Deniz Tufur
PhD Candidate in Political and Social Sciences, European University Institute
My dissertation, supervised by Simon Hix, studies how housing shapes political behaviour: how mortgage holders and renters respond when housing costs rise, and how voters react when financial investors become the owners of the homes they live in. I work with panel data, survey and conjoint experiments, and causal inference designs.
Other research interests: gender and political representation · gender and environmental politics · economic voting in the United States · Turkish elections · democratization
Research
Under review
Why did the UK Conservatives lose support during the cost-of-living crisis, even among their core voters? Using the British Election Study panel, I show that rising mortgage rates pushed Conservative mortgage holders away from the party, especially where lower local house prices left them with the least housing wealth to fall back on.
Working papers
Europe's housing stock is energy inefficient and a major source of emissions. Neither individual landlords nor the state have the capital to renovate it, so European governments are turning to investors to finance the work. Survey experiments in Ireland and Denmark show that voters care not only about the costs of renovation, such as rent increases, but also about who carries it out. Private equity involvement lowers support in both countries, while pension funds are penalised only in Ireland. Support therefore depends on how legitimate a type of investor is in a given country, regardless of the consequences of what it does.
In Ecuador's 2023 referendum on halting oil extraction in Yasuní National Park, men and women voted in separate booths, which lets us study gendered environmental voting with real election results. Using booth-level results and a difference-in-differences design, we find that women are more likely than men to support halting extraction where local economies are not tied to oil. In oil-producing regions the gender gap disappears, because material dependence on the industry overrides value-based differences. Gender gaps in environmental voting are therefore not fixed but shaped by local economic context, which helps explain why they are often weaker in the Global South. An original online survey (N ≈ 1,500) and in-person interviews in oil-producing regions and Yasuní (N = 300), currently under way, probe the individual-level mechanisms.
Most MP candidates in closed-list PR systems never make it to parliament. Is that where their political careers end? This study of Türkiye follows the parliamentary candidates of the five main parties who were not elected into local politics. The municipal back door is gendered: men are promoted to mayoral races, while women are relegated to the municipal councils, which hold far less power than the mayor.
How much renters' housing costs matter in party competition depends on who renters are. In Germany, they are a large, middle-income group close to the median voter. In the U.K., they are fewer and further below the median income, which gives parties less reason to court them. Do British renters under rent pressure still turn to Labour, the party with a long record on rental housing? Using UK Household Longitudinal Study panel data (2013–2022) and two-way fixed effects models, we find that both higher rents and higher rent pressure (rent as a share of household income) are associated with greater support for Labour.
Why did the Baltic states become democracies while most other former Soviet republics did not? The paper asks whether a history of nationhood helps explain these diverging paths.
Work in progress
Did the local cost of living shape how Americans voted for president? We combine local price and labour-market data for around 3,100 counties with election results, the Cooperative Election Study and the American National Election Studies (ANES).
Results from individual ballot boxes for eleven national and local elections in Türkiye. The database lets researchers study Turkish electoral politics at a very fine geographic level.
Education
Teaching & Service
Research Assistantships
Contact
European University Institute, Florence, Italy
Deniz.Tufur@eui.eu
LinkedIn ·
Bluesky ·
Substack
Blog
A second look at the Financial Times chart of 112 rent-control studies. Sorting the studies by policy design shows that rent freezes and permanent caps drive most of the harm (86% of their results are harmful, against 52% for other designs), so the real question is what kind of a rent-control regime a city wants.
Does informing voters about corruption make them punish politicians at the ballot box? Drawing on cases from Brazil and Turkey and on research from elsewhere, I argue that information leads to accountability only when several conditions hold at once, and any one of them can break.
Across 330 Turkish polls from 2015 to 2024, every family of pollsters overstated support for the CHP, and the industry badly underestimated İmamoğlu in the 2024 Istanbul election. I predict that if the polls overstate the CHP again, Yeni Parti may do better than they expect.