Taxation of married couples in Germany and the UK: One-Earner Couples Make the Difference

International comparisons of tax liabilities are a crucial instrument for the analysis of tax systems, since they reveal feasible alternatives for tax reforms. In the public debate, the focus is usually on the tax rates. International comparisons of personal income taxes are difficult because tax schedules are typically progressive, i.e. the average tax rates increase with income. Moreover, there are significant differences in what, and how, income is taxed, depending, for example, on whether it is earned income, capital income, or income from benefit receipt, as well as with respect to the definition of taxable income. One further dimension that is important to consider when comparing the distributional effects of income taxation is the treatment of married couples. Whereas several countries treat married spouses as two different tax units (i.e. individual taxation), other countries tax married couples as a single tax unit (i.e. joint taxation). In a progressive tax system this difference is of central importance (Steiner & Wrohlich, 2004, or Immervoll et al., 2007). Germany (DE) is a prominent example for a country with joint taxation. In Germany married households jointly file their income and thus benefit from full income splitting (Ehegattensplitting)¹ In contrast, in the United Kingdom (UK) married couples are in principle taxed like single households.

In this paper we propose a comparison of tax liabilities for these two countries in order to highlight the effect of the different treatment of married couples in the income taxation. We have chosen the UK and Germany for several reasons. First, despite several differences in the tax and transfer system (see, e.g. Haan, Myck, 2007, for a comparison of in-work tax credits in the two countries) the general progressive tax schedule is very similar in the two countries. Second the recent labor market performance in the two countries is similar, in particular female labor force participation in Germany is by now almost as high as in the UK (see OECD, 2013). And finally in the European context, Germany on the one hand and the UK on the other hand belong to the leading economies, which is why this comparison is particularly relevant.

A comparison of income tax schedules between countries is only meaningful if effective income tax liabilities across the distribution of gross incomes, as well as several socio-economic characteristics, are taken into consideration. For an empirical analysis of such effective tax liabilities, representative information on the distribution of the tax base is needed. Such analyses can be conducted using the European tax and benefit microsimulation model EUROMOD. Using EUROMOD, we compare effective average income tax rates for three groups of tax units, namely single households, one-earner married couples, and two-earner married couples, for Germany and the UK.

Merely looking at tax schedules and marginal tax rates (MTR) does not give a full picture of the actual tax liabilities. Comparison of tax burdens for the three groups between countries needs to be based on effective average tax rates (EATR). Applying the concept of effective tax rates, we go beyond marginal tax rates, as observed in the tax schedules, and account for effects of tax allowances, exemptions, deductions, tax credits as well as taxation rules for families, related to the number of children and marital status. For a comparable application of effective average tax rates, see, for example, Bach, Corneo, and Steiner (2012). Effective income tax progressivity in Germany is analyzed by Lang, Nörbaß, Stahl (1997). A comparative analysis of trends in national personal income tax systems across a large number of countries is conducted by Sabirianova Peter, Buttrick, and Duncan (2010). A comprehensive analysis of the entire tax-transfer treatment of couples in several European countries applying an older program version of EUROMOD is conducted by Immervoll et al. (2009).

There exists an important literature on the optimal taxation of married couples (e.g. Alesina et al., 2011, Boskin & Sheshinski, 1983, or Immervoll et al., 2011). In general these papers suggest that strict joint taxation of married households, which imposes higher marginal tax rates for the secondary earner, is not optimal. In contrast, given the higher elasticities of the secondary earner (in general the wife), Alesina et al. (2011), and Boskin and Sheshinski (1983) show that gender based taxation with lower marginal tax rates for the secondary earner would be optimal, i.e. more efficient in terms of labor supply. The academic discussion has in fact impacted the political debate. Even in Germany several parties now propose to abandon strict joint taxation with full income splitting and to gradually switching to individual tax schedules for married partners (see SPD, 2013, and Bündnis 90/Die Grünen, 2013).

Our empirical analysis complements the mainly theoretical literature by carefully demonstrating the different distributional effects of joint and individual taxation. Comparing effective tax liabilities for married couples between Germany and the UK, we find that tax burdens are, in general, significantly lower for the majority of tax payers in Germany, compared to the UK. The most important difference is found for one-earner married couples. This is largely because married couples are taxed jointly in Germany, whereas they are taxed like singles in the UK. Benefits of joint taxation are particularly large for one-earner couples, and they are found along almost the entire income distribution. Our results support the findings of previous studies that analyze the distributional and fiscal effects as well as the working incentives of joint taxation for married couples in Germany (Bach, Geyer, Haan, & Wrohlich, 2011, Dearing, Hofer, Lietz, Winter-Ebmer, & Wrohlich, 2007, or Steiner & Wrohlich, 2004).

Generally, the systems of taxation of married couples applied in the OECD countries today can be distinguished into three different approaches:

1. Individual taxation

2. Joint taxation with full income splitting

3. Joint taxation with limited income splitting

In the first system, the spouses are assessed individually for taxation, i.e. their incomes are taxed separately. Each of the spouses has an individual marginal tax rate that is independent from the other spouse’s income. This system is applied in the UK since 2001, which was when the UK abolished joint taxation of married couples. Before 2001, there was a Married Couples Allowance (MCA), under which an age-dependent tax-free allowance for the non-working spouses was granted. In 2000, this allowance was grandfathered in for couples with a spouse born before 1935, so that only these couples still benefit from the MCA today. See the EUROMOD country report for the UK for details (Sutherland, Tumino, & Zantomio, 2012).

In the second system, married couples are taxed jointly with full income splitting. The spouses sum up their individual incomes and divide them by two. The couple’s average taxable income is the tax base, and the resulting tax liability is then doubled. This averaging of incomes over spouses can significantly reduce the joint tax liability of the couple if there is relevant progressivity in the tax schedule. The tax relief compared to individual taxation is particularly great for couples with large income differences, in particular one-earner couples. It is smaller for two-earner couples, and it is zero in case both spouses have identical incomes. This system is applied in Germany for married couples (Ehegattensplitting) but not to cohabiting couples, which is why it can be labeled a marriage penalty (see Brewer et al., 2010).

In the third system, married couples are generally taxed jointly, but the extent to which their incomes are added up is limited. Systems two and three can potentially be extended to dependent children in the context of family taxation. In these systems, applied for example in France, children are also considered in the process of splitting taxable incomes across family members, i.e. incomes of children are assessable together with their parents’ incomes. The third system and the systems of family taxation are not subject to analysis in this paper. See Steiner and Wrohlich (2008), for simulations of these systems in the case of Germany.

This section introduces the microsimulation model and the data used for our analysis. Additionally, the grouping of tax units is described, and the definitions of taxable income and gross income in the context of effective tax liabilities are explained, as applied in the simulations.

A comparison of the income tax schedules for Germany and the UK reveals significant differences in both the schedule and in the resulting tax liability (Figure 1). The curves for the marginal tax rates differ greatly between the formula function in Germany and the step function in the UK. Figure 1 compares the schedules for individual taxation, i.e. singles and non-married couples in Germany.

While the British schedule starts after a tax-free allowance of about 7,475 euros⁸ of taxable income with a marginal tax rate of 20 percent on the first step, the German schedule increases after a tax-free allowance of 8,004 euros in a linear-progressive function relatively quickly, ranging from an initial tax rate of 15 percent up to a rate of more than 25 percent in the first progression zone (including the 5.5 percent solidarity surcharge)⁹. Thus, the marginal tax liability from the German schedule exceeds the liability from the British schedule at a taxable income of about 11, 500 euros.

Figure 1

Download asset Open asset

From there, the marginal liability in Germany continues to increase, whereas in the UK it remains constant at 20 percent until taxable income reaches about 39,000 euros; at which point the MTR in the UK jumps up to 40 percent. The rate at this income level is slightly higher in the UK than in Germany, because the German schedule is still in its second progression zone, which reaches, and maintains a tax rate of 44% at an income of about 53,000 euros. Thereafter, the schedule in Germany remains constant up to 250,000 euros taxable income. The UK tax rate increases to 50 percent¹⁰ at a taxable income of about 166,000 euros¹¹.

Thus, the UK’s tax rate is higher than Germany’s for all incomes above 166,000 Euros. This holds true, even for incomes greater than 250,000 Euros, the point at which Germany imposes a so-called “rich tax”, an add-on to the top MTR, such that the MTR for these top incomes lies at slightly above 47 percent.

As a result of these MTRs, the tax schedule in Germany already exhibits much greater progressivity in the lower tails of the income distribution than in the UK. This becomes obvious when comparing the curves for the average tax rates between the countries (Figure 1). The average tax rate is higher in Germany across almost all of income distribution. Only for high incomes, above a taxable income of about 325,000 euros, is the average tax liability greater in the UK.

This picture of the average tax rates resulting directly from the income tax schedules neglects several relevant tax allowances and deductions, which reduce the tax liability for many tax units, in particular those with higher incomes, and importantly, which significantly vary between the two countries (such as lump-sum amounts for work-related expenses, child allowances, or special expenses for old-age provision). The most relevant of these allowances and deductions are accounted for using EUROMOD (also see the previous section). This allows us to derive effective average tax rates by relating the income tax liabilities to the gross income instead of to the taxable income (where the latter would again result in the average tax rates plotted in the schedules in Figure 1).

In this section we present results on the effective tax liabilities from EUROMOD, comparing Germany with the UK. We start with empirical evidence across all groups of tax units, and then differentiate between the three groups of tax units as outlined in Section 2.1.

5.1 Across all taxpayers

When comparing effective average tax liabilities between Germany and the UK, across the distribution of gross incomes (Figure 2), a picture evolves that can only partly be explained by the tax schedules from Figure 1.

In particular, at the center of the income distribution, it is apparent that the effective tax liability in the UK is much greater than in Germany. The effective average tax rate around the median gross income in Germany (20,970 euros) is about 2.5 percent. For the UK, the effective average tax rate is 7 percent at the median income (17,276 euros)¹². At first sight, it may appear odd that the tax rate is so much higher in the UK. It should, however, be noted that employment incomes include all social security contributions (employee and employer). Social security contributions are much greater in Germany than in the UK. Also income-replacing benefits, such as contributory unemployment benefits, parental-leave benefits, or sickness benefits, are higher in Germany. Furthermore, pension income is fully taxable in the UK, whereas in Germany it is only partly taxable with a share depending on the retirement year, due to a long-term phasing in of full pension taxation. These are the main reasons for the significantly higher effective average tax rates at the lower and medium incomes. In the UK the effective average tax liabilities remain greater than in Germany until the 90th income percentile. At the 90th percentile, the effective average tax rate is about 15 percent in both countries.

Figure 2

Download asset Open asset

The picture changes for higher incomes. Starting at a gross income of about 55,000 euros, the effective liability in Germany exceeds the liability in the UK, and it remains about 3-4 percentage points higher until the very highest incomes. As a result, on average over all groups of tax units together, the effective average tax liability is somewhat greater for the 10 percent of highest incomes in Germany than it is for those in the UK. This also means that for about 90 percent of all taxable individuals, the effective tax liability in Germany is less than in the UK, and for the majority it is significantly lower. This stands in great contrast to the picture that evolves from the simple comparison of tax schedules (Figure 1), where the suggestion is that average tax liabilities are greater in Germany, across large parts of both the lower and central parts of the income distribution. However, we have not seen the entire picture, since the presentation for all groups is not yet detailed enough to understand which group is responsible for these differences. We will come to that in the following subsection.

5.2 Cross-country comparison by groups of taxpayers

If we now look at the differentiate picture of the effective tax liabilities by the three groups of tax units (Figure 3), an explanation for the observed picture between the two countries (Figure 2) appears. While in the UK, due to individual taxation rules for almost all tax units, effective tax liabilities are nearly identical for all three groups (Figure 4 in the Appendix), there is great heterogeneity across German groups, which is because married couples in Germany apply joint taxation (Figure 3).

Figure 3

Download asset Open asset

Figure 3 compares effective tax liabilities for the three groups in Germany and the average of all tax units in the UK. Among the three groups in Germany, effective average tax rates (EATR) are highest, already at the median income for the group of singles and individually-assessed couples (group 1 “Individually-assessed”). The significantly lower effective tax liabilities in Germany mainly stem from the inclusion of the tax-free employer social security contributions in the gross income concept. This remains until the 90^th percentile. Comparing this to the UK, while EATRs are somewhat lower in Germany for large parts of the income distribution (until 40,000 euros) and only slightly greater for higher incomes, the overall picture between the two countries is comparable for this group. This is not the case for jointly taxed couples in Germany.

For the group of Two-Earner Couples in Germany, the effective tax liability between the 30th income percentile and the 90th income percentile is, on average, up to 6 percentage points lower than for the Individually-assessed. This group consists in this area of the income distribution of couples with large income differences. They benefit from the regulations of joint taxation because they have different incomes. Beyond the 90th percentile, these benefits vanish for this group because the tax schedule is largely flat in this area, such that there are no benefits resulting from progressivity anymore. Among the top incomes, the EATR is thus almost identical for the Two-Earner Couples and the Individually-assessed.

For the group of One-Earner Couples in Germany, the beneficiary effect from joint taxation is even more prominent (Figure 3). This group has a significantly lower EATR than the Individually-assessed across the entire upper half of the distribution (and also slightly below the median); it is up to 7 percentage points lower. Below the median, One-Earner Couples have, at some incomes, higher EATRs than the Individually-assessed and the Two-Earner Couples. The line for One-Earner Couples in Figure 3 fluctuates slightly at low incomes. This is a composition effect, and it is caused by some extraordinary One-Earner Couples. If low-income earners who are close to retirement, and potentially work on reduced hours, have spouses who are already retired, these married couples are grouped among the One-Earner Couples. The individual incomes of these earners are below the tax-free allowance. Still, they have EATRs greater than zero because their spouses have pension income, which becomes since 2005 increasingly taxable (see page 12). As Figure 3 plots each spouse separately with their share in the joint taxable income, the spouses on low individual income are placed in the lower tails of the income distribution.

In general, this picture nicely points out the merits of joint taxation regulations for One-Earner Couples in Germany. These rules generate the greatest tax benefits for One-Earner Couples and couples with highly unevenly distributed incomes, in particular in the upper tails of the income distribution.

Figure 3 also reveals the reason for the significantly lower EATRs in Germany compared to the UK in the average picture across all groups, which we saw earlier in Figure 2. The Two-Earner Couples in Germany face lower EATRs, up to 8 percentage points lower, than in the UK until the 90th income percentile. One-Earner Couples in Germany also face a lower EATR of up to 8 percentage points across almost the entire income distribution. For Two-Earner Couples in Germany the splitting effect of joint taxation vanishes at about the 90th income percentile because the tax rate is flat in this area, while for the One-Earner Couples these benefits remain until the 99th percentile, where the EATR is still almost 5 percentage points lower.

We show that Germans face higher effective tax liabilities when compared to UK citizens for two of the three groups analyzed. These are the Individually-assessed tax units and the jointly-assessed Two-Earner Couples tax units, but this is only true at the upper tails of the income distribution, beyond about 50,000 euros gross income. However, in this part of the income distribution only about 15 percent of all tax units are affected by these higher effective tax rates. For the majority of Individually-assessed tax units and jointly-assessed Two-Earner Couples, effective tax liabilities in Germany are not greater than in the UK, and for the latter group in the center and the upper part of the income distribution they are even slightly lower. Moreover, One-Earner Couples in Germany face significantly lower effective tax liabilities than in the UK, across almost the entire range of the income distribution. This is counter to the picture gained from a quick look at the tax schedules, where effects of joint taxation rules are not apparent.

Thus, it can be concluded that a proper empirical cross-country comparison of the effective income tax liabilities should account for the differentiated tax units and the effective tax rates imposed on them. In the case of Germany, the analyzed groups of tax units are defined by individual and joint taxation regulations. This is because in Germany joint taxation rules for married couples heavily distort the picture of tax liabilities across the income distribution, whereas in countries with pure individual taxation, such as the UK, this is not the case.

The merits of joint taxation for married couples have long been the subject of political debates in Germany. Moreover there is empirical evidence that these tax benefits for one-earner married couples are responsible, among others, for the relatively low participation of women in Germany’s labor force, as joint taxation significantly increases the marginal tax rates of married women, thereby lowering their incentives to work (Steiner & Wrohlich, 2004).

A conclusive policy recommendation would thus be to significantly reform joint taxation rules for married couples in Germany. This would significantly enhance female labor force participation (Bach et al., 2011). By completely abolishing joint taxation rules, or at least largely limiting the tax benefits for married couples, income tax revenue could be significantly raised. The additional revenue could be used to lower marginal tax rates at the lower end of the income distribution. The effects of bracket creep in the tax schedule, which result from nominal income increases over time and the progressivity of the tax schedule, could be mitigated on top of that (Bach, 2012).

1. 1
   Gender-Based Taxation and the Division of Family Chores

   1. A Alesina
   2. A Ichino
   3. L Karabarbounis

   (2011)

   American Economic Journal: Economic Policy 3:1–40.

   • Google Scholar
2. 2
   Abbau der kalten Progression: Nicht die einzige Herausforderung beim Einkommensteuertarif. DIW Wochenbericht No.12/2012

   1. S Bach

   (2012)

   DIW Berlin: German Institute for Economic Research.

   • Google Scholar
3. 3
   Effective Taxation of Top Incomes in Germany

   1. S Bach
   2. G Corneo
   3. V Steiner

   (2012)

   German Economic Review 14:115–137.

   • Google Scholar
4. 4
   Reform des Ehegattensplitting: Nur eine reine Individualbesteuerung erhöht die Erwerbsanreize deutlich. DIW Wochenbericht No.41/2011

   1. S Bach
   2. J Geyer
   3. P Haan
   4. K Wrohlich

   (2011)

   DIW Berlin: German Institute for Economic Research.

   • Google Scholar
5. 5
   Optimal tax treatment of the family: Married couples

   1. M Boskin
   2. E Sheshinski

   (1983)

   Journal of Public Economics 20:281–297.

   • Google Scholar
6. 6
   The Mirrlees Review: Dimensions of Tax Design

   1. M Brewer
   2. E Saez
   3. A Shephard

   (2010)

   Means-testing and tax rates on earnings, The Mirrlees Review: Dimensions of Tax Design, eds, The Institute for Fiscal Studies, London.

   • Google Scholar
7. 7
   Zeit für den grünen Wandel

   1. Bündnis 90/Die Grünen

   (2013)

   Teilhaben, Einmischen, Zukunft schaffen. Wahlprogramm. Beschlossen auf der Bundesdelegiertenkonferenz vom, 26–28. April 2013, in, Berlin.

   • Google Scholar
8. 8
   The distributional effects of austerity measures: A Comparison of six EU countries. EUROMOD Working Paper Series: EM 06/11

   1. T Callan
   2. C Leventi
   3. H Levy
   4. M Matsaganis
   5. A Paulus
   6. H Sutherland

   (2011)

   The distributional effects of austerity measures: A Comparison of six EU countries. EUROMOD Working Paper Series: EM 06/11.

   • Google Scholar
9. 9
   Why are mothers working longer hours in Austria than in Germany? A comparative microsimulation analysis

   1. H Dearing
   2. H Hofer
   3. C Lietz
   4. R Winter-Ebmer
   5. K Wrohlich

   (2007)

   Fiscal Studies 28:463–495.

   • Google Scholar
10. 10
    Automatic stabilizers and economic crisis: US vs. Europe

    1. M Dolls
    2. C Fuest
    3. A Peichl

    (2012)

    Journal of Public Economics 96:279–294.

    • Google Scholar
11. 11
    Measuring Income in Household Panel Surveys for Germany: A comparison of EU-SILC and SOEP. SOEP Papers

    1. J Frick
    2. K Krell

    (2010)

    Measuring Income in Household Panel Surveys for Germany: A comparison of EU-SILC and SOEP. SOEP Papers, DIW Berlin.

    • Google Scholar
12. 12
    Apply with Caution: Introducing UK-Style In-Work Support in Germany

    1. P Haan
    2. M Myck

    (2007)

    Fiscal Studies 28:43–72.

    • Google Scholar
13. 13
    Average and marginal effective tax rates facing workers in the EU: A micro-level analysis of levels, distributions and driving factors. EUROMOD Working Paper Series. No. EM6/04

    1. H Immervoll

    (2004)

    Average and marginal effective tax rates facing workers in the EU: A micro-level analysis of levels, distributions and driving factors. EUROMOD Working Paper Series. No. EM6/04.

    • Google Scholar
14. 14
    Welfare reform in European countries: A microsimulation analysis

    1. H Immervoll
    2. H J Kleven
    3. CT Kreiner
    4. E Saez

    (2007)

    The Economic Journal 117:1–44.

    • Google Scholar
15. 15
    An Evaluation of the Tax-Transfer Treatment of Married Couples in European Countries. EPRU Working Paper Series 08-03, Economic Policy Research Unit (EPRU)

    1. H Immervoll
    2. H J Kleven
    3. CT Kreiner
    4. N Verdelin

    (2008)

    University of Copenhagen. Department of Economics.

    • Google Scholar
16. 16
    Optimal tax and transfer programs for couples with extensive labor supply responses

    1. H Immervoll
    2. H J Kleven
    3. CT Kreiner
    4. N Verdelin

    (2011)

    Journal of Public Economics 95:1485–1500.

    • Google Scholar
17. 17
    On Income Tax Avoidance: The Case of Germany

    1. O Lang
    2. K-H Nörbaß
    3. K Stahl

    (1997)

    Journal of Public Economics 66:327–347.

    • Google Scholar
18. 18
    EUROMOD Country Report for Germany, 2007–2010

    1. R Ochmann
    2. F Fossen

    (2011)

    ISER at the University of Essex.

    • Google Scholar
19. 19
    Taxing Wages 2011–2012 - Special Feature: Average personal income tax rate and tax wedge progression in OECD countries

    1. OECD

    (2013)

    OECD Publishing.

    • Google Scholar
20. 20
    Wie progressiv ist Deutschland? Das Steuer- und Transfer system in Europāischen Vergleich“. EUROMOD Working Paper No. EM 01/08

    1. A Peichl
    2. T Schafer

    (2008)

    Wie progressiv ist Deutschland? Das Steuer- und Transfer system in Europāischen Vergleich“. EUROMOD Working Paper No. EM 01/08.

    • Google Scholar
21. 21
    Global Reform of Personal Income Taxation, 1981-2005

    1. K Sabirianova Peter
    2. S Buttrick
    3. D Duncan

    (2010)

    National Tax Journal 63:447–478.

    • Google Scholar
22. 22
    Das Wir entscheidet. Das Regierungsprogramm 2013-2017. Beschlossen am 11. April in Augsburg

    1. SPD

    (2013)

    Das Wir entscheidet. Das Regierungsprogramm 2013-2017. Beschlossen am 11. April in Augsburg.

    • Google Scholar
23. 23
    Household taxation, income splitting and labor supply incentives. A microsimulation study for Germany

    1. V Steiner
    2. K Wrohlich

    (2004)

    CESifo Economic Studies 50:541–568.

    • Google Scholar
24. 24
    Introducing Family Tax Splitting in Germany: How Would It Affect the Income Distribution, Work Incentives, and Household Welfare?

    1. V Steiner
    2. K Wrohlich

    (2008)

    Fiscal Studies 64:115–142.

    • Google Scholar
25. 25
    EUROMOD: the European Union tax-benefit microsimulation model

    1. H Sutherland
    2. F Figari

    (2013)

    International Journal of Microsimulation 6:4–26.

    • Google Scholar
26. 26
    EUROMOD Country Report for the United Kingdom, 2006–2011

    1. H Sutherland
    2. A Tumino
    3. F Zantomio

    (2012)

    ISER at the University of Essex.

    • Google Scholar
27. 27
    Redistributive effect and progressivity of taxes: An international comparison across the EU using EUROMOD. EUROMOD Working Paper No. EM 05/04

    1. G Verbist

    (2004)

    Redistributive effect and progressivity of taxes: An international comparison across the EU using EUROMOD. EUROMOD Working Paper No. EM 05/04.

    • Google Scholar
28. 28
    Generierung der Einkommensvariablen in der deutschen EU-SILC-Erhebung LEBEN IN EUROPA. Wirtschaft und Statistik 2/2012

    1. J Vitlina
    2. U Sikorski
    3. H Minkel

    (2012)

    German Federal Statistical Office.

    • Google Scholar

Author details

1. Stefan Bach

   Germany Institute for Economic Research (DIW Berlin), Berlin

   For correspondence

   sbach@diw.de

2. Peter Haan

   Germany Institute for Economic Research (DIW Berlin), Berlin

   For correspondence

   phaan@diw.de

3. Richard Ochmann

   Germany Institute for Economic Research (DIW Berlin), Berlin

   For correspondence

   rochmann@diw.de