Royal Society of Arts (RSA) – Tomorrow’s Investor Programme

Evidence to DWP Select Committee
Inquiry into Collective Pensions

We are delighted that the DWP Select Committee has decided to launch an investigation into the introduction of CDC pensions.
For the past seven years, the RSA, Britain’s oldest independent think tank, has sponsored a programme known as Tomorrow’s Investor. One of its key recommendations has been that Britain should allow the introduction of collective pensions, a choice denied in this country, despite their success elsewhere in the world. During those years significant support was developed for this reform, including from the CBI and TUC.
We were delighted when the government responded with the 2015 Pensions Act. This gave the government powers to allow “Defined Ambition” pensions including CDC. The Act was passed with full cross party support. However the secondary legislation to make the legislation a reality has not been written. In statements made at the time, the government indicated that they remained committed to the policy and delay was the result of resources and priorities. Therefore Parliament has agreed the introduction of CDC and similar pensions. The issue is to establish a regulatory framework to allow them to operate effectively.
Parliament has agreed to the introduction of CDC pensions. The issue is to establish a regulatory framework to allow them to operate effectively.
Studies that have modelled CDC pensions suggest they can achieve a profound improvement in pension outcomes for those who seek “a predictable income from the time they retire until the time they die”.  Indeed they suggest that, for the same payment, CDC will provide a 30% higher pension, and one that is more predictable. Citations of these studies are given in Appendix 1, and an explanation of how this large increase comes about is given in the response to your questions.
The recent announcement by Royal Mail of their desire to introduce collective pensions is indicative of both the immediate need and demand for CDC, and means that early action by the government to implement the structures, which Parliament has sanctioned, should be a priority. But CDC also has larger relevance in creating a structure, which can provide effective pensions to all British people.
For this reason we see three opportunities for CDC, which the Committee may wish to consider.

1. The most urgent is to allow Royal Mail and other likeminded employers to introduce CDC pension if this is supported by their workforce.
2. The second would be to establish a system to allow qualified bodies to establish a CDC structure for pensions in payment, (decumulation). This is an important compliment to “Freedom and Choice”, offering pensioners an effective default choice other than annuitisation.
3. Finally the Committee may wish to consider how a low cost national pension system for pension saving might be established to allow every British citizen to provide effectively for themselves in retirement.

We see these reforms as being supportive to the successful introduction of automatic enrolment which addresses the issue of pension savings (accumulation), but not (as noted in point 2), to pension decumulation.
In this note we aim to address the issues posed by the Committee.

In Section 1 we offer a short introduction.
In Section 2 we address the questions the Committee has asked.
In Section 3, we raise and address other considerations in effective design of CDC pensions.
Finally in Appendix 1 we summarise and provide links to the major studies previously undertaken to look at  CDC, and the conclusions they have reached.

1: Introduction

The RSA and Collective Pensions

The work of the RSA Tomorrow’s Investor Programme aimed to investigate what would be the best framework for a private pension system in the UK. It began with a series of “citizen juries” who received evidence on how the system currently worked, and were asked to adjudicate where reform mighty best be merited. The work was overseen by an advisory board chaired by Sir John Banham, a former Director General of the CBI, and included inter alia a former deputy head of the FSA. The work itself was lead by David Pitt-Watson, together with Dr Hari Mann and the staff of the RSA.
Most jurors concluded that what they wanted was a pension system into which they could save their money, which they could trust would provide them with a retirement income. Some wanted greater involvement and choice. But for most, the desire was to “give their money to someone they could trust” which would then provide an income from retirement until death. In other words, a trustworthy default pension system which would maximise income in retirement.
Sadly in the UK that requirement is not met. Currently those that do not have a DB pension, are on retirement presented with a large sum of money. With this they can buy an annuity. But currently annuities are very expensive, and indeed the cost of an annuity is volatile—hence the saver cannot predict how much will be needed for a certain retirement income.

The alternative of a “pension drawdown” is an effective one for those who, however long they live, have no fear that they will ever run out of income. However for those who are not so privileged, and who wish to secure an income for life, drawdown is not an effective alternative, since they do not know how long they will live.
CDC is a way to resolve these difficulties. Indeed when the British DB system was originally established, it might better have been described as “defined ambition”; that is it had some flexibility to deal with the uncertain storms which might affect the affordability of a pension, in particular returns and longevity. However, a combination of the promises associated with taking” pension holidays”, and subsequent legislation made the system rigid, and ultimately unattractive to employers and potentially unaffordable.

2. Questions asked by the Committee

The Committee has therefore asked

Benefits to savers and the wider economy

Would CDC deliver tangible benefits to savers compared with other models?

Yes. Many studies have modelled CDC outcomes compared to the alternative of annuitisation, and some to the alternative of drawdown. (They are summarised in Appendix 1)

They conclude that CDC will provide a 30%+ uplift relative to saving and then purchasing an annuity, and a considerably larger upside than a drawdown pension, if the pensioner wants to be sure of not running out income if they live to a ripe old age.

[Note, some suggest that a comparison with drawdown is unfair, since on the early death of the pensioner, any remaining assets will be inherited. This is an important point. CDC is designed to “provide an income from retirement until death”.

If the pensioner has enough assets that they have no concern that they will run out of income, there are still gains from CDC but they are more modest that for those who need to ensure their income will not run out before they die.]

Note that two of these studies have been commissioned by the government; one from the Government Actuary and one from the Pension Policy Institute.

So CDC provides better pensions in terms both of outcome, and of predictability for the saver. The Committee may find it helpful to understand why CDC creates these very significant benefits. In the illustrative example below, we lay out the logic of CDC and its advantages.

Illustrative Example

How CDC Pensions can give better outcomes

Jo is saving for a pension, and wants to have enough to provide for herself from retirement until death. She intends to retire at 65, but of course, does not know how long she will live for. Her actuary tells her that there is only a very small chance she will live beyond 100, so if she wants to be sure of an income until that time, she needs to find enough for 35 years of retirement. Jo is told that to achieve this she will need to set aside £350 per month, which will be invested in a balanced set of securities. The income secured cannot be guaranteed, but it is designed to meet the cost of a decent living in retirement. This approach is known as DC plus drawdown.

However, Jo also understands that her likely life expectancy, together with everyone else in her company is 85. That means that if they were able to save for a pension collectively, she and her colleagues would only have to save enough for 20 years. As with any insurance, those who die young might be said to be “subsidising” those who live to an old age. But everyone knows they will be provided for until the day they die. This reduces the cost a lot. If saving for 35 years costs £350 per month, savings for 20 years might cost £200 a month. Individual savings is thus 75% more costly, if the collective invests in exactly the same way as the individual drawdown would. This approach is known as CDC. Note that unfortunately Jo cannot access such a pension, because, although it has the power to do so, the government has not yet written the necessary regulations.

Instead Jo must buy an annuity. An insurance company provides these. To be sure that the payment is “certain”, the insurance company invests only in “safe”, low return securities. Once in retirement these provide a very predictable income, albeit that annuities often do not offer protection against inflation. That is not a big concern today, but might well be in the future. Equally problematic, the cost of an annuity varies with the cost of “safe” securities, so before retirement, it is difficult for Jo to determine how much to save, and what to invest it in. The different investment approach in the run-up to retirement, the investment underpinning the annuity, and the cost of annuity provision (including the profits and reserves for the insurance company), all add up. As a result, Jo needs to set aside £270 a month, 35% higher than CDC. This approach is known as DC plus annuity.

Note that these figures are developed only for this simplified example. However they are consistent with the extensive studies quoted in Appendix 1.

How would a continental-style collective approach work alongside individual freedom and choice?

Although fewer freedoms exist in continental countries with CDC, it is perfectly compatible with freedom and choice (F&C). Indeed one might argue that CDC is its perfect compliment. The aim of F&C was to avoid pension savers having to commit their savings to expensive annuities. What it did not do was to provide an alternative way for them to secure “an income from retirement until death”, save for drawdown, which, as the example shows, requires “over saving”.

CDC can be made flexible. However there is one caveat. Excessive freedom for savers to move, or cash in their savings might not be helpful either in protecting savers from the unscrupulous, or more particularly in encouraging long-term investment. (see below)

Does this risk creating extra complexity and confusion? Would savers understand and trust the income ‘ambition’ offered by CDC?

It is very important that the nature of CDC is understood. Pension outcomes are reasonable ambitions, not promises, and pensions in payment can be varied, albeit that this is rare. In Holland, for example, a flexible system of pensions had successfully provided DB-type benefits for two generations. It was not until the 2008 financial crisis that pensions in payment needed to be were reduced, to avoid intergenerational unfairness. Despite the fact that reductions were relatively small (maximum 6%, average 2%), this created considerable understandable dissatisfaction. Dutch policy makers are aware of the benefit of CDC, and the reductions are modest compared to the upside from CDC (See Question 1). So no Dutch political party is suggesting abandoning CDC. However all would argue the need for clear communication. (See design considerations below)

In terms of complexity and confusion, our citizen juries and later focus groups by the IPPR, suggest that CDC is fully understandable, and that the adjustments made to pensions in Holland would be well within the acceptable range in response to the Global Financial Crisis. We would note the current system of DC and pension freedoms can create a situation where very complex decisions need to be made on retirement. These may be a greater source of confusion.

Converting DB schemes to CDC

Could seriously underfunded DB pension schemes be resolved by changing their pension contract to CDC, along Dutch lines?

This is an entirely separate question. Many would argue that DB pensions are a historic promise, which must now be met, and that one cannot rewrite historic contracts without the consent of all the parties to the contract. That seems a forceful argument.

However, going forward CDC is an appropriate alternative where DB plans are closing. It is important that this is not confused with the resolution of the problems associated with DB.

How would this be regulated and how would the loss of DB pension promises to scheme members be addressed?

See above

Regulation, governance and industry issues

How would CDCs be regulated?

CDC pensions have a structure which resembles DB but without the pension promise. Their governance is critical, and should only be undertaken by trustees whose sole responsibility is to the beneficiaries. (They can then hire fund managers etc. should they choose to do so). (See detailed Q&A below)

Given their structure and trustee governance, the Pension Regulator would therefore seem the appropriate regulatory body.

Is there appetite among employers and the UK pension industry to deliver CDC?

The Committee will be aware that Royal Mail and the CWU are now vocal in their request for CDC. It is rare for an employer to come out publicly in this way. Reform to pension systems is not something, which employers wish to broadcast. However in the lead up to the 2015 Pension Act, the RSA took a delegation of eight pension funds to see the Minister privately to press the point that they would wish to see CDC introduced, and that it was a practical, and better alternative than DB, or DC.

Would CDC funds have a clearer view towards investing for the long term?

CDC creates a permanent source of capital, which has an appropriate long-term focus, and a “risk budget” focussed on real (post inflation) returns. Long-term returns are therefore the significant driver of their investment approach. Indeed the Committee may wish to note the much larger allocation to infrastructure of Dutch pension funds, and the direct investment programmes of Canadian funds. Such allocations are rarer in UK DB funds, and individual DC accounts do not invest in this way.

While CDC does not solve all the problems of short term investing, cross country comparison suggest they would offer substantial advantages over an individualised system. However, there may be some trade-offs to be made between this consideration, and a sensible level of “freedom and choice.

3. Other considerations about CDC design

Measuring Solvency and adjusting expectations and payment

It is critical the actuaries are fair in their assessment of what pension in payments are affordable.

Some have argued, and grabbed headlines to say that any unguaranteed investment is dangerous.  Of course, no guarantee is without its dangers, and the alternative of investing only in gilts creates huge costs, and most likely leaves the pensioner fully exposed to inflation. But there are technical issues which need resolved, (eg. The way solvency should be calculated, and the gates between which pensions may be increased, or where they must be cut).

Intergenerational Transfer

If the actuarial estimate is over generous, ex-post benefits will be seen to have taken from the young to the old, and vice versa. This problem could be avoided by designing funds for different age cohorts. (A suggestion from an FT journalist). But that is clumsy. Ambachsheer, would advocate two pots, for accumulation and decumulation. He notes that the former could be pure DC. While the RSA might not agree that this is the ideal, it is a fairly modest issue relative to the need to allow collective pensions.
Unfairness to people with lower life expectancy.
This is a big issue in all pension design, and one we often ignore. State pensions, for example have this characteristic. This means that those with a shorter life expectancy (typically poorer people), get less benefit that those who live for longer. Like intergenerational transfer, it can be overcome by designing the pension pool, so that as far as possible the collective serves people with similar life expectancy. However this can create clumsiness. And in extensive discussions with the TUC and others who might represent the interests of people with lower life expectancy, they have been clear that this should not stand in the way of the creation of a collective system.

Costs and Scale

Are very important for all pension systems. Many charges are hidden from the beneficiary, and issue that is now hopefully being addressed. A pension investment might expect to achieve a 3-4% real return. A 1% per annum charge will reduce the possible pension by 25%, 2% pa will reduce it by 50%, 3% will take two thirds. So it is critical for any pension system that costs are held down.
Costs are dependent on scale and on marketing effort. Therefore creating very large CDC funds, which are linked into low cost marketing such as Automatic Enrolment, is an important part of the implementation design if full benefits are to be achieved.
Could CDC Plans be consolidated? Could benefits be transferable?
This should be possible. Since they are DC, and have no employer guarantee, they can readily be managed together. It would help if standard terms were established to encourage consolidation vehicles. CDC can also allow transfers, (along the lines suggested by Cardano). However there are trade-offs here, principally to do with investment (see below).

Governance

CDC gives discretion to the executive of the plan about benefits paid, but also about other distributions. If that can be used against the beneficiary’s best interest, this can create problems. These are illustrated in the “with profits” life polices, where savers money was used by private companies to profit the sponsor. Trustee governance does not mean putting private fund managers out of business, simply that they are the agents of the sponsor.
Who would be permitted to run a CDC pension?
Those able to demonstrate competent trustee governance and able to keep costs low.  (See Governance and Costs).

What legislation would be needed to establish CDC?
It should be secondary legislation only. We have commissioned legal advice on this that suggests a comparatively short time, perhaps a few months, would be needed to allow a pension plan like the Royal Mail to go ahead. Private discussions with government agencies also back this up.

How would it relate to the existing system, especially auto enrolment?
Quite elegantly. In particular there is a gap in pension provision at the point of retirement. Freedom and choice made it possible for pensioners not to buy an annuity, but since there is no alternative, left them without a default for decumulation. There is an acute need for such a default, and a decumulation CDC would provide such a service. In the meantime the Committee will be only too aware of the mis-selling taking place, and the certainty that many will run out of income in their old age.

Who would sponsor such a plan?
Big companies, especially those closing DB plans are immediate possible customers, and have lobbied for change behind the scenes. Royal Mail is now public in doing so. (See below). But the fullest benefits will be gained by creating national, low cost, trustee governed institutions.

How would the money be invested?  In particular might it support infrastructure etc?
One attraction of CDC is that it can (subject to the flexibilities discussed above) create a permanent pool of capital, which would be ideal for supporting infrastructure and other real productive investments. Indeed a comparison of the investment portfolios of Dutch and British pension funds, shows many more of these types of investment being undertaken by pension funds in Holland.

Is anyone interested in establishing a CDC pension system?
One objection to working on a CDC structure has been the argument that no one would want to use it. Recent developments at Royal Mail show this is not so. Indeed a Christmas strike has been averted by the Royal Mail’s agreement with the Union, to lobby the government that they should at last write the secondary legislation to give effect to the 2015 pension act.

This pension scheme which affects 140,000 people should on its own justify the limited effort needed to give effect to the 2015 Act. But we understand there are other pension funds, (USS for example) which are similarly positioned. And the big prize would be to create a national system to allow everyone to save into a CDC arrangement, or similar.