Florian Blank, Camille Logeay, Erik Türk, Josef Wöss and Rudolf Zwiener
The figures are compelling: Austrian men who retired in 2013 after paying social security contributions throughout lengthy employment careers received an average net pre-tax state pension of €1,557, paid 14 times a year.1 Meanwhile, recently retired German men with a similar employment history have to make do with an average of €913 a month (Figure 1).2 There is also a large disparity in the figures for women, for whom there are also differences compared to men in terms of the number of years of contributions. OECD projections for people currently starting their first job – based on ideal “textbook” employment careers – forecast that average earners in Austria who remain employed for 45 years will receive gross pensions equivalent to 78.1% of their average earnings, whereas in Germany they will receive just 37.5% (state pensions only).3
Better retirement provision for Austrians
1 The figures for Austria include the two additional (13th and 14th) monthly payments.
Source: WSI/IMK/AK Wien, Graphic: bit.do/impuls0209 Data: bit.do/impuls0210.
Based on the projected contribution rates and particularly the trend in non-wage labour costs, the pension reforms introduced in Germany around the turn of the millennium sought to reduce the benefit level of state pensions and rely on funded private pensions to make up the shortfall. However, this system has since become the target of sharp criticism, since Germany’s pension system now often fails to provide people with a secure retirement.4 While Austria’s state pension system has also undergone numerous reforms, these reforms have actually further developed and stabilised state pensions. Private and occupational pensions play only a secondary role in the Austrian government’s policy strategy.
One of the central arguments in the German pension reform debate was that rising contributions to state pensions and the higher labour costs that they entail would be detrimental to German industry in particular and to employment developments in general.5 However, this paper will use the example of Austria to question this hypothesis of a trade-off between sustained economic growth and guaranteeing people a decent standard of living in their retirement within a pay-as-you-go (PAYG) system against the backdrop of an ageing population.
A comparison of Germany and Austria is of particular interest because it can be understood as a kind of “natural experiment”. The two countries share common geographical borders and very similar cultures, have similarly structured social security systems, are EMU members, and have open economies. The development of the Austrian system can be seen as an answer to the question of what would have happened if Germany’s pension reforms had not abandoned the goal of protecting people’s standard of living. As such, this analysis complements other studies on this topic.6 The article is structured as follows: first, the different reform paths are described, especially from the 2000s onwards. Next, the first part of the hypothesis – performance of the pension systems – is addressed, followed by the second part – economic growth performance. The last section contains our economic policy recommendations.
Pension reforms in Germany and Austria: opposite paths
At the start of the new millennium, Germany’s pension system underwent a fundamental reform comprising a number of different measures.7 The laws enacted by the coalition government of the Social Democratic Party and the Greens abandoned the goal of protecting people’s standard of living during retirement. This was done in order to curb the projected rise in the contribution rate in the context of an ageing society, keeping it to no more than 22% (through 2030). The reason for doing so was the desire to achieve a reduction and long-term stabilisation of businesses’ labour costs with a view to strengthening German industry and limiting supposed harmful effects on labour demand in general. Accordingly, the pension adjustment formula was modified in order to bring about a progressive reduction in pension benefit levels. The legislation stipulates a minimum level of 43% (net before tax) for 2030; in 2001 the corresponding level was 52.6%. The government’s plan was that people would use private or occupational pension schemes to make up the resulting shortfall in their pension provision. In other words, the idea was for funded private pensions to partially replace the state PAYG system. The decision to adopt this policy was undoubtedly influenced by the fact that, at the time, there was a widespread assumption that investing in the capital markets could deliver higher returns than the PAYG system, especially against the backdrop of an ageing population. Based on these envisioned efficiency gains, the use of private and occupational pensions was promoted through taxpayer-funded subsidies. Other reforms affecting the PAYG system were also passed, for instance the progressive taxation of state pensions and the raising of the retirement age from 65 to 67. The result of all these reforms was a significant decline in the state pension scheme’s replacement rate. Our focus in this paper is on whether the promotion of private pension schemes was in fact necessary to sustain economic growth. This is where the comparison with Austria comes in.
Austria’s pension system also underwent numerous reforms, but all of them were related to the PAYG system.8 While the introduction of the new “pension account” legislation in 2005 did involve substantial changes to the system, the fundamental goal of ensuring that the state pension system protects people’s standard of living remained in place. The target for state pensions in Austria is based on the 80/45/65 rule: an 80% gross replacement rate9 for people who have paid 45 years of social security contributions and who retire at 65.
Today, the Austrian state pension system can be said to encompass the entire working population. Compulsory contributions to the state pension system were introduced for the majority of self-employed people as long ago as 1958. Systematic measures were subsequently taken to close the loopholes, meaning that virtually all self-employed people now have to pay compulsory contributions. In addition to these state pension reforms, civil servants’ pensions were gradually brought into line with the state pension system regulations in the 2000s. The system also provides for a taxpayer-funded, means-tested minimum income for pensioners, known as the Ausgleichszulage. Means-tested social welfare benefits are provided to people with insufficient income and no pension entitlement. In Germany the only assistance available to both groups – i.e. people with no pension entitlement and people whose pension income is too low – are the means-tested social welfare benefits known as the Grundsicherung im Alter. This benefit is less generous than Austria’s Ausgleichszulage both in terms of its level and its qualification criteria.
Replacement rates are very low in Germany compared to other countries
The differences in the reforms implemented by the two countries are reflected in the levels of their state pensions. The OECD publishes regular forecasts of future pension levels in its member countries.10 The projections are based on current pension regulations, including reforms that have been approved but not yet implemented, as well as countries’ tax and social security contribution profiles. They thus provide a means of comparing the pensions that a “textbook” young employee might expect to receive based on current legislation.
In the 2015 OECD report, both the pension level and the replacement rates for Germany were revised down significantly compared to the 2013 report.11 Germany’s deferred taxation arrangements are included in the calculations for the current report for the first time. The report found Germany to have very low replacement rates compared to other countries, whereas Austria is among the top-ranked countries. While the pension level for future “average earners” in Austria is projected to be 78% gross and 92% net, the corresponding values for Germany are 37.5% gross and 50% net.
Table 1 highlights the key figures relating to contribution-based state pensions in Germany and Austria. Up to the income threshold,12 there is no difference in the gross replacement rates for the “textbook” person that the calculations are based on.
Gross replacement rates in the OECD pension model (compulsory state systems only)
|Men and women2||0.5||1.0||1.5||0.5||1.0||1.5|
|Gross replacement rate3||37.5||37.5||37.5||78.1||78.1||77.6|
|Net replacement rate4||53.4||50.0||49.0||92.1||91.6||88.9|
|Gross replacement rate3||69.9 (69.7)||59.0 (58.8)||54.4 (54.2)||64.8 (64.4)||52.7 (52.3)||47.5 (47.1)|
|Net replacement rate4||80.7 (80.4)||70.9 (70.7)||66.4 (66.2)||74.5 (74.1)||63.0 (62.6)||58.2 (53.6)|
1 Multiple of average earnings across the economy as a whole. 2 In Germany and Austria, there is no difference between replacement rates for men and women. Across the EU28 and OECD34, small differences are found. 3 % individual average gross earnings. 4 % individual average net earnings.
Source: OECD: Pensions at a Glance 2015 – OECD and G20 Indicators, Paris 2015, Tables 6.1 and 6.7.
While these statistics do tell us something about the relative performance of different pension systems, they are of very limited use for predicting real future pension replacement rates. Even the very low gross replacement rate of 37.5% projected for Germany – which is equivalent to the gross pension level for average earners – is unlikely to be achieved by a lot of people. Many employees do not actually pay social security contributions for the full 45 years. In addition, people generally earn significantly less at the start of their careers and more in the years leading up to their retirement. Accordingly, even if a person earned an average income across their working life as a whole, the actual gross and net replacement rates – based on their final salary prior to retirement – could be up to ten percentage points lower than the figure produced by calculations based on them earning the same average income each year.13
The total replacement rate in Germany remains significantly lower than in Austria even if the calculations include the assumption that the “textbook” individual will take out a private pension.14 Even if we accept the OECD assumptions of a high nominal net rate of return of five per cent in the accumulation phase and a constant four per cent contribution of a person’s income to their supplementary private pension throughout their working life, the 50% gross replacement rate achieved through the combination of state and private pensions in Germany remains well below the Austrian rate of 78%.
There are further arguments as to why the difference between Germany and Austria regarding the OECD replacement rate is likely to be understated: welfare benefits such as survivors’ pensions, disability pensions and rehabilitation benefits are not provided – at least not to the same level – by private pensions; the assumption of a nominal net rate of return of five per cent is unrealistically high;15 and experiences with the entirely employee-financed private Riester pension have so far been disappointing owing to their high costs, lack of product transparency, falling returns and inadequate coverage.16
The reasons for Austria’s high pension levels and their sustainability
How can we explain the fact that pension levels in the Austrian system are currently so much higher and are projected to remain so in the future? Two of the key variables are the contribution rate and the ratio of contributors to beneficiaries. This ratio is influenced both by the retirement age and by demographic and labour market trends. The higher replacement rates in Austria do indeed partially reflect a higher contribution rate (22.8% compared to 18.7% in Germany).17 This also makes it possible to pay higher state pensions. State contributions to the pension system are a further factor, although their share of pension expenditure is in fact similar in Germany and Austria.18
The demographics are also more favourable in Austria. In 2015 the dependency ratio – the ratio of over-65s to those aged 20-64 – was 30.3% in Austria and 35.3% in Germany. Nevertheless, demographic trends and the ageing of a country’s population can only serve as a rough guide. The key factor for the pension system is the ratio of contributors to beneficiaries. Ultimately, in order to achieve a sustainably funded pension system, the number of people in (well-paid) jobs with mandatory social security contributions needs to be as high as possible. However, this ratio is not included in the OECD report.
However, the European Commission’s Ageing Report forecasts that the rise in the ratio of pensioners to contributors in Austria will remain significantly below that in Germany, despite Austria continuing to have a lower statutory retirement age.19 In addition to the slightly higher projected rise in employment rates for Austria, this can mainly be put down to the pronounced decline expected to occur in Germany’s working age population.
The sharp increase in the number of elderly people as a percentage of the total population means that spending on pensions is set to rise in Austria, too. The European Commission estimates that government pension expenditure in 2013 amounted to 13.9% of GDP. This figure is expected to climb to 14.7% by around 2040. Thereafter, the proportion of GDP spent on pensions is forecast to level off, falling slightly again by 2060. This equates to a 0.5 percentage point rise over the forecast period as a whole. Consequently, the European Commission rates the trend for Austria as “broadly stable”. This modest rise is significantly lower than the European Commission’s forecast for Germany of a 2.7 percentage point increase to 12.7% of GDP by 2060.20 Moreover, the figures for Austria include the minimum income payments for pensioners and are based on higher coverage ratios. It is important to bear in mind that these calculations only cover spending on state pensions. In other words, the figures for Germany significantly underestimate the total expenditure, since they do not include private and occupational pensions.21
The fact that civil servants’ pensions in Austria were already brought into line with the state pension system regulations several years ago will substantially reduce the burden of funding the future pension system. Of the 13.9% of GDP spent on pensions in 2013, a quarter went to civil servants’ pensions. The latest spending forecasts indicate that expenditure on civil servants’ pensions will fall from its current level of 3.5% of GDP to just 0.9% by 2060.22
Simulations for Germany calculate that changing over to a pension system that encompasses the entire working population would have positive long-term effects.23 The introduction of compulsory social security contributions for civil servants and the self-employed would allow the net before-tax pension level to be kept at the 2000 level of 52.6% until 2060, while contributions to the state pension scheme would not rise above 26%. On the other hand, failure to implement this reform would result in the pension level falling to around 41% and the contribution rate rising to almost 27.5% by 2060.
High replacement rates have not impaired economic growth
The justification given for the German pension reforms that led to a reduction in the level of state pensions was that contributions were “too high” and could threaten Germany’s competitiveness.24 However, a comparison of the two countries reveals that Austria’s economy has actually outperformed Germany’s in many respects over the past 16 years. Austria’s GDP rose by 22% in real terms between 2000 and 2015, whereas Germany’s rose by just 18% over the same period.25 Productivity and the employment rate (both measured in hours) have also grown more strongly in Austria than in Germany (Figure 2).
Economic development and competitiveness indicators for Germany, Austria and the eurozone
Note: Eurozone 12 countries include the 12 eurozone members as of 2001: Germany, Austria, the Netherlands, Finland, Luxembourg, Belgium, France, Ireland, Spain, Italy, Portugal, Greece.
Sources: Eurostat; AMECO; authors’ own calculations.
Labour costs in Austria have increased more than in Germany. Hourly labour costs in the private sector are now more or less the same in both countries.26 Data from the OECD’s “Taxing Wages” database show that the higher rate of pension contributions in Austria forms part of a social security contribution system in which employers pay significantly higher contributions than employees and indeed than employers in Germany.27
The convergence of labour costs in the two countries since the start of EMU can be largely attributed to the fact that wage increases in Germany have remained below those in Austria. The labour market and social security reforms introduced in Germany at the beginning of the last decade were primarily aimed at reducing labour costs. These measures had the effect of stifling domestic demand in Germany – even though price inflation was higher in Austria, private consumption in Germany rose by just 11% in real terms over the 16-year period, compared to 17% in Austria.28 Economic growth in Germany was almost entirely driven by exports, which more than doubled in the period in question. However, Austria also achieved real growth of 82%. The upshot is that Germany has accrued a current account surplus that now stands at almost nine per cent of GDP, whereas Austria’s current account surplus has remained virtually unchanged at between two and four per cent. In the final analysis, Austria has chosen to maintain a strong welfare state and has opted not to impose government restrictions on wage increases. Meanwhile, Germany has cut back its welfare state whilst at the same time substantially reducing the contribution burden for businesses and people with high incomes.29 The economic data indicate that Austria’s policies have delivered better results. Moreover, our interpretation of both countries’ economic development is ultimately backed up by the change in German economic policy undertaken in the wake of the major recession of 2008-09. This involved taking active measures over a period of time to stabilise demand, resulting in a return to stronger wage growth and the normalisation of employment and economic growth trends.30
Economic policy recommendations
Despite numerous reforms, Austria’s pension system continues to be based on state pensions to which employers contribute more than their employees. In Germany, on the other hand, the state pension replacement rate has been and continues to be cut in a bid to reduce non-wage labour costs. One important consequence of this is the transfer of part of the responsibility for pensions and their funding to private individuals who are now expected to provide for their retirement themselves. While private pension plans (e.g. the Riester pension) do not normally benefit from employer contributions, joint funding systems are still widespread among occupational pension schemes. The new occupational pensions financed by employees through earnings conversion have – at least to some extent – been exposed as a cost-cutting measure for businesses, since they enable many companies to get out of paying employer social security contributions. The earnings conversion approach also results in lower pensions.31 To all intents and purposes, Germany has been left with a system that has abandoned the goal of protecting people’s standard of living. In the future, even average earners who remain in full-time employment throughout their working lives will have to work for much longer in order to obtain a state pension high enough to meet their basic subsistence needs.32
The comparison of Germany and Austria reveals that there was in fact no economic imperative for Germany to cut back state pensions and promote funded private pensions as a partial substitute. Since Germany was reasonably competitive compared to other countries at the start of the 21st century, there was no need to free employers from the obligation to match their employees’ contributions. The view that the financial burden of the welfare state was the main cause of the labour market depression of the 2000s is not supported by the comparison of macroeconomic data with Austria.
Consequently, any suggestion that the German government should promote private top-up pensions even more strongly in the future should be regarded as highly suspect, especially since the current system has not been systematically evaluated. A statutory evaluation requirement similar to those that already exist for labour market policy instruments is essential for a system that profits from government subsidies. In its absence, it is likely that the benefit of any doubt will favour the interests of businesses (including pension providers) rather than those of beneficiaries. In fact, what the German government really needs to do is contemplate gradually raising the level of state pensions again. In the meantime, government support for private pensions should be scaled back. Austria’s track record demonstrates that better results are achieved if government revenue is used to strengthen and secure state pensions.
At present, it would appear that occupational pensions are a valuable means of topping up the state pension in Germany to a level that provides workers with the necessary degree of protection. This is especially true because collectively organised occupational pensions are more efficient than purely private pension plans. However, there is considerable variation among different industries and companies in terms of the availability and implementation of such pensions. Unlike in Germany, Austrian employers are obliged to cover at least 50% of the cost of occupational pensions. A similar requirement for employers to contribute at least 50% to the funding of occupational pensions in Germany would also provide the basis for a sustainable earnings conversion system, as long as it was not used as a pretext for reducing the level of state pensions. However, it is still necessary to ask whether capital market risks, possible extra costs arising from the involvement of private pension providers, regulatory requirements and distribution problems mean that it would nevertheless be better and cheaper to focus on strengthening state pensions. The option of a state pension system encompassing the entire working population should also be closely examined in Germany. It would be necessary to clarify exactly what the appropriate transitional measures would look like and how funding and protection priorities could be squared with each other.
Finally, it is important to remember that any assessment of the total economic cost of pension provision needs to take the total amount of money spent on pensions into account. The Austrian system, with its higher contributions, only appears to be significantly “more expensive” if the costs of private and occupational pensions in Germany are not included in the calculations. A pension system’s future viability also depends on the existence of a consensus regarding both the benefits that the system should deliver and its overall cost. There is no definitive measure of the maximum acceptable size for a state pension system. Austria has comfortably outperformed Germany on a range of economic indicators. It provides a model for how the organisation of retirement provision in Germany could be significantly improved by extending the state pension system to encompass the entire working population, ensuring high pension levels and providing a taxpayer-funded, means-tested top-up allowance for people with low pensions.
The more the level of state pensions falls in the future, the more young people will be forced to provide for their own retirement by taking out private pensions that do not benefit from employer contributions. In doing so, they will be obliged to rely on a pension industry that is highly driven by commissions and focused on profits as well as on financial markets, whose unpredictable returns could possibly be low for many years. Rather than being a question of young versus old, pension reforms should be about creating fair conditions in order to ensure that young and old alike are well provided for in their retirement.
- 1 Authors’ calculations based on Hauptverband der österreichischen Sozialversicherungsträger: Statistische Daten aus der Sozialversicherung: Pensionsversicherung Berichtsjahr 2013, Vienna 2014, Table 32 (PV der Unselbständigen. Erstmalige Pensionsneuzuerkennungen gegliedert nach Pensionsarten).
- 2 Deutsche Rentenversicherung Bund: Rentenversicherung in Zeitreihen, Berlin 2015, p. 124.
- 3 OECD: Pensions at a Glance 2015 – OECD and G20 Indicators, Paris 2015, Table 6.1.
- 4 W. Schmähl: Von der Rente als Zuschuss zum Lebensunterhalt zur “Zuschuss-Rente”, in: Wirtschaftsdienst, Vol. 92, No. 5, 2012, pp. 304-313.
- 5 G. Schröder: Regierungserklärung des Bundeskanzlers “Mut zum Frieden und zur Veränderung” (“Agenda 2010”), 14 March 2003.
- 6 C. Logeay, V. Meinhardt, K. Rietzler, R. Zwiener: Macroeconomic Consequences of the Funded Pension System – Illusions and Realities, IMK Report No. 43e, November 2009.
- 7 For detailed information on the pension reforms in both countries and on the Austrian state pension system, see F. Blank, C. Logeay, E. Türk, J. Wöss, R. Zwiener: Alterssicherung in Deutschland und Österreich: Vom Nachbarn lernen?, WSI-Report No. 27, Düsseldorf 2016. Chapters 2 and 3 and Appendix 2 are particularly relevant here.
- 8 Ibid. Chapter 3 provides an overview of the transitional rules.
- 9 Based on average lifetime earnings.
- 10 OECD: Pensions at a Glance 2015 …, op. cit.; OECD: Pensions at a Glance 2013 – OECD and G20 Indicators, Paris 2013; F. Blank et al., op. cit., Chapter 4.
- 11 OECD: Pensions at a Glance 2015 ..., op. cit.
- 12 Approximately 1.55 times the average full-time earnings in Germany and 1.49 times in Austria. See OECD: Pensions at a Glance 2015 ..., op. cit. for current OECD values.
- 13 V. Meinhardt: Modellrechnungen zur Bestimmung der Alterseinkünfte auf der Basis von Erwerbsverläufen, IMK Study No. 36, Düsseldorf 2014.
- 14 OECD: Pensions at a Glance 2015 ..., op. cit., Tables 6.4 and 6.10.
- 15 E. Türk, D. Mum: Weit überzogene Renditeerwartungen in der kapitalgedeckten Alterssicherung. Warum die OECD und die Europäische Kommission ihre Renditeannahmen deutlich nach unten korrigieren sollten, in: Wirtschaft und Gesellschaft, Vol. 41, No. 2, 2014, pp. 257-274.
- 16 H. Joebges, V. Meinhardt, K. Rietzler, R. Zwiener: On the Path to Old-Age Poverty, Assessing the Impact of the Funded Riester Pension, IMK Report No. 73, Düsseldorf 2012; F. Blank: Die Riester-Rente – Überblick zum Stand der Forschung und sozialpolitische Bewertung nach zehn Jahren, in: Sozialer Fortschritt, Vol. 60, No. 6, 2011, pp. 109-115.
- 17 It should be noted that Germans are expected to invest a further four per cent of their gross wages in private pension schemes. This means that the overall pension effort today is already close to the Austrian contribution rate.
- 18 F. Blank et al., op. cit., Table 3.
- 19 European Commission: The 2015 Ageing Report, Economic and budgetary projections for the 28 EU Member States (2013-2060), European Economy 3/2015, Brussels 2015.
- 20 Ibid.
- 21 Assessments of pension systems’ financial sustainability should only be based on robust long-term projections such as the ones published by the European Commission. The “pension indices” published by funded private pension product providers or consulting firms that are closely associated with them do not constitute a robust basis for this type of assessment.
- 22 A large part of this reduction is due to the transfer of spending to the state pension system owing to the major changes in the structure of the working population. An equally large part is due to the measures to bring civil servants’ pensions in line with the state pension system regulations.
- 23 M. Werding: Alterssicherung, Arbeitsmarktdynamik und neue Reformen: Wie das Rentensystem stabilisiert werden kann, Studie im Auftrag der Bertelsmann Stiftung, Gütersloh 2013.
- 24 G. Schröder, op. cit.
- 25 All of the following figures are based on Eurostat national accounts data (quarterly, seasonally and calendar adjusted).
- 26 A. Herzog-Stein, H. Joebges, T. Niechoj, U. Stein, R. Zwiener: German labour costs have risen only moderately. European comparison of trends in labour and unit labour costs in 2014 and the first two quarters of 2015, IMK Report No. 109e, January 2016. This assertion is based on the most recent available data for labour costs in industry from 2014, when the corresponding figures were €31.80/hr in Germany and € 31.70/hr in Austria.
- 27 OECD: Taxing Wages 2016, Paris 2016, OECD Publishing.
- 28 As in the case of real GDP, the gap between the two countries for this indicator reached a high of nine per cent in 2011 and has been shrinking ever since.
- 29 A. Herzog-Stein, F. Lindner, R. Zwiener: Is the supply side all that counts? How Germany’s one-sided economic policy has squandered opportunities and is damaging Europe, IMK Report No. 87e, November 2013.
- 30 It was only in 2014 that the number of hours worked by people in employment surpassed the previous high from 2000 on a long-term basis.
- 31 W. Schmähl, A. Oehlschläger: Abgabenfreie Entgeltumwandlung aus sozial- und verteilungspolitischer Perspektive, Berlin, Münster 2007, LIT Verlag; F. Blank: Die betriebliche Altersversorgung durch Entgeltumwandlung: Regulierung, Verbreitung und verteilungs–politische Aspekte, in: Vierteljahreshefte zur Wirtschaftsforschung, Vol. 83, No. 4, 2014, pp. 129-142.
- 32 J. Steffen: „Fürsorge Break-even“ der gesetzlichen Rente, Arbeitnehmerkammer Bremen, 4 August 2011.