Price announcement algorithm

Consumer and Market Authority

This algorithm helps the ACM find price announcements in messages that might say something about tacit or explicit alignment. This information by itself does not trigger action, but it can be a reason for inspectors to further investigate a possible infringement.

Last change on 4th of October 2024, at 13:36 (CET) | Publication Standard 1.0

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

General information

Theme

Economy

Begin date

2022-6

Contact information

[e-mail]

Responsible use

Goal and impact

This algorithm is designed to counteract tacit or explicit alignment. Tacit coordination means that firms organise behaviour without agreement or contact by looking at each other and doing the same thing. Explicit coordination, on the contrary, involves clear and direct agreements together on, for example, price fixing.

The algorithm can help find statements in messages used to regulate behaviour between companies. For example, a message in which an ice cream vendor complains about higher energy costs and indicates that he feels forced to pass them on from the start of the summer holidays. Mutual organising behaviour can be visible in messages about price announcements and can cause problems because it reduces competition. As a result, customers have to pay higher prices and have less choice. By detecting and addressing abuses in competition, companies compete more fairly, creating more choice for customers and better prices.

If the algorithm gives information that there might be collusion, the ACM will investigate further. If the suspicions are correct, the ACM can take action. This could be by providing additional explanations, issuing a warning, issuing a binding instruction, imposing an order under penalty, or imposing a fine. With this algorithm, the ACM contributes to a healthy economy by making markets work well for all people and companies, now and in the future.

Considerations

It is a lot of work to read one by one all messages that might contain price announcements. With this algorithm, the ACM can better and more quickly find information that may indicate alignment.

Human intervention

The information provided by this algorithm is assessed by inspectors and further investigated if necessary. So there is always human intervention because someone checks the information. The information provided by the algorithm does not in itself lead to action being taken, but may be a reason for the ACM to have inspectors further investigate a possible infringement of the Competition Act.

Risk management

The information provided by this algorithm is one of the many sources the ACM uses to monitor. Inspectors always check that the information is relevant and reliable.

Operations

Data

Online resources

Technical design

The price announcement algorithm includes a model that searches messages to determine whether they contain a price announcement. This model has been trained once so that it can recognise new links indicating price announcements based on that training itself. The training was conducted based on sentences that supervisory staff identified as price announcements. The model then finds out in what way parts of that sentence relate to price announcements. The model then predicts a value between 0 and 1: a score of 1 means that the message is likely to contain a price announcement, while, for example, a score of 0.1 indicates that there is no price announcement in the message. Based on a (user-defined) cut-off value, messages are displayed for further investigation by inspectors.