Want more Insights from SAS? Subscribe to our Insights newsletter. Or check back often to get more insights on the topics you care about, including analytics, big data, data management, marketing, and risk & fraud.
Small-time cheats and organized crime: Benefits fraud re-examined
Are benefit agencies missing the threat from organized crime by wasting time on small-time cheats?
Losses from benefits fraud continues to grow. In the UK alone, preliminary government estimates indicate that £1.1 billion was lost to benefits fraud in FY2014-15. It’s an experience mirrored globally – and the scale of losses underlines the urgency for authorities to tackle the issue.
Today, agencies typically concentrate on small-time benefit cheats, many of whom are inexperienced and relatively easy to catch using conventional low-tech methods. However, because of the low return on investment in time and resources, many agencies struggle to cover their administration costs with this approach to benefits fraud. Even successful investigations can fail to return revenue back to their departments. In fact, it could be said that if they don’t update their detection strategies to tackle the larger organized frauds, welfare departments may be financially better off if they investigated no fraud at all!
Not only is the scale of benefits fraud typically much greater, but gang members are often involved in other nefarious activities, adding to the net yield the authorities could potentially bring in.
The rewards of shifting their focus to organized gangs could be significant. Not only is the scale of benefits fraud typically much greater, but gang members are often involved in other nefarious activities, adding to the net yield the authorities could potentially bring in.
Unfortunately, however, the problem of organized benefits fraud is likely to get worse before it gets better. Progressive schemes, such as the UK’s Universal Credit and Working Tax Credits, which switch the focus from unemployment benefits to top-up benefits for the low-paid, are making it harder for authorities to determine whether someone is legitimately entitled to aid. Such schemes not only enable fraudsters to cover their tracks, they also incentivize unscrupulous employers to cash-in on their employees’ top-up status.
On balance, these new welfare arrangements are a positive. Indeed, schemes like Universal Credit, designed to release people from the benefits trap, could significantly help the poor and vulnerable. However, there is a shortage of new counter-fraud strategies to match the new threats these systems create and, where such approaches have been implemented, they have often been misdirected.
Approaches to combatting benefits fraud
In an attempt to prevent benefits fraud, many agencies have focused on increasing the complexity of the application processes of these new schemes. Ultimately though, a complicated application process won’t reduce the need for post-application fraud detection investment. A large-scale operation will easily navigate through a complicated application process, while small-scale fraud doesn’t usually occur during the application process, but further down the line as claimants’ changing circumstances go unreported.
To tackle the problem, the key for benefits agencies is to balance a centralized analysis function that creates a holistic view of activity, with local teams that act upon the information and use their judgment on individual cases, feeding back suggested improvements.
Through the use of analytics tools, the central team would then concentrate on collecting relevant benefits data and running analysis on it, using network link analysis to unveil connections between individuals, businesses and activities and identify linkages among seemingly unrelated claims, enabling it to pinpoint corrupt businesses working together to defraud benefit agencies.
Peer group analysis can then help identify businesses behaving differently from similar organizations, perhaps in terms of the volumes of transactions conducted or revenues generated compared to employee count. It is important that local investigation teams focus on yield and prioritize investigations that will deliver the greatest financial returns.
Data sharing with other government departments can further enhance results. Historically though, agencies have been hampered by a lack of information sharing between departments due to political issues, and to concerns about privacy and security. However, the work carried out by a benefits department’s central analysis team can counter these reservations. Organized gangs are likely to be involved in additional offenses that may fall under the remit of other agencies. If the analysis team unveils evidence upfront, it will be easier to persuade other departments to get involved.
Whatever the precise make-up of the team, once the data gathering and analysis is complete, the focus will switch to generating high-quality packets of evidence to pass on to local investigative teams to pursue and close benefits fraud cases.
Complementing investigatory expertise
Analytics technology shouldn’t be viewed in isolation, but as complementary to the knowledge and expertise of local investigators. For example, while initial analysis may highlight an influx of staff at a previously quiet agricultural business as a potential risk, an experienced local investigator would know that it is hop-picking season across the area, helping explain the influx.
This combination of a data-driven analytics approach, coupled with local expertise, can be used to improve benefits fraud targeting while helping green-light genuine claimants, thus narrowing the field of investigation and enabling agencies to focus on the higher-risk cases.
Through this approach, agencies can rapidly gauge which cases need to be prioritized and processed through full investigations, thus strengthening the opportunity of early detection and prevention of organized, large-scale benefits fraud. Ultimately, everyone benefits … except the fraudsters.