Presentation to the Monitoring Committee of the Parliamentary Assembly of the Council of Europe
Mister chairman, esteemed colleagues,
Let me start by expressing my appreciation for your interest in the childcare allowance case in the Netherlands. Substantial indications of systemic or institutional discrimination are the linking pin with the Senate’s parliamentary investigation committee on the effectiveness of anti-discrimination legislation.
To be clear: our report does not discuss the childcare allowance case directly. The question the Senate wanted to address is why our legislation seems unable to curb this wide spread and systemic discrimination. As you may know, the first article in our constitution explicitly bans discrimination on any ground. We also have specific laws against discrimination on more specific grounds. We have implemented a system for complaints and local institutions to address individual cases. In short, we have extensive policies against discrimination. And yet… discrimination not only persists despite our legislation and policies, but sometimes because of them. And it is highly detrimental to our citizens.
The committee looked specifically at discrimination in the domains of the labor market, education, social security, and police; four domains with a different degree of governmental influence. In each domain we selected specific issues in discrimination that would help us understand the dynamics so that we can improve the legislative process. In the domain of social security, we looked at two issues: first, the role of algorithms in detecting unlawful use of social support and second, the fact that certain groups tend to avoid the social security system, even if they are entitled to receive support.
The analysis of our investigation highlighted the discriminatory risk of algorithms, especially when prejudice and bias are incorporated in the risk profiles and data sets. Moreover, even relevant and / or seemingly neutral information can contribute to discriminatory use of profiles and data. A combination of postal codes, use of IP-addresses, and phone numbers for example can indicate ethnicity or nationality and thereby result in indirect discrimination.
With regard to the non-use of social security provisions, the complexity of the system and the fact that the government seems to mistrust those use are in need of support, were found to be important factors. This regards especially those with fewer social-economic resources and people with structural or temporary impaired capabilities. Although these criteria are hard to define in law, the outcome can be seen as discriminatory.
In a more general sense, the investigation resulted in six aspects that parliament needs to assess in the legislative process. These aspects can be formulated as questions that parliament should ask for every relevant bill.
First, trust. Does the government trust or mistrusts its citizens? The fundamental attitude should be that people by and large can be trusted and that in varying degrees they need support. If the government displays fundamental mistrust, this will likely result in discriminatory laws and policies.
Second, attention. Does the government display continuous attention for discriminatory processes and outcomes and does it listen specifically to what people need and experience? Lack of attention puts systems above people and easily results in discriminatory laws and policies.
Third, norms and language. Do new laws explicitly refer to antidiscrimination principles and make them concrete? And are implicit norms inclusive enough or do they favor certain groups? Vague and implicit norms can easily result in discriminatory laws and policies.
Fourth, simplicity. Do our laws and policies provide transparent, consistent, and integrated criteria and regulations to citizens and institutions, including educators and social services? The complexity of our laws and policies makes it difficult for citizens to claim the support they need, to execute their rights and to file complaints where needed. It also yields space for bias and prejudice and can therefore result in discriminatory laws and policies.
Fifth, leadership and accountability. Does the government explicitly make institutions and organizations responsible to curb discrimination and to arrange accountability structures? And do our policies provide for the necessary skills and professional space to use and account for discretionary power and hardship clauses? Failure to do so, especially in situations of unclear norms or conflicting political demands, may result in discriminatory laws and policies.
Sixth, clear and effective complaint procedures. Are the possibilities for citizens to complain about certain decisions clear, accessible, and effective? It is not enough to have procedures in place, if people cannot realistically use them. Moreover, this should not be the only safeguard because then only the well resourced citizens are able to use them which actually increases the risk of discriminatory laws and policies.
Trust, attention, norms, simplicity, leadership and accountability, and clear procedures. Obviously, these principles for parliamentary assessment of laws and policies are not a foolproof remedy for discrimination. They are, however, an important instrument in addressing the systemic and institutional dimension of discrimination. They clarify how our legislative processes willingly or unwillingly can result in discrimination, and they show what we can do to reduce that. Even more, they are just principles for good laws and good policies for all our citizens.
Thank you mister chairman.