Revolution on the horizon: Will the Dutch pension system still set the bar for sustainability, adequacy and integrity?
Insights

Revolution on the horizon: Will the Dutch pension system still set the bar for sustainability, adequacy and integrity?

With the adequacy, sustainability and even integrity of the world’s pension systems being tested to the limit, the Netherlands has achieved more than most in meeting the demographic, economic and financial challenges that have compromised many other pension systems over the past couple of decades. However, with the dramatic economic and financial effects of the Covid pandemic now also bearing down on both public and private sector finances, we explore whether the Netherlands, in reshaping its second pillar, can retain its pre-eminent position in the global pension system rankings.

Biggest isn't always best

The global pensions market has assets of around $56 trillion, of which the Netherlands, the world’s sixth largest pensions system by assets, accounts for around $1.9 trillion (3.6%).¹ However, big in pensions doesn’t always translate into best. Although considerably smaller than the US market
(62% of global pensions assets) and noticeably smaller than the Japanese (6.9%), UK (6.8%), Canadian (5.9%) and Australian (4.4%) markets,² the Netherlands has a trump card – it consistently tops the authoritative Mercer CFA Institute Global Pension Index Report (MCFAGPI),³ with a much coveted A-grade based on its robust integrity (or security), adequacy and sustainability credentials. While Canada and the – fast-growing – Australian pensions market, with their B-grades, can realistically aspire to achieving an upgrade by building greater levels of adequacy and sustainability into their systems, the US and UK, both rated C+, and Japan, with its lowly D, each have some way to go before an A-grade becomes a tangible possibility.

Pension systems are under pressure

However, while pensions systems across the world may be as disparate as the people they serve, without exception they all have one thing in common: they are under pressure. Globally, people
are living longer, while fertility rates decline. To compound these demographic headwinds, most developed economies are now seeing their second wave baby boomer generation, of the early to mid-1960s, beginning to retire. Moreover, recent sub-par global economic conditions, which will likely result in a further prolonged period of historically low, even negative, interest rates and modest real investment returns, has added further complexity to the conundrum.

Alongside this, is the global shift from collective passivity to individual responsibility, as State and employer paternalism cedes to savers increasingly taking a great deal more personal responsibility for their retirement and financial futures and so shouldering more investment, longevity and decision-making risk. Taking all of this together, you have a situation where governments worldwide are grappling with the problem of how to engineer pension systems that are both sustainable and equitable for current and future generations of retirees.

A uniquely generous pension system

While some countries, like the Netherlands, have better prepared for these seismic demographic, economic and choice architecture shifts than others, many of these others continue to buckle under the strain. Indeed, even against this backdrop, the Dutch pension system remains uniquely generous in many respects.

Unlike most other pension systems, with the notable exception of Japan, it is principally characterised by a very well-developed second pillar of fully funded defined benefit (DB) schemes,4 set up as non-profit organisations and operating independently of the sponsor. These schemes are offered by around 90% of employers, via industry-wide and corporate plans, with near-universal employee coverage.5 In addition, about 10 professional group pension funds are made available to specific professions. Employer and employee contributions are set by collective labour agreements, while accruals are typically based on average lifetime earnings.6 Moreover, given the second pillar’s central tenets of risk sharing and equity between stakeholders, all members of a pension fund pay the same contribution while each participant accrues the same fixed percentage of their pay each year as a future pension entitlement.
Crucially, when combined with a reasonably generous first pillar, “basic” flat-rate, pay-as-you-go public pension, the Algemene Ouderdomswet (AOW),7 the resulting replacement rates are far above OECD averages,8 with old age poverty rates far below that of most other countries.9 Indeed, the gap between disposable income before and after retirement is considerably smaller than that for almost any other country. Moreover, to encourage engagement, employees receive an annual pension overview (uniform pensioenoverzicht, UPO), so they can keep tabs on exactly what pension benefits they have built up to date. However, the one fly in the ointment remains the gender pensions gap which, in the Netherlands, remains stubbornly high at over 40%, versus an OECD average of 25%.10

As in most other developed pension markets, the first two pillars are supplemented by a third pillar representing individual voluntary pension saving. In the Netherlands, this is mainly for those who
do not, or have not been able to, accrue a pension in the second pillar, whether because their employer is not affiliated to a pension fund, they are not in permanent employment or they are self-employed. This saving is principally directly through banking and insurance products, such as annuities, single premium policies and life insurance policies. Rather disconcertingly, and while not limited to the Netherlands, with the rise of the gig economy there are now increasing numbers of
freelancers and contractors who accrue little or no pension in the second or third pillars and who
will therefore rely heavily on the AOW in retirement.

RISK SHARING – STILL THE SECOND PILLAR PANACEA?

Defined benefit (DB)

As noted above, the collective sharing of risk, to ensure fairness between all stakeholders (comprising employer, employees and pensioners) though principally within and between all generations of member, is central to the second pillar of the Dutch pension system. However, even in a defined benefit (DB) context, this risk sharing isn’t applied in a totally equitable and symmetrical manner in either the good or bad times. For instance in the bad times, some stakeholders take a greater hit than others in restoring the financial viability of the fund, whether by:

  • The employee and/or employer being required to increase their pension contributions;
  • Limited or no indexation being applied to member benefits;
  • No catch up indexation for past missed indexation being applied;
  • Reducing pension entitlements (as a last resort).

Moreover, combining a uniform contribution rate with a uniform accrual rate isn’t the equitable panacea it appears to be, in that it leads to an inequitable value transfer from low to high earners, from the short lived to the long lived and from those early in their careers to those later in their careers – where the value of each pension accrual increases significantly with age.

Defined contribution (DC)

For defined contribution (DC) schemes, this central tenet of risk sharing all but disappears. As the ultimate amount of pension depends on the contributions paid during the accrual phase,11 the return on these contributions and the terms on which the resulting capital sum is annuitised at retirement, the employee bears both the investment risk and the interestrate risk. A typical individual DC scheme uses an age-related contributions table, set by the Dutch tax authorities, such that contributions increase with age.

DC is popular mainly with small and medium-sized companies and start-ups, because the costs of the pension plan and the amount of individual contributions are easy both to ascertain and considerably lower than for DB, while the framework is flexible. That said, individual DC schemes are relatively rare in the Netherlands. Rather, DC is often combined with DB. An individual may, for example, accrue DB benefits up to a certain income level, beyond which DC benefits are accrued.

Collective defined contribution (CDC)

Collective Defined Contribution (CDC) schemes set a target or “ambition” level of benefits to be paid as a pension for life with variable increases. Based on a fixed level of employer contributions, set for a fixed term of at least five years, after which they are reviewed, a CDC scheme, unlike DB, has the scope to redefine the benefits it offers if circumstances change, without the employer being required to pay shortfall contributions.12 As there is no sponsor covenant, there are no guarantees. So, whereas the amount of pension is based on salary and service years, just as it is for DB, if the fixed contribution rate and the investment returns prove to be insufficient to support this accrual, then the pension benefit will be necessarily lower than originally intended.
By pooling money into a single fund, with a desired long-term asset mix, CDC pension arrangements allow employees to pool the risks of investing and longevity, achieve lower costs through economies of scale and permit greater and longer-term exposure to growth and illiquid assets. Crucially, CDC, through employee risk sharing, also provides intergenerational fairness by not allowing one cohort to benefit significantly at the expense of another. CDC therefore provides a half way house between the generous pension provision, collective passivity and certain outcomes of DB and the greater individual responsibility, less generous and less certain outcomes of DC.

Preserving sustainability and risk sharing

The Netherlands three-pillar pension system, with its clear and simple rules forged through political and stakeholder consensus, is also supported by pension-saving being seen as the norm, pension assets that have almost doubled over the past decade to over 200% of GDP13 and the active promotion of older age labour force participation, given the seemingly inexorable rise in life expectancies.14 Moreover, over the past couple of decades the Dutch pension system has shown considerable capacity to adapt its three pillars to changing circumstances and the need to preserve the sustainability of its system.
Indeed, to help preserve the sustainability of the first pillar, since 2013 the AOW retirement age has increased from age 65 to 67 today and will increase broadly in line with increases in average life expectancy from 2022.15 The sustainability of the second pillar came into sharp focus in 2003, following the collapse in funding levels resulting from the dot.com bust. Unlike in the Anglo-Saxon pension markets where, in response, final salary DB plans increasingly become displaced by DC, the Netherlands instead sought to improve the risk management, hence the sustainability, of its DB plans. By moving to a stricter regime of valuing DB liabilities16 and adopting more stringent funding level, or coverage ratio,17 requirements, most DB plans moved from final salary to lifetime average earnings as the basis of accruing benefits and introduced the conditional indexation of accruals. Others, however, moved to CDC.
The global financial crisis of 2008/09 provided yet another severe test for the country’s DB plans, with the subsequently dramatic declines in the risk-free interest rate driving up the value of scheme liabilities to unprecedented levels. With many schemes breaching the stricter funding level requirements, for some this meant imposing conditional indexation to its fullest extent in an attempt to restore the financial viability of the fund, while for others it meant taking the more extreme action of cutting benefits. In an effort to make DB plans more resilient to financial market (and, indeed, longevity) shocks, and to prevent further cuts in pension benefits resulting from abrupt changes in funding ratios, a smoothed funding ratio and provision to rectify scheme underfunding over a 10 year period was introduced.18
This regulatory rethink helpfully coincided with the Dutch Central Bank19 accelerating DB consolidation, principally among corporate schemes, resulting in the circa 1,000 schemes that existed in 2000 becoming about 200 today – vastly improving scheme governance and slashing administration and service costs in the process. Combined, these measures should, in the face of a future financial crisis, culminate in fewer pension cuts. However, restoring full indexation and backdating past missed indexation has proved more difficult, given the high threshold funding level required.

Reshaping the second pillar

Of course, the next really big challenge for the Netherlands is to further reshape the second pillar to take into account not only unrelenting demographic headwinds, changing lifecycle and career patterns but also ultra-low/negative interest and, more latterly, the economic devastation wreaked by the Covid pandemic. Consequently, this reform will not simply be an evolution of the system, that has existed since the 1950s, but a genuine revolution in pension design aimed at making the system future-proof, with the collective sharing of risk remaining central to the system’s ethos.

After almost a decade of debate, the end game, which will see all DB pension schemes move to a CDC-like system, will centre around two new contract options: the new pension contract (NPC) and the improved defined contribution plus (improved DC+) scheme.

The former by far represents the biggest philosophical shift from the accepted norm, in that the long-held central principles of combining a uniform contribution rate with a uniform accrual rate for all and guaranteeing the level of member benefits will disappear. Rather, the NPC will convert all DB member accruals to an individual capital entitlement, or a notional account, within a single collective investment pool. Of course, the question of who will absorb the existing deficits of those DB schemes that remain underfunded must be addressed.

Going forwards, an annual benefit, which can be adjusted up or down, will then be allocated to the member’s notional account throughout their lifetime, i.e. throughout both the accruals and decumulation phases. Based on a system of age-specific allocation of returns, these benefits will be determined by a combination of fixed contribution payments,20 annual investment returns adjusted for an actuarially-determined return or deduction reflecting interest rate movements and transfers to or from an investment solidarity reserve, which would seek to smooth the intergenerational effects of investment sequencing risk.

Capped at 15% of total fund assets, this solidarity reserve will be funded by up to 10% of total contributions and 10% of “excess” investment returns, i.e. those generated above a defined threshold return. Of course, the rules of exactly how and when this reserve could be distributed among the membership would have to be transparently determined in advance. Indeed, from 2026, as part of a very transparent process, members should be able to compare projected returns by age cohort.

Additionally, those older pension members disadvantaged by the move to degressive accrual will need to be compensated, either by the pension plan or by the employer. This will take the form of either a payment or an additional, tax-incentivised, pension contribution, albeit capped at 3% of the employee’s pensionable earnings, for a 10-year period, ending in 2036 latest or when the employee changes jobs (with the new employer compensating them on the basis applied to its own workforce). Depending on who funds this compensation and how, this could result in a substantial cost to employers and may see employers revising their salary structures accordingly.

More evolution than revolution is the design of the improved DC+ scheme which will replace the current age-dependent DC contribution structure with a flat-rate contribution (though age-related contributions in existing DC schemes will continue) and the default of annuitisation at retirement
with the continued investment of capital (though members can opt out of the latter if annuitisation is preferred). Of course, given the lower costs and governance requirements associated with DC, the improved DC+ scheme may well be an attractive alternative to the NPC for many corporate and smaller sector DB funds.

WHAT ARE THE LIKELY ASSET ALLOCATION SHIFTS FROM TRANSITIONING TO THE NPC?

The implications of moving to the NPC structure on pension fund investment will, undoubtedly, be acute, given the shift in focus from coverage ratios to investment returns and from managing regulatory capital to managing economic capital.

While diversification will remain integral to investment strategy, in all likelihood asset allocation will be shaped by three fundamental factors:

  1. Given the age-specific allocation of returns, the emphasis will shift to creating investment strategies that take into account the risks associated with each age cohort.
  2. Cash flow driven investment (CDI) is likely to take precedence over liability driven investment (LDI), resulting in some traditional hedging assets being sold in favour of shorter-dated credit, though interest rate hedging will remain a cornerstone of investment policy.
  3. A longer investment timeframe should favour a greater allocation to risk assets and perhaps more illiquid investments, not least to private markets.

Although the exact size and timing of the resulting asset shifts are difficult to call at this early stage, especially against the backdrop of an ever greater focus on integrating ESG and climate risks into investment decision making, one thing is for sure, they will be seismic.21 Indeed, anecdotal evidence suggests that these asset allocation shifts have already begun.

So can the Dutch pension system retain its pre-eminent position?

As we know, pre-eminence in pension system design derives from three broad factors: adequacy, sustainability and integrity (or security). As noted earlier, these are the metrics upon which the influential annual Mercer CFA Institute Global Pension Index Report (MCFAGPI) is based. Pre-eminence also derives from equity through risk sharing – a mainstay of the Dutch pension system since its inception.

By moving from DB to a CDC-like system, while the sustainability box is almost certainly ticked, adequacy and integrity could potentially be called into question. However, in making this assessment, there are three crucial factors to bear in mind:

  1. Adequacy: The current pension benefits formula of seeking to achieve an income replacement ratio of 75% of the median wage for 40 years contributions and 80% for 42 years will remain unchanged.
  2. Integrity: Applying investment returns adjusted for interest rate movements and transfers to or from a solidarity reserve to individual notional capital accounts should ensure each member receives a fair asset share throughout the accruals and decumulation phases.
  3. Adequacy and integrity: Members disadvantaged by the new arrangements should be adequately compensated.

The last two points additionally address the intergenerational fairness question of sharing the burden of paying pensions across the generations.

With the reforms having been submitted to the Dutch parliament in June 2020, following the signing by the social partners (government, unions and employers)22 of a framework agreement a year earlier, the legislation and accompany regulation is expected to be finalised by the end of this year, with the transitioning taking place between 1 January 2022 and 1 January 2026, latest. In addition, it is expected that by 1 January 2024, the social partners would have reached the necessary decisions regarding each new pension arrangement, the transfer of accrued rights and the compensation payable to older cohorts.

Of course, given the complexity associated with agreeing the most favourable terms for employers and employees, the transitioning of accruals and the terms of the compensation, much of the detail has yet to be determined. Indeed, employers will be required to publish a transition plan that includes information on the available choices, the scheme’s rationale and supporting calculations. Not only will the social partners have to agree with the proposed pension scheme arrangements put forward by the employer, but they and the pension scheme board of trustees can demand alternative provisions if the transfer is considered detrimental to particular member cohorts. In short, none of this will be easy.

Conclusion

Ultimately, this reform of the second pillar must be seen in the context of the need to future-proof the Dutch pension system against a world increasingly characterised by rapid economic and social change, shifting lifestyles and career patterns, overwhelmed public finances and the inability of the state to sustainably support structural demographic challenges.

When evaluated within this frame, and taking on board the generationally fair outcomes that should result from the reforms, the Dutch pension system should retain its pre-eminence, though agreeing the terms and the transition process, as already noted, won’t be without its challenges. In fact, those other pension systems facing exactly the same headwinds as the Netherlands, albeit with bigger adequacy and sustainability issues to resolve, would do well to observe the structure of these reforms and the process that’s lead to where the Dutch pension system is today. Indeed, all eyes will now be the implementation of these reforms and the resultant outcomes. However, even before these reforms have fully played out, it would seem that once again, the Dutch pension system is about to become the poster child for sustainability, adequacy and integrity.

The contributions of Rob Schreur, Martijn Vos, Anita Joosten and Prosper van Zanten are gratefully acknowledged.

8 April 2021
Chris Wagstaff
Chris Wagstaff
Head of Pensions Investment & Education
Share on twitter
Share on linkedin
Share on email
Key topics
Related topics
Share on twitter
Share on linkedin
Share on email
Key topics
Related topics

PDF

Revolution on the horizon: Will the Dutch pension system still set the bar for sustainability, adequacy and integrity?

1 Thinking Ahead Institute. Global Pension Assets Study 2021. pp.5 and 21.
2 Thinking Ahead Institute (2021). op.cit. p.21.
3 Mercer (2020), Mercer CFA Institute Global Pension Index, Mercer, Melbourne.
4 Thinking Ahead Institute (2021). op.cit. p.16. Split by assets under management, the Dutch pension system is 94% DB and 6% DC, while the Japanese market is 95% DB and 5% DC.
5 While membership of a pension scheme in the Netherlands isn’t mandatory (although sector-wide funds can, with approval from the Ministry of Social Affairs and Employment) stipulate compulsory membership) around 75% of employees accrue their second pillar benefits via an industry-wide pension fund. However, employer participation is typically mandatory for those employers operating in these industries, thereby allowing employees to move between employers within the same industry without impacting their pension rights. An employer can only opt of an industry-wide pension fund if they are not subject to mandatory participation and can offer a better corporate pension plan to its employees than the incumbent industry-wide scheme. Although corporate schemes account for about 80% of schemes by number, industry-wide schemes are, as already noted, far more dominant in terms of assets under management and, by the number of active members. The five biggest industry-wide plans, which account for more than 50% of Dutch pension assets, are: ABP (Government employees and education), PFZW (Care and welfare), PMT (Metal and technical workers), BpfBouw (Construction) and PME (Metal/electro manufacturing). Corporate plans include those sponsored by household names such as Shell (SSPF), Heineken and KLM.
6 Average earnings schemes usually feature conditional indexation, whereby pension entitlements pre and post retirement are revalued each year in line with price inflation or, to a lesser extent, the wage increase in the sector, unless the financial position of the fund is judged by the pension fund board, against regulatory funding requirements, as being unable to provide indexation. See: OECD (2019), Pensions at a Glance 2019: OECD and G20 Indicators, OECD Publishing, Paris.p.137.
7 For each year an individual lives or works in the Netherlands, they build up 2% of a full public pension under the General Old Age Pensions Act (Algemene Ouderdomswet, AOW). After 50 years of Dutch residency the individual is entitled to receive the full AOW pension at the AOW retirement age. Currently, this is equivalent to 29% of gross 2018 average earnings, against an OECD average of just over 20% for other such non-contributory first pillar benefits and just under 15% for contributory. The AOW pension is adjusted twice a year in line with wage inflation. See: OECD (2019). op.cit. p.135.
8 Gross replacement rates are 70.9% for the average male earner versus 49% OECD average, and 73.5% and 70.1% for low and high earners respectively, versus the OECD average of 60% and 44.7%. See: OECD (2019). op.cit. p.147.
9 The Netherlands has the world’s third lowest old age poverty rate at c.3% of those aged 65+, compared to an OECD average of c.13%. See: OECD (2019). op.cit. p.23.
10 OECD (2019). op.cit.p.22 The gender pensions gap is the difference between the pensions of men and women at age 65 as a percentage of men’s pensions.
11 Employees tend to contribute about one-third and the employer two-thirds of the overall contribution.
12 Thereby providing the employer with more predictable costs than for DB.5
13 Thinking Ahead Group (2021). op.cit. pp.12 and 25. The exceptional growth in this ratio over the past decade, which quantifies the level of assets set aside to pay future benefits, enables the Netherlands to lessen intergenerational inequality through risk sharing. Only Canada and Australia come anywhere close to this ratio at 192.5% and 174.8%, respectively. Additionally, the Netherlands hasn’t succumbed to the temptation of many governments during the Covid pandemic of allowing temporary access to accrued pensions (or temporarily reducing the level of compulsory contribution rates), which will likely have a longer-term material impact on the adequacy, sustainability and integrity of pension systems.
14 The Netherlands has experienced amongst the highest growth rates of employment among the 55-64 year old cohort since 2000. See: OECD (2019). op. cit p.24.
15 Linking the statutory retirement age to life expectancies applies to those born after 30 September 1957 – the year in which the AOW was created. Source: European Commission Employment, Social Affairs and Inclusion, Netherlands Retirement Pension.
16 The practice of applying a fixed 4% discount rate to valuing liabilities was replaced by applying a nominal risk-free interest rate.
17 The coverage ratio is that between a scheme’s assets and the liabilities and is used to express the financial viability of the pension fund.
18 Since 1 January 2015 the, so-called, policy funding ratio determines the financial health of Dutch pension funds. Taken as the average of the funding ratio of the last 12 months, it compares the assets of a pension fund to its liabilities. Should the index reach 100%, the fund is able to pay all pensions. Should the ratio become too low, funds are forced to cut their pensions. The policy funding ratio is reported to the Dutch Central Bank (De Nederlandsche Bank NV, or (DNB) on a quarterly basis. In Q320, the five largest Dutch pension funds reported the following policy funding ratios to the DNB: ABP (95.8%), PFZW (96.5%), PMT (97.6%), BpfBouw (112.4%) and PME (96.9%). Source: Statistica, 8 January 2021.
https://www.statista.com/statistics/580417/policy-funding-ratio-of-the-five-largest-pension-funds-in-the-netherlands/
19 The DNB both authorises pension providers, on the basis on having sufficient financial assets and a fit and proper board of trustees, and closely monitors their financial and management operations.
20 Unlike CDC, these fixed contributions (which, over time, can be re-evaluated) wouldn’t be determined on the basis of a targeted or an “ambition” level of benefits. Additionally, tax eligible annual pension contributions would be capped at between 30% and 33% of pensionable earnings until 2035 and then reviewed every five years.
21 See: Reforms will see Dutch pensions transfer €500bn of assets. Natalie Tuck. European Pensions. 30 June 2020.
22 The Dutch social partnership model gives trade unions and employers a decisive say in areas like pensions policy.

Important information

The research and analysis included on this website has been produced by Columbia Threadneedle Investments for its own investment management activities, may have been acted upon prior to publication and is made available here incidentally. Any opinions expressed are made as at the date of publication but are subject to change without notice and should not be seen as investment advice. Information obtained from external sources is believed to be reliable but its accuracy or completeness cannot be guaranteed.

Related Insights

15 April 2021

Edward Al-Hussainy

Senior Interest Rate and Currency Analyst

The spectre of rising inflation

With the help of Edward Al-Hussainy, Senior Interest Rate and Currency Analyst, and Adrian Hilton, Head of Global Rates and Emerging Market Debt, the Eye of the Needle team tackles the hot topic of potentially rising inflation, why it might happen, what it would mean for consumers, companies and investors, and how central banks might go about tackling it.
Read time - 2 min
12 April 2021

Fixed Income Desk

In Credit - Weekly Snapshot

In Credit Weekly Snapshot - April 2021

Our fixed income team provide their weekly snapshot of market events.
Read time - 3 min
12 April 2021

Guillaume Langellier

Credit Research Analyst

Office-working: long-term symptoms or full recovery?

With the advent of the Covid-19 pandemic, millions of office workers worldwide were forced to work from home (WFH) for the majority of 2020.
Read time - 5 min

You may also like

Investment approach

Teamwork defines us and is fundamental to our investment approach, which is structured to facilitate the generation, assessment and implementation of good, strong investment ideas for our portfolios.

Funds and Prices

Columbia Threadneedle Investments has a comprehensive range of investment funds catering for a broad range of objectives.

Our Capabilities

We offer a broad range of actively managed investment strategies and solutions covering global, regional and domestic markets and asset classes.