Sector Overview Pensions 2022

Publication: 25 May 2022 The state of affairs in the pension sector Experience with premium agreements is still limited Sector Overview Pensions – 2022 Report

2 Sector Overview Pensions – 2022 Contents Introduction Sector Overview 2022 page 3 01 Size of the second pillar pension sector page 9 Highlights page 5 02 Accumulation phase page 17 03 Payout phase page 21 04 Characteristics of the scheme page 25 05 Choices during the accumulation phase (premium agreements) page 36 06 Choices at retirement date (premium agreements and payout agreements) page 40

3 Sector Overview Pensions – 2022 Introduction Sector Overview 2022 The pension system reform is approaching rapidly. This reform entails, among other things, a transition to premium agreements. This has a major impact on pension providers, employers, employees, and pensioners. The insights from this second Sector Overview Pensions ('Sector Overview') can be valuable for pension professionals when making implementation choices or designing pension schemes. The AFM therefore urges them to read this report and use the insights in decision-making. This enables pension providers and social partners to compare the characteristics of their own scheme and the choice options for participants with sector-wide insights. This can help them in setting up and evaluating their own pension scheme and service delivery.

Following the adjustment of the Dutch pension system, all participants will build up pension via a premium agreement. The focus of the Sector Overview is therefore on premium agreements; this type of agreement brings different risks for participants than the currently common payout agreements. The premium agreement is new not only for many participants but also for many pension providers. Many of the characteristics that determine the design of premium agreements are therefore also relevant for pension providers who will execute this type of scheme in the new pension system for the first time. Think, for example, of the investment freedom that is or is not offered to participants. Is it worth offering opt-out investing if only a small portion of participants use it sector-wide? Pension providers will also have to make choices regarding the offer and design of the payout(s) they wish to offer. Previous AFM research shows that not all pension providers take sufficient account of the characteristics and needs of participants, such as the risks they want and can take, when developing variable pension products. They also do not always guide participants well in choosing between a variable or fixed pension payout. This can have far-reaching financial consequences for participants. Good choice guidance is therefore even more important in the new system.

This Sector Overview was created based on data provided by all Dutch pension funds, pension insurers, and premium pension institutions (the 'pension providers'). The insights in this Sector Overview are based on information from the reporting year 2020 (reference date 31 December 2020). The figures in this report are, like last year, based on the provided data. For the sake of efficiency and robustness of the sector, it was decided to leave the requested data points unchanged. The AFM plans to keep the reporting unchanged until the transition to the new pension system, i.e., until the reporting year 2023.

4 Sector Overview Pensions – 2022 When the legislation for the new system is final, the AFM will align the reporting with it and consult the sector.

In this report, annual developments in the explanations have been included insofar as the difference with last year is relevant and material. From next year, the AFM expects to publish figures that include annual developments. Data from three consecutive years will then be available. During the creation of this Sector Overview, deviations in the data were found for some providers that indicated reporting errors in 2019. As a result, some data points for the reporting year 2019 included in this Sector Overview may differ from last year. In case of a large deviation, this is explained in the text.

The AFM continues to monitor developments and trends in the sector. The AFM uses the insights to execute its supervision in a risk-based and efficient manner. With this publication, the AFM shares the information from and about the pension providers with the pension sector, so that they can take note of the sector-wide insights. This publication is not traceable to individual pension providers.

Below you will first find some highlights from the Sector Overview. In the following chapters, the AFM goes deeper into the findings from the Sector Overview.

5 Sector Overview Pensions – 2022 Highlights Size of the second pillar pension sector (Chapter 1) The total pension assets have increased from € 1,645 billion to € 1,836 billion. Despite the corona crisis which caused stock markets to fall in March 2020, the year 2020 was a good investment year for most pension providers. Dutch pension providers execute pension schemes for approximately 410,000 employers in 2020 and managed a total of € 1,836 billion in assets at the end of 2020. Of this, € 1,220 billion is in the accumulation phase and € 616 billion in the payout phase. Pension funds are responsible for the largest part of this assets, 90% of the pensions in the accumulation phase and also nearly 90% of the ongoing payouts are held with pension funds. The pension assets (in the accumulation phase) amounting to € 1,220 billion are largely managed by the ten percent largest providers; they manage the pension of more than eight out of ten participants in the accumulation phase. In total, this concerns nearly 21 million pensions. These pensions are built up by active and former participants. One person can build up multiple pensions, both with one provider and with different providers.

While pension assets increased significantly in 2020 compared to 2019, the number of Dutch second pillar pension providers decreased from 206 to 190. The pension providers that stopped second pillar pension are mainly pension funds (fifteen in total, of which eleven enterprise pension funds) and one insurer. The pension providers that are still active consist of 171 pension funds, twelve insurers, and seven premium pension institutions ('PPIs'). On closer inspection, the sector seems to be developing towards a small group of providers, mainly pension funds, which are responsible for a large part of the Dutch pension assets.

Accumulation phase (Chapter 2) More than two-thirds of pension funds still do not execute premium agreements. How much experience do pension providers have with executing premium agreements? This is a relevant question because after the system reform, all participants will build up pension via a premium agreement.

There are seven million active participants who build up pension in the Netherlands, of whom one in five does so via a premium or capital agreement. The group of premium and capital agreements is referred to as 'premium agreements' in this report where applicable. The total value of premium agreements is still limited compared to payout agreements. Only 6% of the total value of pensions in the accumulation phase (€ 1,220 billion) has been built up via a premium or capital agreement. This means that in 2020 almost all value in the accumulation phase was built up within a payout agreement (€ 1,146 billion).

The largest part of the built-up value of premium agreements (70%) is held with the ten insurers that execute premium agreements. Of these insurers, seven actively offer premium agreements. Less than one-third of the pension funds currently have experience with offering premium agreements: of the 171 pension funds, 53 funds offer a premium agreement and only 45 funds currently actively offer a premium agreement. The system change will therefore mean that two-thirds of the pension funds will execute a type of agreement they have not executed before.

Payout phase (Chapter 3) Less than 6% of ongoing old-age pensions from premium agreements consist of variable payouts. The sector has limited experience with executing pension payouts that result from a premium agreement. The number of ongoing payouts from a premium agreement stands in contrast to the number resulting from a payout agreement; 171 thousand versus 3.8 million.

Also, the value of payouts resulting from premium agreements is a fraction of the value resulting from payout agreements. Of the total € 616 billion in ongoing payouts, € 513 billion is an old-age pension payout resulting from a payout agreement and € 13 billion results from a premium agreement. This means that only 2% of the value of all ongoing payouts comes from a premium agreement.

Another point of attention is the limited experience of pension funds with variable payouts. Of the 53 pension funds that offer a premium agreement, only eight offer a variable payout. The size of these variable payouts is very limited compared to the total value of all ongoing payouts; less than 6% of the value of payouts from a premium agreement is a variable payout. In other words, 0.1% of the value of all ongoing payouts is a variable payout.

The thirteen pension providers that offer variable payouts all expect (almost) that the share of variable payouts in the portfolio will increase strongly from 2023 onwards. Here too, the expectation is that the system change will have a major impact on the vast majority of pension providers and participants.

At the end of 2020, nearly € 2.4 billion in pension value had not yet been claimed by 370,000 participants, meaning that the average value of an unclaimed pension is approximately € 6,500.

Characteristics of the scheme (Chapter 4) Seven out of ten inactive participants with a premium agreement have built-up pension assets under € 10,000 and thus below the buy-out threshold. The question is whether the current design of premium schemes generally leads to an adequate pension when the majority of inactive participants with a premium agreement have a built-up value below the buy-out threshold of € 10,000.

For participants with a pension below the buy-out threshold, the pension is paid out in a lump sum at retirement date and no monthly pension payout is received. Because in the current system premium agreements are not always a basic scheme, but also occur as net or excess schemes, it cannot be stated that a low built-up value within a premium agreement leads to disappointment, but when a large part of participants has a built-up value under € 10,000, this is a signal to evaluate the scheme. With regard to the upcoming system change in which pension schemes become premium agreements, it is important to pay attention to this.

Pension schemes are designed very differently and the characteristics of the scheme, such as the basis for the premium or the way disability is covered, are highly determinative for the amount of the pension and the risks participants face. Even with premium agreements, the amount of the annual pension premium (partly) determines the amount of the final pension. For 216,000 participants with only a premium agreement as a basic scheme, the annual premium does not increase as the participant gets older; the annual premium is a fixed percentage of the pension basis, meaning that for this group there is a regressive pension build-up. Of this group of participants, 77% have an annual premium lower than € 1,000. This can lead to more disappointment regarding the final amount of the pension.

There are nearly 1.3 million active participants who have a premium agreement as a basic scheme without having an excess scheme with the same pension provider. More than a quarter of these participants (336,000) have a capped salary. Of this latter group, approximately 29% have an annual premium lower than € 1,000 and 28% have an annual premium between € 1,000 and € 2,500. For nearly eight out of ten participants within these groups, it applies that they have a built-up value smaller than € 10,000, and thus lower than the buy-out threshold. Such a characteristic of the scheme thus seems to be an indicator that the pension build-up (for many participants) is limited.

When designing the scheme, one can also think of partner pension. With premium agreements, a certain partner pension occurs proportionally more often. With a certain partner pension, the partner must be known to the pension provider at the time of the participant's death for the granting of the partner pension. Nearly 30% of all participants with a partner pension in a premium agreement have a certain partner pension, compared to 3% in a payout agreement. If this participant has not registered their partner, it may be that there is no right to partner pension when the participant dies. For nearly a quarter of all active participants, the partner pension is also covered on a risk basis. In the event of a change of employer, this is disadvantageous for these participants. The covered partner pension on a risk basis expires upon termination of the employment relationship. A risk here is that participants are not aware that the partner pension expires upon termination of the employment relationship, which can lead to considerable disappointment.

Choices during the accumulation phase – premium agreements (Chapter 5) Less than one in ten participants opt for variable. In the new pension system, variable payouts play a much larger role than now. It is up to pension providers to guide participants sufficiently in this.

Currently, the fixed payout is the default payout in most cases, and very few participants deviate from this default pre-selection choice; by far the majority of participants, more than 9 out of 10, currently opt for a fixed payout. The question is to what extent participants have consciously chosen this and whether the pre-selection choice aligns with the characteristics and preferences of the participant.

In light of the above, it is striking that the differences between pension providers are large; in some pension providers, no less than 90% of participants choose to opt for a variable payout, even when the default is fixed.

A small two-thirds of participants in a premium agreement have investment freedom. These participants have the possibility to choose between one or more investment mixes defined by the pension provider in the form of lifecycles (profile investing) or to choose the investments themselves (opt-out investing). Participants with a premium scheme at pension funds have less often investment freedom (19%) compared to participants at insurers (88%) and PPIs (94%). This is because pension funds with many participants in a premium scheme do not offer this.

It is striking that participants who have investment freedom hardly use it and mostly invest according to the default investment profile. Of the small group of participants with investment freedom at pension funds, only 8% choose a different profile than the default investment profile and only 1% for opt-out investing. This raises the question of whether it is worth offering opt-out investing if only a small portion of participants use it.

Choices at retirement date – premium agreements and payout agreements (Chapter 6) Participants make little use of the choices they have at retirement date, such as a high-low structure. Participants can make a number of choices at retirement date regarding the pension payout they will receive, such as variation in payout amount after pension starts (high-low or low-high structure) or converting partner pension into a higher old-age pension and vice versa (exchange options). Other choices at retirement date, such as part-time retirement, are not part of this Sector Overview.

Most participants who reached retirement date in 2020 and had the possibility to vary in the payout amount of the pension did not use this. Around 36% made use of the high-low structure and almost no one (approximately 1%) chose a low-high structure. The question is whether participants have no need for varying the payout amount, or if there are other reasons such as unnecessary barriers in the choice environment.

At pension funds, participants with a premium agreement at retirement date had less often the choice to vary in the payout amount than at insurers or PPIs. For example, most participants with a premium agreement at pension funds (89%) had no choice to vary in the amount of the payout, while this was only 7% at insurers and PPIs.

Also, the exchange options, for example converting partner pension into a higher old-age pension, were hardly used by participants who retired in 2020. More than two-thirds of participants did not use this. When a participant has no partner at retirement date, the question is whether it is wise not to exchange the built-up partner pension for old-age pension. This certainly applies when the scheme determines that the partner situation at retirement date is decisive for whether or not a partner pension is paid out. In that case, the AFM expects the pension provider to ensure adequate guidance and prevent a pensioner from having a partner pension that will never be paid out.

9 Sector Overview Pensions – 2022

1. End of 2020, there were 190 pension providers of second pillar pensions in the Netherlands For this Sector Overview, the AFM asked all pension providers established in the Netherlands to fill in the supervisory report. Most providers of second pillar pensions are pension funds, as shown in Figure 1a.

On the reference date 31 December 2020, there were a total of 190 Dutch pension providers of second pillar pensions active, of which the vast majority are pension funds (90%). Furthermore, there are twelve insurers that execute second pillar pensions and seven PPIs.

The number of pension providers has decreased from 206 (end of 2019) to 190 (end of 2020). In 2020, there were thus sixteen fewer pension providers than in 2019, of which fifteen pension funds and one insurer.

The number of pension providers, by type of pension provider. The largest part of the pension funds consists of enterprise pension funds (Figure 1b). Compared to all types of pension providers, the number of enterprise pension funds has decreased most strongly compared to 2019. Where there were 123 enterprise pension funds in 2019,

Figure 1a: Number of pension providers 01 Size of the second pillar pension sector

10 Sector Overview Pensions – 2022 01 the number in 2020 shrank to 112. This is a decrease of 9%.

Of the total 171 pension funds that submitted a report, six were in liquidation. The pension funds in liquidation had not yet transferred all existing rights and obligations to the new pension provider at the end of 2020. Pension funds in liquidation that had transferred the pensions by the end of 2020 did not report and are not part of this reporting.

The number of pension funds by type of pension fund. 2. The total value of second pillar pensions end 2020 is € 1,836 billion The AFM distinguishes two phases in this Sector Overview: (1) the accumulation phase, i.e., the period until retirement date during which an (former) participant builds up pension; and (2) the payout phase, i.e., the period during which a pensioner receives a pension payout. The total value of the accumulation phase amounts to € 1,220 billion, of which the largest part (90%) is built up at pension funds. By value, the AFM understands in this report the value of the pensions in euros. For a pension based on investments (such as with a premium agreement), the value is the number of participations times the price. For a pension claim (such as with a payout agreement), it concerns the technical provision. This concerns premium, capital, and payout agreements. See Figure 2a.

The total value of second pillar pensions in the accumulation phase, by type of pension provider. Figure 1b: Number of pension funds Figure 2a: Value in accumulation phase (in billions €)

11 Sector Overview Pensions – 2022 01 The built-up value in billions per decile and the share of each decile expressed as a percentage of the total value in the payout phase. Each decile represents the successive 10% of pension providers by size by value.

In the above Figure 2b, it is made clear how the total value in the accumulation phase is distributed among the different pension providers. The group of the 10% largest pension providers (by value, decile 1) manages more than 77% of the built-up value. Compared to last year, this is a small decrease, namely from 79% in 2019 to 77% in 2020. In this decrease of two percentage points, it must be taken into account that the first decile, i.e., the 10% largest pension providers, consists of fewer pension providers this year and that this thus has an effect on the size of the deciles. Where last year the first decile consisted of the twenty largest pension providers, the first decile in 2020 consists of only nineteen pension providers. This is of course because the total number of pension providers in 2020 (190) has decreased compared to last year (206). This explanation also applies to Figure 2d explained below.

The total value of all ongoing payouts together amounts to more than € 616 billion, of which nearly 90% is held with pension funds. See Figure