Current Research

Welfare Effects of Property Taxation

Co-Author: Max Löffler

We investigate the welfare implications of property taxation. We apply a sufficient statistics approach that accounts for the distributional effects of tax changes at the household level within a spatial equilibrium framework. We show that equity effects are driven by price adjustments in the housing and labor markets, while efficiency is determined by changes in public goods. Using microdata and exploiting 5,500 municipal property tax changes in Germany, where assessed housing values remain constant, we find that 83 percent of the tax burden is passed through to rental prices, with modest labor market effects. Simulations of the welfare effects of property taxes reveal that the price effects of property tax hikes are regressive. Despite low efficiency costs of the tax, it becomes distributionally neutral only if public good preferences are very high.

Real Effects of Firm Property Taxes 

Co-Authors: Jarkko Harju and Teemu Lyytikäinen and Erkka Silvennoinen

In many countries commercial property taxes are an important source of local revenue, yet little is known about their real effects on firms. Using Finnish administrative microdata that allow exact measurement of firm-level property tax payments, we exploit more than 1,000 municipal tax rate changes in an event study design to estimate the real effects of property taxation. Firm-level property tax payments respond nearly one-for-one to statutory rate changes, while the underlying tax base is relatively inelastic. We find that higher property taxes reduce total capital stock by lowering building investment. In contrast, firm scale measures show limited average responses, and entry effects are modest. These results suggest that commercial property taxes are rather revenue-efficient but distort firms’ capital composition.

The Effect of Social Housing on the Dynamics of Gentrification

Co-Authors: Maximilian Günnewig-Mönert, and Andreas Mense

We study the consequences of expiring social housing contracts on neighborhood dynamics in Berlin. When affordability covenants expire, subsidized units return to the private market, potentially altering local housing market dynamics through higher prices, increased mobility, and changes in neighborhood composition. We exploit the staggered, contract-length–driven roll-off of Germany’s social housing program between 2010 and 2020 to quantify the effects of these expirations. Using high-resolution administrative data and a difference-in-differences event-study design that accounts for heterogeneous treatment timing, we show that the average loss of 43 regulated dwellings per tract increases advertised rents by 4–5%, triggers additional renovation activity, and induces selective in-migration of higher-income, lower-risk, and more ethnically native residents. These results highlight how the structure and duration of housing subsidies shape long-run affordability, gentrification, and urban inequality.

Housing Market Effects of Renter Protection

Co-Authors: Joscha Reiners

This paper estimates the effects of Berlin's Milieu Protection Areas (MPAs; Milieuschutzgebiete) on housing-market outcomes linked to displacement pressure. MPAs are place-based regulations that restrict selected modernization measures and rental-to-condominium conversions in designated neighborhoods. We exploit staggered MPA adoption across Berlin's planning areas in an intensity-weighted event-study design and combine rental listings from ImmobilienScout24 with administrative apartment transaction data from the Gutachterausschuss Berlin. MPAs reduce advertised rents in treated areas, with effects emerging gradually and concentrated in the pre-war housing stock. We find no robust effects on sale prices, but conversion-related transaction activity declines significantly. Mechanism results indicate that MPAs shift the composition of rental offers toward less upgraded units: the share of high-condition listings falls, equipment quality declines, and modernization-related amenities become less common. The findings suggest that MPAs moderate rental-market pressure not through direct price regulation, but by constraining upgrading and conversion.

Price Passthrough without Intertemporal Substitution: Evidence from Germany's Temporary VAT Cut

Co-Authors: Günter Beck, Muzammil Hussain, Xavier Jaravel, and Sebastian Kessing

We evaluate the effectiveness of an unconventional fiscal policy implemented during the first wave of the Covid-19 pandemic in Germany, where VAT was temporarily cut from July 2020 to January 2021. We estimate the causal effects of the policy on consumer prices and quantities purchased using granular scanner data, covering both durable and non-durable goods with over 100 million transactions. Using other European countries as control groups for Germany, together with a complementary research design that analyzes shifts in the product-level distribution of price changes over time within Germany, we find sizable pass-through of the VAT changes to retail prices: 65 percent for durables and 91 percent for non-durables. However, we do not observe significant quantity responses for either durable or non-durable goods. We find that even in a short time window around the policy shock, there is little change in quantities purchased even for durable goods, despite strong price incentives to delay consumption for these goods. Using a simple model, we show that the empirical estimates imply a tight upper bound on the elasticity of intertemporal substitution (EIS). Our baseline specification rules out an EIS above 0.08.

Survivor Benefits, Self-Insurance, and Gender Inequality: Evidence from Sweden

Co-Authors: Nadja Dwenger, Johannes Kochems, and Martin Nybom

We study how women respond to reductions in public insurance against widowhood, with a particular focus on forward-looking adjustments before widowhood. We exploit a cohort-based reform in Sweden that sharply curtailed survivor benefits in 1990. Applying a regression discontinuity design to population-wide administrative data, we study labor market outcomes, wealth, and retirement over more than three decades. Treated women delay retirement by two to three months and accumulate modestly more net wealth, but show little earnings response during prime working ages. After an early and unexpected partner death, women in treated cohorts increase employment and earnings substantially more than control cohorts, consistent with the larger income loss they face under the reformed benefit schedule. In a simulation, we show that forward-looking behavioral adjustments offset roughly two-thirds of the reform-induced decline in expected lifetime consumption, yet fall short of full compensation: the net present value of expected consumption falls by 14 percent of baseline annual income. Accounting for the disutility of additional labor supply implies welfare losses of around 50 percent of baseline annual utility. The insurance cost of increased consumption risk upon widowhood is small, in part because women self-insure through precautionary wealth accumulation.