Which of the company’s valued employees is dissatisfied and will quit soon? Which employee has potential and should be trained further, which is unmotivated and should be notified in the performance review that s*he has no future in the company if s*he does not change her*his attitude? For some, this represents a chance for greater satisfaction on the part of companies and employees, for others it is a dystopia of surveillance and control: companies rely on systems for algorithmically controlled, automated decision-making or preparation (algorithmic/automated decision-making, ADM) in human resources management. We want to find out which functionalities they offer to the users – here the companies – and to those affected – the employees -, what information the staff members have about the deployment of the systems and what right they should have to have a say in their implementation.
The introduction of automated data evaluation mechanisms, which assist in the recruitment, management and performance evaluation of employees, has led to imperceptible and lasting changes in the conventions regarding employment criteria, working hours and workload, and last but not least in the expectations of employees and employers in the labour market. ADM technologies are used to optimise production planning and controlling in the sense of “scientific business management”; at the same time, the logic of quantification is a mandatory prerequisite for their application. There are indications from other areas that algorithms consolidate existing distortions (e.g. sexist or other forms of discrimination) and thus make it more difficult to achieve the diversity desired by society. On the other hand, these technologies – if well done – could also make more comprehensible and objective decisions than individuals.
- Access to the data and the final analysis results can change the balance of power in a significant way, in different directions. Is pure “knowledge of power” produced for the management, or do employees have the possibility not only to get to know the evaluations concerning themselves, but also to understand or even verify the results using alternative systems?
- Who has access to and control over data that is collected – the company, the business that provides the ADM system, the workers’ council? Who should have it? How can a meaningful regulation of such systems be designed?
- Does the deployment of ADM in operational control lead to a fairness gain or loss for the employees?
- What does the deployment of ADM mean for the autonomy of employees and for co-determination?
In order to make robust statements about the impact of ADM on employee autonomy and co-determination, information must be gathered about the systems available to companies and how they are being used. On the basis of this exploration, tangible informational, legal and ethical analyses can be carried out in order to find out where the systems actually work, how their application itself and the results of their application can be legally classified, whether there are any regulatory gaps and how these results can be communicated to workers’ councils, but also to a wider public.
Project period: 01.01.2018 – 31.12.2019
Project lead: Matthias Spielkamp
funded by Hans Boeckler Stiftung