BookmarkSubscribeRSS Feed

Predictive analytics in the public sector

Started ‎10-15-2013 by
Modified ‎10-06-2015 by
Views 1,190

How the Revenue Irish Tax and Customs uses data mining for better audit selection
Have you ever been audited for taxes? I hope the answer is “no” but those of you who have appreciate the tremendous effort it takes – both for you personally and the government agency conducting the audit. It’s not an activity taken lightly. So shouldn’t audits be “hand-picked” based on taxpayer data?

 

 

Many tax authorities are turning to analytics to carefully select those to be audited based on a number of factors. And this is just one example of how analytics supports effective strategies for the public sector. Revenue, the Irish Tax and Customs Authority, has found that analytics can reduce costs, increase yields and improve its service to tax payers.

Duncan Cleary, Senior Statistician at Revenue, explains the agency’s use of data mining to better target audits in this technical paper. I had the chance to ask him a few more questions about the project as well.

I understand your analytics team uses both Revenue data and data from other sources to develop predictive models. What other sources do you draw from?
     Cleary: We are increasingly leveraging third party data sources to tackle risk. Examples include interest reporting on deposits from banks and financial institutions, payments made by government bodies to third parties and merchant acquirer data on payments to merchants by credit card/ debit card.  We keep under review what other information may be of use and seek it, including new legislation, as required.

 

Your paper focuses on how analytics is being applied from a compliance perspective, and it references a paper on analytics efforts with a customer (tax payer) service focus. Can you give a brief idea of Revenue’s analytics strategy improve tax payer service?
     Cleary: Tax authorities tend to use analytics for compliance purposes but of course there are multiple possible applications on the service side. We have used segmentation of our case base to better understand the profile of the taxpayer population and we have analyzed large scale taxpayer survey data with segmentation to better understand the results. We also look at online and other channel taxpayer contact behavior with a view to improving service and reducing overuse of more costly contact channels. Much of this work informs our strategies for improving services to our taxpayers.

What’s next for Revenue – what other parts of the agency plan to build out an analytical approach?
     Cleary: Up next and on-going is consolidation and expansion of our award winning Real Time Risk (RTR) framework, which rates risk cases and events to prevent fraud and error, expanding RTR to Customs risks, increasing the use of Social Network Analysis, increasing the use of Spatial Analytics/GIS, and using Hybrid approaches to tackling risk. There is no area of Revenue’s work that could not be assisted, in some way, by using the power of analytics.

Do you have other questions for Cleary on his work? Ask them by commenting on this article. Also note that Cleary will present at Analytics 2013, Oct. 21-22, in Orlando, FL.

Version history
Last update:
‎10-06-2015 03:08 PM
Updated by:
Contributors

SAS Innovate 2025: Register Now

Registration is now open for SAS Innovate 2025 , our biggest and most exciting global event of the year! Join us in Orlando, FL, May 6-9.
Sign up by Dec. 31 to get the 2024 rate of just $495.
Register now!

Free course: Data Literacy Essentials

Data Literacy is for all, even absolute beginners. Jump on board with this free e-learning  and boost your career prospects.

Get Started

Article Tags