The InRa study (“Institutionen & Rassismus” / “Institutions & Racism”) is one of the most comprehensive investigations of institutional racism in German federal agencies to date. Conducted within the framework of the Research Institute Social Cohesion, it brought together 23 individual research projects that examined the issue from five complementary perspectives: the attitudes of agency employees, the experiences of those affected by racial discrimination, the transmission of discriminatory knowledge through institutional routines, the effectiveness of existing complaint mechanisms, and the historical and international context of racist discourses.
The study surveyed employees at four major federal agencies — including the Federal Office for Migration and Refugees, the Federal Police, the Customs Administration, and the Federal Employment Agency — and conducted an online survey among Muslims, 80 percent of whom reported experiences of discrimination in their interactions with public authorities. Notably, access to some agencies proved difficult, and not all institutions were equally willing to cooperate with the research — a finding the researchers consider significant in itself.
The study’s release prompted defensive reactions from some of the agencies involved, sparking a broader public debate about whether German institutions are willing to confront structural racism within their own ranks.
In the following interview, Prof. Dr. Gert Pickel, who led the overall InRa study, and Prof. Dr. Holger Lengfeld, who directed the employee survey, discuss key findings, methodological challenges, and the political responses their work has provoked.
There is no shortage of research on racism. What sets your study apart?
Gert Pickel: What makes the InRa study distinctive is that it does not examine racism in institutions from a single perspective but from five complementary angles: the attitudes of employees in federal agencies, the experiences of people subjected to racial discrimination, the transmission of racist interpretive patterns through routines and everyday knowledge within agencies, the effectiveness of existing complaint mechanisms, and the historical and international contextualization of knowledge systems and discourses on racism. We studied racism in institutions using the full range of methods available to the social sciences and humanities, and we observed, surveyed, accompanied, and critically examined different actors involved in processes of racial discrimination. In our view, only this methodological breadth can do justice to the complexity of the phenomenon.
Were there any findings that particularly surprised you?
Gert Pickel: Not necessarily surprising, but remarkably clear-cut were the results of an online survey among Muslims. It showed that 80 percent of respondents had experienced discrimination in their interactions with public agencies. Between 40 and 50 percent reported experiencing discrimination specifically at job centers, social welfare offices, and immigration authorities. Racism manifested itself in individual attitudes of specific employees, in agency practices and the exercise of discretion, and in the handling of complaints — in other words, on all three levels: individual, institutional, and structural. These findings carry even greater weight given that gaining access to some agencies was fraught with considerable obstacles and only succeeded where there was a willingness to participate in the research. This willingness — and this, too, is a finding of our study — is very unevenly distributed across Germany. Together with the finding that institutional racism is strongly dependent on regional context, this makes clear that it is possible to deal with racial discrimination in a sensitive and considered manner, but that some actors resist doing so — in part for ideological reasons. What matters, however, is that even under these conditions — in institutions that were fundamentally willing to cooperate — racial discrimination was demonstrable.
In the report, you describe racial discrimination not as an isolated incident but as a structural risk. What concretely distinguishes this structural understanding from the individual prejudices of specific employees — and which examples from the study make this distinction particularly clear?
Gert Pickel: A structural risk is expressed in the systematic unequal treatment of different groups. For example, when Sinti and Roma or Muslims are denied services that other citizens receive. In institutional racism, a key element is organizational culture. When senior employees tell newer colleagues that Sinti and Roma can never be trusted, or that Muslims should always be asked to submit additional documentation because that is standard practice within the agency, these are structural features. This is also reflected in the absence of clearly designated complaint mechanisms or a mission statement opposing discrimination.
The InRa employee survey — one component of the overall study — found that the level of discriminatory attitudes among agency employees was, on average, in the lower third of the measurement scale. Does that mean, in your view, that there is no racism problem in the public sector?
Holger Lengfeld: The answer depends on one’s perspective. What our analyses show is that employees at the four agencies we studied — the Federal Office for Migration and Refugees (BAMF), the Federal Employment Agency, the Customs Administration, and the Federal Police (Bundespolizei) — differ only marginally from the general population in their levels of discriminatory attitudes. So it is not worse, but neither is it better. If you take the position that no one in the public sector should hold prejudices against particular groups, you will be disappointed by our results. But if you assume that federal employees come from the middle of society, you will assess the findings differently. Which of these assessments is correct is something we cannot resolve scientifically. What is clear is that both perspectives identify a point at which society should take action.
Some of the agencies examined have responded critically to the findings, citing among other things the low response rates in the surveys. How do you respond to the charge that the study paints a distorted picture?
Holger Lengfeld: The statements issued by two agencies attempt to cast doubt on the validity of our findings. I find this puzzling. With the support of agency leadership, we invested considerable effort in outreach — briefing management, staff councils, and data protection officers, carefully designing the questionnaire and the survey procedures, sending reminders during the data collection period, and applying ex-post weighting to adjust the responses to the underlying populations. All of this was done by the book and based on extensive experience. The participation rates we achieved are therefore satisfactory — including at the Federal Police, since most officers do not have their own desk computers, are frequently out in the field, and have little time for such a survey. At the BAMF, we conducted a full census, which is always the gold standard for employee surveys. The Federal Police did not allow this, which is why we drew a carefully designed random sample of stations. Of course, every survey has potential sources of error, but I cannot see why ours should be exceptionally large. What I would have preferred is for the agencies in question to engage with the substantive findings. Perhaps they will do so once they read our separately published study with all the details.
The team behind the employee survey also examined differences between agencies in the extent of ethnic discrimination. What are the key factors at play?
Holger Lengfeld: In our study, we looked at four agencies with fundamentally different mandates. Two of them — the Federal Police and the Customs Administration — we classify as control-oriented organizations. Their task is to enforce public order. Employees working in this context regularly find themselves in tense situations when interacting with foreign nationals — for example, when carrying out border checks or enforcing customs regulations. We found that these contacts with the people being controlled are frequently experienced as stressful and negative, because there is often tension between those doing the controlling and those being controlled. This experience in the course of professional duties can reinforce prejudice. By contrast, the other two agencies — the Federal Employment Agency and the BAMF — enable a larger proportion of applicants to receive the services they have applied for. Here, we found significantly more positive interactions between employees and applicants, along with lower levels of prejudice. In short: the nature of an agency’s mandate can affect the level of prejudice — especially when the majority of an agency’s employees share similar experiences.
One of your central recommendations is to extend the General Equal Treatment Act (Allgemeines Gleichbehandlungsgesetz, AGG) to cover the relationship between the state and its citizens. Given the sensitive nature of some security and administrative policy areas: how realistic is this reform — and what do you say to agencies that fear overregulation or constraints on their operational capacity?
Gert Pickel: The reactions are indeed mixed. While some point to overburdening and the constraints you mention, others take the criticism on board. We were also told that in a post-migration society, institutions must be open to people with a migration background — on normative grounds, but also because of impending labor shortages. The extent to which such reforms are implemented, mission statements adopted, or reporting mechanisms established depends not least on the leadership of the respective agencies. Where leaders commit to measures against racism, this will permeate the organization. At the same time, legislators must signal a willingness to take action against racial discrimination. Racism must be recognized as racism, and it must be understood that this is not a matter of individual sensibilities on the part of citizens, but of structural and institutional problems — and a violation of the principle of equal treatment.
Interview: Sarah Lempp