Algorithm Policy differences

Dutch Tax and Customs Administration

This page contains information about the 'Policy Differences' algorithm. The algorithm supports Tax Administration staff in monitoring differences between the declared tax wage in employer data and employee data in the policy records. The Policy Differences algorithm helps Tax Administration employees respond to possible differences.

Last change on 2nd of July 2025, at 11:04 (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/algoritme-polisverschillen/

Responsible use

Goal and impact

The Tax Authority checks withholding agents for correct and complete declaration and payment of payroll taxes. Payroll taxes consist of wage tax, national insurance contributions, employee insurance and the income-related Healthcare Insurance Act contribution. Withholding agents (such as an employer, pension fund or benefits agency) deduct the wage tax and national insurance contributions from their employee's pay and, in some cases, the income-related healthcare insurance contribution. They remit this to the tax authorities. The employee insurance contributions and usually the income-dependent healthcare insurance contribution are not deducted from the employee's salary, but are paid by the employer to the Tax Administration.

A withholding agent must periodically submit a wage tax return to the Tax and Customs Administration. In the Tax and Customs Administration's process, the messages of these wage tax returns are accepted and processed in a mass automated manner.

The payroll tax return consists of a collective part (employer data) and nominative part (employee data). The Tax and Customs Administration stores the collective part with employer data of withholding agents in its records. The UWV puts the nominative part with employee data in its records. Some data of the wage declaration message are included in both the collective and the nominative part, such as the taxable wage. Of such data, the total of all nominative entries (employees) should be equal to the collective entry.

A wage declaration message may consist of an initial (complete) declaration or another type of declaration (correction messages, supplementary declarations). If there are errors in these declarations, this may lead to a difference between the employer data and employee data.

Considerations

Periodically, the Tax and Customs Administration submits the collective wage sums with employer data for each withholding agent per period to the UWV. The UWV then compares the wage sums from the employer data per period with the total wage sums from the employee data per period.

It then delivers it to have the differences corrected by the withholding agent so that the correct amounts of wages, payroll taxes and the correct data are recorded. This contributes to the use of the same data between Tax Authorities and the UWV. Among other things, this is important for the pre-completed income tax return.

A general description of this the operation of the Wage Return Chain is published on the "Wage Return Chain" site. (https://www.loonaangifteketen.nl/)

Based on the clues from the algorithm, the handler can assess the signal. It is the Tax Administration's handler who ultimately determines whether the wage return contains inaccuracies and can contact the withholding agent.

Assessing signals are important for the withholding agent to achieve correct recording of data from the payroll returns. The algorithm can support an employee of the Tax Administration in doing so. Based on the differences indicated by the algorithm, the assessment is therefore more careful, efficient and uniform.

Assessing signals are important for the withholding agent to arrive at a correct registration of data from the wage declarations.

Human intervention

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

The operation of the algorithm always involves human intervention. The algorithm detects and selects. It is the Tax Administration employee who makes the decision whether to contact the withholding agent.

Risk management

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

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

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

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. Social Insurance (Funding) Act:
7. General provisions Citizen Service Number Act:
8. Archives Act 1995:
9. Structure of Implementation and Income Act:

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 financiering sociale verzekeringen:: https://wetten.overheid.nl/BWBR0017745/
• Wet algemene bepalingen Burgerservicenummer:: https://wetten.overheid.nl/BWBR0022428/
• Archiefwet 1995: : https://wetten.overheid.nl/BWBR0007376/
• Wet Structuur Uitvoeringsorganisatie en inkomen:: https://wetten.overheid.nl/BWBR0013060/2020-04-01

Elaboration on impact assessments

• Privacy and AVG

The use of data is tested 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 Administration regularly examines whether the data used are still necessary and may therefore be used.

• Use of special personal data

The Policy Differences algorithm does not use special personal data.

• Equality and non-discrimination

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 (LH number)
• Payroll tax return details

Links to data sources

• Identificerende gegevens (LH-nummer): Belastingdienst
• Aangiftegegevens Loonheffingen: Belastingdienst

Technical design

The algorithm consists of selection rules created by content experts based on laws, regulations and expertise.

The algorithm is not self-learning. That means it does not evolve as it is used.

External provider

The algorithm was developed by staff from the Tax Office and UWV and is also maintained by them.