Currently on the Federal, State, and local level, the government is seeking to constrain budget deficits while still providing their constituents with the same services. This morning in the Cannon building I learned how the creative use of public record data can save state governments money. I had the opportunity to listen to a discussion hosted by Representative Tom Price (GA-6) and Lexis Nexis about how the state of Georgia has started to use commercial data analytics services to prevent tax refund fraud and use the savings to help fix their budget problems.

Andy Bucholz, Lexis Nexis Risk Solutions and Doug MacGinnitie, Commissioner of the Georgia Department of Revenue led the discussion entitled Combating Income Tax Refund Fraud Using Preventative and Investigative Tools: Lessons Learned From Georgia. Doug started off by noting that he became more concerned about the issue a few years ago when his wife had her identity used fraudulently for a tax refund.

Identity theft for government payments is one of the largest and fastest growing types of crime in the country. In 2006 the state of Georgia blocked 32,987 fraudulent tax refund returns worth some $26,900,000. By last year those numbers had risen to 158,462 fraudulent returns and $98,700,000 in payouts by the state. This increase is related mainly to two things:

1. The online migration of government services such as tax returns and applying for unemployment. This allows people to apply multiple times and in multiple states from one place all with little actual risk of being caught.
2. Sending tax refunds out on pre paid debit/EBT cards or electronic payments. Once funds are withdrawn from these cards, they are hard to trace.

In any year the state of Georgia processes around 4 million tax returns, 3 million of which receive some type of refund. All of this is done in less than 4 months. This means it is hard to spend a lot of time looking at any individual tax return to assess its risk of being fraudulent. How could the state of Georgia reduce the tax fraud rate in these circumstances? On its own, it tried to create and use fraud detection rules based on multiple filings that used the same name, address, and Social Security Number. This helped to prevent some fraudulent tax refunds but not enough.

Starting in 2011 the state of Georgia began using a data analytics service from Lexis Nexis to help flag potential fraudulent tax refunds. Lexis Nexis used public record data from across the states to identify potential fraudulent returns by flagging returns where there was a sudden change in address or change in the number of dependents. Other information such as incarcerations in other states was used as red flag indicators as well.

These potentially fraudulent applicants for a tax refunds were automatically mailed a letter asking them to confirm their identity online. This was done by asking them a series of questions that only they would know such as which of these addresses did you live at 8 years ago? Or what type of car do you own currently? All of which Lexis Nexis is able to confirm through public records data. If they were unable to pass this identity check, that was a good sign that they were not who they pretended to be.

Last year Georgia spent just over $3,000,000 on this data analytics service while saving over $23,000,000 in fraudulent tax return refunds. Georgia believes they are ahead of most states in preventing tax return refund fraud and hope that being ahead of the curve will prevent future fraud. The fraudsters will go to other states because it is easier to commit the fraud there!

One of the concerns Georgia had initially when trying to solve this problem was that by cracking down on tax refund fraud they would drastically slow down processing the real returns. Real returns accounted for 96% of the returns filed in the state and a processing delay could adversely affect their citizens who depended on the returns to be able to cover day to day living expenses. But this fear did not materialize. On average Lexis Nexis was able to flag or confirm a tax return in only five extra minutes and without having to force the state to use additional employees to do it.

Georgia has gotten good at blocking tax refund fraud but that it is still hard to actually prosecute people for it. As a result, the fraudsters really have nothing to lose and the number of fraudulent claims will continue to increase in the future. Georgia will try to stay in front of the problem so that they can protect the state, individuals within the state, and the state’s tax payers and use the savings to continue to provide the government services their citizens expect. By continuing to use good data analytics services, the state is also reducing the incidence of identity theft – which takes 12-18 months to completely repair. As David LeDuc pointed out in a recent blog this is a major and growing problem.

The states of South Carolina, Louisiana, and Connecticut are also using Lexis Nexis to prevent tax refund fraud from happening in their states so that they can continue to provide for their constituents. The use of data analytics for this worthwhile purpose is only one example of how data can be used for the public good.

Denys Emmert is the Public Policy intern at SIIA. He has a degree in marketing and political science from Florida State University.