Big Data, Analytics, and Tax Fraud by D. José Borja Tomé at Big Data Spain 2017

For 25 years the Spanish Tax Agency has intensively used data in order to help taxpayers comply with their obligations and combat fraud. The speech will discuss the way information is managed at the agency, and how some challenges have been faced by the use of big data technologies and advanced analytics.

https://www.bigdataspain.org/2017/talk/big-data-and-tax-fraud

Big Data Spain 2017
16th - 17th November Kinépolis Madrid

In order to illustrate the use of big data, projects such as the analysis of invoicing for the control of VAT, or the study of relations between all legal and natural persons in order to unveil hidden structures, will be presented.

Transcript

2. Big data, analytics, and tax fraud Big Data Spain 16 November 2017, Madrid José Borja Tomé Subdirector General de Tecnologías de Análisis de la Información e Investigación del Fraude
3. Analytic tools at AEAT # Users Analytic capabilities Aggregated analysis Simple network Analysis Advanced models Teseo Market tools Zújar (DW) Search engines Standard reports Individual query # Users
4. Organization + Governance Governance Of Information Completeness Semantic Quality Security Integration IT Analytic project leaders Model factory Business Analytic project leaders IT operational Apps staff
5. Departamento de Informática Tributaria LIBROS REGISTRO FACTURAS Se informa sobre ciertos datos de las facturas (NO SE TRANSMITEN LAS FACTURAS) LIBROS REGISTRO S.I.I. Se informa sobre ciertos datos con trascendencia tributaria (DESAPARECE LA OBLIGACIÓN DE PRESENTAR EL MODELO 340 Y 347) INVOICES INVOICE RECORD BOOK RECORD BOOKS Certain data about invoices are transmitted (NOT the invoices themselves) Certain tax relevant data are transmitted (models 340 and 347 are no longer required)  HELP TAXPAYERS: 1st phase: Tax data + crosschecks 2nd phase: Help in filling the VAT tax return  IMPROVE CONTROL: New control model with immediate and complete information.
6. Departamento de Informática Tributaria Graph Analysis. Participation in companies. Results: wealth calculation Participation Value Value of part. 6,62% 20.000.000 € 1.323.933,92 € 11,03% 10.000 € 1.103,46 € 11,82% 10.000 € 1.182,32 € 53,77% 10.000 € 5.376,66 € 71,92% 10.000 € 7.191,90 € 90,00% 10.000 € 9.000,00 € Total 1.323.933,92 €
7. Calculation for ALL individuals and companies at “n” levels of depth Aggregation of indirect participation through multiple paths Combined with other property such as real estate and financial assets Time evolution Family aggregation Results: wealth calculation • Hidden assets • True ownership • Figureheads • Screen companies • Fraud netwoks • Unexplained wealth increases Fraud Collection
8. Departamento de Informática Tributaria Advanced Analytics at AEAT
9. Departamento de Informática Tributaria Study and Reports on the VAT Gap in the EU-28 Member States: 2017 Final Report CASE – Center for Social and Economic Research (Project leader) Institute for Advanced Studies (Consortium leader) In consortium with CPB IFS DIW IPP DONDENA PWC ETLA ISER IEB Warsaw, 18 September 2017
10. Departamento de Informática Tributaria Tax Policies in the EU 2016 Survey (European Commission)
11. Departamento de Informática Tributaria What helps us thinking different? • Information (access for everybody needs governance) • People: smart, passion, confident, expose business to IT+ analytics and vice versa, diversity (multiple approaches, different minds) • Management: set goals, seek agility (bimodal), give trust (security, failure) Most needs can be satisfied with less sophisticated technologies Some lessons
12. Big data, analytics, and tax fraud Big Data Spain 16 November 2017, Madrid José Borja Tomé Subdirector General de Tecnologías de Análisis de la Información e Investigación del Fraude