Who benefits from the preferential treatment of business property under the German inheritance tax?

Abstract

This paper contributes to the recent debate on the tax privilege for business property introduced by the German Inheritance Tax Act of 2009. We shed light on the tax effects of the preferential treatment based on micro-level data of realized business transfers from the German Inheritance Tax Statistic. As the tax exemptions are tied to employment expenses of the transferred business, we combine information from the tax statistic with a simulation of future employment expenses. We then compare the effective tax rates under a preferential treatment to the ones under a non-preferential treatment alternative. Results show that a large portion of business successions is untaxed under the current German inheritance tax law and that average effective tax rates approach zero. Moreover, the preferential treatment levels the differences between the effective taxation of small and large transfers and leads to an almost equal effective tax rate across all tax brackets. We also demonstrate that tax reductions are considerably larger in times of economic growth where the requirements for a preferential treatment are more likely to be met.

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2
Fig. 3

Notes

  1. 1.

    The presented results are based on FDZ (2012) and own calculations.

  2. 2.

    To ensure privacy of the tax filings that is protected by German law, the data is not publicly available. Therefore, our analyses are conducted in cooperation with the Research Data Centre of the German Federal Statistical Office and the Statistical Offices of the Länder.

  3. 3.

    As mentioned before, about 64 % of the transfers would not have to pay inheritance tax even in the absence of any preferential provisions for business transfers, e.g., due to tax exemptions on the personal level.

  4. 4.

    For the relevant subsample of firms that could potentially benefit from a preferential treatment, mean effective tax rates increase from 0.22 % to 0.65 % and the standard deviation rises from 1.03 % to 1.99 %.

  5. 5.

    As a recent example, see the US administration’s plan for a business tax reform that envisions the elimination of seemingly unjustified subsidies and a broader tax base as a precondition for a more effective tax system (The White House 2012).

  6. 6.

    The methodological difference is in line with Friedman’s postulate that assumptions of a theory do not need to be realistic but “sufficiently good approximations for the purpose in hand” (Friedman 1966, p. 15). In that sense, the assumptions made in Simons et al. (2012) are sufficient to answer the question which one of the two tax alternatives should be chosen. However, a quantification of the tax effects requires a more extensive and more detailed consideration of the potential determinants.

  7. 7.

    In the following, we use the term inheritance tax for the German inheritance and gift tax. With some minor exceptions, e.g., a special tax exemption for children after the death of their parents, transfers after death and inter vivos transfers are treated equally under German law. In the US, the term estate tax is more frequently used. For a detailed description of the US estate tax system see, for example, Batchelder (2009).

  8. 8.

    We address possible implications of these assumptions in Sects. 5.4 and 5.5. The results suggest that our main results are not materially affected.

  9. 9.

    For a detailed explanation of the computation see Simons et al. (2012).

  10. 10.

    Simons et al. (2012) assume proportional tax rates in their analysis. Analytically, they show that this assumption should not materially affect the choice between the standard and the option plan (Simons et al. 2012, p. 22). However, to allow for an accurate quantification of the resulting effective tax rates we use the exact tax schedule including the hardship provisions granted by the German inheritance tax code.

  11. 11.

    Please note that the effective tax rates in this study refer to the value of a specific transfer, not to the entire wealth transferred to a person. A potential problem of using the transfer-specific statistic of 2011 arises if more than one taxable transfer occurs within 10 years, but not all transfers are already recorded in the statistic. In this case, we would underestimate the effective tax rates. Since there are little economic incentives for such a behavior we do not expect this issue to materially affect our results. For practical problems in determining the value of transferred business see, e.g., Müller and Sureth (2011).

  12. 12.

    Note that the information provided in Table 1 allows us to extend the study of Simons et al. (2012), e.g., in that we are able to base our analysis on the empirical incidence of the 15 cases described in Simons et al. (2012, p. 10).

  13. 13.

    The indices represent the corresponding paragraphs of the German inheritance tax code.

  14. 14.

    As in many European countries, close relatives can use additional tax exemptions compared to third-party transfers [for an overview see (Næss-Schmidt et al. 2011; Eurostat 2012)].

  15. 15.

    While the non-preferential treatment requires an upfront tax payment, both preferential treatment alternatives lead to a partial deferral of tax payments. Thus, assuming a positive interest rate would lead to even higher differences between the non-preferential and the preferential treatment. Moreover, the effective tax rates shown can be interpreted as upper bounds since a positive interest rate would lead to a lower net present value of tax payments.

  16. 16.

    The respective intervals are chosen based on the classification of the publicly available version of the German Inheritance Tax Statistic of 2011.

  17. 17.

    We use the VLLMS version of the database, accessed through WRDS. Last access was in October 2015. While Amadeus offers firm-level data, it is not possible to link individual transfers recorded in the inheritance tax statistic to a specific firm due to data protection under the German law.

  18. 18.

    We use the median of the empirical distribution to mitigate the influence of outliers.

  19. 19.

    In contrast to the study of Simons et al. (2012) who employ both a binomial and a Markov model to approximate the development of employment expenses we use the empirical distribution based on German companies included in Amadeus. This allows for a more detailed and more descriptive analysis.

  20. 20.

    While this assumption may be theoretically important, it is practically irrelevant for our simulation since negative draws never occur. The same holds for the alternative specification reported in Appendix C.

  21. 21.

    To mitigate the influence of outliers, we only consider firms with total assets between 1 million euro and 151,000,000 million euro and truncate the growth rates at the bottom and top 1 %-level. Figure 4 in Appendix A suggests that the normal distribution used in our simulations is the most reasonable approximation of the empirical distribution.

  22. 22.

    In the resulting subsample, we truncate employment growth rates at the bottom and top 1 %-level.

  23. 23.

    Using the same set of transfers for both the preferential and the non-preferential treatment rules out selection bias.

  24. 24.

    Effective tax rates are computed as the percentage share of tax payments in the value of transferred property before any tax exemption, i.e., \(T_i / X\), for \(i \in \{ OP, SP, PRIV \}\).

  25. 25.

    Slight deviations regarding the number of runs are rounding differences.

  26. 26.

    The percentage is taken from a survey among German family businesses conducted by the Ifo Institute in 2014 (Potrafke et al. 2014, p.21).

  27. 27.

    The estimated fiscal revenue under the preferential treatment is calculated by multiplying the average tax payments of 47,841 euro (14,283 euro) that result under the standard plan (option plan) with the number of eligible transfers (14,800 \(\cdot\) 0.62). Accordingly, the estimated revenue under a non-preferential treatment results from multiplying the average tax payment of 386,038 euro with the number of eligible transfers (14,800 \(\cdot\) 0.62).

  28. 28.

    Our data also shows that especially inter vivos transfers which are typically smaller than transfers after death are excluded from the analysis.

  29. 29.

    We jointly consider the effect of multiple factors in Sect. 5.2, where we extend our analysis to a multivariate setup. Thereby, we are able to analyze the effect of a variation in one factor while holding all other factors constant.

  30. 30.

    Note that the effective tax rates under a non-preferential treatment remain unchanged compared to Table 4 as tax provisions for private property do not provide any tax reductions based on the development of employment expenses.

  31. 31.

    According to the our approach in the previous sections we assume r to equal zero.

  32. 32.

    For example, if we assume the average equity ratio for German SMEs in 2012 of 27.4 % (Kreditanstalt für Wiederaufbau 2013), a business with initial employment expenses of 800,000 euro would only meet the threshold if its property value exceeded 55.089 million euro.

  33. 33.

    This time period equals the one used to determine the parameters of our simulation. We again truncate the distribution at the top and bottom 1 %-level. The exact values for the 25-, 50-, and 75 %-quantiles of the distribution are 0.52, 1.28, and 3.19, respectively. These ratios remain fairly stable over time. Note that the market value of equity could be higher. We refer to shareholders’ equity to be consistent with our simulation and since market values are only available for relatively large listed companies.

  34. 34.

    Since we conjointly include the runs for the option and the standard plan in our estimations, we suppress the subscripts used in the previous sections.

  35. 35.

    As tax payments cannot be negative by definition, we use a censored regression model instead of standard OLS regressions to avoid estimation bias.

  36. 36.

    The variables \(\Delta _{(1)-(2)}\) and \(\Delta _{(3)-(4)}\) provide the differences between the coefficients shown in columns (1) and (2) or (3) and (4), respectively.

  37. 37.

    We use shareholders’ equity as provided in the Amadeus database as a proxy for firm size since this variable can arguably be linked to the size of the transferred property X.

  38. 38.

    Detailed results are available upon request.

  39. 39.

    In addition, please also note that our main analyses in Sects. 4.1 and 4.2 are based on a subsample that mostly consists of medium-sized and large transfers.

  40. 40.

    For the development of the current reform draft and the earlier parliamentary discussion see, e.g., BMFI (2015b), Deutscher Bundestag (2015b), and Deutscher Bundestag (2015a).

  41. 41.

    To qualify as a family business, the reform draft requires that the company agreement includes provisions to restrict distributions to the owners, ties ownership rights to family membership, and provides monetary incentives for personal continuity.

  42. 42.

    Specifically, the 90th and the 95th quantile in Panel C of Table 1 suggest that less than 100 transfers exceed the projected size thresholds.

  43. 43.

    See Simons et al. (2012) for an analysis of the optimal choice of a respective treatment alternative.

  44. 44.

    For the mining industry, growth rates of wages and salaries are only provided until 2008 by the national accounts.

References

  1. Batchelder L (2009) Estate tax reform: issues and options. Tax Notes, NYU Law and Economics Research Papers No. 09-02

  2. BMFI–Bundesministerium der Finanzen (2015a) Entwurf eines Gesetzes zur Anpassung des Erbschaftsteuer-und Schenkungsteuergesetzes an die Rechtsprechung des Bundesverfassungsgerichts. Gesetzentwurf der Bundesregierung, 8.7.2015 [Draft of the German Federal Cabinet for an amendment of the Inheritance and Gift Tax Law to comply with the jurisdiction of the German Federal Constitutional Court]

  3. BMFI–Bundesministerium der Finanzen (2015b) Referentenentwurf des Bundesministeriums der Finanzen. Entwurf eines Gesetzes zur Anpassung des Erbschaftsteuer-und Schenkungsteuergesetzes an die Rechtsprechung des Bundesverfassungsgerichts, 6.1.2015 [First Draft of the German Federal Cabinet for an amendment of the Inheritance and Gift Tax Law to comply with the jurisdiction of the German Federal Constitutional Court]

  4. BVerfG–Bundesverfassungsgericht (2014) Urteil vom 17. Dezember 2014 (1 BvL 21/12) [Decision of the German Federal Constitutional Court]

  5. Destatis–Statistisches Bundesamt (2011) VGR der Länder–Arbeitnehmerentgelt, Bruttolöhne und-gehälter in den Ländern [German National Statistical Office–National accounts–gross wages and salaries]. Wiesbaden

  6. Deutscher Bundestag (2015a) Antwort der Bundesregierung auf die Kleine Anfrage der Abgeordneten Lisa Paus, Dr. Gerhard Schick, Dr. Thomas Gambke, weiterer Abgeordneter und der Fraktion Bündnis 90/Die Grünen) [Response of the German Government to the Minor Interpellation of the parliamentary group of the Green Party]. In: Bundestags-Drucksache 18/5055, 6/1/2015:1–8

  7. Deutscher Bundestag (2015b) Kleine Anfrage der Abgeordneten Lisa Paus, Dr. Gerhard Schick, Dr. Thomas Gambke, Britta Haßelmann, Anja Hajduk, Dieter Janecek, Markus Kurth, Dr. Tobias Lindner, Beate Müller-Gemmeke, Corinna Rüffer und der Fraktion Bündnis 90 / Die Grünen) [Minor Interpellation of the parliamentary group of the Green Party]. In: Bundestags-Drucksache 18/4849, 5/6/2015:1–5

  8. Eurostat–European Commission (2012) Taxation trends in the European Union. Data for the EU member states, Iceland and Norway

  9. FDZ der Statistischen Ämter des Bundes und der Länder (2012) Erbschaft- und Schenkungssteuerstatistik 2011 [Research Data Centre of the German Federal Statistical Office and the Statistical Offices of the Länder - Inheritance and gift tax statistics 2011]. Wiesbaden

  10. Friedman M (1966) The methodology of positive economics. In: Friedman M (ed) Essays in positive economics. Chicago University Press

  11. Houben H, Maiterth R (2011) Endangering of businesses by the German inheritance tax? An empirical analysis. Bus Res 4(1):32–46

    Article  Google Scholar 

  12. Kopczuk W (2013) Taxation and intergenerational transfer of wealth. In: Auerbach A, Chetty R, Feldstein M, Saez E (eds) Handbook of public economics. Elsevier

  13. Kreditanstalt für Wiederaufbau (2013) KfW-Mittelstandspanel 2013 [KfW Panel on small and medium entities 2013]. Frankfurt, Germany

    Google Scholar 

  14. Miller D, Steier L, Le Breton-Miller I (2003) Lost in time: intergenerational succession, change, and failure in family business. J Bus Ventur 18(4):513–531

    Article  Google Scholar 

  15. Müller J, Sureth C (2011) Marktnahe Bewertung von Unternehmen nach der Erbschaftsteuerreform? [Approximate market valuation of businesses after the inheritence tax reform]. Zeitschrift für betriebswirtschaftliche Forschung Sonderheft 63(11):45–83

    Article  Google Scholar 

  16. Næss-Schmidt H, Pedersen T, Harhoff F, Winiarczyk M, Jervelund C (2011) Study on inheritance taxes in EU member states and possible mechanisms to resolve problems of double inheritance taxation in the EU. Copenhagen Economics

  17. Pickety T, Saez E (2007) How progressive is the U.S. federal tax system? A historical and international perspective. J Econ Perspect 21(1):3–24

    Article  Google Scholar 

  18. Potrafke N, Kauder B, Reischmann M, Riem M, Schinke C (2014) Die Auswirkungen der Erbschaftsteuer auf Familienunternehmen [The impact of the inheritance tax on family businesses]. Stiftung Familienunternehmen

  19. Simons D, Voeller D, Corsten M (2012) Erbschaftsteuerliche Verschonungsregeln für das Betriebsvermögen? Ein theoretischer Ansatz zur Steuerplanung [Inheritance tax credits for business property - a theoretical approach to fiscal planning]. Zeitschrift für betriebswirtschaftliche Forschung 64(2):2–36

    Article  Google Scholar 

  20. The White House (ed) (2012) The president’s framework for business tax reform. Joint report by the White House and the Department of the Treasury

  21. Zifonun N, Schöffel R (2004) Neue Erbschaft- und Schenkungssteuerstatistik 2002 [The new Inheritance and Gift Tax Statistic 2002]. Wirtschaft und Statistik 9:1028–1035

    Google Scholar 

Download references

Acknowledgments

We thank the Research Data Centre of the German Federal Statistical Office and the Statistical Offices of the Länder for providing us with micro-level inheritance tax data. We also gratefully acknowledge the helpful comments and suggestions from Jochen Hundsdoerfer (the editor), two anonymous referees, participants of the 2nd EIASM Workshop on Current Research in Taxation in Muenster, the 2013 EAA Annual Meeting in Paris as well as seminar participants at Leibniz Universität Hannover, Tilburg University, Wirtschaftsuniversität Wien, and colleagues in the MannheimTaxation (MaTax) ScienceCampus. Dennis Voeller appreciates the financial support from the Fritz Thyssen Foundation and the Julius Paul Stiegler Memorial Foundation.

Author information

Affiliations

Authors

Corresponding author

Correspondence to Dennis Voeller.

Appendices

Appendixes

Appendix A: Empirical distribution of employment expense growth

This figure shows histograms of the empirical growth rates based on firms’ employment expanses obtained from the Amadeus database over the period 2007–2013. Figure 4a shows 10,000 randomly drawn observations from the full sample used to determine the annual means and standard deviations for our base scenario. Figure 4b depicts all 9402 observations used to estimate the distribution parameters of the recession scenario. The figure suggests that the normal distribution used in our simulation is a reasonable approximation of the empirical distribution.

Fig. 4
figure4

Empirical distribution of employment expense growth. This figure shows histograms of the empirical growth rates of employments expenses. Figure 4a is based on 10,000 randomly drawn observations from the full sample used to parameterize the distribution under the base scenario and Fig. 4b depicts all 9402 observations used to estimate the distribution parameters of the recession scenario

Appendix B: Full sample analysis

In this appendix, we present results based on the full sample of all transfers reported in the German Inheritance Tax Statistic, i.e., we do not exclude cases that would lead to no tax payments under a non-preferential treatment. Results depicted in Table 10 are comparable to the ones presented in Tables 3 and 4, respectively. Note that the results for large transfers reported in Panel B in Table 4 are identical to the ones in Table 10 as all transfers with a property value above 2.5 million euro would lead to a tax payment under a non-preferential treatment. While effective tax rates (the percentage of transfers that lead to no tax payment) are lower (higher) compared to the analysis in the main text, our key findings remain unchanged.

Table 10 Simulation results

Appendix C: Alternative specification of the simulation

Simulation

To show the robustness of our results we base this alternative specification on the mean growth rates reported in the German national accounts. Thereby, the base scenario relies on the average annual growth rates of salaries and wages in Germany from 2003 to 2010 as reported in the German national accounts (Destatis 2011). Panel A of Table 11 shows the annual growth rates used in the simulation. As can be seen in the table, a moderate wage increase has been observed in 6 out of 7 years. In one year a slight decrease of wages (0.6 %) occurs. The recession scenario relies on the average development in the German mining industry between 2001 and 2008.Footnote 44 The mining industry exhibited the lowest growth of all industries considered in the national accounts. Wages shrunk in all periods with rates between \(-\)1.6 % and \(-\)6.2 %. For the alternative specification of the simulation, a standard deviation of 10 % is assumed. Therefore, difference in average growth rates per year are more pronounced in this simulation compared to the one presented in the paper.

Panel B of Table 11 reports aggregated employment expenses under both scenarios that result from the simulation consisting of 370,000 individual runs for each preferential treatment alternative and scenario. Under the base scenario, mean and median values of employment expenses lie considerably above the thresholds that have to be met under the two tax credit alternatives. This suggests that under an average economic development a majority of taxpayers is able to fully enjoy the tax reductions offered by the two preferential treatment alternatives. If the less restrictive standard plan is chosen even more than 75 % of the cases exceed the expense threshold. However, this observation changes under a recession scenario. While still more than 75 % exceed the threshold under the standard plan, employment expenses are considerably below the benchmark that is required under the option plan.

Table 11 Simulation of economic scenarios

Full sample and relevant subsample results

See Table 12.

Table 12 Simulation results

The effects of the preferential treatment

See Tables 13 and 14.

Table 13 Simulation results: Base scenario
Table 14 Simulation results: Recession scenario

Endogeneity of employment expenses

See Table 15.

Table 15 Incentives to increase employment expenses

Multivariate analysis

See Tables 16 and 17.

Table 16 Multivariate analysis by economic scenario
Table 17 Multivariate analysis by value of transferred property

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Franke, B., Simons, D. & Voeller, D. Who benefits from the preferential treatment of business property under the German inheritance tax?. J Bus Econ 86, 997–1041 (2016). https://doi.org/10.1007/s11573-016-0807-7

Download citation

Keywords

  • Business transfers
  • Inheritance tax
  • Preferential tax treatment
  • Simulation

JEL Classification

  • C15
  • H30
  • K34