Signal model OB Carousel Fraud (OBCF)

Dutch Tax and Customs Administration

This page contains information about the 'OB Carousel Fraud' algorithm. This algorithm helps the Inland Revenue find possible carousel fraud.

Last change on 8th of April 2025, at 9:18 (CET) | Publication Standard 1.0

• Publication category: Impactful algorithms
• Impact assessment: Field not filled in.
• Status: In use

General information

Theme

Public finance

Begin date

Field not filled in.

Contact information

[e-mail]

Link to publication website

https://over-ons.belastingdienst.nl/onderwerpen/omgaan-met-gegevens/algoritmeregister/

Link to source registration

https://over-ons.belastingdienst.nl/onderwerpen/omgaan-met-gegevens/algoritmeregister/signaalmodel-ob-carrouselfraude-obcf/

Responsible use

Goal and impact

Most businesses file sales tax (also known as VAT) returns. Sales tax returns are filed for a certain period of time, monthly, quarterly or annually. In some cases, the return is not filed, or is filed incorrectly in order to pay less tax or wrongly claim tax back. In such cases, fraud may be involved. A well-known form of VAT fraud is carousel fraud.

In carousel fraud, a trader does not pay VAT to the Tax Authorities, while this trader charges this VAT to his customers. The fraud crosses several EU countries via a complex multi-company supply chain. At least one company does not remit the VAT received. Because the fraud is so complexly set up, the tax authorities and well-intentioned traders are misled. The entrepreneur keeps the VAT received for himself, thus disadvantaging the government and society. Because the goods or services are often traded in a circle, this is called carousel fraud.

The Tax and Customs Administration wants to spot possible carousel fraud as early as possible so that timely action can be taken to prevent damage. To this end, the OBCF algorithm, among others, is used. The results of the algorithm are used by experienced staff as an aid to assess whether carousel fraud may be involved.

Incidentally, as in the above, the abuse does not always take place in a circle. Using the algorithm, other forms of Intra-Community VAT fraud can also be detected.

Considerations

Using the deployment of the OBCF algorithm, experienced staff can detect possible carousel networks at the earliest possible stage. This allows fraud to be prevented or mitigated as early as possible. OBCF assigns a score to companies based on certain data known to indicate an increased likelihood of possible involvement in carousel fraud. This allows the officer to conduct quicker and more targeted investigations and to spot possible carousel fraud more quickly.

Human intervention

Human intervention in the Tax Administration context implies that a competent and knowledgeable employee plays a substantial role in decision-making.

Human intervention is always involved in the operation of the algorithm. The algorithm assigns a score that is used by employees as an indication when making a treatment recommendation. A high score does not automatically lead to treatment. It is the Inland Revenue employee who makes the decision.

Risk management

The General Administrative Law Act (Awb) requires the government's actions to be transparent and lawful. The Tax and Customs Administration observes the general principles of good governance when applying and developing algorithms.

Conditions, a quality framework, have been drawn up by the Tax and Customs Administration for the development of algorithms. This contains rules and agreements that are followed during algorithm development. The conditions of the National Audit Authority are leading in this respect. At set moments, the Tax and Customs Administration checks whether the algorithm still meets the quality requirements.

The algorithm uses data collected under various tax laws. As required by the AVG, no more data is used than necessary.

The business rules are reviewed periodically and adjusted if necessary to remain compliant with laws and regulations.

Legal basis

• General Administrative Law Act (Awb)
• General State Taxes Act (Awr)
• Citizen Service Number (General Provisions) Act
• Turnover Tax Act 1968
• General data protection regulation (AVG)
• General Data Protection Regulation (Implementation) Act (UAVG)
• Archives Act 1995.
• Regulation (EU) No 904/2010 on administrative cooperation and combating fraud in the field of value added tax.

Links to legal bases

• Algemene wet bestuursrecht (Awb): https://wetten.overheid.nl/BWBR0005537
• Algemene wet inzake rijksbelastingen (Awr): https://wetten.overheid.nl/BWBR0002320/
• Wet algemene bepalingen burgerservicenummer: https://wetten.overheid.nl/BWBR0022428/
• Wet op de omzetbelasting 1968 : https://wetten.overheid.nl/BWBR0002629/
• Algemene verordening gegevensbescherming (AVG): https://eur-lex.europa.eu/legal-content/NL/TXT/HTML/?uri=CELEX:32016R0679
• Uitvoeringswet algemene verordening gegevensbescherming (UAVG): https://wetten.overheid.nl/BWBR0040940/
• Archiefwet 1995.: https://wetten.overheid.nl/BWBR0007376/
• Verordening (EU) nr. 904/2010 betreffende de administratieve samenwerking en de bestrijding van fraude op het gebied van de belasting over de toegevoegde waarde.: https://eur-lex.europa.eu/legal-content/NL/TXT/HTML/?uri=CELEX:02010R0904-20240101

Elaboration on impact assessments

The use of data should be assessed against the General Data Protection Regulation (AVG). Reviewing personal data reveals any privacy risks and allows appropriate measures to be taken.

The AVG prescribes that no more data should be used than necessary. This is called data minimisation. The Tax Authority regularly reviews whether the data used is still necessary and therefore may be used.

The algorithm is assessed in line with applicable non-discrimination principles for direct and indirect discrimination. Processing as little personal data as possible reduces the risk of direct discrimination. Employees involved in developing and managing the algorithms receive training on data protection and bias.

Operations

Data

• Identifying data entrepreneur (including BSN)
• Company details
• Turnover tax return details
• Input data
• Trade and company data in other EU Member States
• Payment data and collection data
• Vehicle data

Links to data sources

• Identificerende gegevens ondernemer (o.a. BSN): Basisregistratie Personen (BRP)
• Bedrijfsgegevens: KvK (Kamer van Koophandel)
• Aangiftegegevens Omzetbelasting: Belastingdienst
• Invoergegevens: Douane
• Handel- en bedrijfsgegevens in andere EU-lidstaten: Europese Belastingdiensten
• Betalingsgegevens en invorderingsgegevens: Belastingdienst
• Voertuiggegevens: RDW

Technical design

The algorithm uses business rules created by content experts based on laws, regulations and expertise. The algorithm assigns a score based on these business rules. The results of the algorithm are used by a group of experienced staff dedicated to investigating possible carousel fraud.

The algorithm is not self-learning. This means that it does not evolve while being used.

External provider

The algorithm was developed by staff at the Inland Revenue and is maintained internally.