A consultant's perspective
Russell WrightSenior Vice PresidentDefined Contribution
Find me on LinkedIn
Drop me an email
Anran Chen (PhD)Senior Vice PresidentALM and Investment Strategy
Arash NasriSenior Vice PresidentGlobal Assets
As consultants to DC governance bodies, pension providers and wealth managers, we’re increasingly asked about portfolio construction and optimal withdrawal rates for retirement income strategies. Retirement income planning has been called ‘the nastiest, hardest problem in finance’ by one Nobel Laureate, William Sharpe.[1] So it’s understandable it can feel like a daunting task.
If a young Sharpe had been advising another Nobel Laureate, Albert Einstein, maybe the famous proverb of compound growth being the most powerful force in the universe* (Einstein’s 8th wonder of the world), would be partnered with a maxim about the most destructive force in the universe – sequencing risk.
Destroying your portfolioSequencing risk is the risk that the order of investment returns and withdrawals are unfavourable. When selling investments for withdrawals, poor returns at the start can have a huge impact on the ability for the portfolio to last the required period. Whereas poor returns later, have far less impact.
So how destructive is sequencing risk? Here’s an example:
We’ll use the performance of a portfolio invested 50% in FTSE100 equity and 50% in GBP corporate debt from 2000 to 2020.
Unlucky Ursula is going to experience these returns in the actual order they happened. Whilst Lucky Larry is going to experience the same annual returns, but in reverse. We’ll keep everything else the same (a pension pot of £100,000 and £8,000 withdrawn at the start of each year). Let’s see how this plays out…
As you might have guessed, things don’t work out well for Unlucky Ursula. She experiences three years of poor investment performance at the start and ends up running out of money by year 15.
Lucky Larry, on the other hand, has better fortunes. His pot lasts the full 20 years and he’s withdrawn far more money as a result.
So, same starting pot, same withdrawals, same average returns, but very different outcomes.
Full disclosure, there’s no evidence Einstein said this (we’ve tried to find it!) – but we couldn’t resist a pop-art-inspired image.
People having either good or bad outcomes based on a lucky or unlucky order of investment returns doesn’t feel either right or fair. So how do we deal with this?
Rethinking retirement portfolio constructionGiven that investing in equities usually offers the highest expected return over the longer term, most accumulation portfolios have the largest allocation to this asset class. One could argue that equities should still be the dominant allocation in a decumulation portfolio with potentially 30+ years to live post-retirement. However, in retirement, the priority isn’t necessarily about maximising growth, but minimising the risk of running out of money. Within a drawdown portfolio, this can be achieved in a number of ways, including:
Reduce the equity allocation The thinking with this option is that reducing the equity allocation within the portfolio will reduce the volatility and the possibility of running out of money as regular income is taken. This is the primary way to reduce sequencing risk, but compromises on the potential upside of equity returns.
However, as was seen in 2022 during the Liz Truss premiership, other asset classes that have historically been less volatile than equities (e.g. government bonds) can experience sudden, sharp falls in value. Your choice of alternative asset classes is therefore extremely important.
Traditionally, these alternative asset classes have been more contractual in nature, i.e. fixed income, but when interest rates and inflation are both rising, both the valuation and volatility of these assets can feel more like equities.
There are therefore two ideas that we think will start to play more of a part in decumulation portfolios. Firstly, introducing an allocation to less liquid assets. This may sound counterintuitive at a time when people need liquidity to fund withdrawals. But, if these assets are held by a master trust or DC provider that can manage the day-to-day liquidity needs, the less liquid assets provide a buffer against daily volatility in asset valuations. Plus, they are typically less correlated to listed equities, so provide additional diversification benefits as well.
Secondly, something needs to be done to ensure unusual, but devastating, sudden market falls are defended against. Tail-risk protection strategies are used widely in fund management (although not traditionally by DC or wealth clients). Choosing managers who hold certain types of derivatives (e.g. options) can provide explicit down-side protection and could be held against both equity and bond portfolios.
Active portfolio management The use of a more active, dynamic approach to portfolio management during decumulation needs to be considered.
Investment strategies used in decumulation have typically followed a similar approach to that in accumulation. So for workplace pensions, this includes heavy use of passive management and a longer-term asset allocation outlook. As the impact of downside risk is so much more accentuated in decumulation and with asset returns inherently uncertain, having the ability to adapt quickly to changing market conditions would provide fund managers and governance bodies with a mandate to better support DC members.
Even if we’re asked to advise primarily on a retirement portfolio, we’ll also want to consider more creative solutions to tackle the problem.
One approach is to adjust withdrawal rates depending on investment returns (for example, taking less out after a bad year). This strategy can work well for people receiving ongoing financial advice. But it’s challenging for a pension scheme to implement as individual circumstances and needs can’t be taken into account. However, notifying these individuals to consider changing their withdrawal rates via a text message or email could nudge them in the right direction.
It’s also important not to ignore annuities just because they can’t always be neatly included as part of a portfolio (particularly as rates are looking more attractive again). One solution that deserves attention is investing some of the assets that would have gone into bonds (in a typical mixed equity / bond portfolio) into an annuity instead. The guaranteed income produced could mean more exposure could be justified to the equity allocation, giving greater opportunity for growth in the long term.
The challenges to this approach include a) the clunky nature of the annuity buying process, b) poorer rates for smaller annuity purchases and c) product providers or wealth managers being left to justify what could look like an undiversified remaining portfolio to the regulator. We’ve already seen some innovation to deal with this last point, with the likes of Just and Canada Life providing solutions that keep the annuity within the portfolio. And the first two points could be tackled with combined efforts from the industry.
We’re also excited by deferred annuities. Using a small proportion of a portfolio to buy an income that will start paying out once you reach an agreed age would help manage longevity risk. However, it’s hard to convince people to buy an annuity that pays out immediately, let alone one they might not live long enough to see any income from. Still, it’s worthy of consideration in a bid to achieve a sleep-at-night solution for drawdown.
Exploring new solutionsGoing back to Einstein and Sharpe, as brilliant outside-of-the-box thinkers they found new ways to solve difficult problems – given the scale of the retirement income challenge and the destructive nature of sequencing risk, it’s this type of thinking we desperately need across the industry. While there’s no silver bullet, there is hope now it’s an active conversation.
