Mind the personal funding gap

Managing the personal funding gap in a defined-contribution plan requires knowing how much money people will need in retirement. Having estimated that requirement, we can estimate that gap to likely account size and offer suggestions on how to reduce that gap. In the following pages, we examine how this gap can be reduced.

How big is the gap?

Our calculations are based on the Association of Superannuation Fund of Australia’s (ASFA) ‘comfortable’ income target for a single person. How close are retirees to this standard? According to the Australian Bureau of Statistics (ABS), many people are facing a shortfall when they stop working. ABS estimates for 2013-14 for 55- to 64-year-olds place the average balance at $255,000 and the median at just $106,000, well short of ASFA’s comfortable target of $545,000.

We think, however, that the ASFA assumptions may understate the savings required for retirement. We have calculated a ‘lower growth’ target that assumes lower investment returns. This results in a bigger savings shortfall.

Figure 1 – The target for savers may be higher in a world of lower-growth expectations

Source: ASFA, Schroders. *Investment returns based on the allocations of an average balanced fund and our return forecasts.

Another issue when judging the funding gap is deciding what happens to consumption as people age. Most people assume it declines. However, this is not necessarily the case. In Figure 2, we see that those over 85 consume about only 10% less than those at age 65. This result is due to the fact that inflation boosts the costs of consumption for everyone, a fact that further increases the funding gap.

Figure 2 – Australian consumption shows a mixed picture as people age

Source: ASFA retirement standard, 2016

Filling the gap in pre-retirement – possible options

Trying to get more return from the assets is likely to involve more risk. When this risk is increased and how much it is increased can alter the outcome. The options we have evaluated (apart from higher contributions) are: 1, the average Australian balanced approach; 2, the average Australian life-cycle approach; 3, a life cycle with higher allocations to equities at the start; and 4, dynamic investing.

Balanced versus life-cycle approaches

In Figure 3, we compare the average balanced approach against the average life-cycle approach. Over 40 years, both result in similar outcomes, as less than 50% of people reach the ASFA comfortable target under either approach. We focus on the bottom 75% because bad outcomes can put families closer to the breadline. Getting great outcomes (top 25%) can be regarded as providing for extra luxuries in retirement.

Figure 3 – Balanced and life-cycle approaches result in identical outcomes over 40 years

However, if we look at this in the last 10 years before retirement (Figure 4), when sequencing risk is important, we can argue that a life-cycle strategy is superior to the balanced strategy because the spread of outcomes is tighter and, specifically, the outcome for the bottom 5% is better.

Figure 4 – Comparing balanced with life cycle just before retirement tightens outcomes

More growth assets earlier

In the UK, it is common to have high equity allocations early in the life cycle to take advantage of the return premium from equities while individuals can bear the volatility given the time they have until retirement.  Increasing the equity allocation from 89% to 100% at the start provides a marginal improvement for the median balanced investor but makes the bottom 5% of individuals worse off. (See Figure 5).

Figure 5 – Increasing equity allocations at the start of the life cycle hurts the bottom 5%

Being dynamic and outcome-oriented

One of the issues with the balanced approach is its tether to bonds in an environment where bonds may not provide the returns of the past. This is particularly true of life-cycle approaches in the last years before retirement when the allocation to bonds is typically higher than a balanced approach.

Bonds are used in both approaches to reduce volatility of the portfolio, which is particularly important in the later stages of saving. Rather than asking a manager to produce excess returns, what if they are asked to reduce the relative volatility? Instead of setting a benchmark-plus target, the portfolio could be given a return target of exceeding inflation or cash returns. Importantly, the allocations should not be tethered to bonds. The manager should be given a broad range of asset classes and be instructed that allocations be based on valuations rather than rigid allocation rules. This would ensure the manager is not forced to hold significant holdings in unattractive asset classes.

In Figure 6, we show that outcome-oriented approaches can move the spread of outcomes upwards without significantly increasing the spread of outcomes compared with a balanced or life-cycle approach.  This is a result of using a wider number of asset classes and the skills of an active manager to allocate between asset classes. We have chosen to consider inflation-related outcome-oriented portfolios because the prices of the items needed in retirement increase with inflation. The two outcome-oriented portfolios aim to return CPI+5%, a return similar to the historic returns of Australian equities, but assume the manager can reduce the volatility by one-third (to 9%) and one-half (to 7%). The result is an increased Sharpe ratio from 0.44 for equities to 0.55 and 0.71 for the two outcome-oriented portfolios.

Figure 6 – Outcome-oriented dynamic portfolios can improve the outcomes for all

These results for outcome-oriented portfolios are only achieved if all the assets are managed using this approach. If only a small proportion is managed in this manner, the impact will be dampened so that there will be little difference between the existing approach (balanced or life cycle) versus a modified approach.

Conclusion

Future long-term returns on assets are likely to be significantly lower than those experienced over the past 30 years. Worsening long-term returns on assets mean that the personal underfunding gap is likely to widen if Australians want to maintain a standard of living in retirement similar to pre-retirement. Closing the underfunding gap will likely be a combination of increasing contributions and changing historical investment habits. Approaches with static or rising allocations to bonds (balanced and life cycle respectively) are susceptible given the outlook for bonds long term. From an investment perspective, the underfunding gap can be reduced by using a dynamic outcome-oriented approach on a large proportion of assets.

Download to read the 23-page version of this paper.