated letter to remind claimants to report changes in circumstances – but that simplification techniques are used to make sure the letter is as effective as possible.
Overall, the research resulted in over £52,000 in overpayments identified and over £11,000 in underpayment identified across all councils.
This has been a great example of effective local authority collaboration to find more effective ways of tackling fraud and error. Capita has provided end-to-end support throughout this project, including the design of the solution that enabled the successful award of FERIS funding right through to delivering the study with our teams. It has given us evidence-based guidance on how we can make enhancements to our housing benefit correspondence to support the reduction of housing benefit fraud and error.
Head of collections and housing benefits, London Borough of Harrow