The German real estate transfer tax: Evidence for single-family home transactions

https://doi.org/10.1016/j.regsciurbeco.2018.08.005Get rights and content

Highlights

  • We study the effects of real estate transfer taxes in Germany.

  • Higher real estate transfer tax produces massive anticipation effects.

  • A one percentage point higher tax rate yields approximately 7% fewer transactions due to the lock-in effect.

  • Our findings prove the necessary condition for inefficient market distortions due to real estate transfer taxes to be present.

Abstract

This paper uses recent data for single-family home purchases to study the effects of the German real estate transfer tax. We aim to separate the tax's anticipatory effects from its lock-in effects on real estate transactions. The data indicate that an increase in the transfer tax is negatively correlated with the number of transactions that take place in the market for single-family homes. We estimate that an increase in the transfer tax produces enormous anticipation effects: 41% more transactions for the month just before and 42% fewer transactions just after. In addition, a one-percentage-point increase in the transfer tax yields approximately 7% fewer transactions due to the lock-in effect.

Introduction

The impact of real estate transfer taxes1 on the residential housing market is a controversial subject in both political debates and scholarly research. On the one hand, some authors consider the positive effects of transfer taxes that result from less speculation in the real estate market (see Catte et al., 2004). However, these findings are empirically ambiguous (see Fu et al., 2013; Aregger et al., 2013). They must also be placed in the context of potentially larger economic distortions because, on the other hand, adverse effects might accompany real estate transfer taxes: lower labor mobility, mismatched housing situations (for example, retired households might be discouraged from downsizing, leading to misallocation of the housing stock; see Glaeser and Luttmer, 2003) or an inefficient shift from owner occupancy towards rental housing. In contrast to previous studies, we focus on effects on the timing of transactions. We show that the anticipation of a transfer tax increase leads to massive distortions regarding the transaction date (the anticipation effect). Moreover, higher taxes reduce overall market activity (the lock-in effect).2 This proves the necessary condition for any of the abovementioned adverse economic distortions to be present. We exploit a unique new dataset on real estate transfer tax increases at different times in different states in Germany. We conduct a two-way fixed effects regression using a full survey of actual single-family home transactions and clearly quantify the effects of a change in the tax rate on the number of transactions, which reflect the market reactions.

Despite the economic relevance, there is only a small body of literature that focuses on the effects of real estate transfer taxes on the housing market.3 One of the first studies addressing the effects of an increase in the transfer tax was undertaken by Benjamin et al. (1993). The authors find that the sales prices of properties within Philadelphia decreased relative to properties outside of the metropolitan area.4 In a more recent study, the impact of Toronto's transfer tax has been studied by Dachis et al. (2012). As the real estate market did not anticipate the tax change, the authors only study lock-in effects and estimate a decline in transactions and a decline in prices. Davidoff and Leigh (2013) analyze the price effects of an increase in the transfer tax in Australia. They find that an increase in the transfer tax leads to a drop in house prices and decreases the number of transactions. Kopczuk and Munroe (2015) examine the effects on house prices of a 1% real estate transfer tax on residential transactions valued at over $1 million (the price notch) in New York and New Jersey. They find evidence of significant bunching just below this price notch. Similarly, Slemrod et al. (2017) find evidence of manipulative sorting around the price notch but not around the time notch (i.e., the date of the tax change). Compared to previous studies, the authors examine a rather small tax change, which could explain the limited response. With regard to European real estate transfer taxes, Best and Kleven (2017) study the impact of a tax holiday in the U.K. and show that there is bunching just below the price notches5 in addition to distortions involving the volume and timing of transactions. The same unanticipated stamp duty tax holiday was studied by Besley et al. (2014), who find that it led to significant decreases in sales prices and increases in transactions. In the case of Germany, Petkova and Weichenrieder (2017) use annual indices of property transactions and average prices to study the effects on prices and on the number of transactions for single-family homes and apartments separately.6 They find significant tax effects on the number of transactions only for single-family homes.

We are one of very few studies to quantify both anticipatory and lock-in effects due to changes in the tax rate. We include dummy variables before and after the tax increase to capture when transactions are pushed ahead of the tax increase. By controlling for this bunching around the tax increase, we can measure the lock-in effect of the tax increase. Second, due to the unique institutional setting, we can benefit from regional variation in tax levels and tax increases in Germany for our empirical analysis as transfer taxes are levied at the local level. Additionally, German states impose a flat tax rate and, therefore, no price notch, which is why we are able to perfectly isolate the time notch effect, as no manipulation around the price notch needs to be accounted for. Third, we contribute to the quite limited empirical literature on non-Anglo-Saxon areas. We believe that markets behave differently depending on the role of information and friction. In continental European housing markets, bank lending practices are more conservative, homeownership and residential mobility rates are lower, housing supply tends to be more rigid and tenant-landlord regulations are comparatively strict (Andrews et al., 2011). Fourth, our micro data on single-family home transactions provide a powerful basis for assessing the adverse effects of real estate transfer taxes in various German states over the 2005–2015 period. The results of our study indicate that a transfer tax increase is negatively correlated with the number of transactions that occur in the market for single-family homes.

Section snippets

Institutional background

To highlight the economic relevance of real estate transfer taxes and to provide support for our empirical strategy, we now present relevant institutional background facts on real estate transfer taxes.

Real estate transfer taxes are commonplace and an important source of government revenues in many OECD countries (Andrews et al., 2011).7 Nonetheless, there is significant

Conceptual framework

In this section, we refer to the theoretical framework proposed by Slemrod et al. (2017) to distinguish among the various economic effects of increases in the real estate transfer tax. Slemrod et al. (2017) address both a price and a time discontinuity (or notch) for when new taxes take effect. As there is no price notch in Germany, we limit our analysis to the time notch. We will describe a simplification of the model used in Slemrod et al. (2017, p. 142). We do not account for the bargaining

Data

In the following sections, we empirically investigate the conceptual framework described above. Figures on real estate transactions in Germany are scarce. Other studies on the effects of real estate transfer taxes have relied on data aggregated at the federal level and on an annual basis, which are quite crude and empirically unfounded (see Rheinisch-Westfälisches Institut für Wirtschaftsforschung, 2012), or on annual indices that are not consistent across states or over time (see Petkova and

Empirical strategy

All German states began with the same real estate transfer tax levels at the outset of our observed time frame. State governments have been authorized to independently set their own tax rates since September 2006. Whenever a state changes its tax rate, the remaining states function as control groups. In our sample, many states have raised their tax rates by different amounts and at different times. To measure causal effects, we need to rule out reverse causality of tax increases and transaction

Robustness checks

For robustness exercises, we apply a number of different specifications to our model that are summarized in Table 3, Table 4. First, we consider different lengths for the anticipation period by comparing the baseline regressions with two dummy variables before and after tax increases with the case of one and three dummy variables (column ‘Robustness (1)’ in Table 3). Including only one month before and after a tax increase potentially creates omitted variable bias that we can quantify by

Conclusion

The real estate transfer tax is a major part of all transaction costs in property purchases in Germany. An increase in the tax rate makes real estate acquisitions significantly more expensive. We conclude that the announcement of an increase in the real estate transfer tax leads to a significant reaction in the housing market: Many market participants accelerate planned transactions to take advantage of the lower tax rate. Moreover, a drastic drop in transactions can be observed after the tax

References (33)

  • M. Best et al.

    Housing market responses to transaction taxes: evidence from notches and stimulus in the UK

    Rev. Econ. Stud.

    (2017)
  • Bundesinstitut für Bau-, Stadt- und Raumforschung

    Indikatoren und Karten zur Raum- und Stadtentwicklung, Ausgabe 2014, Durchschnittliches Haushaltseinkommen in € je Einwohner

    (2015)
  • Bundesministerium der Finanzen

    Endgültige Abrechnung des Länderfinanzausgleichs für das Jahr 2012

    (2015)
  • Bundesministerium für Verkehr, Bau und Stadtentwicklung (BMVBS)

    Dokumentation des Expertengesprächs Wohnungs- und Immobilienwirtschaft, Deutschland zukunftsfähig gestalten

    (2012)
  • Bundesrat

    Grunderwerbsteuerbefreiung bei Fusionen von Wohnungsunternehmen in den neuen Ländern

    (2004)
  • P. Catte et al.

    Housing Markets, Wealth and the Business Cycle

    (2004)
  • Cited by (20)

    View all citing articles on Scopus

    We are grateful to Marcel Thum, Thiess Büttner, the participants of the Homes-uP International Meeting, the Urban Studies Seminar at the University of Glasgow, the European Real Estate Society Annual Conference, the 15th Finanzwissenschaftliches Seminar in Berlin, the Macroeconomic Seminar at the University of Barcelona and two anonymous referees for very helpful comments and discussions. We would like to thank the staff at the Property Valuation Committees in Berlin, Brandenburg, Bremen, Rhineland-Palatinate, Saarland and Saxony-Anhalt for providing access to the data and for their support of this project. Financial support from the Leibniz-Gemeinschaft is gratefully acknowledged. All errors are our own.

    View full text