Neighbourhoods won’t be improved by banning the unemployed

Hans Koster, Jos van Ommeren 27 July 2022



One of the most important principles in modern societies is that residents can choose to live where they like. Article 13 of the Universal Declaration of Human Rights asserts that: “Everyone has the right to freedom of movement and residence within the borders of each state”. Only under extreme circumstances do countries introduce laws that violate this principle, such as with sex offenders and criminals.

Recently, however, a range of countries, including Sweden, Denmark, and the Netherlands, have introduced laws that prohibit disadvantaged individuals from moving into public housing units in specific areas, for example because they are unemployed or have too low an income. The general idea behind these laws is that a high concentration of disadvantaged residents in a neighbourhood results in the formation of ghettos and unwanted spatial disparities in the standards of living (Diamond and Moretti 2022).

This raises two questions: Are these laws effective? And do they have undesirable side effects? Our new study for the Netherlands (Koster and Van Ommeren 2022) sheds light on both of these questions. We show that the laws are not particularly effective in attracting households with higher income levels to targeted neighbourhoods. What is perhaps more important is that we also show that the laws have significant negative side effects. These side effects come into existence because a negative stigma is created by these laws, as evidenced by lower house prices in targeted neighbourhoods. A negative stigma effect makes sense. Given prominent advertising of targeted deprived neighbourhoods in the media, who wants to live in a neighbourhood that is now commonly known to be a bad place to live?

Exclusionary policies do not improve social mixing

Our study analyses the effects of the Dutch Act on Extraordinary Measures for Urban Problems. This legislation, introduced in 2003, allows local governments to prohibit unemployed households from moving into public housing in targeted neighbourhoods. The Act was first implemented in Rotterdam, the second largest city of the Netherlands, in 2006, followed by other cities later on.

In the Netherlands, about 30% of the population lives in public housing, but this percentage is much higher in targeted neighbourhoods, where it is almost 60%. Particularly for low-income households, who are more likely to be unemployed, public housing is important.

Using register data for the whole population and using an event-study type of analysis focusing on areas close to the borders of treated areas, we show in Figure 1 that the Act reduces unemployment in targeted areas somewhat. After implementation of the Act in specific deprived neighbourhoods or streets, the share of unemployed is reduced by, on average, 0.4 percentage points per year that the Act is implemented. Hence, after five years, the reduction in the share of unemployed is about 2 percentage points, which is about one-sixth of the average unemployment rate in these neighbourhoods.

Figure 1 The effects of the Dutch Act on Extraordinary Measures for Urban Problems on unemployment

The Act appears largely ineffective in changing the demographic composition of neighbourhoods by improving social mixing. If policymakers hoped that the Act would have substantially improved the incomes in deprived neighbourhoods, then they will have been disappointed, as the average incomes of those living in public housing in targeted neighbourhoods increased by a meagre 1.5%. Other measurements of social composition (such as the share of migrants, the share of single households or the average level of education) are even less affected by the Act.

In other words, we find that the policy did lead to a ‘mechanical’ drop in the share of unemployed, but neither changed, nor improved, the social composition of neighbourhoods.

Exclusionary policies may lead to stigma

At the same time, due to the prominent advertising of targeted deprived neighbourhoods, a stigma may have been created. The presence of place-based stigma due to the announcement of the targeted neighbourhoods has been overlooked in the discussion on whether policies should be people- or place-based (Einiö and Overman 2016, Garcilazo et al. 2010, Gill 2010).

To investigate the presence of a stigma effect, we focus on changes in house prices, which is a standard, and precise, way to investigate changes in perceived attractiveness of neighbourhoods. More specifically, the idea is that if prospective homebuyers think that the Act has improved a neighbourhood (for example, because of improved social mixing), then this would increase house prices. On the other hand, if prospective buyers think that the Act has created a negative stigma, because of many negative reports in the media about the neighbourhood where they live, one would expect prices to drop.

The results show that house prices drop by about 3-5% after the Act has been announced. We show this by comparing changes in prices over time of houses that are close to (within 100 metres) but on different sides of neighbourhood borders (see Figure 2 for an example for the city of Rotterdam).

Figure 2 The Dutch Act on Extraordinary Measures in parts of Rotterdam

Of course, evidence of a stigma effect based on one specific Act does not imply that the presence of a stigma effect due to prominent advertising of targeted areas holds more generally. We show that this stigma effect is of a similar magnitude in two other national place-based policy programmes, adding to the external validity of the findings.

Exclusionary policies should be banned

In sum, we show that laws that violate the right of the freedom of movement by banning the unemployed from certain areas are largely ineffective, because they do not materially change the demographic composition of a neighbourhood. Furthermore, these laws seem to create a negative stigma effect of targeted neighbourhoods. Hence, all inhabitants living in targeted neighbourhoods – the poor in public housing as well as the rich in private housing – are likely worse off at the end of the day.


Diamond, R and E Moretti (2022), “Geographical differences in standard of living across US cities”,, 17 March.

Einiö, E and H G Overman (2016), “Effectiveness of place-based policies: UK evidence”,, 7 April.

Garcilazo, J E, J O Martins and W Tompson (2010), “Why policies may need to be place-based in order to be people-centred”,, 20 November.

Gill, I (2010), “Regional development policies: Place-based or people-centred?”,, 9 October.

Koster, H R Aand J N Van Ommeren (2022), “Neighbourhood Stigma and Place-based Policies”, CEPR Discussion Paper 17132.



Topics:  Poverty and income inequality Welfare state and social Europe

Tags:  social housing, public housing, ghettos, The Netherlands, social mixing, exclusionary policies

Professor of Urban Economics and Real Estate, Vrije Universiteit Amsterdam

Professor of Urban Economics, Vrije Universiteit Amsterdam


CEPR Policy Research