Improving audit efficiency
Estonian Tax and Customs Board uses SAS to detect and prevent tax fraud
Boasting the highest per-capita GDP of any of the former Soviet republics, Estonia achieved rapid economic success after asserting autonomy and turning to the open market. The tiny Baltic country's new way of doing business provided new opportunities for fraudsters looking to bilk the government. Thanks to a prowess for fraud detection, the Estonian Tax and Customs Board makes it hard for the tax cheats to win.
A favorable business climate requires effective government – and the Estonian Tax and Customs Board plays a leading role in making sure it stays that way. The board is charged with ensuring the receipt of budget revenue from state taxes and customs duties, fostering legal economic activities, carrying out national tax and customs policies, ensuring tax compliance, and enforcing customs regulations.
We're correctly analyzing the data to target fraud or mistakes in 80-90% of the filings that we audit. That makes our team much more productive.
Deputy Director General of Core Processes
Among fraudsters, activity primarily involves claiming unearned VAT rebates and circumventing income taxes through illegal "underground" salaries. According to Ivar Laur, Head of the Analysis Division, and Anneli Nappus, Head Analyst of the Tax and Custom Board's Intelligence Department, the board needed a cost-effective way to pursue suspicious activity while avoiding audits of law-abiding taxpayers.
"We knew that we were losing significant revenue," Laur says, "but the tools we were using – primarily Microsoft Access and Microsoft Excel – were not adequate for the analyses we needed."
"They were inflexible, slow, and error-prone," he adds. "The cheaters would find loopholes, and it would take us a while to fix those gaps. We needed to be able to change our risk models and features fast enough to keep pace with the cheaters. That's when we began to use SAS® Enterprise Guide."
More creative models
Every month, the board receives nearly 25,000 tax filings that require immediate evaluation using a dynamic set of criteria.
"Without SAS, we would have had to manage a constant cycle of software adaptation – which would have been expensive and time-consuming," says Nappus. "With SAS, we can access live data from multiple Oracle databases. We don't need to make data requests – we can explore the information and learn more about the data to develop more creative approaches and models to identify fraudulent patterns."
Faster, more accurate decisions
Nappus and Laur report that, by using SAS Enterprise Guide to build and evaluate fraud-detection models, Estonia is improving its ability to pinpoint fraudulent filings.
"Our decisions are much faster and much better," says Laur, "because they are based on data, not hunches."
Before SAS, the team had to rely on hunches and gut feelings. With SAS, they now take decisive action quickly and confidently. "We're identifying investigation targets and detecting risks as soon as they happen and initiating audits to get to the truth," Nappus says. "Previously, we could only look for basic irregularities on a smaller set of filings. Now, we can ask more sophisticated questions and analyze a much broader range of filings, so we have greater and more thorough coverage – and it's faster, too."
Precise audit selection
Previously, the audit team selected its investigation targets. Today, however, the analysis team is charged with finding fraud candidates. If the wrong cases are chosen – and the audited filings are legitimate – law-abiding taxpayers are needlessly bothered, the audit team is unproductive, and the fraudsters continue to elude authorities.
"If our team only has time to audit, say, 100 taxpayers, you want there to be a very high likelihood of fraud so that there's revenue recovery and a criminal is apprehended," Nappus says.
Greater revenue recovery
"SAS is our tool for determining our risks," says Egon Veermäe, Deputy Director General of Core Processes. "Before we implemented this system, we were spending too many resources on taxpayers who hadn't made mistakes. As many as 70-80 percent of our audits yielded no problems or recoveries. Today, with SAS among other tools, we're correctly analyzing the data to target fraud or mistakes in 80-90 percent of the filings that we audit. That makes our team much more productive and allows us to focus on more important things."
"When we compare what we did before vs. what we do now, we've noted significant time savings," Laur says. "Our team is able to review and analyze more data than ever – about five times as much as we did five years ago – more risks, more fields, more ability to discover more possibilities."
"This is much more advanced than what other European countries are using," Laur adds. "When a new risk type emerges, we have immediate access to the data and can instantly modify our models. We can see how big our risk exposure is and react accordingly. The universality and flexibility of SAS help us adapt in a fast-changing environment. Our declaration forms change every year, and it would be hard to keep up using other tools or systems."
Detect and prevent tax fraud with greater accuracy and efficiency
Before, 70-80% of audits yielded no evidence of fraud or error; now, 80-90% of cases do.
The results illustrated in this article are specific to the particular situations, business models, data input, and computing environments described herein. Each SAS customer’s experience is unique based on business and technical variables and all statements must be considered non-typical. Actual savings, results, and performance characteristics will vary depending on individual customer configurations and conditions. SAS does not guarantee or represent that every customer will achieve similar results. The only warranties for SAS products and services are those that are set forth in the express warranty statements in the written agreement for such products and services. Nothing herein should be construed as constituting an additional warranty. Customers have shared their successes with SAS as part of an agreed-upon contractual exchange or project success summarization following a successful implementation of SAS software. Brand and product names are trademarks of their respective companies.