Customer Success Stories
Federal Public Service Finance uses SAS analytics to master huge fiscal database
Analysing tax returns, verifying import and export customs declarations, consolidating and checking the data from dozens of local tax collector's offices - those are just a few of the jobs that the Belgian Federal Public Service (FPS) Finance has entrusted to SAS software for reasons of speed, efficiency and flawless execution.
fotograaf: Jan Locus
As you can imagine, the FPS Finance is a government institution that collects huge amounts of data from a variety of sources. For a start there's the Collection and Recovery Department of the FPS Finance that collects money from taxpayers and checks on late payers. It also recovers funds due from non-paying individuals and companies, using all possible legal means if necessary To give you an idea of the magnitude: every year, the department handles information related to over three million income tax returns, which results in one of the largest databases in Belgium. The department also manages the analysis of real estate for patrimonial documentation, the analysis of all tax returns from legal bodies as well as private citizens and the VAT, and the risk evaluation of import and export customs declarations for merchandise.
A perfect fit
"For the past few years, we have been keen users of SAS Enterprise Guide and SAS Enterprise Miner, mainly for risk management in the various fiscal domains", says Dierk Op 't Eynde, Coordinator Data Mining at FPS Finance. "The reason is very simple: they are excellent solutions for data analysis and data mining, two crucial missions for our administration considering the size of all the fiscal data to be analysed, the fast-changing Belgian legislation and European directives. And the nice thing about it all is that the SAS tools fit perfectly with the IT standards of the FPS Finance".
It used to be quite different altogether. Efficiently managing and exploiting vast databases of information is a challenge for any system, and processing times used to be quite lengthy which wasn't exactly conducive to higher efficiency or service quality. That's why in 2010 several specialised services of the FPS Finance switched to a SAS environment for data mining, risk analysis and performance management. The ICT department created a business analytics environment in which a data warehouse was directly linked to a risk analysis environment, used by business analysts from several departments for a variety of operations, ranging from data mining over risk analysis to ad hoc inquiries. "Currently we have some 45 users supported by two administrators", Dierk Op 't Eynde explains. "Recentely, we engaged 20 (!) new, highly skilled dataminers in view of the high expectations from our policy makers regarding the handling of fiscal fraud. Almost all of them succeeded the SAS exam Certficatied Predictive Modeler ".
Fighting fraud and more
Using SAS software to combat fiscal fraud is a major undertaking, says Dierk. "It is our duty as FPS Finance to guarantee a correct and fair tax collection, and that means that every taxpayer should pay what he or she is legally due, neither more nor less. Fighting fiscal fraud is a major focus here. But apart from handling fraud detection and tax collection those analyses lend themselves perfectly for improving the Customer Relationship Management of the FPS Finance with its citizens and for creating statistics, forecasts and simulations regarding all fiscal earnings. So our use of SAS solutions goes far beyond the fair collection of taxes, a more efficient deployment of tax inspectors or policy supporting information for audits. Specific analyses will enable us to recognize our stakeholders and enhance our service to citizens and companies. Our strategy is evolving towards a more compliant and shared approach that reflects the various risks and is primarily aimed at preventing abuses".
Apart from monitoring the tax collection, risk analyses are carried out on a regular basis to identify which individuals or companies represent a risk of not meeting their fiscal liabilities, and which are likely to become insolvent or bankrupt. Here, data mining is put to good use to spot various forms of fraud or check on figures that might help predict bankruptcies at an earlier stage. Whereas previously it took the analysts up to half a day to produce a scoring list, they can now obtain this in a few minutes, simply because the SAS tools are better at handling huge volumes of data.
One of the things that Dierk Op 't Eynde appreciates very much is the fact that the huge range of analytical options offered by SAS is accessible via user friendly wizards. "on the one hand it means there's a fairly short learning curve for new users, while on the other more experienced users enjoy the added value of taking the next step to proper development. Of course strictly within the well-defined standards for nomenclature, documentation, and version control".
The same strict rules and regulations apply to data quality. "Let's not fool ourselves", Dierk insists: "no single administrative databank is perfect on quality. That's why, in order to get as close to perfection as possible, we have initiated two strategic programmes. On the one hand there are major efforts to modernise operational applications and to make historical versions of the data available for exploitation by building a data warehouse, while on the other hand we are decidedly making great efforts to make development, databases and infrastructure more uniform and standardised".
Back in 2010, the SAS implementation happened at a time when the federal authorities were raising the bar on ICT security - databases, for example, had to be more centralised and the access to certain files became more restricted. Privacy rules and legislation were also reinforced, and as result of all those changes a complete review of the operational methods was needed. Tasks and access rights were allocated in a different way and the organisation of the services had to be modified as well. Luckily SAS allows user access rules and profiles to be clearly and easily defined. Dierk Op 't Eynde: "It is obvious that every data stream and every analysis has to comply with the law for safeguarding privacy. Within the FPS Finance we have developed the necessary procedures and technical measures to meet those requirements: a privacy commission, a coordinating consulting body on data mining, a detailed RACI Matrix for the analysis of risk management, an independent "third party" for Coding/Decoding, hiding column identification data - often the name and the address of the tax payer, logging the data mining environment, and so on".
When asked for a typical example of a SAS implementation at FPS Finance, Dierk refers to the estimation of the venal value of a property using a multiple regression model. "This involves combining various indicators such as geographical, constructional and economic variables, resulting in realistic price brackets which in principle can be calculated for any property, in its current state as well as for its past value". In addition to all the very specific implementations of SAS in the fiscal context, the software also comes in very handy when the department has to answer ad hoc questions from members of Parliament - being able to provide fast and precise information is of the utmost importance then!
And last but no least, our organisation bought the SAS products Social Network Analysis, Dataflux and Visual Analytics… The best is yet to come.
Copyright © SAS Institute Inc. All Rights Reserved.
Federal Public Service Finance
Efficiently manage and exploit vast databases of information to guarantee fair and correct tax collection, prevent abuses, and improve services to citizens and companies.
SAS® Enterprise Miner
SAS® Enterprise Guide