Work and Pensions CommitteeWritten evidence submitted by Aon Hewitt
Why do we need a New Type of Pension Scheme?
Employees like to be able to plan for retirement and this is much easier when they can predict the size of their retirement pension. Employees would like defined benefit (DB) pensions. But employers also want predictability of their pension contributions and pension costs, and are no longer prepared to tolerate violent swings and large deficits. Employers are increasingly saying they will only support defined contribution (DC) plans. Can we square the circle between the wants and needs of employers and employees? We believe we can, through defined aspiration schemes. That is why we strongly support the Pension Minister’s call for the introduction of new schemes of this type.
Until fairly recently, employees of many medium and large companies could obtain predicable pensions by joining and accruing benefits in a final salary pension scheme. These schemes provide benefits based on a member’s salary close to and service up to date of leaving. Since salaries are generally stable, the pensions paid by final salary schemes are predictable. However, many employers have closed their schemes to new members. A significant proportion are now going one step further and are closing their final salary scheme to future accrual of existing members (freezing their plans).
There are a number of reasons why companies have done this:
The cash contributions payable to final salary schemes cannot be known in advance because they depend on unknowns such as life expectancy, future salary increases and future investment returns. Predictions will be made for these unknowns to enable planning of contribution levels. However, the predictions will undoubtedly provide to be wrong, leading to volatile contribution requirements particularly when investment markets are volatile.
The cost of providing £1 per annum of pension has significantly increased. There have been two drivers for this (a) low interest rates and (b) longer life expectancy. When many final salary schemes were set up the normal retirement age for male members was 65 and an employee might be expected to live only ten years in retirement. Life expectancy has been increasing at a steady rate of three month increase for every elapsed year, so that same pension promise might now have to be paid for twenty years. The reduction in interest rates over the past twenty years—when “real” interest rates, net of inflation have dropped from 4% per annum to 0% per annum now—has substantially increased liabilities of defined benefit schemes.
Legislation has required final salary schemes to provide inflationary linked pension increases. Even thought the compulsory indexation has reduced over time, past benefits remain linked to high rates of indexation, effectively fully index-linking the benefits.
Since 2004 solvent employers have effectively been required to ensure all the accrued benefits will be paid in full.
The accounting requirements require that any deficit in a final salary scheme has to be reflected in a company’s balance sheet position.
A number of companies—predominantly larger companies—have introduced defined benefit schemes that do not conform to the final salary design. These plans—career average plans, cash balance plans—offer less risk to the sponsoring employer, but are still defined benefit plans for accounting purposes, and as such have to be reported under one of the pensions accounting standards FRS17 or IAS 19. This exposes companies to the balance sheet volatilities, and cash uncertainties that they have been reluctant to accept.
The vast majority of companies now provide retirement provision (for new entrants, and increasingly as above for existing employees) via DC schemes because all of the other alternatives suffer from most or all of the problems affecting final salary schemes.
However, there are a number of major problems with DC schemes:
The benefits provided to members are highly unpredictable, as the Pensions Minister has pointed out.
The focus is on the size of the DC pot and not the amount of pension that it will purchase.
It hard to educate members to a sufficiently high level to be able to make informed investment decisions.
Investment policies may be sub-optimal.
We believe that there is scope for a new type of scheme design which sits between the existing designs which can address these issues. This is why we support the introduction of defined aspiration schemes.
Defined Aspiration Schemes
The term defined aspiration scheme can cover a wide range of possibilities of risk sharing mechanisms. The various types evidenced by the Pensions Minister can all be provided under current legislation, yet have not proven popular with employers. We believe that a specific type of defined aspiration scheme—a collective DC scheme along the lines found in Holland—could reinvigorate pensions saving and attract significant employer interests.
Collective DC is normally used to describe a pension scheme where the employer and employee contribution rates are fixed as for conventional DC. However, for a collective DC scheme all the assets are pooled rather than each member having an actual or notional pot of money earmarked for his benefits. A collective DC scheme can be thought of as a defined aspiration scheme because it aspires to (but does not promise) a certain level of pension.
The initial amount of defined aspiration pension is set at the level which is expected to be provided based on the contributions payable by and in respect of the member. The aim is that the amount of the defined aspiration pension should be adjusted each year in line with inflation. The age at which the pension could be paid would reflect expected increases in future life expectancy.
However:
If investment returns are better than expected, higher increases could be provided both pre and post retirement, but
If investment returns are worse than expected or life expectancy improves by more than expected, lower increases will be provided to defined aspiration pensions and/or the central age from which the pensions can be paid may be adjusted. In extreme circumstances benefits for members may have to be reduced in order to ensure that the cost does not increase beyond the original contributions paid.
It would be possible to design such a pension so that the possibility of a cut in benefits applied even to pensions that had come into payment. We believe that the prospect or reality of cutting pensions in payment will be unacceptable and feel that greater certainty can be offered to pensioners by buying a level annuity when each member reaches retirement.
Inherent in the design of a defined aspiration scheme is that there is never a surplus or deficit. Consequently, there is no question of the employer contributing towards a deficit. Nor is there any question of any surplus being returned to the employer. With these parameters it is clear that the scheme constitutes a defined contribution scheme for the purposes of the employer’s company balance sheet. There are no FRS17 or IAS19 problems with a defined aspiration scheme.
A defined aspiration scheme has a number of common elements with a conventional defined contribution scheme. But the key distinguishing features are:
Benefits are expressed in pension terms for members. Repeated evidence indicates that members underestimate the amount they would need in an account, in order to provide their retirement income.
Benefit volatility close to retirement is progressively reduced by spreading the risk/reward flowing from investing the assets of the scheme from the members approaching retirement to the younger members. In a conventional defined contribution scheme, there is no sharing of risks/rewards between the members.
This sharing of risk/reward removes the need to invest in less volatile assets as a member approaches retirement. This should, on average, lead to at least a 10% higher level of retirement income for members of a defined aspiration pension scheme, compared to a conventional defined contribution scheme.
Investment policy is conducted on an aggregate basis, without the need for individual member involvement, and potentially delivering economies of scale and access to the best expertise available.
Comparison with other Types of Pension Scheme
The diagram below summarises how collective DC/defined aspiration schemes compare with other types of pension scheme:
Defined benefit schemes include final salary and career average salary schemes impose a heavy burden on the scheme’s sponsor because it bears all the investment and longevity risks. Cash balance plans materially reduce the investment risk and take away the longevity risk, but pass these over to members, who generally are ill equipped to deal with them. Under current conditions, retiring members under a cash balance plan would be purchasing annuities at very disadvantageous rates, because of low interest rates cause by the quantitative easing programme.
Conditional indexation schemes reduce some of the risks to the scheme’s sponsor by not having guaranteed increases to benefits either before or after they come into payment. If investment returns are poor, or members live longer than expected, the level of increase is cut back. This enables the sponsor’s contribution rate to be more stable than with a defined benefit scheme. Such schemes are similar to defined aspiration schemes, but there is a significant difference. In a conditional indexation scheme, the level of increase before the benefit comes into payment cannot become negative, but in a defined aspiration scheme it can. This lack of guarantee in a defined aspiration scheme is necessary in order for the Sponsor’s contribution rate to be fixed.
The level of volatility of benefits for those approaching retirement in collective DC/defined aspiration schemes is lower than traditional DC schemes because risks and rewards are passed from the members approaching retirement to the younger members. The differences between defined aspiration schemes and traditional DC schemes are explored more thoroughly in the next section.
Comparison with DC
A typical investment policy followed by a member of a traditional DC scheme is a “lifestyle” policy. Under this approach the assets are invested wholly in equities or other reward seeking assets until say five or 10 years before the member’s expected retirement date. The reward seeking assets are then gradually sold and replaced with assets whose value moves closely in line with the benefits “purchased” by the member’s money purchase pot at retirement. Typically, the investments held immediately prior to retirement will be 75% long dated bonds and 25% cash.
We believe that defined aspiration scheme can offer potentially significantly better outcomes for members than conventional DC schemes, and most importantly without asking members to make complex investment decisions that experience shows they are unwilling or unable to take.
Investment Efficiency
The trustee of a defined aspiration scheme could be required under the terms of the trust deed to invest the assets of the scheme largely in return seeking assets with little or no investment in government bonds and cash. The rationale for this is that, by pooling of investment risk collectively, there is scope for investing the scheme’s assets with a longer term investment horizon than a member of a defined contribution scheme could sensibly adopt in relation to his own retirement account. Over the longer term, since equities and other reward seeking assets are expected to give higher returns than long dated government bonds and cash, a defined aspiration scheme could deliver higher benefits. Modelling by the Government Actuaries department concluded:
“CDC (Collective DC) schemes do appear to exhibit superior performance on average when compared to conventional DC schemes. In theory this improvement is in the order of 20 to 25%, but in the simulation it is as high as 39% for some members.”
Much of this improved return arises because investment in higher risk assets is assumed to continue into retirement. Our view, noted earlier, is this leads to unacceptable risks of pension cuts for members in receipt of pensions. A more realistic view for the out performance would be of the order of 10–15%—a meaningful increase for the average member.
There are further arguments around investment efficiency:
A collective arrangement will represent a larger pot of money, and will hence attract the best investment advice and expertise available. This can include superior governance arrangements, with full time investment professionals in place watching markets on a daily basis, as well as developments such as medium term asset allocation, and de-risking strategies that are more difficult to accommodate in individual DC arrangements.
The larger investment pools will be able to access a wider range of better performing asset managers and some of the more esoteric asset categories, which might be excluded from conventional DC arrangements, either because of minimum investment requirements, or by the sheer size of fees relative to smaller investment pools.
Whilst both of these matters can be—and indeed are being—addressed in progressive designs of DC schemes, using delegated growth funds and other innovations, there is often a lag whilst new ideas are taken on board, and this can lead to inferior outcomes for members.
Expenses
The administration expenses associated with running a traditional DC scheme are quite high due to the need to maintain individual member accounts. It would not be necessary to maintain such accounts for a defined aspiration scheme, although other expenses would be incurred (eg valuation fees). We expect defined aspiration schemes to be found amongst larger employers, and groupings of employers, and so the expenses for such a scheme will be lower than DC arrangements.
Cost of investment training and advice
In order for members to make informed investment decisions in a traditional DC scheme it is necessary to provide them with investment training and advice. Despite all of this advice, education and support most members do not take active investment decisions—because these are complicated decisions. Most will follow defaults that are used by the scheme—be it a global passive equity fund, delegated growth fund or whatever. There is little to be gained by encouraging or cajoling the vast majority of members to take complex investment decisions. However, for a trust based defined aspiration scheme the investment decisions will be made by the trustees—or more likely by their outsourced chief investment officer or delegated manager—and the associated costs would be much lower.
Risk and reward transference
Under a traditional DC scheme each member has his own “pot of money” and there is no transfer of risk and reward between members. Under a defined aspiration scheme benefit volatility can be reduced for members approaching retirement by transferring the volatility to younger members. For example, a 25% rise/fall in the stock market might have a small effect on the expected benefits for a member approaching retirement, but could be designed to increase/decrease the expected benefits for young members by more than 25%.
Choice of annuity
Under a traditional DC scheme a member may choose to purchase an annuity with no or a high level of pension increases. A defined aspiration scheme would not generally offer this choice. Instead, the aim would be to provide inflationary increases with bigger or smaller increases being paid in practice depending on the investment returns achieved. In practice, again, experience indicates that most DC members will choose a non-increasing annuity, since this offers the biggest immediate income. But this higher immediate incomer comes at the price of exposing the pensioner to the effects of inflation on their retirement income—an increasing concern as life expectancy continues to increase.
Benefits and Potential Problems with Defined Aspiration Schemes
Defined aspiration schemes have a number of advantages compared to other pension schemes:
as for traditional DC schemes, the sponsor’s contribution rate is fixed. Our research confirms that very few sponsors are comfortable in setting up a new pension scheme under which its contribution rate might vary. Any replacement scheme has to be defined contribution, from the employer’s perspective.
But unlike traditional DC schemes:
predictability of benefits close to retirement can be achieved without needing to invest in lower returning assets and hence, on average, the benefits provided are expected to be larger; and
members do not need to make investment decisions, which can be taken by professionals in the field.
Defined aspiration schemes therefore have the potential to be the scheme of choice in the future. However, such schemes are not without their problems and the key problems are discussed below.
The UK pensions legislation governing defined contribution schemes is much less onerous than that governing defined benefit schemes. We do not believe that any employers would be prepared to set up a defined aspiration scheme if there is a chance it will be deemed to be a defined benefit scheme. Pensions legislation would need to be amended to reflect the unique characteristics of defined aspiration schemes.
Even if the initial legislation for defined aspiration schemes is of a light touch nature, there is the risk that the legislation is amended over time and becomes rather more onerous. We have already seen this with the legislation covering defined benefit schemes—first came the obligation to provide inflation protection for benefits for early leavers, then the obligation to provide inflation protection to benefits in payment, then the imposition of a debt on the employer if it wishes to wind-up a defined benefit scheme and finally the need to pay levies to the Pension Protection Fund. There are already worrying signs of more onerous legislation governing traditional DC schemes—we believe that there is a need to protect defined aspiration schemes from “legislation creep”.
It is critical that members are made aware that benefits quoted to them are aspirational and the actual benefits paid could be higher or lower. Benefit reductions will be rare, but will require sensitive handling, and careful pre-positioning. The legislation needs to make it clear that benefits can be reduced in extreme circumstances and that trustees and employers will be protected in these circumstances.
The flip side of the issue above is that members need to have it clearly and regularly explained to them that benefits are aspirational only—not guarantees. This was a major concern of the DWP when CDC schemes were discussed previously, and was we understand one of the major reasons for their opposition. It is perhaps a more mature debate that is now required to say that an aspiration is better than a guarantee—because guarantees are too expensive and so will not be available. This will need a degree of national debate.
Higher returns and hence benefits can only be achieved if defined aspiration schemes invest in reward seeking assets. Some trustees may feel uncomfortable with the degree of volatility that this provides to the benefits for members. In order to balance the needs for reasonable benefits and reasonable returns we believe that it would be appropriate for a defined aspiration scheme’s trust deed and rules to specify a minimum and maximum level of expected volatility.
Defined aspiration schemes could be designed so that the effective sponsor contribution in respect of each member is a fixed percentage of salary (as for most traditional DC schemes) or is a percentage of salary which increases with age (so that the pension expected to be provided based on each year of service is similar). One or other (or both!) of these designs might fall foul of the age discrimination legislation.
Defined aspiration schemes would not be able, under current legislation, to be used as the qualifying scheme for automatic enrolment purposes. Change to the legislation would be required.
Conclusions
We strongly support the introduction of the type of defined aspiration scheme described in this paper—modelled heavily on the Dutch collective DC scheme.
The key features are:
A defined contribution scheme, from the sponsor’s perspective. We see little or no appetite amongst our corporate clients to accept DB liabilities any more.
A benefit expressed in pension terms for the member, and one which does not require the members to take long-term, complex investment decisions, which they are unable or unwilling to take.
A pension scheme design which sits in the spectrum between final salary, career average and cash balance on the one hand, and pure DC on the other hand. The superior DC schemes at present, with their emphasis on better member outcomes could increasingly come to resemble defined aspiration schemes.
Legislation to facilitate defined aspiration schemes would need to address two key issues—the potential (however remote) for benefits to be reduced to keep the cost constant and the aspirational rather than guaranteed nature of the pension.
13 April 2012