In the Netherlands the transfer of pension rights has been a practice since more than 25 years. Since mid 1994 workers have the right to transfer the value of the pension to their new pension fund or insurance company.
This experience stimulates VB to embrace the proposals and to endorse the objective to remove obstacles to labour mobility caused by supplementary pension schemes. But the same experience also convinces VB that more is needed than just a few principles to make this a success.
Within The Netherlands there is a legal basis for the transfer of the value of the pension rights to the scheme of a new employer. The main reason why the transfer of money became popular was the final salary scheme. Most pensions were based on the final salary just before retiring. To maintain the full rights and to benefit from the so called past-service a transfer is profitable. But since most schemes have changed into an career average scheme the most important reason has disappeared. Financially the main reason to transfer the money now is linked to indexation. Workers switching jobs do not have to worry about a loss of their pension. On the contrary, most pension funds have the ambition to maintain purchasing power of the pension accrued. Legally, the indexation of the early-leaver must be equal to the indexation of the pensioner. So only if one can expect that the indexation of the workers in the new fund will be higher than the indexation of the early-leavers in the old fund, it is worthwhile to transfer the money.
The basic principle fo r the success of the system is the actuarial rules. For defined benefit (DB) schemes, where the pension is based on the average or final salary, more is needed for a smooth transfer.
Different pension funds use different assumptions on the discount rate, the life expectancy, the frequency of partnership etc. The actual value of the future pension is recalculated on a uniform basis into a transferable value by the ‘old’ pension fund. This value is transferred.
The ‘new’ pension fund will make the opposite calculation. When the basis of these two calculations is not equal the employee will loose or gain value. The result is that there will be a one-way traffic of money between the European countries. The transfers of the pensions from a country with a low life-expectancy to a country with high life expectancy will ceteris paribus be very unpopular. Also a transfer from a country were a high discount rate is used to a country where a low discount rate is used will be unlikely.
On the opposite side, the transfers to a country with a low life-expectancy and/or a high discount rate will be very attractive. The national transfer is relative easy because the pension schemes are quite similar and all function within only one taxation-regime. These kinds of differences multiply the complexity.
A similar set of assumptions is the condition sine qua non.
Years ago, the European Commission asked the European social partners to negotiate the conditions of portability. They did not succeed. The EU has often confirmed the right of member states and social partners to design their own pension system. The determination of a maximum employment period before acquiring pension rights seems therefore arbitrary. The ‘loss of pension rights’ for this period or the period worked before the minimum age, is not an issue, because rights have never been acquired, nor have been paid for them. Young mobile workers do not have a disadvantage compared to young non-mobiles.
A retirement income mostly is a mix of funded and pay-as-you-go systems and a shared responsibility of government, social partners and individuals. This combination is the basis for a stable old age system in which all kind of risks can be covered. With regard to portability the difference between funding and pay-as-you-go is essential.
The assets of funded schemes can be transferred relatively easy. In a pay-as-you-go-system however, the working generation pays money for the retired generation and in the future they will receive an income paid by the generation working then. These contributions therefore cannot be spent twice. When a right for the portability of pay-as-you-go systems is established, this system will turn in fact into a funded system. In the second pillar, funded as well as unfunded elements are part of the pension scheme.
The EU member states have organised the funded and the unfunded parts differently. Therefore it is essential not only to exclude the first pillar – as it is has been done – but also to exclude the pay-as-you-go elements in the second pillar.
Indexation can be an unfunded part of the second pillar scheme. In that case this can be financed by future contributions or by future returns on investments. The directive describes that member states shall ‘ensure a fair adjustment of dormant pension rights’. It continues: ‘Member states have different instruments for making this adjustment, depending in particular on how the rights of active members develop.’
In the Netherlands, as well as in most other countries there is no automatic adjustment of the pension rights of employees that are loyal to their employer. So there is no reason to favour mobile workers above loyal employees. It seems to be more logical to treat dormant rights like the pensioners right. A fair adjustment should take the unfunded part in account.
The acquired rights have a strict relation to the contribution paid. When a fixed indexation was not a part of the paid contribution it may be not fair to adjust the dormant pension rights in respect of the other scheme members.
The outcome may well be that it is left to the European court to clarify the legislation. This will create uncertainty and very high cost for a long time.
Joos Nijtmans is at VB in the Hague, the Dutch Association of Industry-wide Pension funds