n the discussion about the future of pensions, the concept of ‘solidarity' is often used. Now that the financing of old-age retirement
provision is under pressure, solidarity plays an increasingly important role. On one hand, the question is what groups want or have to show solidarity with other groups. On the other, there is the question of how this
solidarity should be organised.

In the first pillar - a general retirement provision paid from taxes, borne by a large share of the population - most people share risks with one another.

This article focuses on the second pillar, which is organised as a collective, secondary employment benefit through the employer.

Within collective pension schemes, choices were made with regard to solidarity between various groups of participants/ employees, which are valued differently by these groups. Put differently: the costs and benefits of these schemes are not equal for all participants.

As a result, some groups might find the current systems optimal, while others claim adjustments should be made to keep the system sustainable.

Following De Beer, we define solidarity as "a positive way of sharing fate: the fate of one person depends positively on the behaviour of another person. People show solidarity when their behaviour benefits others"1.

The definition is rather broad, as it also covers gift giving, for instance. This also holds for De Beer's description of solidarity as "any behaviour that intentionally benefits others, without there being anything delivered instantaneously in return".

The inclusion of the terms "intentionally" and "instantaneously" provide a direction for further specification. The term intentionally suggests that showing solidarity can be considered a goal: one intends to act in such way that others profit from it.

Solidarity differs from pure exchange of favours, because there is no instantaneous act of giving anything in return. This also shows in the definition of solidarity as "a positive sharing of fate between individuals and groups. This is to say that social relationships are geared towards the strong helping the weak, and the promotion of common interests." Here, the term ‘help' is used. 2

De Beer (2005) distinguishes two types of solidarity: mandatory and voluntary solidarity; and one-sided versus two-sided solidarity.

Voluntary solidarity can be organised informally and be unorganised. It is often called ‘warm solidarity'. Consider for instance care for children, or charity. Sometimes, this solidarity is one-sided, for example in the case of caring for one's children, but sometimes it is two-sided, for example in the case of helping friends.

In the case of voluntary institutionalised solidarity, or "cold" solidarity, one should for example think of fire insurance, where one doesn't know beforehand whether it will be one's own house that burns down or someone else's. This solidarity is two-sided. The one-sided, institutionalised solidarity can be found in social benefit that is paid from taxes.

Empirical evidence suggests that one-sided solidarity is declining, while two-sided solidarity gains importance. 3 The importance of institutionalised (cold) one-sided solidarity in particular is declining in terms of share of gross domestic product. The informal (warm) solidarity, both one- and two-sided, appear to have remained the same. 4

It is remarkable that institutionalised (cold) two-sided solidarity is increasing. This seems to suggest that citizens take a more calculating approach to solidarity: similar to insurance schemes, they increasingly want to know the costs and benefits of solidarity.

Further distinction is between risk sharing solidarity and subsidising solidarity. The latter is ex ante: it is clear beforehand who shows solidarity with whom. In the former case, this only becomes clear ex post.

With risk sharing solidarity, the risks are pooled, as with a fire insurance. In this way, the risks are shared, as in collective schemes. Thereby, the collective fees can cover the collective risks, without this being the case for the individual situation as well.

The same holds for the asset management risks of pension funds. In collective funds, more risks can be taken as compared to individual schemes, even when individuals optimise their portfolios.

In the situation of subsidising solidarity, people refrain from matching the individual fees with their individual risk. This allows for levelling the good risks with the bad risks as, for instance, between employees and retirees, men and women, and healthy and disabled. In these cases it is clear beforehand that a sum of money flows from one group to another.

Subsidising solidarity goes beyond risk sharing, being the risk solidarity which follows from collectivity. Pension funds combine the two types of solidarity, while insurance firms provide for risk solidarity.

The benefits of collective schemes that facilitate risk sharing solidarity are not the subject of debate. However, its quantitative impact of this solidarity is a topic for research. As for risk solidarity which is institutionally arranged for, we indicated earlier that the one-sided type gives way to the two-sided type. People do not mind showing solidarity with others, as long as the solidarity is reversed when they themselves need it.

This demand for security invokes formal schemes, with rights and obligations. From this perspective, the citizen's demand for formal solidarity arrangements can be considered as a sign of insecurity about the future rather than as a lack of solidarity.

This interpretation is consistent with the finding that warm solidarity, albeit one-sided or two-sided, does not seem to diminish. It might even be the case that the security provided by institutional schemes facilitates the participation in informal feelings and acts of solidarity.

De Beer Says: "Voluntary, spontaneous solidarity and compulsory, organised solidarity are therefore no substitutes, but instead complements of each other."

In pension funds, mandatory solidarity is organised through collectivity. Certain aspects of collective schemes imply ex-ante subsidies between different groups of participants such as:

❒ Employees and retirees;

❒ The young and the old;

❒ Men and women;

❒ The healthy and the disabled (if the latter are exempted from paying contributions);

❒ Employed and unemployed (to the extent that the latter retain accrual of pension);

❒ Companies with other companies, who have different demographically different populations of participants.

Not all forms of subsidising solidarity are relevant to all pension funds, or to the same extent. This depends on the characteristics of the pension scheme.

The list above took stock of several costs and benefits of collective pension schemes. The next questions are how much euro is involved, who collects the benefits and who pays the costs of the institutionalised solidarity.

The main benefits of collectivity result from the way risk sharing between generations is organised. This boils down to ex post solidarity with respect to asset management risks.

By spreading the gains and losses on the financial markets over all generations, the collectivity of participants is willing to accept more risks. Moreover, the uniformity of the collective pension product allows funds to offer their services to the participants for a substantially lower fee.

Mandatory, industry-wide pension funds turn out to be most efficient. The comparison between collective and individual pension schemes leads to even larger cost differentials. This holds even more if insurance firms provide the individual pensions, as these organisations have to cover marketing expenses and charge a profit margin on top of operational costs.

On average, insurance firms spend about 25% of contributions on operational costs and profit margins, while collective pension schemes need only 3.5%, which is about seven times less. 5

The main costs of collective pension schemes primarily relate to the limited freedom to adjust the accrual of pension to one's personal situation. The financial policy of the pension fund is targeted to the average participant. Therefore it is not necessarily optimal for all individual participants.

For example, a younger participant may want to invest a larger share of his pension savings in high-yielding equity, while a retiree may prefer the safety and stability of bonds. Moreover, a collective scheme distorts labour market participation and hampers labour market mobility.


ogically, the next question is how the costs of collective pension schemes can be reduced without jeopardising the benefits. There are a number of dilemmas that collective pension schemes will have to face sooner or later, when considering changes to the current system.

Without compulsory participation, individual pension savings can be better matched to the individual situation. Participants themselves know best about their future career, the value of their human capital and their preferred moment of retirement.

All this pleads for less paternalism and more freedom to choose. However, most Dutch citizens indicate they are not eager for more freedom. Instead, they consider having more options as an extra problem to solve, rather than an opportunity for tailor-made financial planning.

Moreover, empirical research from abroad indicates that people tend to make mistakes while acting freely: they save too little and they misallocate the money. Individual systems cannot profit from the substantial benefits of sharing risks across generations. This risk sharing allows participants to take more equity risk, even more than the youngest participant would take in his optimal portfolio. Finally, the costs of asset management are many times higher for individuals as compared to collective schemes. Although the asset mix may not be optimal for all individuals, the higher costs of individual freedom do not compensate for this. Therefore, a collective pension fund seems the best alternative.

Instead of abolishing the compulsory participation, one could consider offering more
tailor-made options within collective schemes. Many pension funds more or less
consider doing so, for instance by allowing different retirement ages and offering complementary pension products.

However, the better match with the individual financial situation which is thus attained should be carefully evaluated against the higher costs, to prevent throwing out the baby with the bathwater. This refers to both the administrative processing of individual arrangements and the adaptation of the asset management. After all, the economies of scale of a uniform product within a collective scheme, are very substantial.

Several forms of subsidising solidarity reflect on several groups. The accompanying flows of money have decreased over the last few years. For instance, the transition from the final-pay system to the career-average system diminishes the subsidy from people with a flat career path to those with a steep career path. There is also less subsidisation of single participants to participants with a partner now that most schemes allow for transferring the money involved to one's own retirement payment.

The European Court of Justice has decreed that women and men have to be treated equally, irrespective of the difference in average life expectancy, which involves higher expected retirement expenses. The subsidy from men to women appears modest, however. Lastly, there is solidarity with the disabled who also receive pension accrual.

An important aspect of collective pension schemes is the uniform contribution and accrual system. This is designed to the average participant, both with respect to the contribution rate and to the retirement payment. In practice, this implies a pay-as-you-go element within collective schemes, so that participants show solidarity with each other over generations.

As long as participants stay within the same industry with the same collective pension scheme, there is no reason for change. However, when labour market mobility leads to many people switching from one pension scheme to another, this may lead to pension payments that are lower than expected. Once participants realise this fact, they may hesitate moving to another employer, thereby limiting labour market mobility.

It takes time and money to adjust this system. Moreover, alternative systems have their own distorting influence on the labour market. For example, if the contribution becomes age-dependent, someone aged 64 has to pay four times as much as someone who is 25 years old. Solidarity between generations would become much smaller.


n the other hand, if the contribution remains the same but the accrual becomes age-dependent, it is far less attractive for the older generation to continue working. They gain very little pension accrual and might decide to withdraw from the labour market

Hardly anyone will be opposed to more transparency. Moreover, the new Dutch Pension Act has led us beyond the point of no return: transparency is the key issue in market supervision. This holds for all financial institutions which are subject to supervision from the central bank.

Moreover, the financial markets authority monitors the communication between pension funds and their participants.

Transparency might endanger the stability of the present system, because lack of awareness fosters solidarity. The more people know more about solidarity, and the more data on the amounts of money involved become available, the more people may begin to question it. Therefore, transparency feeds the debate on the necessity and usefulness of solidarity.

Onno Steenbeek and Fieke van der Lecq are affiliated with the Faculty of Economics of the Erasmus University in Rotterdam, the Netherlands

1 P T de Beer, ‘Hoe solidair is de Nederlander nog?', in: E de Jong en M Buijsen (red), Solidariteit onder druk?, Valkhof Pers, 2005,
pp. 54-79
2 Van Oorschot, 1997, aangehaald in R Beltzer, en R Biezeveld, De pensioenvoorziening als bindmiddel, Amsterdam, Aksant, 2004
3 P T de Beer, ‘Hoe solidair is de Nederlander nog?', in: E de Jong en M Buijsen (red), Solidariteit onder druk?, Valkhof Pers, 2005,
pp. 61-70
4 There is no crowding out of ‘warm' solidarity by ‘cold' solidarity. Therefore, there is no analogy with the notion of extrinsic motivation crowding out intrinsic motivation (Frey, 1997). After all, the warm solidarity does not diminish
5 J A Bikker and J de Dreu, Uitvoeringskosten van pensioenverstrekkers, in: S G van der Lecq and O W Steenbeek (red), Kosten en baten van collectieve pensioensystemen, Deventer: Kluwer, 2006


‘The cost and benefits of collective pensions schemes' - a synopsis


A number of aspects of collective pension schemes imply solidarity between groups of participants. The impact of choices about solidarity for several groups are the focus of the book on which this article is based (Van der Lecq and Steenbeek, 2006).

The first copy of the book ‘Kosten en baten van collectieve
pensioensystemen' (Costs and benefits of the collective pensions schemes) was presented to the Dutch minister of social affairs and employment, Aart.Jan de Geus, on 4 December 2006. It includes
contributions by more than 20 researchers from policy circles and academia. The 10 chapters cover concepts of solidarity, the costs and benefits of solidarity and the macro economic impact of solidarity within collective pension schemes.

The book consists of four parts. Part 1 (chapters 2 and 3) provide an overview of types of solidarity. This refers to several concepts of
solidarity as well as to different groups of participants that show
solidarity with each other. Chapter 2 digresses on types of solidarity within collective pension funds, while chapter 3 compares these with types of solidarity in the healthcare sector. Both in healthcare and in pension schemes, the benefits of solidarity are unequally distributed among the groups of participants. The impact of solidarity differs between the sectors.

Part 2 (chapters 4 to 8) offers quantitative information about the
magnitude of solidarity transfers within collective pension schemes. As far as possible, solidarity is measured in terms of money flows. It turns out that the several types of solidarity have rather different
magnitudes. Careful scrutiny brings the right focus in the debate about the future of collective pension schemes.

Part 3 (chapters 9 and 10) focuses on the compulsory participation of individuals in collective schemes. It also discusses the mandatory
participation of company pension funds in industry pension funds. This is a crucial element of the Dutch second-pillar pension system. Its rationalisations are still valid and the benefits seem to exceed the costs.

In part 4 (chapters 11 and 12) conclusions are drawn. Here, the emphasis is on the macro-economic aspects of solidarity within pension funds. The way risk sharing is organised seems to be the main benefit of collective pension schemes. It allows a pension fund to take more risks, which in turn leads to lower costs and more stable retirement payments. The costs of collective schemes are mainly caused by their distorting effect on the labour market. The uniform contribution and accrual system discourages the labour market participation of your people, while it enhances the participation of the elderly. Although the net sum of costs and benefits of collective systems is not exactly clear yet, it seems that its benefits exceed the costs.