Reducing fraud and error through nudge techniques


Following a successful application of FERIS funding, Capita has delivered a collaborative project with five councils in England to test the impact of using behavioural science, ‘nudge’ techniques on customer correspondence to help reduce fraud and error across housing benefit.

Increasing the number of housing benefit claimants that report a change in their circumstances is an important factor in reducing fraud and error in housing benefit payments.

The Fraud and Error Reduction Incentive Scheme (FERIS) offers financial rewards to local authorities who reduce fraud and error in their housing benefit cases. Capita Local Public Services supported five partnering councils, including London Boroughs of Harrow and Southwark, Wycombe, Vale of White Horse and South Oxfordshire District Councils to apply for FERIS funding.

Following the funding award, we designed a collaborative experiment which tested the effectiveness of small changes to housing benefit correspondence to assess the impact on the number of changes in circumstances being reported and the financial impact on overpayments and underpayments.

The experiment design used best practice principles and frameworks for behavioural science. Four separate nudges, using simplification and social norm principles were designed and applied to housing benefit correspondence. The total sample size of housing benefit customers was then split randomly and evenly across five treatment groups, including a control group (where no changes were applied), so we could measure the impact of the results.

Fast facts

  • Partnering councils: London Borough of Harrow, London Borough of Southwark, Wycombe District Council, South Oxfordshire District Council and Vale of White Horse District Council
  • £52,763.32 worth of overpayments identified
  • £11,596.36 worth of underpayments identified
  • Identified the most appropriate nudge treatments for maximising the number of changes in circumstances reported and financial adjustments
“This has been a great example of effective local authority collaboration to find more effective ways of tackling fraud an error, an issue for councils across the country. 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 experiment 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.”

Fernando Silverio, Head of Collections and Housing Benefits, London Borough of Harrow

The results A total of 305 changes were reported, representing 11% of the total sample size. The results show that all treatments did have a level of impact on the number of people reporting a change to their circumstances compared to the control group; however certain treatments were more successful.

For example:

  • Treatment one, where a standalone communication was introduced generated the highest number of reported changes (15% against the treatment sample size)
  • The control group, using local existing local practices, where no treatment was applied represented the lowest number of changes reported against the sample size at 7%
  • Out of all the treatments applied, treatment three, where social norm principles were used had the lowest impact on changes reported, representing 9% of the treatment sample size (but this was still greater than when no treatment was applied). However these results improved when a combined nudge was applied in treatment four. 

Impact on financial adjustments

The type of nudge used also had a significant impact on the financial adjustments required on the amount of housing benefit a customer receives. Although the number of changes reported by claimants was the highest in treatment one, the impact on financial adjustments was significantly greater in treatment two. This used clearer and more direct language around customers’ responsibility to report changes in circumstances and resulted in £19,000 worth of financial adjustments being identified – over £17,902.9 higher than the control group.


The experiment has demonstrated the importance of using more pro-active measures to encourage customers to report the changes in circumstances. Crucially it provides all five councils with a proven mechanism to ensure the right nudges are being applied to customer correspondence to generate the maximum impact on both the number of changes reported and the impact on financial adjustments. Based on the projections from the results, application of nudges techniques would improve the number of responses and lower the amount of over-payments being made. In turn this will support the reduction of fraud and error and enable the councils to minimise subsidy loss and improve caseload integrity.

“Using our experience in behavioural science techniques, we have been able to design an experiment that has given guidance to our partnering councils on how simple and cost effective changes can make a significant impact on reducing fraud and error across local authority housing benefit teams.”

John De La Rue, Service Development Director Capita