Szabolcs Schmidt is the Head of Unit for the Non-discrimination and Roma coordination at the Directorate-General of Justice of the European Commission. Rachel Waerniers is a policy officer at Unia, the Belgian equality body. TvMR sat down with both of them to talk about data on equality and non-discrimination: what they are, what we need them for and all the work that remains to be done.
Could you both briefly explain your daily work on equality?
Mr. Schmidt I am the Head of Unit responsible for non-discrimination and Roma coordination within the European Commission. The European Union has issued directives combating discrimination in many facets of life on the basis of so-called discrimination grounds: sexual orientation, age, disability, ethnic origin, racial origin, religion and gender.
Ms. Waerniers I work for Unia, the Belgian equality body, which works to fight discrimination and combat inequality on the grounds that are covered in the Belgian non-discrimination acts (Belgium has transposed the European directives broadly, for instance also protecting discrimination grounds such as social origin, health, wealth and civil status, Ed.), except gender, for which Belgium has a separate institution. I work within Unia’s policy service and have been working on an equality data research project for almost two years.
The European Handbook on Equality Data defines equality data as quantitative or qualitative information that is useful for the purposes of describing and analysing the state of (in)equality, its cause or its effects in society. Within the EU’s competence on equality and non-discrimination, which policy and rule-making work has been happening on equality data at the European level?
Mr. Schmidt If you look at the directives themselves, you wouldn’t necessarily find anything about ‘equality data’, but the directives establish the requirement for every EU member state to set up an equality body. These equality bodies assist victims of individual cases of discrimination, but they also make independent surveys and give independent recommendations to their authorities. To do that, they need data. We know from plenty of individual cases that the core European value of equality is not yet achieved, but individual cases alone don’t provide a good overview of structural problems. We generally lack information about the overall equality situation of certain minority groups: to which extent does a certain ethnic group or a sexual minority have problems accessing the labour market or is it being denied certain services?
That’s why the Commission started developing policies to promote the gathering of equality data within the member states. We invited the Fundamental Rights Agency (FRA), Eurostat and representatives of the member states to exchange their needs when it comes to equality data, under the auspices of the Commission’s High Level Group on Non-discrimination, Equality and Diversity. The topic of the collection of equality data is a good example of rules coming into existence in a soft manner. In the end, indeed, a set of guidelines was commonly agreed upon and officialised in the 2016 Handbook you mentioned. Now all member states can use this methodology. We continue to organise best practice exchanges between member states. Actually, nothing really new had to be created. There have always been plenty of data, they were just not being used in the field of equality.
In Unia’s 2021 report on its project called ‘Improving equality data collection in Belgium’, there were quite a lot of references to these EU guidelines. Can you tell us about Unia’s project?
Ms. Waerniers Unia has been producing equality data for a long time. For instance, since 2013, we publish the Socio-economic Monitoring with the Federal Public Service Employment, Labour and Social Dialogue (a report mapping out the situation of persons on the labour market according to their national origin and migration background, based on administrative data from the Crossroads Bank for Social Security.). Unia also commissions (academic) studies on specific topics like housing, the labour market and education, so-called ‘Diversity Barometers’. For the current Equality data project, Unia collaborated with the Equal Opportunities Team of the Federal Public Service for Justice in Belgium and got funding from the EU Commission’s Rights, Equality and Citizenship Programme to get a broad view on where we stand with equality data in Belgium and how to improve the situation.
What were the goals Unia set out to reach with this project?
Ms. Waerniers The project had three objectives. First, we needed an overview of the equality data we already had in Belgium and those that were lacking. We mapped all existing data sources for the “racial” discrimination grounds, philosophical or religious conviction and beliefs and then sexual orientation, gender identity, gender expression and sex characteristics. For the gender data, we were in touch with the Institute for Equality between Women and Men. We based this mapping on the EU guidelines and the mapping tool of the Fundamental Rights Agency (FRA) and adapted it to the Belgian context. The second step was to analyze this mapping. We did a survey with different stakeholders in the Belgian equality data landscape, like civil society organizations, statistical institutions etc. and did interviews with experts on equality data. Based on all this, we wrote reports with recommendations to improve the collection and processing of equality data. Finally, the third objective was to make the existing equality data in Belgium accessible to a broader public so they can be used. We centralized the data in a transparent and searchable online data hub, available to everyone.
What kind of documents am I going to find in this online data hub?
Ms. Waerniers The sources are very diverse. When we started, we didn’t realize the amount of diverse data that already exist, from administrative data sources and national statistics or reports that are based on administrative data to reports of civil society organisations, but also academic research in cooperation with public institutions or civil society organisations. For example, we found that when we look at the so-called racial criteria in Belgium, we have quite a lot of data because there is a lot of administrative information based on a person’s or their parents’ nationality.
On the one hand, EU member states are gathering data using supranational guidelines which were specifically made for that. On the other hand, that same supranational level has issued the GDPR, an important regulation on the protection of personal data. How do these two elements combined allow for the collection of equality data?
Mr. Schmidt The rules are not contradictory. All equality data have to be gathered and processed under strict observance of the respective data protection rules, which, for instance, exclude any type of processing of personal data beyond what is needed to aggregate the data. Sometimes, people are reluctant to provide or gather data. Because these are sensitive data, they think that it is better not to touch on them, but then we would miss out on being able to understand the (in)equality situation.
Ms. Waerniers There is indeed a lot of misconception around the legal framework on personal data and GDPR. A lot is possible within the GDPR framework. That’s the challenge: getting people to understand that this framework protects sensitive data and simultaneously enables the creation of equality data on a more collective and anonymised level. However, through our participatory approach and through the involvement of stakeholders in the project, we helped create awareness right away on the importance of equality data and the possibilities of gathering them within the protective legal framework.
Mr. Schmidt It’s important to work together and build trust with those groups who are particularly affected by inequality, so they know that the collection of data is in their very interest. For instance, a survey could show that in Belgium 60% of the Roma are leaving school early, compared to a much lower percentage in the population in general. It’s easy then to understand why it’s important to tackle that problem in a targeted manner and not enact a general obligation to attend school, for instance. Without the data, the authorities wouldn’t be able to identify the problem correctly.
Ideally, people identify as belonging to a certain group. Take the example above: Roma people would ideally indicate in a survey that they are indeed Roma. Why is self-identification such an important part of qualitative equality data?
Mr. Schmidt Self-identification with certain characteristics or as part of a certain group is generally seen as the soundest way to build equality data, since the identification is not imposed by anyone else. A person is the most credible source of information about themselves.
Ms. Waerniers Basing equality data on self-identification as much as possible is also one of the guidelines in the United Nations’ human rights based approach to data.
Mr. Schmidt However, there are risks. If a person is in an environment that is not very accepting, there is a tendency that they will prefer not to self-identify with some characteristics, not even in an anonymous survey. In some member states, for instance, Roma people easily self-identify as belonging to this ethnic group. In others, the Roma insist that they are citizen of a country and are not part of the Roma ethnic group. There is progress to be made in removing stigma because of belonging to a certain group of people.
Collecting equality data is one thing, using them is the next step. Could you elaborate on what equality data can be used for?
Ms. Waerniers It’s indeed important to emphasize that equality data are not a goal in itself, just a means to an end. We need a precise picture of reality in order to identify what the problem is, formulate new policies and then monitor and evaluate them. Equality data also help make some forms of inequality more visible. For example, Unia’s Socio-economic Monitoring I mentioned earlier makes clear where the problems in the labour market are and informs policy recommendations. We can also use data to communicate or sensibilize more specifically about certain topics. Finally, the individual complaints Unia handles on a daily basis are not only a rich source of data in themselves, but equality data might also help solving these cases, for example by using society-wide statistical data as proof of discrimination in an individual case.
Mr. Schmidt Equality data are important for member states to understand their (in)equality challenges as well as possible and come up with a policy response. In some countries migration will be a big issue, in others it will be religion. Age structures differ from country to country, as does the treatment of people living with disability etc. Most of the policies flowing from equality data will be at the national level, but data show that ethnic discrimination, for instance, is a prominent issue in the entirety of the EU, so here the EU could also build policies on Union-wide equality data.
In summary, EU guidelines exist and are constantly being improved based on how member states can gather equality data to create better policy on equality and non-discrimination. What’s next for equality and equality data?
Ms. Waerniers We would like to follow up on the recommendations in our report. One was to create a more coordinated approach to equality data in Belgium through creating a platform where we bring all the stakeholders together. This platform could solidify the dynamic that we created in our project through the participative and consultative groups. Another recommendation was the need to look further into self-identification based on ethnic or national origin in Belgium. It’s unclear how to use the current administrative data in a standardised way that allows us to compare the data, because the federal entities in our country often use different definitions, for example when it comes to what someone’s ‘origin’ is. I mentioned earlier that we have a lot of data on someone’s administrative status on origin and nationality, but that doesn’t necessarily match a person’s self-identification. Finally, the data hub and project report focused on a limited group of discrimination grounds. We are currently talking with our partners at the Equal Opportunities Team of the Federal Justice Service to create a follow-up project focusing specifically on disability data.
Lots of work to be done. Good luck!
This interview was conducted by Dominique De Meyst, PhD researcher at U Hasselt.