The little things matter
• Ultimately, most people accumulate wealth to spend it at some point
• In the drawdown phase, seemingly small changes can make a big difference to the income stream that can be generated from an investment
Risk management is a key element of active portfolio management. Generating outperformance in down markets (via reducing risk) versus generating the same outperformance in up markets (via increasing risk) can have a materially different journey for the end investor. A portfolio which generates alpha by reducing risk will generally exhibit lower volatility and be exposed to less sequencing risk than the alternative.
In this month’s Market Insight, we demonstrate quantitatively the impact of this on a hypothetical household using our in-house retirement income model. Our retirement income model uses a Monte-Carlo simulation to model the path and distribution of potential fortnightly household investment income streams using a variety of assumptions such as balance, expected returns, volatility, time horizon, drawdown method, household composition and assets and age pension eligibility. In practice the full set of results looks like the charts below. However, we summarise the outcomes into percentiles for ease of interpretation. To interpret a percentile, picture ranking all outcomes worst to best. The 50th percentile is the outcome in the middle, it is the average and most likely. The 25th percentile is the outcome halfway between the absolute worst and the middle. It is a bad outcome, but 75% of the simulations have a better outcome.
We also focus on the left-hand side of the distribution (50th percentile – the median result, and worse) as individual investors tend to care more about not having enough in retirement rather than having too much.
To quantify the relative performance of an actively managed portfolio versus a passively managed portfolio we compare the relative performance of our Strategic 70 portfolio (70% growth assets, 30% defensive assets) versus its Strategic Asset Allocation (SAA) since inception. When reviewing relative returns, volatility and sequencing risk, we exclude the impact of the age pension. We address how the age pension can impact the results at the end of the Insight. All of the analysis is pre-tax, and post inflation (assuming 2.5% inflation). We have adopted a dynamic total return approach to retirement income, where the amount drawn is a function of expected return and residual investment horizon. This is essentially a function which aims to as smoothly as possible draw down capital to zero (or level of desired bequest) over the investment horizon.
The most obvious place to start is investment returns. All else equal, higher returns should lead to improved retirement income, as there is a higher capital base from which to draw on. Since inception, the Drummond Strategic 70 portfolio has outperformed its SAA by 2.3 percentage points per annum. Below, we model the difference in income from two simulations, one with an expected real return of 3.1% (the 10 year real expected return for the Strategic 70 Portfolio in the 2022 SAA Review) and the other of 5.4% (adding the excess return generated since inception versus the SAA). These performance figures in absolute and relative terms are by no means guaranteed, but are simply shown to highlight the potential spectrum of retirement income. At the 50th percentile (the median outcome and yellow line), average income over the forty-year horizon is on average $560 per fortnight higher ($2,060 versus $1,500). At the 5th percentile (a very bad outcome), fortnightly income is on average $300 per fortnight higher.
Using a % based drawdown method rather than a dynamic method also has a large impact. Average income at the median outcome is $60 per fortnight higher, but the final balance is around $350,000 higher ($600,000 versus $250,000).
The impact of higher returns on portfolio outcomes is not unexpected – the whole active investment management industry is largely built around trying to deliver better returns. However, the bumpiness of the journey also matters, particularly if your ride overall is disappointing. Below, we repeat the analysis we undertook for higher expected returns and apply it to lower expected volatility. Again, we use the volatility difference between our Strategic 70 Portfolio and its SAA since inception to provide the alternative path (8.8% vs 10.1%). The impact here is less pronounced than expected returns, but still meaningful. At the median outcome, average income is $40 per fortnight higher in the lower volatility portfolio. However, importantly, in the left-hand side of the distribution (lower percentile values) this impact grows (to $130 per fortnight at the 5th percentile and $85 per fortnight at the 25th percentile). The effect is even positive at the 75th percentile, though less so ($40 per fortnight).
Whenever negative cash flows are introduced, even with the same expected return, a lower volatility portfolio should provide a higher overall income than a higher volatility portfolio. This benefit grows as investment outcomes become worse, demonstrating the benefit of lower volatility to individuals who are concerned with certainty of income. It also highlights a very important point around how active returns are generated. If you add alpha via increasing portfolio risk, some of this alpha will be offset by higher volatility. If you add alpha while also reducing portfolio risk, the portfolio will benefit from both.
A concept which is often written about in retirement income literature is sequencing risk. The point at which an investor moves from accumulation to decumulation is very important and often out of their control. Due to active management, the Strategic 70 Portfolio fell by 3.2 percentage points less than its SAA during the Covid market drawdown in March 2020. Below, we show the impact of beginning a decumulation phase with a 3.2% lower balance on income. Though 3.2% does not seem like a lot – in this scenario it is a starting balance of $968,000 versus $1,000,000, it has a substantial impact on the potential income generated from the investment. At the 50th percentile average fortnightly income is $50 higher than with the lower starting balance. This highlights the benefit of preserving capital even if only a little when an investor is close to the decumulation phase.
The age pension is a very important consideration when modelling retirement income, particularly for lower balance households. It in effect puts a floor under the income of a household, reducing the consequence of investment failure. Below, we show the difference the age pension makes to our hypothetical household over the 40 year horizon. Income across the board is higher as the pension kicks in once asset balances fall below the means test threshold. The point at which this relativity maxes out (to the value of the full age pension) occurs earlier in the lower percentile simulations, reflecting lower investment balances and thus, earlier eligibility for a full pension.
The interplay between investments and the age pension is an important conversation for individual investors to have with their trusted advisors. For those with lower balances it will likely be the most important retirement consideration. For those with high balances, it may be irrelevant. In addition, individuals should weigh the likelihood that the age pension will still be available at its current rate with its current eligibility criteria in the coming decades given the poor outlook for the Federal Budget balance and rising dependency ratios.
Adding Up the Little Things
The above analysis plays a key role in portfolio management and investment philosophy at Drummond Capital Partners where we strive to deliver our clients a better investment journey. The best outcome for our investors is for us to add most of our outperformance by reducing risk in weak markets. Our proven tactical asset allocation process has been successful in 2018, 2020 and the current 2022 bear market. These outcomes increase expected returns, lower volatility, and improve drawdowns (sequencing risk) for investors. In our opinion, this is a better approach to active management than simply adding structural risk to a portfolio and hoping markets go up.
 This hypothetical household contains married homeowners, eligible for the age pension, with $1,000,000 invested in financial assets, $80,000 in non-financial assets, with a forty-year investment horizon and no desire to bequest any of their wealth.
 Drawdown rate is 6% of the portfolio per annum.
 Recall in the dynamic drawdown approach target final balance is $0, hence most of the impact is felt through income rather than financial balance.
Never miss an update
Enjoy this wire? Hit the 'like' button to let us know. Stay up to date with my content by hitting the 'follow' button below and you'll be notified every time I post a wire.
Not already a Livewire member? Sign up today to get free access to investment ideas and strategies from Australia’s leading investors.
Alex has 16 years’ experience as a portfolio manager and economist. As portfolio manager Alex contributes to our strategic and tactical asset allocation processes, and portfolio construction. Alex previously spent 3 years at Cbus Super as a...