Reforming the reform
The criticism of over-complication and bureaucracy is the one most frequently levelled at German pensions and savings. As one result of that complexity, pickup on the so-called Riester-products in the newly established third pillar of German pensions has been rather low so far with only about 3m contracts signed after one year. As a result, the government is in the midst of a discussion on how to reform the system again, where a proposal is expected to be brought out this fall. In an effort to sail round the complex details of the German system, the current debate can be summarised with five simple questions:
Question 1: Do Germans have to save more for financing their pensions and cope with the expected lowering of payments from the state system? While the answer to that question may be quite obvious to most Anglo-Saxons, it is only since the last reform by former secretary of labour Walter Riester as of 2001 that parties in Germany would unanimously answer with a clear ‘Yes’. This change of paradigm can be qualified as the biggest achievement of Riester and his reform.
Question 2: Why have the so-called Riester-products not been successful in the market? Answer: While the current state of the economy and its impact on incomes and the propensity to save may certainly have played a role, the high number of product requirements is probably the biggest hurdle to acceptance. On the side of the investor, the products lack transparency, and in particular the requirement to have a large portion of the payouts in form of annuities led people to continue to prefer more flexible products, like plain vanilla life insurance contracts. Also, the limitation of contributions to E525 per year make Riester-products a negligible form of savings to deal with, for investors as well as for sales people on the provider side.
Question 3: Why is the system so complicated, and, why are different pensions vehicles each treated differently in product requirements and in taxation? The German system encompasses five different vehicles in occupational pensions (Durchführung-swege), plus the Riester-products, plus life insurance and annuities in the third pillar, and many variations and combinations thereof like deferred compensation arrangements and work-overtime savings (Zeitwertmodelle), and each treated differently. But, to make a long answer short, these regulations have been developed over more than 30 years of smaller pension reforms, and cleaning them up should thus be at the core of the expected ‘reform of reforms’.
Question 4: Would it make sense to include other forms of savings, like wealth accumulation savings (Vermoegensbildung), savings for own housing (Bausparzulagen), and company stock investments (Vermögensbeteiligungen), which all have been put in a pensions context at some point in the past? Answer: If the promotion of pension saving has the highest priority, integrating these other savings forms would make sense on the background of tight fiscal budgets. After all, an integration would not mean to abolish the tax breaks for these other savings, but rather to extend the time horizon of the accumulation phase to the end of the working life and dedicate these savings to pensions. And, any earlier payout there would be possible, but simply result in additional tax payments.
Question 5: With positive answers to the previous four questions, how can it be done in practice? The answer to that is perhaps simpler than expected: While at present all regulations refer to individual types of products and vehicles, most people at the same time are invested in several savings forms, eg, they save for Vermögensbildung, plus have a Riester-product, plus one life insurance, and are perhaps beneficiaries of several occupational pension plans with one or several employers. This pile of different pension savings is clearly intransparent for anyone who is not a top-level pension expert, and on the portability aspect the possibilities to reallocate capital over time are very diminished. While it seems rather difficult if not impossible to harmonise regulations for all products and vehicles in question, it should easily be possible to define a set of five mega-rules for the basic requirements deemed necessary for pension savings:
o Rule 1: The first and very basic rule would require to bring all tax favoured investments within the umbrella of one pension savings account per individual or beneficiary, the Altersvorsorge-konto. This year for the first time individuals will receive a statement detailing what they can expect from the pay-as-you-go state system on retirement. Following on from that, each individual should establish one personal pensions account for tax sheltered products with a bank, insurance company, asset manager or third party provider like a custodian. The account would serve to administer all contributions and payouts for funded pensions, it could be submitted with the income tax declaration year by year, and tax-efficient payouts should not occur before the pensions age. In the first place, the introduction of the account as such would quite naturally increase transparency, and provide ongoing information to individuals on their level of pension savings, similar to the account statement of the pay-as-you-go.
o Rule 2: Another rule can define the tax treatment of capital gains, and also the mechanics of how to combine contributions from pre-tax income with contributions from after-tax income (Ertragswertbesteuerung), in particular while the system is in transition towards deferred taxation. The account keeper could earmark contributions from pre-tax income versus those from taxed income, where the guiding principle for tax treatment would be to avoid any double-taxation. Also, in case of an introduction of the already discussed taxation of capital gains on ‘normal’ savings, the tax-sheltered Altersvorsorgekonto would pose additional incentives for pensions savings.
o Rule 3: This rule would set upper limits for contributions. In a pragmatic approach, one could simply add the tax-breaks granted now for all products and vehicles in question, which would then lead to a tax-neutral result for the reform. Depending on details, and respecting the need to work with transition periods, the limit for combined tax-free contributions (Altersvorsorge-pauschale) could be in the area somewhere between E5,000 and E10,000 pa.
o Rule 4: A fourth mega-rule can reflect the government’s wish to define a minimum standard of safety, and perhaps also a minimum coverage of biometric risks. In that, it is sub-optimal to ask every product to provide all guarantees. Rather would it be more efficient to define a minimum safety pension level per beneficiary, respectively per account in the first place. After discounting, the minimum level of safety would be established by requiring a minimum capital invested or minimum contributions per account for its bottom layer, the Base Pension (Basisrente). This could amount to for instance half the Altersvorsorge-pauschale in terms of annual contributions, or an absolute minimum of eg, E2,500 pa. The theoretical foundation for that level would be set by the contributions required to achieve a minimum payout from funded pensions, as a supplement to the expected lower level of state pensions.
Actual safety would then be achieved by allowing only safe products, or products with return guarantees and biometric coverage, into the Base Pension. At the core of this layer could thus be the Riester-products, then also other products currently qualifying for pensions savings, like life insurance and company DB plans, or plans with a minimum guarantee, or annuities. Within the layer of Base Pension, the advantage over the present system would come from the increase in transparency with the harmonised account reporting, and also from a more flexible allocation of investments over time: Each individual could choose from these different products and vehicles, and substitute one with the other without loosing the tax-breaks on the single product, which is currently the case.
o Rule 5: Once the base pension is filled to target per account, a second layer or support pension savings (Aufbaurente) can be dedicated to achieve a higher pension level, with the goal to provide an adequate standard of living upon retirement. While individuals can opt to invest in base pension products also for that layer, the support pension would allow for a variety of investments, eg, registered funds and pensions vehicles in the second pillar with DC arrangements. Employer company shares could also be integrated, as well as real estate investments and venture capital. After all, investors with long-term horizons are in the best position to finance productive enterprises like venture capital, if in a well-diversified form to cope with the associated risks. In particular for the layer of support pension, diversification arrangements should set some limits for the investments, for instance could the risks associated with investing in shares of the employer company be restricted by setting upper limits per account. A last rule should therefore deal with diversification aspects, and also with disclosure requirements on the account and the respective underlying products.
Again, advantages over the present system are twofold in the surplus pension part: First, a variety of savings forms can be integrated into the pension system, assuming that is the social target with the higher priority. At the same time, risks associated with some of those products can be diversified relative to the total pension capital per person, which would hardly be possible via single product regulations. The increased transparency would also help to cope with risks, and, in combination with disclosure requirements, ultimately provide a sound basis to introduce DC arrangements in Germany.
The overall account would have a structure very similar to the traditional setup of an insurance fund, with one layer of capital set aside to cover the guarantees, and another layer – the free capital – to build a surplus and achieve adequate positive returns. And, again, capital could be reallocated between the two layers once preferences or other factors change, with the sole requirement to keep up the minimum level for the base pension. The Altersvorsorgekonto approach would thus help to combine advantages of traditional collective schemes – like unified reporting, capital efficiency and risk pooling – with the advantages of individual schemes – like high flexibility and portability.
This reform proposal has recently been adopted by the Green Party, and could very soon be brought to the Bundestag. On the other hand, the commission for reform of the social system led by Professor Bert Rürup of the Darmstadt Technical University has also published an initial proposal for pension reform, which could be very much along the same lines. The biggest difference currently is that Rürup suggests not to grant any tax breaks for what is described as ‘support pension’, and puts high emphasis on annuity-like products. This proposal thus lacks the integrational aspects of the Altersvorsorgekonto approach, with its dedication of many forms of savings toward pensions. As a consequence, the overall system would remain segregated and more complicated, except if the government was willing to abolish tax breaks for all forms of current non-pensions savings.
It also seems questionable if politicians today should, based on very specific theories, impose rather narrow definitions for pension savings, which will ultimately determine the income levels of many generations in the future. Rather could a more liberal approach as described here also sets standards for the cross-border treatment of pensions in the EU – a system with deferred taxation, perhaps after a transition period. Again, expecting all EU member states to harmonise their taxation standards would probably mean asking too much.
But one could try to apply the principle ideas behind the Altersvorsorgekonto to the pan-European level: If all states could agree on a pensions accounting similar to that approach, and on only one mega-rule, cross-border taxation could be put on a sound basis in compliance with free movement of capital and people. The agreement could, along the lines of the second rule, be such that payouts which stem from savings made from taxed income will not be taxed, whereas payouts from tax-exempt savings will be. If then a German citizen would wish to relocate to a country with a different set of pension products and regulations, they could show their Altersvorsorgekonto account statement which has been earmarked for savings from pre-tax and after-tax income to the local tax authorities. Local taxation would then only apply for pension payments from income that has not already been taxed. Conversely, pension income arising from contributions previously taxed in one country could be made tax-exempt in another country, even if the new domicile applies deferred taxation for the local system.
Again, the administrative tools to provide the information for cross-border movements would be pension savings accounts, which offer much higher flexibility than the regulation of products and vehicles. Adding up account statements and the resulting taxations may even serve as a basis for transfer payments from one EU state to another, if that is deemed necessary. At the end of the day, a German pension reform along these lines may even have blueprint characteristics for other EU countries – which some people in Europe may find a rather surprising development.
Peter Koenig has worked in asset management for nearly 15 years, most recently with Morgan Stanley in Frankfurt. He is chairman of the investment committee of ABA, the German pensions association