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.
Taking pre-emptive action to stem the tide of VAT fraud losses
by Jérôme Bryssinck, Advisory Fraud Solutions Manager, SAS
Governments worldwide are growing increasingly concerned over the huge losses incurred by organized VAT fraud. According to a VAT Gap study published by the European Commission in 2016, EU countries alone lost an estimated €159.5 billion in VAT revenues due to non-compliance or non-collection during 2014, equivalent to 14.06 percent of total expected VAT revenue of 26 member states.
Organized VAT fraud drains money from national reserves; makes fair competition difficult and leads to restrictions on legitimate businesses. Government tax departments therefore urgently need to root it out and eliminate it.
Hybrid analytics technology will allow tax departments to focus on the most urgent organized VAT cases and networks.
Unfortunately, most have not had the tools to optimally address VAT fraud until recently. It’s a complex challenge. The majority of organized VAT frauds are planned and executed quickly. Early detection is therefore crucial. Yet, identifying this kind of fraud is difficult and requires not just an understanding of the linkages between the businesses involved, but also the analysis of millions of taxpayers, VAT returns and intra-community transactions.
Traditional methods of tackling organized VAT fraud demonstrate significant shortcomings. Skilled staff can be subject to attrition and government cuts, and in many countries are limited and continuing to decrease in number. Equally, manually gathering and preparing data for fraud detection is time-consuming, laborious and error-prone. Linked to this, the inability to perform sophisticated analysis of transactions in order to identify suspicious companies, relationships and behaviours makes authorities susceptible to organized crime, no matter how skilled their existing detection teams.
Driven by the lack of effective technology in place, some countries are creating new processes and regulations to specifically address the problem of VAT fraud. However, by adding extra controls, or asking bona fide traders for additional information, they end up creating greater operational burdens, adding costs to the economy and rendering trading more difficult.
An integrated approach
Organized VAT fraud is a fast-moving crime and therefore solutions need to detect it before significant damage is done. Governments need to take a hybrid analytics approach, enabling them to blend risk analysis and anomaly detection to quickly pinpoint suspicious cases.
They can run analysis to ascertain how much VAT money is at risk in each case and determine which cases to focus on. They can close the net on the criminals through a combination of anomaly detection to highlight changes in behavior and new types of fraud, as well as social network analysis to unveil hidden connections between firms.
Having these analytical techniques working in tandem not only enables investigators to raise detection levels, it also helps them control alert volumes and reduce false positive rates. Empowered with this information, they then have options in terms of next steps, including monitoring the activities of the suspected business; de-registering VAT numbers; or commencing a criminal investigation.
Hybrid analytics technology will allow tax departments to focus on the most urgent organized VAT cases and networks. Implementing this approach allows them to process cases faster and more accurately, and collaborate more closely with government agencies, thus detecting and preventing VAT fraud before it's too late.
Most government investigators are highly-skilled, but without the technology running the analytics upfront, they’re struggling to stay on top of the workload. In moving from a reactive to a more proactive way of operating, and thereby driving a more preventive approach, the ability to prioritize the highest risk cases is key– and hybrid analytics technology supports this way of working.
The approach also allows departments to increase their ROI per investigator, by tackling higher-value cases first and conducting more efficient, accurate investigations. Additionally, it enables them to reduce their ratio of false positives, leading to a hit rate as high as 80 percent.
Making effective use of this technology helps businesses shift the risk-reward ratio. The ability of hybrid analytics to detect potentially fraudulent transactions and identify more perpetrators will ultimately act as a deterrent, with the enticement of making large sums of money counteracted by the increased likelihood of being caught.
Rather than opting for inflexible black-box solutions that cannot be adapted or fine-tuned as required, it’s important that tax departments choose agile, flexible solutions that analysts can adapt and tailor as needed. They can use the solution for organized VAT fraud today, but tomorrow could potentially use it in other areas, such as tackling direct tax or opportunistic VAT fraud.
In other words, the same technology that enables businesses to detect and confront the scourge of organized VAT fraud will in the future allow government departments to tackle a much wider spectrum of nefarious activities across the whole fraud spectrum and help drive a broader preventative approach.