Business Analytics Blog

Business Analytics Blog

Opinions expressed on this blog reflect the writer’s views and not the position of the Capgemini Group

How analytics is used to combat tax and welfare fraud by Alexa Morrison and Jon Chadwick

How analytics is used to combat tax and welfare fraud

With a cross-government budget cap and the recent announcement of further cuts [bbc] the UK Government continues to be challenged to maintain service levels with fewer resources.

Amid this austerity, the taxpaying public is sensitive to stories of both high value tax evasion and avoidance, [bbcguardian] and lower level benefits ‘cheats’ [hertsad, mirror]. So the Government tax and welfare bodies are under pressure to minimise revenue loss and effectively do more with less.

One of the tools open to the Government is Analytics, which can help to target the right cases with limited resources to improve the efficiency with which the Government can predict, prevent and respond to tax and welfare fraud and error.

  • Predict: by minimising revenue leakage through stronger predictive and post-processing capabilities. By monitoring transactions and building customer insight, you can reduce risk, improve compliance and put in place effective debt collection operations.
  • Prevent: by using analytics to identify possible fraud and errors before they happen through anomaly identification, ensuring procedure is followed, and influencing user behaviour to reduce non-compliant activity.
  • Respond: by initiating more cost-effective interventions based on risk profiling, and utilise insight from previous cases to make risk rules more effective.

 

Tax and welfare fraud and error costs UK taxpayers over £37bn per annum

The UK Government estimates that the tax gap was £34bn in 2012/13 (gov) and benefits overpayments were £3.2bn in 2014/15 (gov).  So, how can similar analytical methods be applied to fraud detection to target fraud interventions more effectively?

In a previous blog, Cops-and-robbers-in-the-digital-era, we talked about using social network analysis to predict, prevent and respond to fraud, so in this post, we look at some of the other ways analytics is being used to combat fraud and other crimes.

Predict: Predictive Policing

Police in the U.S. have responded to rising crime by improving their crime prediction algorithms so they can target where their police officers should focus their efforts. ‘Predictive policing’ identifies high crime rate areas – in 500-square foot “boxes” – through real-time algorithms:

  • Predictive algorithms have helped reduce crime in several US cities by >30% (bbc)
  • Visualisation of dynamic crime hotspots led to a day without crime on Feb. 13 in LAPD’s Foothill Division. (nbc)

 

Prevent: Fraud Detection as a Service

The US state of Iowa identified the need to improve their fraud prevention strategy to combat the increasing sophistication of fraudulent attacks on their unemployment benefit system. As the fraudsters had become more tech-savvy, so the government needed a more effective response.

Since Iowa did not have the analytics capability in-house, they decided the least cost, least risk and quickest option was to buy in fraud detection as a service (govtech).

The cloud-based analytics solution analyses benefits claims and alerts agencies to problems that require follow-up investigation. Because the analysis can take place upfront, at the point of a claim coming in, anomalies can be detected early to prevent pay-out of fraudulent benefits.

Respond: Using Real-Time Information

HMRC announced that tax credit fraud had fallen to an all-time low in 2013-14 (gov, ft). They attributed this in part to the introduction of ‘Real Time Information’ for PAYE income tax – where employers are now required to report salaries when they are paid, rather than annually. This means HMRC now have a richer and more up-to-date data set, from which to spot fraud and error more easily and earlier. The constant flow of data also reduces peaks in processing, so HMRC can operate with more efficiency.

The UK Chancellor recently announced the ambition to extend real-time tax systems, so all annual tax returns move to online tax accounts by 2020 (bbc). This will provide an even richer source of data to support fraud interventions.

It is clear that analytics is proving to be the most cost effective way to predict, prevent and respond to this huge challenge.

About the author

Jonathan Chadwick
Jonathan Chadwick
Jon has worked for 18 years as an analytical consultant in the UK, USA and Europe for a diverse range of sectors, most recently Financial, Oil & Gas and Government. Jon has extensive experience in benefits realisation, modelling, business analytics, portfolio management and change management. Jon devised and created Figure It Out.

Connect with me

Leave a comment

Your email address will not be published. Required fields are marked *.