Abstract
This Selected Issues paper examines the long-term issues with pension expenditures in the Netherlands. The paper highlights that the public pension for a single person is equal to 70 percent of the (statutory) minimum wage. The minimum wage and public pensions thus move in lock-step; they are both by law indexed to contract wages in the private sector. This paper examines the structural policies of the Netherlands. Real wages and employment growth are also analyzed.
I. Pension Expenditures: Long-Term Issues 1/
The pension system in the Netherlands comprises two elements: a basic pension provided by the public sector, financed on a pay-as-you-go basis; and supplementary pension schemes in the private sector, which are fully funded and have built up large reserves. The public pension is provided to all residents, while the supplementary pension system covers more than 80 percent of employees and many self-employed. Table 1 provides a historical overview of revenues and expenditures of these two components of the pension system to illustrate the different financing mechanisms and the amounts involved. This paper discusses in turn the long-term impact of public pensions on the public finances, and long-term aspects of supplementary pension plans that impinge on the public finances. Included in the section on public pensions is a discussion of alternative long-term public debt scenarios that incorporate the increased demographic burden of public pensions.
Pension Revenue and Expenditure
(In percent of GDP)
Pension Revenue and Expenditure
(In percent of GDP)
| 1970 | 1975 | 1980 | 1985 | 1990 | 1994 | |||
|---|---|---|---|---|---|---|---|---|
| Public pensions | ||||||||
| Revenue | 4.4 | 5.5 | 5.6 | 5.6 | 5.7 | 5.3 | ||
| Expenditure | 4.3 | 5.3 | 5.7 | 5.4 | 5.7 | 5.4 | ||
| Balance | 0.1 | 0.2 | –0.1 | 0.2 | –0.1 | –0.1 | ||
| Supplementary pensions | ||||||||
| Revenue | 6.8 | 8.7 | 9.0 | 9.4 | 8.6 | 8.8 | ||
| Investment income | 1.7 | 2.5 | 3.7 | 5.5 | 5.4 | 5.4 | ||
| Premium income | 5.1 | 6.2 | 5.6 | 4.3 | 3.2 | 3.4 | ||
| Expenditure | 2.8 | 3.4 | 3.2 | 3.7 | 4.3 | 4.8 | ||
| Balance | 4.0 | 5.3 | 5.8 | 5.7 | 4.3 | 3.9 | ||
Pension Revenue and Expenditure
(In percent of GDP)
| 1970 | 1975 | 1980 | 1985 | 1990 | 1994 | |||
|---|---|---|---|---|---|---|---|---|
| Public pensions | ||||||||
| Revenue | 4.4 | 5.5 | 5.6 | 5.6 | 5.7 | 5.3 | ||
| Expenditure | 4.3 | 5.3 | 5.7 | 5.4 | 5.7 | 5.4 | ||
| Balance | 0.1 | 0.2 | –0.1 | 0.2 | –0.1 | –0.1 | ||
| Supplementary pensions | ||||||||
| Revenue | 6.8 | 8.7 | 9.0 | 9.4 | 8.6 | 8.8 | ||
| Investment income | 1.7 | 2.5 | 3.7 | 5.5 | 5.4 | 5.4 | ||
| Premium income | 5.1 | 6.2 | 5.6 | 4.3 | 3.2 | 3.4 | ||
| Expenditure | 2.8 | 3.4 | 3.2 | 3.7 | 4.3 | 4.8 | ||
| Balance | 4.0 | 5.3 | 5.8 | 5.7 | 4.3 | 3.9 | ||
1. Public pensions
a. Characteristics and trends
The General Old-Age Pensions Law came into force in 1957 and covers the entire resident population, irrespective of nationality, employment, or marital status. Since 1919, compulsory old-age insurance had been in effect, but it was gradually perceived to be too limited as it only applied to wage earners and benefits were fixed in nominal terms; the view took hold that social insurance should provide the entire population with social security and freedom from want. The General Old-Age Pensions Law thus stipulates that any resident age 65 or older is to receive a public pension (with pro-rating for years of non-residence between ages 15 and 65). 1/
The public pension for a single person is equal to 70 percent of the (statutory) minimum wage, and for each person of a married or unmarried couple—opposite or same sex—the pension is equal to 50 percent of the minimum wage. 2/ The minimum wage and public pensions (as well as other social benefits) thus move in lock-step; they are both by law indexed to contract wages 3/ in the private sector—although escape clauses have allowed the minimum wage and social benefits to be frozen in nominal terms during 1984-89 and 1993-95. As in many other countries, public pensions are financed on a pay-as-you-go basis (i.e., contribution rates are adjusted so that revenues keep pace with expenditures on an annual basis). Contributions are paid as a percentage of wages (the contribution rate is currently 15.4 percent), up to a ceiling that is indexed to the cost of living. In 1995, total public pension outlays amounted to 5.3 percent of GDP (or 9 percent of gross wages).
In a comparison across the European Union (EU) of the net present value of current liabilities of the basic pension system relative to GDP, the Netherlands was found to face the largest burden [Kuné et. al. (1993)]. Three factors play a role in the more negative outcome in the Netherlands:
The public pension in the Netherlands, as noted, is universal. The cost effect of this may be illustrated by comparing the Netherlands and West Germany, where the scheme is limited to only those in the labor force. Kuné et. al. report a lower net present value of liabilities in West Germany than in the Netherlands, despite a higher average annual old-age basic public pension (ECU 7,320 vs. 6,560 in 1990). 4/
The benefit level appears to be relatively generous: the minimum wage has been relatively high compared to the minimum wage in other countries and to the average wage in the Netherlands. Nine of the twelve EU countries are reported to have average old-age basic public pensions lower than the Netherlands.
The only other country in the EU12 that provides every resident with a basic flat-rate pension is Denmark, but its retirement age is set higher at 67, with an earnings test up to age 70. Thus, despite a nearly identical average pension, the net present value of basic public pension liabilities in Denmark are estimated by Kuné et. al. to be less than half of those in the Netherlands.
While a net present value calculation provides a useful summary statistic for inter-country comparison, it does not provide insight in the evolution of the public pension burden over time, which is important in terms of policy analysis and design. The aging of the population in the coming decades is expected to lead to upward pressure on public pension expenditures. The old-age dependency ratio is currently lower than in neighboring countries, but is projected to increase more rapidly and reach a higher peak some 40 years from now. 1/ According to the latest projections by the Dutch Central Bureau of Statistics, the number of persons aged 65 and older may rise from 21 percent of the working-age population (aged 20 to 64) in 1995 to 41 percent in 2035. However, such projections are subject to substantial uncertainty. This is illustrated by the fact that, compared to the demographic projection of three years earlier, the birth and immigration rates have been raised, leading to a somewhat lower projected old-age dependency rate in the coming decades. Current projections could underestimate the greying of the population to the extent that the rise in life expectancy is greater than assumed.
Numerous studies have been carried out in the Netherlands in the past 15 years to gauge the evolution of public pension costs as the greying of the population proceeds. Early investigations considering solely the rise in the dependency ratio reported an alarming rise in that cost over time, 2/ but later studies have come to the conclusion that the increase is manageable, amounting by 2030-40 to at most some 4 percentage points of gross wages or roughly 3 percentage points of GDP [Ministry of Social Affairs and Employment (1987), WRR (1993), and Jansweijer (1996a)]. 3/
This “worst case” projection of pension costs incorporates relatively cautious projections for economic growth and the aging of the population; at the other end of the spectrum, projections incorporating a smaller aging effect and faster labor supply growth/economic growth could in principle increase total resources available to pay public pensions and mitigate the rise in the relative burden of public pensions.
However, the range of outcomes in these later studies hinge critically on a joint assumption regarding (a) maintenance of the current system of indexation to contract wages and (b) the size of wage drift. 1/ As long as wage drift—the difference between average and contract wages—is positive, the indexation of public pensions to contract wages leads to slower growth in the public pension than in the average wage, thus helping to contain the cost of public pensions relative to gross wages and GDP. In the three major studies cited above wage drift is assumed to amount to 1 percent a year, and this is a key factor offsetting the influences of demography, universality, relative generosity (at present), and earliness of retirement age. However, wage drift has actually been very low in recent years (see Table 2). The studies do not provide a convincing case that it will increase. 2/
Wage Drift
(Average annual rate, in percent)
Wage Drift
(Average annual rate, in percent)
| 1981-85 | 1986-90 | 1991-94 | |
|---|---|---|---|
| 1.0 | 0.9 | 0.1 |
Wage Drift
(Average annual rate, in percent)
| 1981-85 | 1986-90 | 1991-94 | |
|---|---|---|---|
| 1.0 | 0.9 | 0.1 |
It should also be noted that in these recent studies, the ratio of the public pension to the average wage declines sharply over the next 40 years by some one third, 1/ which could lead to pressure for costly upward changes in the indexation mechanism or ad hoc increases in public pensions. If public pensions were to keep pace with average instead of contract wages, either ex ante or ex post, the future burden of public pensions relative to gross wages or GDP would be higher than suggested, 2/ unless there were offsetting factors. 3/
In sum, while projections prepared over the past few years have been less alarming than earlier studies based solely on the rise in the dependency ratio, it would not appear prudent to view the so-called “worst-case projections” emerging from these recent studies as an upper limit to the possible rise in public pension expenditures relative to gross wages or GDP.
b. Policy options
The central policy problem posed by the possible pension outlook is the risk that, unless action is taken now, either contributions may need to be raised—while the revenue burden is already high—or sizeable fiscal transfers would be needed to balance the public pension scheme, which could inflate the public debt ratio. This could be addressed by entitlement reform, compensating fiscal adjustment, or a combination of the two.
(i) In view of the possibly significant rise in public pension expenditures over the next 40 years, and the desirability of avoiding last-minute and drastic changes in system parameters, options for pension reform in the next few years are worth examining. The simplicity of the public pension scheme clearly limits the choices: the indexation rule is crucial, as discussed above, and the retirement age is an obvious variable that affects total spending. The current level of the public pension also determines spending, but if changes were desired in this level they could be made over time through appropriate choice of indexation rule. Finally, the universality of the public pension system should no doubt be seen as a continuous policy goal that is to be safeguarded by maintaining the financial viability of the system, rather than a policy variable that may be changed to influence cost. 1/ In terms of specific reform measures:
Limiting the indexation of public pensions would reduce the growth of spending but could well run into income distribution constraints as the position of retirees dependent solely on the public pension declined relative to others. A further complicating factor is that slowing the growth of the public pension would imply a compensating increase in supplementary pensions, as these promise a defined total—public and supplementary—pension as a percentage of the last earned wage (prorated for years worked). The possibly substantial cost of the increase in supplementary pensions would be borne by employers and employees, i.e., would be paid out of gross wages. Some de-linking of public and supplementary pensions would be desirable to make reduced indexation of public pensions a fully viable option (see section 2 below for a more detailed discussion).
Raising the retirement age has been recommended in WRR (1993) as an option for long-term cost control. Raising the retirement age would not only lower public pension spending but also potentially raise the contribution base. However, the latter depends on the labor market behavior of those close to retirement. The participation rate of people 60 to 64 years old is currently just 14 percent due to the attractiveness of disability and early retirement schemes at these ages (in the EU in 1994, only Belgium, Luxembourg and France had lower participation rates in this age group). Since benefits provided by these schemes are higher than pension benefits, a higher retirement age that only resulted in people remaining longer in the disability or early retirement schemes would be counterproductive. In addition, the implication for supplementary pensions under current rules might again be that they would need to rise to compensate for the loss of the public pension. These problems could be addressed however: if public pensions could be adjusted without affecting supplementary pensions, and the participation rate of older people could be increased, an increase in the retirement age would appear to be the primary option to help control public pension costs.
(ii) Given the uncertainties surrounding possibilities for pension reform, and the inherent credibility problems attached to any policy that would need to be maintained for a long period (such as moderation of pension indexation), a prudent approach would be to begin to prepare the ground for an increase over time in public pensions by creating room in the public finances.
One element of such an approach could be to include pension income (public and supplementary above a certain minimum) in the contribution base for public pensions, thus increasing the funds available for public pension expenditures. However, the incentive effects on migration of retirees and/or pension funds, especially in the context of a monetary union without a unified taxation system, would have to be carefully examined.
A more reliable approach would be to create favorable public debt dynamics early on, so as to be able to ultimately absorb higher public expenditures without significant increases in revenue relative to GDP. To illustrate the mechanics of such an approach two long-term scenarios are shown in Chart 1 to demonstrate the benefits of early fiscal adjustment in the context of a coming demographic shock to the public finances.
NETHERLANDS: Long-Term Fiscal Scenarios
(In Percent of GDP)
Citation: IMF Staff Country Reports 1996, 080; 10.5089/9781451829303.002.A001
source: staff calculations.NETHERLANDS: Long-Term Fiscal Scenarios
(In Percent of GDP)
Citation: IMF Staff Country Reports 1996, 080; 10.5089/9781451829303.002.A001
source: staff calculations.NETHERLANDS: Long-Term Fiscal Scenarios
(In Percent of GDP)
Citation: IMF Staff Country Reports 1996, 080; 10.5089/9781451829303.002.A001
source: staff calculations.In the medium-term, with noninterest expenditures rising by less than 1 percent a year in real terms on average, the general government budget could be balanced by the year 2000, or the revenue ratio could be cut by about 1 ½ percentage points of GDP between 1997 and 2000. Because of implied interest savings, the difference between the two scenarios by the year 2000 regarding the deficit amounts to nearly 2 percent of GDP, but regarding the revenue ratio to little more than 1 ½ percent.
The rationale for proceeding to balance the budget by 2000 (and achieving a small surplus soon after) emerges from the longer term projections, which draw on earlier work by the staff to illustrate how tax cuts can be sustainable and thus credible to economic agents in a scenario that also stabilizes the debt ratio at a prudent level. 1/ Thus, in the more ambitious long-term scenario, a small surplus is achieved early in the next decade. As a result, the debt ratio continues to decline sharply, and accompanying interest savings allow the ratio of revenue to GDP to be reduced. When demographic costs begin to mount, only a small increase in the revenue ratio turns out to be necessary to prevent a ballooning of the debt ratio. By contrast, in the absence of early deficit reduction, the debt and interest burden remains high in the alternative scenario, and the low revenue ratio in the early years is not sustainable once the demographic burden hits in force. The revenue ratio in this scenario ends up at a significantly and lastingly higher level than in the scenario with early deficit reduction.
Assumptions underlying these scenarios include the following. Noninterest expenditure, excluding any demographic effect, is assumed steady relative to GDP from early in the next decade; to ensure comparability of the scenarios, noninterest expenditure is assumed the same in both. Active structural policies are assumed to mitigate the effect on potential growth of a declining share in the total population of people of working age. A similar assumption is used or explored in other studies [Jansweijer (1996a), SM/94/100], and helps mitigate the future burden of public pensions relative to GDP. In addition, -demographic effects on the budget are assumed to be limited through policy changes to ultimately no more than 5 percent of GDP. This amount is intended to capture not only the additional cost of public pensions, but also the net increase in other demographically dependent expenditure categories. 1/
In this connection, two of the other long-run influences, potentially on opposite sides of the ledger, deserve mention [Hebbink (1996)]:
As the share of younger people in the population declines, spending on education relative to GDP may fall. However, projections of the number of older people in the population are more reliable than of the number of younger people, thus possibly giving expectations of savings due to fewer younger people a greater degree of uncertainty than projections of increased pension costs.
With an aging population health care spending relative to GDP may rise. Some Dutch studies have concluded that the demographic effect on health care expenditures relative to GDP may be negligible, as the purely per capita effect is projected to keep pace with the rise in GDP per head. However, it is acknowledged that the relative price of health care is likely to rise, although not because of demographic factors [WRR (1993)]. For the G-7 countries, the OECD has projected a demographic effect on health care spending of 1 to 3 percentage points of GDP [Hebbink (1996)]. It would appear prudent to project some rise in health care spending relative to GDP in the Netherlands in addition to the likely increase in public pension expenditure. 2/
2. Supplementary pensions
This section discusses the link between public and supplementary pensions, long-term trends in the supplementary pension system, tax-like features of supplementary pension premiums, and policy implications.
a. Key features
The supplementary pension system includes about 80 sectoral pension funds, a public sector fund (ABP) that is among the largest pension funds in the world, and more than 1,000 company pension funds. They serve, respectively, 2 ½ million, one million, and ¾ of a million workers (including part timers). 3/ Many of the company pension funds are small and have taken out pension contracts with life insurance companies. The company funds provide pensions that are more generous than those of sectoral funds because of the labor union consent that is required for dispensation from the sectoral pension funds. The Pension Law mandates that pension funds are fully funded but does not prescribe the computation method. Pension reserves of pension funds and insurance companies are on the order of 100 percent of GDP (Table 3).
Reserves of Institutional Investors
(In percent of GDP)
Algemeen Burgerlijk Pensioenfonds (pension fund for civil servants and teachers).
Reserves of Institutional Investors
(In percent of GDP)
| 1991 | 1995 | |
|---|---|---|
| Private pension funds | 45.7 | 53.2 |
| ABP 1/ | 30.4 | 30.8 |
| Insurance companies | 28.4 | 37.0 |
| Social security funds | 3.2 | 0.7 |
| Investment companies | 11.1 | 17.4 |
Algemeen Burgerlijk Pensioenfonds (pension fund for civil servants and teachers).
Reserves of Institutional Investors
(In percent of GDP)
| 1991 | 1995 | |
|---|---|---|
| Private pension funds | 45.7 | 53.2 |
| ABP 1/ | 30.4 | 30.8 |
| Insurance companies | 28.4 | 37.0 |
| Social security funds | 3.2 | 0.7 |
| Investment companies | 11.1 | 17.4 |
Algemeen Burgerlijk Pensioenfonds (pension fund for civil servants and teachers).
Supplementary pension funds provide benefits under defined-benefit plans, with the promised total pension at age 65 generally equal to 70 percent of pay in the last year or several years for an employee with a complete work history (at least 40 years); the pension is prorated for years worked. 1/ A number of changes over the years have resulted in effective increases in pension benefits. Married women and part-time workers at one time were not eligible to participate in pension funds, but such restrictions have now all been lifted. Benefits have also been extended to widowers, and since 1992 pension indexation rights of “sleepers” in a pension fund (former participants who have not yet retired) are by law identical to those of retirees. 1/
The pension that is promised is inclusive of the public pension-defined as the public pension of a couple by some pension funds, of a single person by others. The higher the public pension, the less supplementary pension is built up (the supplementary pension “tops up” the public pension). Premiums are paid only over the income that is deemed not covered by the public pension. In response to the repeated nominal freezing of public pensions in the last 15 years, in collective bargaining agreements at some companies a notional public pension has been substituted for the actual one. 2/
The premium rate charged by a pension fund is determined uniformally each year for all premium-paying participants so as to increase reserves to the level required to pay for future pension benefits; the premium computation assumes that current salaries remain unchanged. In this system, a young employee pays a pension premium (the uniform rate) that is more than necessary to cover purely his/her own pension, while an employee close to retirement pays less. 3/ This implicit transfer between generations represents a pay-as-you-go element in the private pension system (and is the reason for the need to increase the uniform pension premium when the old-age dependency ratio rises).
The real interest rate (defined as the difference between the nominal interest rate and nominal wage growth) also is an important determinant of pension premiums. Since current reserves cover liability for pensions resulting from current salaries, it follows that the growth rate of reserves (the interest rate) would need to keep pace with the growth rate of wages to avoid future shortfalls in reserves (with unchanged premium rates). 4/ In the 1980s, low wage growth and high interest rates increased the reserves of pension funds while premiums could be lowered and benefits made more generous (Table 1). However, real interest rates were negative in the 1960s and 1970s. If, on the basis of the longer-term historical experience future real interest rates were assumed to be low or negligible, the full brunt of the aging of the population would fall on pension premiums.
b. Long-term projections
While pension funds hold substantial reserves, and are fully funded in the sense that current pension liabilities could be paid out of these reserves, the increase in supplementary pension obligations over the next several decades is projected to outstrip that of premium receipts at current premium rates—or, to put it differently, premium rates would have to rise to pay for promised pension benefits. This results from the uniform-premium financing method combined with projected demographic and economic developments.
Recent studies have projected premium increases for supplementary pensions ranging from a doubling to a quadrupling in 40 years [Ministry of Social Affairs and Employment (1986), WRR (1993), and Jansweijer (1996a)], with the real interest rate assumptions explaining much of the variation in premium projections. 1/ Three factors contribute to these major premium increases in the period ahead:
First, it is projected that the public pension continues to be indexed to contract wages and as a result will slip relative to average wages. Supplementary pensions, as currently designed, will pick up much of the slack, 2/ which implies higher premiums (note that if public pensions were to keep pace with average wages, the “hit” would be taken by the public finances).
Second, the aging of the population leads to higher premiums because of the pay-as-you-go element in the system, as explained above.
Third, this effect is reinforced by the practice of pension funds to ignore indexation of pensions for retirees and “sleepers,” which is promised contingent on the availability of sufficient reserves; it is assumed (hoped) that interest rates will be high enough to generate such funds. In the projections, no reserves are being formed to pay for indexation, while the aging of the population increases the number of pensioners relative to contributors. Assuming that the full promised pensions are to be paid, premium rates for workers would need to rise to finance this (on a pay-as-you-go basis).
c. Policy implications
At the outset, it should be noted that the Government holds a latent tax claim on the reserves of pension funds and life insurers, because pension premiums are deductible for income tax purposes, while pension benefits are taxed. This has typically not been incorporated in the demographic picture, and provides a (small) offset to future revenue needs. In addition, significant assets (more than 30 percent of GDP) have been built up by the Government in the pension fund for civil servants and teachers (ABP) to finance the supplementary pensions for these public employees. The extent of funding, including for civil servants, is larger in the Netherlands than in a number of other countries [Bovenberg and Peterson (1992)], 1/ and is a factor helping to limit the effect on the public finances of the demographic shock.
There are, however, indirect ways in which supplementary pensions, though provided by the private sector and fully funded, may be relevant to an assessment of the public finances—and of incentives affecting the labor market—because of the de facto link with public pensions (supplementary pensions top up public pensions) and the “collective,” tax-like, features in their design. These characteristics are particularly germane in the context of public pension reform: a cutback in public pension benefits (aimed at controlling outlays in the context of adverse demographics) that led to an increase in supplementary pension premiums (total benefits being little changed) could in effect be viewed as closely akin to a tax increase. In general, to the extent that supplementary pension premiums are viewed as taxes, they constitute adverse labor market incentives.
The collective character comes about because supplementary pensions are provided on a defined-benefit basis, premiums are not determined on an individual but group basis (by company or by sector), and participation in a sectoral pension fund is declared mandatory for all employers in the sector by the Minister of Social Affairs and Employment. 2/ Given these features, supplementary pension premiums, while calibrated in the aggregate to generate pension reserves that cover current pension liabilities, may not be fully viewed by individuals as saving, and to that extent may add to the already high wedge between labor costs and take-home pay. Indeed, the Central Planning Bureau includes supplementary pension premiums in its calculation of tax wedges. 1/
If future workers were to balk at paying the high supplementary pension premiums that would be required on current projections, it would place an undue burden on the public finances to provide offsetting tax relief. Instead, supplementary pension funds would have to lower premiums and cut pension indexation for retirees and sleepers (as pension contracts specify that this indexation is subject to availability of funds). Such suspensions of pension indexation would likely not accord with expectations of the population, and should be avoided by timely reductions in supplementary pension obligations. Changes in this direction are beginning to be made, with one large company currently planning to switch to a scheme in which an individual’s pension is set on the basis of his/her average instead of last wage, thereby effectively lowering pensions. 2/ Such changes move in the direction of a system with lower starting pensions, while at the same time safeguarding the ability to maintain indexation of pensions for retirees and sleepers.
The Government has indirect levers at its disposal to influence developments: it could change the conditions on the tax deductibility of pension premiums to encourage those pension funds that attempt to limit future liabilities, and it could no longer declare binding participation in pension funds that did not mitigate future pension increases sufficiently. Similarly, it could encourage de-linking of supplementary and public pensions to strengthen the self-standing private nature of the funds. Furthermore, the individual insurance elements of supplementary pensions could be strengthened. Indeed, the Government intends to change limitations on tax deductibility of pension premiums to accommodate a greater variety of pension plans, and it intends to allow more competition among pension funds, which could lead to more choice of plans for individual workers.
3. Conclusions
Available projections suggest a rise in premiums for public and supplementary pensions from around 17 percent of gross wages currently to some 35-40 percent in about four decades (implying even higher actual rates since the base is narrower than gross wages). To avoid substantial worsening of incentives in the labor market, a combination of policy options appears attractive:
Indexing public pensions at less than the rate of increase of wages, or increasing the retirement age, would reduce the cost of public pensions.
To the extent that supplementary pension premiums are perceived as taxes, efforts could also be directed at restructuring the design of the supplementary pension plans to underscore their individual insurance content. To avoid causing compensating increases in supplementary pensions, de-linking of supplementary pensions from public pensions is desirable. In any case, the public finances should not take on the burden of offsetting higher supplementary pension premiums with lower taxes.
Reductions in the overall level of pensions, now often set at 70 percent of the last wage, would slow projected increases in premiums.
Most importantly, as noted above, a prudent preparation for the rising cost of public pensions would involve durable reductions in the public debt through the pursuit of early fiscal consolidation.
References
Aaron, H.J., “The Social-Insurance Paradox,” The Canadian Journal of Economic and Political Science (32), pp. 371–7.
Bakker, Bas B., Saving in the Netherlands: Sources. Trends, and Mobility, 1993.
de Beus, D.J., “Pensioendoelstellingen, Structuren, en vrije Markt,” Economisch Statistische Berichten 81 (4055), p. 372, 1996.
Bikker, J.A., “Demografische Onevenwichtigheid, Nationale Besparingen en Lopende Rekening,” Maandschrift Economie 58 (3), pp. 202–22, 1994.
Bovenberg, A.L., “Een Nieuwe Fiscale Behandeling van Pensioenfondsen,” Economisch Statistische Berichten 78 (3895), pp. 91–93, 1993.
Bovenberg, A.L., Overvloed en Onbehagen: over Sparen en Investeren in Nederland, 1991.
Bovenberg, Lans, and Carel Petersen, “Public Debt and Pension Policy,” Fiscal Studies 13 (3), 1992.
Hebbink, G.E., “Vergrijzing, Overheidsfinancién en Besparingen: een Overzicht van Recente Literatuur,” Onderzoeksrapport WO&E no. 451/9601, De Nederlandsche Bank, 1996.
Huijser, A.P., and P.D. van Loo, The Ageing Population. Pensions and Contractual Savings, Monetary Monographs No. 5, De Nederlandsche Bank, 1986.
Jansweijer, R.M.A., Gouden Bergen. Diepe Dalen: De Inkomensgevolgen van een Betaalbare Oudedagsvoorziening,” Wetenschappelijke Raad voor het Regeringsbeleid, Voorstudies en Achtergronden, no. V92, 1996a.
Jansweijer, R.M.A., “Te Hoge Verwachtingen rond de Oudedagsvoorziening,” Economisch Statistische Berichten 81 (4040), pp. 24–9, 1996b.
Kuné, Jan B., Wilfried F.M. Petit, Aggie J.H. Pinxt, “The Hidden Liabilities of the Basic Pensions System in the Member States,” Centre for European Policy Studies, 1993.
Ministry of Social Affairs and Employment, “Analyse Vergrijzingsproblematiek,” unpublished, 1996.
Ministry of Social Affairs and Employment, “Gespiegeld in de Tijd; de AOW in de Toekomst,” Rapport van de Commissie Financiering Oudedagsvoorziening, 1987.
Ministry of Social Affairs and Employment, Social Security in the Netherlands, 1990.
WRR (Wetenschappelijke Raad voor het Regeringsbeleid), Ouderen voor Ouderen: Demografische Ontwikkelingen en Beleid, Rapporten aan de Regering, no. 42, 1993.
Prepared by Frank Lakwijk.
Residents are also eligible for other benefits: any old or disabled resident of the Netherlands is eligible to receive benefits linked to the minimum wage, and any resident is eligible to receive benefits related to family size and reimbursement for medical expenses arising from serious illness or long-term disability.
Single parents receive 90 percent of the minimum wage, and retirees with a partner under 65 may receive a supplement.
Contract wages are average wage scales in collective bargaining agreements.
It should be noted that the comparison presented by Kunē et. al. presents a biased picture of liabilities because the cost of providing income support to persons not covered by the basic pension system is not included in the projections.
See Kingdom of the Netherlands • Netherlands - Selected Background Issues, SM/94/100 (4/22/94), page 8.
See Huijser and van Loo (1986), pp. 26-27.
Also, the current financing mechanism could cause contribution rates to double, partly because the ceiling on wage income subject to contributions is indexed to prices, not wages [Ministry of Social Affairs and Employment (1996)]: it is known that an adjustment of this mechanism is needed in due course (even if the overall cost of public pensions is manageable), and this adjustment is assumed in the discussion that follows.
In addition, the studies took into account that the intended equality between the public pension and the minimum wage relates to the net (after tax) amounts; the gross (before tax) public pension is determined by adding the applicable tax and social contribution amounts to the targeted net pension. Any increase in the contribution rate for public pensions cuts into net contract wages and thus, through the indexation mechanism, the net minimum wage. The net public pension falls in turn, allowing a reduction in the gross public pension given that retirees do not pay the contribution rate for public pensions. As a result, the contribution rate, which is related to gross pensions, can be somewhat reduced. This restraint of public pension expenditures can amount to a few percentage points of gross wages over the time period considered.
To the contrary, the retirement of older and thus better-paid baby boomers would tend to pull down the average wage, while at the same time contract wages might begin to rise faster, reflecting emerging scarcity of labor.
For example, in the base scenario of Jansweijer (1996), the public pension of a couple falls from 54 percent of net household income of young workers in 1995 to 39 percent in 2040.
Ministry of Social affairs and Employment (1987) and Jansweijer (1996a) project that the cost of public pensions would be several percentage points of gross wages higher in this case.
Recent preliminary calculations by the Ministry of Social Affairs and Employment (1996) suggest that public pension expenditures may rise by no more than 1 ½ percentage points of GDP (or 2 ¾ percentage points of gross wages) during 1995-2035 in a scenario with low economic growth and wage drift of just ¼ of a percent a year, but details on the mechanism that produces this outcome are unclear.
Hebbink (1996) notes that in all extant reform proposals the maintenance of a collectively financed basic pension is assumed.
See SM/94/100
The 1995 Annual Report of De Nederlandsche Bank (April 1996) mentions a figure of 5 percentage points of GDP for the expenditure increase resulting from demographic developments in the areas of public pensions and health care.
Health care insurance as part of the social security system currently amounts to about 7 percent of GDP. This insurance is compulsory and implies transfers between young and old age groups that are likely to rise as the population ages.
There are also pension funds for some 36,000 self-employed, including medical specialists.
Some pension funds define benefits on the basis of average salary. Note that even at the -ABP, where long-term participation might be expected to be the rule, the average number of years of participation at retirement was 29 for men and 20 for women in 1994, instead of the “standard” 40 years [Jansweijer (1996b)]. The prorating of the pension implies that, for example, women retiring from the ABP received on average about 35 percent of their last salary.
The loss of real value of built-up pension rights was seen an impediment to job change.
In one case, the notional public pension is computed by indexing the public pension in a base year with the contract wage increase of the company. A renewed freeze of the public pension would thus not affect this company’s supplementary pensions.
Consider a salary increase of f. 100 for a young and older worker. The pension premium (perhaps 10 percent) would be the same amount for each, but would have many more years to earn a return in the case of the young worker, who is hence, from the individual’s point of view, overcharged.
This holds in a model with certain restrictive assumptions, such as a stable population structure and no pension indexation. If the interest rate exceeds the rate of wage growth, capital funding for the financing of old-age pensions would be desirable, while pay-as-you-go financing would be preferable if the reverse held [Aaron (1966)].
To illustrate, Jansweijer (1996a) projects a premium rise from 7 percent of gross wages in 1995 to 26 percent in 2040 in his base scenario that includes a real interest rate declining to -0.4 percent in the long term. If the real interest rate instead were assumed to be 3 percent, the premium rate would still more than double [Jansweijer (1996b)].
The compensation by supplementary pensions is incomplete, because it only applies to employees, and only occurs fully for those with a complete work history (of 40 years).
The higher funding compared to other countries may be a reflection of higher pension benefits.
See SM/94/100, pages 41-44, for a discussion of the practice of declaring binding collective bargaining agreements.
Defined-contribution plans, where the pension equals accumulated contributions including interest, are more individual in nature. However, disadvantages of such plans include the risk of low real interest rates for prolonged periods and high transaction costs [Bovenberg (1993), de Beus (1996)]—the greater the emphasis on individuality, the higher administrative costs are likely to be. The choice of pension system involves difficult trade offs between considerations of return, incentives, degree of compulsion, and equity.
Current workers are grandfathered.
