Project 4.1

Governments adopt policies for specific migrant groups. Such policies can help migrant groups thrive. However, they can result in negative responses from other groups who might perceive such measures as preferential treatment at their expense, thereby threatening social cohesion at large. In this PhD project, we aim to analyse how and under which circumstances migrant-specific policies facilitate the integration of migrant groups while simultaneously upholding social cohesion in and between all societal groups. The project will, amongst others, study the effects of tax benefits for knowledge workers on housing dynamics and social cohesion in the Netherlands throughout time.

Governments sometimes adopt national policies to support specific groups, such as migrants. These include providing tax benefits to high-skilled labourers, giving refugees priority in housing or ensuring basic needs of irregular migrants. Tailoring support to such groups can be important for these groups to thrive. Moreover, local accommodation and adaptation of these policies to these groups may be more effective than a single one-size-fits-all approach. At the same time, group-specific policies might also result in negative responses from other groups, who might perceive such measures as preferential treatment at their expense, thereby threatening social cohesion at large. In this project, we aim to disentangle which local circumstances and practices facilitate the integration of various migrant groups targeted in national or sub-national policies while simultaneously upholding social cohesion in and between all societal groups. We plan to use detailed longitudinal observational data at the municipality level to identify areas subject to more or less intervention to support migrants and compare outcomes between municipalities.

We are interested in establishing a set of stylized facts that describe current and historical patterns of social cohesion in relation to national and subnational policies that affect the integration of migrants and refugees as well as the response of the majority group. The unit of analysis is primarily at the municipality level, as this provides variation needed to identify consequences of policies for both minority and majority groups, depending on the timing of their implementation or the type of policies enacted. We envision the following:

For research question 1, we aim to use CBS (micro)data on the 30 per cent rule and housing prices and construct a neighbourhood segregation measure. We will relate this to data on housing price changes, voting behaviour and survey responses regarding trust in government while controlling for contextual variables.

For research question 2, various scenarios are possible, depending on the methodological skills and interests of the candidate. For example, we could develop a typology of local policy responses to identify variation in local policy practices over time for a sample of municipalities. This can be done via content analysis or web scraping methods for data collection. Following Tjaden and Spörlein (2024), we could distinguish between policies that target specific identity groups separately (e.g., refugees) or policies that mix these groups with other groups (e.g., other groups in need of social housing). Another option is to collect information about how much budget is allocated to specific groups in comparison to more general policies in a given municipality.

Once local policy practices have been categorized, we could connect temporal variation in municipal-level policies with contemporaneous CBS microdata on social cohesion/integration at the municipal level. Alternatively, observational or survey data can be analysed to assess the efficacy of policies that affect these groups and the municipalities in which they reside, while controlling for context-related factors. For causal inference, we can employ applied statistical methods such as difference-in-differences, instrumental variables and/or regression discontinuity models, which suit the longitudinal nature of our municipal-level dataset.