Selection module Actual Return on Investment (selection module OWR)

Dutch Tax and Customs Administration

Concerns an algorithm with selection rules that assesses whether Actual Return Returns (OWRs) submitted by taxpayers can be directly processed into an Income Tax assessment without the intervention of a Tax Administration staff member, or whether further assessment by a staff member is required. The algorithm will be used when processing OWRs for the years 2017 to 2024.

Last change on 19th of December 2025, at 12:51 (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

12-2025

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/

Responsible use

Goal and impact

The algorithm was developed to quickly, carefully and efficiently assess Actual Return Returns (OWRs) submitted by taxpayers and determine whether they can be processed immediately or require manual assessment by a Tax Administration employee. By deploying the algorithm, OWRs for the years 2017 to 2024 can be processed faster, providing citizens with faster clarity on their Income Tax assessment and making the assessment more systematic and less error-prone than full manual processing.

Considerations

When deploying the OWR selection module, consideration was given to whether the assessment of all Opgaaf Werkelijk Rendement (OWR) submissions would be entirely manual or supported by an algorithm with selection rules. The alternative, requiring Tax Administration staff to manually collect and assess the relevant data for all OWRs, would make the process less efficient and more error-prone.

The deployment of the algorithm supports a systematic and accurate assessment of OWRs over the years 2017 to 2024. As a result, OWRs can be processed faster and citizens get clarity faster. The algorithm determines the opinion based on the relevant data and supports Tax Administration employees to perform their work carefully.

In applying the algorithm, diligence, transparency and lawfulness are basic principles. The algorithm is not self-learning, does not use special personal data and does not process more data than necessary. The selection rules are based on laws, regulations and tax expertise and are periodically assessed and adjusted if necessary. In addition, provision is made for human intervention by an authorised and expert Tax Administration employee, especially when the processing may result in a deviation from the OWR.

Human intervention

The results of the algorithm are used to support the processing of the Statement of Actual Return (OWR). Based on selection rules, the algorithm assesses whether an OWR can be accepted and processed directly into an Income Tax assessment, or whether further assessment is required.

When the algorithm ejects an OWR, it is processed by a competent and expert Tax Administration employee. In doing so, the algorithm specifies the issues to which the clerk should pay attention. The clerk reviews the OWR and can check the outcome and, if necessary, adjust it. Human intervention occurs when the handling may lead to a deviation from the OWR.

Risk management

The algorithm is rule-based and not self-learning. The selection rules are drawn up by content experts and are periodically reviewed and adjusted if necessary to remain compliant with laws and regulations.

When deploying the algorithm, only data necessary for the purpose is used, in line with the principle of data minimisation. No special personal data are processed and the use of personal data is assessed against the General Data Protection Regulation (GDPR).

In addition, human intervention is ensured. OWRs ejected by the algorithm are reviewed by an authorised and expert Tax Administration employee, who can check the outcome and adjust it if necessary.

Legal basis

1. General State Tax Act:
2. General Administrative Law Act:
3. General Data Protection Regulation:
4. General Data Protection Regulation Implementation Act:
5. Income Tax Act 2001:
6. Legal Restoration/ Bridging Act box 3:
7. Box 3 Counter Evidence Act:
8. International (tax) treaties:
9. Citizen Service Number (General Provisions) Act:
10. Archives Act 1995:

Links to legal bases

• General State Tax Act:: https://wetten.overheid.nl/BWBR0002320/
• General Administrative Law Act:: https://wetten.overheid.nl/BWBR0005537/
• General Data Protection Regulation:: https://eur-lex.europa.eu/legal-content/NL/TXT/HTML/?uri=CELEX:32016R0679
• General Data Protection Regulation Implementation Act:: https://wetten.overheid.nl/BWBR0040940/
• Income Tax Act 2001:: https://wetten.overheid.nl/BWBR0011353/
• Legal Restoration Act/ Box 3 Bridging Act: https://zoek.officielebekendmakingen.nl/stb-2022-534.html
• Box 3 rebuttal scheme law: https://zoek.officielebekendmakingen.nl/stb-2025-195.html
• International (tax) treaties: https://www.rijksoverheid.nl/documenten/circulaires/2023/03/09/verdragen-op-het-gebied-van-directe-belastingen
• General provisions citizen service number act:: https://wetten.overheid.nl/BWBR0022428/
• Archives Act 1995:: Archiefwet 1995: https://wetten.overheid.nl/BWBR0007376/

Elaboration on impact assessments

A Data Protection Impact Assessment (DPIA) has been carried out on the business process within which the OWR selection module is deployed. The use of personal data within the algorithm was tested against the General Data Protection Regulation (AVG), identifying any privacy risks and taking appropriate measures. It was assessed that no more data is used than necessary (data minimisation).

In addition, the algorithm is assessed in line with the applicable principles of diligence, transparency, equality and non-discrimination, according to the principles involved in impact tests such as a DPIA and IAMA.

Operations

Data

1. Income tax return details
2. Personal data

Links to data sources

• Income tax return details: Belastingdienst
• Personal data: Belastingdienst en Basisregistratie Personen (BRP)

Technical design

The algorithm consists of a set of fixed selection rules drawn up by content experts based on laws and regulations and tax expertise. The OWR selection module processes the Opgaaf Werkelijk Rendement (OWR) submitted by taxpayers in combination with available income tax return data and personal data from the Tax Administration's systems.

Based on the selection rules, the algorithm assesses whether an OWR can be processed without uncertainties. If all relevant rules are met, the OWR can be processed directly into an Income Tax assessment or an objection can be dealt with in accordance with the OWR. If the selection rules identify an uncertainty or possible enforcement gap, the OWR is ejected for manual processing.

The algorithm is not self-learning and does not apply machine learning. The outcome consists of a recommendation to process the OWR automatically or submit it to a Tax Administration employee, indicating the issues that require further attention.

External provider

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