Analysis at the gate (for income tax returns)

Dutch Tax and Customs Administration

Every year, the Tax Administration receives more than 10 million income tax returns, which must be finalised within three years. Therefore, the Tax Administration uses an algorithm to detect returns containing possible errors.

Last change on 6th of May 2025, at 11:38 (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

2019

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/analyse-aan-de-poort-bij-aangiften-inkomstenbelasting/

Responsible use

Goal and impact

Every year, the Tax Administration receives more than 10 million income tax returns, which must be finalised within three years.

Therefore, the Tax Authority uses an algorithm to detect returns containing possible errors by citizens. This is done with relevant data already known to the Tax Authority and the data included in the return by the taxpayer.

The Tax and Customs Administration corrects these errors in income tax returns at an early stage in contact with citizens. As a result of this rectification, citizens have quicker clarity with a final income tax assessment that includes a correct aggregate income.

Considerations

For citizens, it is important that income tax returns are processed quickly and carefully. The algorithm can support the Tax Administration in this regard.

The algorithm contributes to the systematic and accurate detection of errors in income tax returns.

Mistakes made by citizens can have a major impact on an income tax assessment, resulting in payment and debt problems.

This algorithm has a service character by recovering possible mistakes made by citizens and preventing unnecessary payment and debt problems among them.

The algorithm determines the possible mistake based on the relevant data. A content expert checks this error. The alternative is for an employee to manually collect and review the relevant data for all returns. This would make the process more error-prone and less efficient.

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 involved in the operation of the algorithm. The algorithm determines the advice based on the relevant data. A content expert checks this advice.

Risk management

A risk with this algorithm is that a declaration may be wrongly flagged as possibly incorrect, causing a citizen to unfairly experience additional scrutiny or delay in processing.

There is also a risk that an error could go undetected, resulting in an incorrect assessment being imposed. This can lead to financial consequences for citizens, such as payment problems or subsequent corrections.

These risks are mitigated by:

• the use of carefully designed selection rules, based on laws, regulations and experience,
• the fact that any possible error is always checked by a content expert, and,
• regular evaluation and adjustment of the operation of the algorithm.

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. Payroll Tax Act 1964:
6. Income Tax Act 2001:
7. Citizen Service Number (General Provisions) Act:
8. Archives Act 1995:

Links to legal bases

• Algemene wet inzake rijksbelastingen: : https://wetten.overheid.nl/BWBR0002320/
• Algemene wet bestuursrecht: : https://wetten.overheid.nl/BWBR0005537/
• Algemene verordening gegevensbescherming: : https://eur-lex.europa.eu/legal-content/NL/TXT/HTML/?uri=CELEX:32016R0679
• Uitvoeringswet algemene verordening gegevensbescherming: : https://wetten.overheid.nl/BWBR0040940/
• Wet op de Loonbelasting 1964:: https://wetten.overheid.nl/BWBR0002471/
• Wet Inkomstenbelasting 2001:: https://wetten.overheid.nl/BWBR0011353/
• Wet algemene bepalingen burgerservicenummer:: https://wetten.overheid.nl/BWBR0022428/
• Archiefwet 1995: : https://wetten.overheid.nl/BWBR0007376/

Elaboration on impact assessments

The use of the data should be assessed against the AVG.

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

Operations

Data

1. Income tax return details
2. Company data
3. Third-party income
4. Wage, pension and benefit data
5. Personal data

Links to data sources

• Aangiftegegevens inkomstenbelasting: Belastingdienst
• Bedrijfsgegevens: Kamer van Koophandel (KvK)
• Inkomsten van derden: Opdrachtgevers of uitbetalers
• Loon-, pensioen en uitkeringsgegevens: Werkgevers, uitkeringsinstanties en pensioenuitvoerders
• Persoonsgegevens: Basisregistratie Personen (BRP)

Technical design

The algorithm consists of fixed selection rules drawn up by content experts based on laws and regulations and tax expertise.

As soon as an income tax return arrives, the algorithm compares the citizen's completed data with data already available at the Tax Administration. Based on this, it assesses whether there might be an error.

If a possible error is flagged, the algorithm generates a signal for further assessment. These signals are then assessed by a content expert. The algorithm itself does not apply corrections; it only assists in the detection of possible errors.

The algorithm is not self-learning and does not evolve during use. The rules are evaluated periodically and adjusted where necessary.

External provider

The algorithm was developed by Tax Administration staff and is also maintained internally.