This week we're exploring the relationship between immigration, housing policy, and support for the far right. How does the expansion of social housing affect political attitudes? And what are the implications for public policy?
Across many democracies, far-right movements are gaining momentum — a trend that worries policymakers, researchers, and citizens alike. A common explanation points to material hardship: when people feel left behind economically and socially, they often turn to radical political alternatives. One critical dimension of this hardship is housing — especially the lack of affordable and secure homes. Could building more affordable housing help reduce support for far-right parties?
New research provides a nuanced answer. It finds that expanding access to social housing does seem to lower far-right support — but only in areas with low immigration. In communities where immigration is already high, the effect reverses.
To unpack why this is happening, and what it means for policymakers, host Prof Alan Renwick speaks with Dr. Gloria Gennaro, Lecturer in Public Policy and Data Science at UCL’s Department of Political Science. Dr. Gennaro shares insights from her latest study, exploring how housing policy, economic insecurity, and social dynamics intersect with political behavior.
Mentioned in this episode:
[00:00:05] Alan Renwick: Hello, this is UCL Uncovering Politics, and this week we're exploring the relationship between immigration, housing policy, and support for the far right. How does the expansion of social housing affect political attitudes? And what are the implications for public policy?
Hello, my name is Alan Renwick, and welcome to UCL Uncovering Politics, the podcast of the School of Public Policy and Department of Political Science at University College London.
The far right is on the rise in many democracies, and that is a source of widespread concern. One source of support for the far right is likely to be material hardship. Many people's lives are tough. They see mainstream parties as having failed them, so they seek more radical alternatives. One aspect of such hardship is a lack of adequate housing. It would seem, then, that increasing the provision of affordable housing should be one way of countering the far-right wave.
New research investigates that idea and finds a mixed picture. Another key focus of far-right politics is, of course, opposition to immigration. The research finds that increasing access to social housing does reduce far-right support in areas where existing levels of immigration are low, but it has the opposite effect where immigration is already high.
So why might that be, and what are the implications?
To explore these questions, I'm delighted to be joined by the author of the study, Dr Gloria Gennaro, who is Lecturer in Public Policy and Data Science here in the UCL Department of Political Science. Gloria, welcome to UCL Uncovering Politics. I believe this is your first time on the show, so it’s wonderful to have you and to hear about your research today.
Let’s start with the big picture — the background to your research. What do we already know about the factors that lead to support for the far right? I gave a few hints in the introduction, but it would be great to hear your perspective and overview of what the literature tells us.
[00:02:17] Gloria Gennaro: Yes, of course, Alan. Thanks for having me on the podcast.
I think we actually know quite a bit about the possible drivers of support for the far right. You rightly mentioned two major strands of literature. The first is on immigration. Many researchers have investigated, using robust methods, the conditions under which immigration increases support for the far right. For example, a sudden influx of immigrants from very different backgrounds tends to act as a catalyst for anti-immigrant sentiment and far-right support. There’s quite a lot of evidence on that.
The second research tradition looks at the welfare state — and more recently, at its retrenchment. When citizens feel left behind by public services, including housing but not only housing, they often turn to the far right. This can be a form of protest against mainstream parties, or simply a response to the generous promises that far-right parties often make in the area of welfare.
Now, if you put those two strands together, you arrive at what I think is the main hypothesis in the space I’m working in — what I call the “competition hypothesis”. This suggests that rooted citizens, or native residents if you prefer, may express more negative sentiments towards immigration when they feel they are in competition with immigrants for access to public services.
It’s a very intuitive idea. Public services are constrained — especially housing, which is a finite resource. People perceive it as a zero-sum game: if I get housing, you might not. In those conditions, it’s more likely that people will feel they’re competing with others for the same services. And when that “other” is someone who doesn’t belong to your in-group — someone who doesn’t look like you — you may end up attributing the problem to that diversity.
So essentially, people compete for resources, and this can create anti-immigrant sentiment when immigrants are perceived to be benefiting from those resources.
Now, the key question for me — which I think remains unanswered in the literature — is what happens if, instead of retrenching the welfare state and increasing competition, you actually try to expand it? When I started this project, I expected that expanding the welfare state — in this case, increasing the provision of housing — would reduce competition and improve attitudes towards immigration.
But the answer turns out to be much more complex. It really depends on the underlying structure of conflict in those societies. That’s where the paper comes in: it shows that expanding the welfare state doesn’t automatically mitigate intergroup conflict. It depends on how you do it.
[00:05:59] Alan Renwick: So just to clarify that initial hypothesis — the idea is that if you expand the amount of public or social housing available, you’re reducing competition. There’s more to go round, so people feel less like others are getting something they’re not. That should reduce the tension that fuels support for the far right, and therefore reduce far-right support itself. That’s the core hypothesis you started with, is that right?
[00:06:37] Gloria Gennaro: That’s correct — that was my initial hypothesis. What I find in the research, and what makes a lot of sense in hindsight given what we know about the politicisation of immigration and how context shapes our response to the welfare state, is that the competition hypothesis holds true when you expand the welfare state in a context where rooted native citizens perceive it to be for them.
In that case, expansion alleviates concerns around affordable housing or marginal competition with immigrants. Ultimately, the additional housing is perceived as benefiting themselves or their friends. This is what I find in relation to a reduction in support for the far right in areas that are essentially homogeneous — in my study, that’s areas in France where rooted French citizens predominantly live.
[00:07:50] Alan Renwick: So, areas where there isn’t already a high immigrant population.
[00:07:54] Gloria Gennaro: Exactly. Where the population is composed primarily of rooted French citizens, with no immigration background or immigrant descent.
[00:08:04] Alan Renwick: Hmm. Whereas you’re finding the opposite in areas that do already have a high immigrant population.
[00:08:10] Gloria Gennaro: That’s correct. The explanation is that when you have a population structure with high levels of immigration, combined with very hostile political rhetoric — as was the case in France during the period under study — increasing the provision of public housing, which was perceived at the time as a very immigrant-centred welfare benefit, actually reignites this latent conflict.
So you get more public housing, but people think, “Ah, this isn’t for me or my in-group or my friends — it’s for foreigners.” That’s the alternative hypothesis, and it’s what I test in the paper.
[00:09:01] Alan Renwick: Hmm. Let’s explore that hypothesis and your findings in more detail shortly. But before we do, it would be really interesting to hear more about the research itself.
You mentioned that you focus on France — I suppose that’s partly because France has had a strong far right for quite some time, dominated by what’s now called the National Rally, formerly the National Front, led by Marine Le Pen. But also, France’s housing policy allows you to implement what we often refer to as a quasi-experimental research design to capture the effects of the policy.
Could you take us through the design of the study and explain what features of housing policy made that possible?
[00:09:53] Gloria Gennaro: The key challenge in studying the relationship between housing and political attitudes is that places with different political preferences often have very different levels of public housing or welfare provision. For example, municipalities that have consistently elected left-wing mayors may have more public housing than those that haven’t. So it’s difficult to claim a causal link between housing expansion and political preferences, because of what we call endogeneity — the two factors are co-determined.
I focus on France not only because of its political context, but also because, starting in 1999 and more fully in 2000, the country implemented a fairly draconian public housing policy aimed at expanding provision across most urban municipalities. Eligible municipalities were required to ensure that 20% of total dwellings were offered under public housing conditions.
This policy applied to urban municipalities with at least 3,500 inhabitants — or slightly fewer in Île-de-France, the region around Paris. So there’s a clear population threshold: municipalities above it were subject to the policy and had to increase public housing, while those below it were untouched.
This creates what we call a natural experiment — something that resembles a randomised experiment. It’s as if public housing were randomly distributed across municipalities. By focusing on municipalities close to the 3,500 population threshold, I’m comparing places that are very similar in many respects — population size, geography, economic conditions — but differ only in whether they were required to implement the policy.
That gives me leverage to compare outcomes in terms of political preferences.
[00:12:47] Alan Renwick: Great, thank you. I think it’s the first time in quite a while that we’ve had someone on the podcast talking about a natural experiment, so it’s really helpful to walk through that logic clearly.
If listeners imagine a randomised controlled trial — like a drug trial — you’d have a group of participants, and you’d randomly assign some to receive the treatment and others not. Then you’d compare outcomes between the groups, knowing that any difference is due to the treatment, because the allocation was random.
We can’t do that with housing, of course — we can’t randomly allocate it. But in France, we happen to have a policy that almost randomly assigns municipalities to receive the higher housing target or not. So you can treat it as if they were randomly allocated and compare the outcomes.
I suppose the question is: is this really entirely random? If it’s larger municipalities that are being “treated” — as we might say — with the policy, and smaller ones that are not, could there be differences arising from that which we ought to be concerned about?
[00:14:15] Gloria Gennaro: Yes, you're absolutely right. If we could randomly allocate housing, we wouldn’t need to worry about differences between municipalities — randomisation would ensure they were roughly the same. But that’s not the case. We can’t randomly allocate housing, so we have to run a lot of tests to make sure the municipalities are actually similar.
In the paper, I measure this similarity using census data — looking at social demographics and a range of characteristics, including political preferences before the policy was implemented. By restricting the sample to municipalities very close to the population threshold, I find that they are indeed very similar across a wide range of observable characteristics.
I should also mention — without getting too technical — that there’s a second dimension of variation used in the study: time. Municipalities gradually become eligible as they cross the population threshold. So I can compare not only municipalities above and below the threshold, but also the same municipality before and after it becomes eligible. This allows me to control for characteristics that won’t change just because the population has increased slightly — things like geography, for example.
The combination of these two approaches gives me a sensitive approximation of what, in an ideal world, would be a randomised controlled trial.
[00:16:11] Alan Renwick: Great, that’s really clear — thank you. And how big is the treatment here? You’ve said the policy is implemented in some places but not others. There’s often scepticism around housing targets — they’re announced but not realised. Do we actually see a significant difference between the two groups of municipalities in terms of the amount of new public housing built over this period?
[00:16:42] Gloria Gennaro: Yes, we do. And that scepticism is grounded in reality — there was a lot of local debate around this policy. In particular, municipalities with vocal right-wing mayors were strongly opposed and tried to avoid meeting the target.
However, the policy includes fairly strong enforcement mechanisms. Municipalities that don’t comply are subject to substantial fines. What we see in the data is that, on average, the adoption of public housing is significantly higher in treated municipalities — especially as time goes by. Of course, it takes time to build or convert housing into public housing, but over time we see a clear difference.
Non-compliance does exist, but it’s spread across all types of municipalities and isn’t correlated with immigration levels. So that should give us some reassurance that, at least in the data, this isn’t a confounding factor.
[00:18:05] Alan Renwick: Great. So that’s the research design. And we’ve already given away the punchline — the core result is that in municipalities subject to the policy, support for the far right increases in areas with high levels of immigration, and decreases in areas with low levels of immigration.
How should we interpret that? You’ve touched on it already, but could you go further in explaining what’s going on?
[00:18:47] Gloria Gennaro: Yes, it’s an excellent question. The result is stark, but also difficult to interpret. We observe diverging trends in how municipalities respond to the policy, depending on pre-existing immigration levels.
This could be due to several factors. My hypothesis is that in areas with high immigration, the arrival of public housing reignites local conflict over welfare. But there are other possibilities. Perhaps those municipalities differ in other ways, or maybe new residents arriving with the public housing contribute to changes in aggregate voting outcomes.
These are reasonable assumptions, and I test them. The data show some demographic changes associated with public housing, but not enough to explain the divergent voting trends. We also see that both immigrant and native residents benefit equally from improved access to public housing — I have data on pending housing demands, which serve as a measure of congestion, and the situation improves for both groups.
So those alternative explanations don’t really hold. What does seem to hold is an analysis of how public housing is framed in local newspapers — which I use as a proxy for how it’s discussed in the community.
In that part of the paper, I build an index of association between immigration and public housing — looking at how often they’re discussed together in the same newspaper articles, and how frequently they’re discussed overall in a locality.
Using this simple measure, I find that the same public housing policy not only drives far-right voting, but also increases the association between immigration and public housing in local discourse. That’s what makes me confident in saying that what’s happening here is a shift in how people perceive public housing and immigration — particularly in areas with a history of local migration.
[00:22:00] Alan Renwick: Yes, I found that part of the paper really remarkable. You managed to analyse newspapers across France and obtain local circulation data — at least down to the departmental level — so you could see not just the national discourse, but also what’s happening locally.
And through that, you show that in areas with high levels of immigration, there’s a strong connection being made between housing and immigration in newspaper coverage. Whereas in areas with low immigration, that connection isn’t made to the same extent. So you’re seeing a difference in how the policy is being talked about in different parts of France — and the hypothesis is that this difference in discourse is driving the divergent voting outcomes.
[00:23:01] Gloria Gennaro: Yes, that’s correct. I look at all newspaper outlets in France — some are local by nature, others are national — but I’m able to construct a local measure of public discourse using circulation data.
So I know, for example, how many copies of Le Figaro are sold in a specific department. I then calculate a locally weighted average of the framing measure, weighted by circulation. That tells me whether a newspaper that takes a strong stance on immigration and public housing is widely read in a particular area.
This forms the basis of my measure of chauvinist framing at the local level.
[00:23:57] Alan Renwick: Great. So we have these results — really fascinating findings. What are the implications? And particularly, what should policymakers take away from this? Should they be cautious about expanding welfare or housing provision in diverse communities? Or should they be thinking in a different direction?
[00:24:20] Gloria Gennaro: I really hope that’s not the message people take from the paper. The main takeaway is that whenever you implement a public policy involving the allocation of new resources — whether that’s expansion, as in this case, or reallocation from one area to another — the reaction on the ground depends heavily on local conditions.
Understanding those local conditions beforehand is crucial if you want to build support for your policy. In the specific case of public housing, the issue seems to be around chauvinist framing circulating locally. So the best step would have been to clarify how public housing is used — by immigrant populations and by rooted natives — and to highlight that the main beneficiaries of this policy were, in fact, French citizens, not immigrant residents.
If anything, this policy — which seems to have triggered an extreme anti-immigration reaction in certain locations — should have had the opposite effect.
[00:25:47] Alan Renwick: Yes, that’s very clear. And it is striking that, as you say, the whole population benefits from the provision of social housing. Yet somehow, in these cases, the local discourse doesn’t reflect that. Changing that discourse is always difficult, of course — it’s hard to know how to shift public narratives around a policy like this. But that does seem to be the better way forward in response to these findings.
[00:26:15] Gloria Gennaro: Yes, and of course different policies will require different kinds of provisions. But paying attention to local conditions is especially important for policies with a strong physical impact — those that are highly visible, salient to the population, and perceived as zero-sum.
These are policies where people feel they’re competing for a finite, limited resource. Public housing is one example, but others include food stamps — anything that’s perceived as subject to competition tends to be particularly sensitive in this regard.
[00:26:59] Alan Renwick: So this research is now published, and as is often the case with academic work, that means you actually conducted it some time ago. Are you continuing to research in this area? Is there further work that will deepen our understanding?
[00:27:11] Gloria Gennaro: Yes, I’m continuing to work on housing and immigration along similar lines. At the moment, I’m looking at urban renewal policies and how they might affect intergroup conflict between newly arrived immigrants and the existing population.
I’m also exploring cases where local populations are resettled in response to new migrants arriving. But all of that is still in progress.
[00:27:47] Alan Renwick: Well, we very much look forward to having you back on the podcast once that research is complete and published. Gloria, thank you so much. This is important and thought-provoking work — helping us think through the underlying mechanisms, the effects of policy, and the broader implications. It’s been incredibly valuable having you on the podcast.
We’ve been discussing the article Immigration, Public Housing, and Support for the French National Front by Gloria Gennaro, currently available on Early View in the Journal of Politics. As ever, we’ll include details in the show notes for this episode.
Next week, we’ll be looking at new evidence from Ukraine on how exposure to war affects people’s commitment to democratic principles. To make sure you don’t miss that or any future episodes of UCL Uncovering Politics, just subscribe. You can do so on Apple, Google Podcasts, or whatever podcast provider you use. And while you’re there, we’d love it if you could take a moment to rate or review us too.
I’m Alan Renwick. This episode was produced by Eleanor Kingwell-Banham and Kaiser Kang. Our theme music is written and performed by John Mann. This has been UCL Uncovering Politics. Thank you for listening.