DC Debate Q3: Getting the right amount of risk in defaults

Can members tolerate significant volatility in the first years of retirement, or should schemes control risk to avoid opt-outs?

John Reeve: We all know that, in investment terms, volatility is inextricably linked with high returns. A mathematical assessment therefore suggests that those who can reasonably accept volatility (the young with many years to retirement) should take risks and benefit from the higher returns.

However, we also have to accept that investments are the only area in which risk is seen as a good thing. In no other walk of life do we encourage people to take risks. Given this it can be no surprise that the less sophisticated investor (which describes most of us) sees volatility as a bad thing. Falls in value invariably lead to a reduction in savings rates rather than the increase that should follow a fall in price.

It therefore makes absolute sense to encourage the majority of new savers into ‘safe’ investments with low volatility to build confidence and encourage the savings habit. Once this confidence is built and savers are hopefully more educated, more risk can be encouraged and returns will hopefully increase.

Many defaults have well diversified growth portfolios that will help manage risk, but remain significantly exposed to equity market risk

Alistair Byrne, State Street Global Advisors

In the meantime, a change in our approach to savings is important. We should embrace behavioural finance and stop talking about ‘volatility’ and ‘risk’ – both words with negative connotations. Instead we should look for more positive language that will encourage savings.

Jonathan Parker: As with many aspects of finance, the theory often butts heads with the realities of human behaviour. People feel losses in their savings much more acutely than equivalent gains and the absolute worst outcome would be for young people to be scared off from saving due to falls in the value of their account.

Workplace pensions do have a number of features that act as a counterbalance to the behavioural concerns.

In particular, members’ own money only forms a part of total contributions (there is employer money and tax relief as well) and knowledge of account balances is extremely low. People may not even realise they have lost money, although there is some evidence that this can change during periods of market stress.

The answer to this is to understand your members and specifically their tolerance to risk. In addition, communicate regularly with them, particularly if markets are experiencing a period of volatility.

Our experience is that reassuring communications that reinforce the importance of continuing to save and remind members that pensions are long-term investments can really help to mitigate any adverse reactions.

Alistair Byrne: Early in their career, members have substantial capacity to bear volatility and may benefit from pound cost averaging when they make regular contributions.

However, most are also risk-averse and made uncomfortable by short-term losses. We think it is important to have some risk management, such as diversification and dynamic asset allocation, to protect members from the worst of volatility rather than rely just on inertia and lack of awareness.

Volatility control becomes more important in the consolidation phase – when investment returns on accumulated savings start to outweigh contributions – and becomes crucial approaching and after retirement, where sequencing risk (the opposite of pound cost averaging) becomes a concern.

Helen Ball: Members will have a different appetite for risk depending upon their own financial circumstances and level of financial understanding. Assuming that they will all have the same tolerance during ‘the first years of saving’ does not necessarily work for all schemes.

A greater degree of volatility is generally accepted in the early years, if the intention is to maximise the peaks of performance. There is less concern about the potential troughs, because there is more time to make up any losses.

However, in auto-enrolment schemes the provision of a good savings experience is important to prevent opt-outs. In those schemes, risk control in the early years is given a high priority, so people are not put off pension saving.

Andy Cheseldine: I think the short answer is yes, they can tolerate significant volatility.

After a full year of contributions, the next 12 months of contributions are likely to add somewhere between 8.3 per cent (month 13) to 4.3 per cent (month 24) to the aggregate fund value each month. Investment fund volatility is likely to pale into insignificance in comparison.

That does not mean we can, or should, ignore volatility. But where it represents potentially rewarded risk and therefore greater long-term growth, that should be our focus.

Avoiding opt-outs is a perfectly reasonable objective. But quality communication, focused on the value of employer contributions and tax relief on top of member contributions can make it clear that pension saving is still the best savings option available.

Are DC default funds diversified enough? Should any changes to strategy be considered, given the maturity of the cycle?

Ball: The level of diversification in a DC default fund will depend on its investment strategy, which has to be reviewed every three years by law. If everything is invested in one kind of asset, such as global equities, the fund may not be well placed if there is a market crash.

In other words, it can be risky to put all of your eggs in one basket. Achieving greater diversification can result in smaller peaks and troughs, minimising the risk of serious losses.

Byrne: Many defaults have well diversified growth portfolios that will help manage risk, but remain significantly exposed to equity market risk.

Dynamic asset allocation and volatility controls can help mitigate losses, and should be considered by schemes that currently have only static strategic allocations to equities.

The bulk of the asset allocation decisions should be stable in a DC environment

Andy Cheseldine, Capital Cranfield Trustees

The more crucial risk mitigation is the glidepath, which makes sure members closer to retirement have overall more defensive portfolios, given they have less capacity to recover from losses.

Cheseldine: I believe that most default funds are already aligned to the needs of their members’ circumstances to the extent that this is possible with a widely diverse group of members.

Some defaults are, in effect, diversified managed funds. Others, at least in their long-term portfolios, are more growth focused – and that usually means a heavy equity bias.

Even this can be quite diversified, with unhedged global funds offering currency exposure (as do UK equity funds given the weight of overseas earnings in the FTSE 100).

Whatever the strategy, I think the bulk of the asset allocation decisions should be stable in a DC environment and trustees should be wary of trying to time markets.

There may be an argument for making proactive decisions based on clear risk indicators that vary over time. However, I would not want to make wholesale allocation decisions based purely on perception of market maturity.

Even if you had significant confidence in your decisions, how would you communicate those decisions to members who have had their investment strategy explained to them in terms of stable allocation frameworks?

Reeve: It seems to me unlikely that any fund will ever be able to withstand all the circumstances that markets can throw at us.

It is also important to remember that default funds should look to include the appropriate level of risk, not minimise risk. Generally, while diversification will reduce volatility, it will also reduce returns.

There is a truism in investing not to invest in anything you do not understand. Diversification adds complication to investment choices. While adding a wider range of investments to a fund might increase diversification, it will also add complexity and may therefore spook savers.

However, I think that the days of the single default fund are numbered.

We are seeing a trend to what might be described as segmented or targeted defaults. This is where the default that is deemed appropriate for any saver is selected by reference to some key factors such as age, size of savings and level of contribution. In this way risk can be better targeted to where it is most appropriate.

Parker: Looking across the market of DC default funds, there is a broad spectrum of approaches taken to investment strategy and asset allocation – equity allocations can vary from 37 to 100 per cent.

Asset class diversification is a sensible strategy and many default funds do now invest across more than just equities during the accumulation period.

Those with mandates to make dynamic asset allocation decisions should of course be considering the relative value of asset classes at whatever stage in a cycle we might be.

However, as we have seen with a number of diversified growth funds, it is incredibly difficult to know where we might be in any particular cycle, and there is a risk of missing out on future upside. Any decisions need to be balanced against the long-term horizons than pension savers have.

A good rule of thumb for any investor is often to sit tight and avoid unnecessary trading activity. With DC typically being a monthly savings product (as distinct from a single premium), the ability to invest future contributions at lower prices is beneficial.

How far from retirement should derisking begin, and what asset mixes best serve the variety of choices made at retirement?

Byrne: We think that a gradual derisking should begin about 20 years before the expected retirement date and a more significant derisking 10 and five years out. This reflects the length of the equity market cycle and the point at which the balance between returns on accumulated savings and future contributions begins to shift.

Retirement itself is uncertain and freedom and choice has led to individuals accessing retirement assets earlier, which would also justify earlier derisking.

‘One-size-fits-all’ is no longer appropriate in the data-rich world in which we now live

John Reeve, Cosan Consulting

Our research suggests members will take advantage of the full range of access options – lump sum, drawdown, annuity – but few will know far in advance what option they will take.

That calls for a ‘keep your options open’ asset allocation, which maintains exposure to growth assets to generate real return while balancing that with lower risk assets such as fixed income to reduce volatility.

Our analysis leads us to favour approximately 40 per cent growth/60 per cent bonds with some risk management and volatility controls overlaid on the growth assets.

Cheseldine:

A great question, but it begs a number of others. Do you know when retirement will be? Will members withdraw all at once or take a tax-free pension commencement lump sum at the earliest opportunity (currently age 55)? Will members continue working after retirement? What are markets going to look like at that time?

Most schemes start derisking between five and 10 years before selected retirement date. I have seen actuarial forecasting that says three years is the optimum from a risk and reward perspective but that most people retire three years before they expected.

That was some years ago, and expectations or practical experience may well have changed, but it indicates around six years might be optimal.

So much depends on the asset allocation before you start derisking, and the target asset allocation at expected retirement date. A lot also depends on the industry the member works in.

In terms of asset classes, the only thing we can say with any confidence is that a single asset class is probably too risky and that the allocation should be driven by what the member expects to use the funds for.

The problem is that even financially sophisticated members are unlikely to have a clear view of their needs until the actual moment of retirement. This is where trustees can add value, understanding both the demographics and expectations of members, then combining that with knowledge of investment risk management and insight into markets.

Ball: The danger in derisking too soon is that members will lose some element of investment return and be forced into making decisions too far away from their retirement date. This places the trustees in a very difficult position, but with support from their investment consultant most are finding their way through this.

Keeping derisking plans under review is an obvious way of making sure that trustees’ choices remain relevant to their members. The asset mixes that follow from that decision will depend on the chosen strategy, with some still remaining faithful to the idea of an annuity as the ultimate default for those members who will make no choice at all about their retirement.

However, we are also seeing more strategies now that involve assets more suited to a drawdown or ‘wealth preservation’ style of investment, which aim to leave the door open to members choosing whether they want to take their benefits in the form of an annuity, cash or drawdown.

Parker: A recent survey of mastertrusts laid bare the spectrum of approaches to derisking. The longest period was 25 years from retirement, while the shortest had no derisking. Some have more than one derisking period, reflecting the importance of age 55 as a point when tax free cash might be taken. So, who is right?

It is difficult to make the case for an individual just starting their working life having the same asset mix as someone about to retire, which suggests that some changes to asset allocation are needed across the span of members careers. Many of our clients begin their derisking 10 to 15 years before an individual’s chosen retirement age.

This reflects a pragmatic approach to having to juggle a number of uncertain factors, including when and how someone is going to start to take their benefits, having an incomplete picture of their pension assets and how capital markets are going to perform.

The design of a ‘universal’ default strategy naturally has to make certain compromises. Although it is still early days, trends as outlined in the Financial Conduct Authority’s Retirement Outcomes Review suggest that drawdown is likely to become the most popular way in which people generate an income in retirement. This means that asset mixes that include 40 to 60 per cent in a diverse range of investments delivering real returns may be those typically favoured by clients.

Reeve: This is another area where segmented or targeted defaults are needed. For those aiming for cash because the benefit is a small part of their pension savings, then it may be that more risk is appropriate and so derisking can begin later.

Where the benefit is larger and forms a major element of the saver’s retirement planning, risk may be less acceptable and so it may be that derisking should start sooner.

While it may be difficult to assess which apply in different circumstances, initiatives like the pensions dashboard and the increasing use of ‘big data’ should help us find more sophisticated solutions that better match savers requirements.

‘One-size-fits-all’ is no longer appropriate in the data-rich world in which we now live. We need to start to use the information available to us to help create more sophisticated solutions to meet savers' changing and varying needs.