The more your money works for you, the less you have to work for money.
- Idowu Koyenikan
I have a mate who is considering retiring early. He has outlined twenty different scenarios to solve the universal problem of retirement – how do you convert a pool of assets into cash to pay for life.
Most investing commentary is focused on accumulating assets. This is the fun part of investing because maybe – just maybe – that next share you buy will be the one that makes all your problems disappear.
Less attention is paid to spending money in retirement. This is a far more complex problem to solve and involves a degree of nuance which isn’t required in the typical ‘get rich quick’ pitch.
I’ve outlined three common approaches retirees use to convert assets to income and scored them based on how they address three key financial issues in retirement:
- Longevity protection is the attempt to plan for a retirement of an unknown length. Earning a high enough return to continue to support withdrawals over the long-term is the key to addressing longevity risk.
- Sequencing risk exists because the spending needs during retirement often necessitates the selling of assets in down markets. Forced to sell low retirees are unable to fully take advantage of market recoveries. The central issue is not the market drop as they periodically occur over time. The real issue is the need to sell during these downturns.
- Lifetime spending refers to maximizing spending during retirement to support the best life possible.
Set withdrawal rate / 4% rule
This is the classic approach to converting assets to income. There are several different variations to this strategy, but the basic premise is a withdrawal rate is used in the first year of retirement and subsequently the dollar amount of the withdrawal is increased in line with inflation.
The theory states that portfolios are liquidated proportionally – if the withdrawal rate is 4% than 4% of each investment is sold. This may or may not reflect reality.
For an investor in a pre-mixed industry super fund there may not be an opportunity to pick what to sell. An investor with a self-managed super fund can pick what to sell. Regardless, selecting assets doesn’t work in the modeling which is used to test the effectiveness of withdrawal rates.
I’ve gone through the modeling in detail in my article on Monte Carlo simulations. A Monte Carlo simulation tests a withdrawal rate against sequencing and longevity risk scenarios based on different returns and inflation.
I will assume a 4% withdrawal rate is used as that is the rule of thumb. I will also assume that the rule is strictly followed.
Longevity protection – 5 out of 10
The baseline scenario I ran in my previous article gave a 78.37% chance of not running out of money in a 30-year retirement. This was a 70 / 30 portfolio split between shares and bonds.
This isn’t a great outcome and given the lack of knowledge about the mechanics of the 4% rule many people are using it for far longer retirements than intended. Strict adherence to the rule puts a retiree at risk if the market performs poorly, if inflation is high and / or if retirements are lengthy. As a result, I’ve given the 4% rule a 5 out of 10 score in addressing longevity risk.
Sequencing risk protection – 3 out of 10
A simple scenario illustrates the pitfalls of strictly following the 4% rule when faced with a severe bear market early in retirement.
A retiree with a diversified portfolio of shares, bonds and cash would see outsized losses in a significant bear market. Common sense suggests a retiree shouldn’t sell low from beaten down shares and instead should spend proceeds from cash and bond allocations. But selling a portion of every asset is what the 4% rule prescribes.
This approach meaningfully increases the impact of sequencing risk. In the modeling I did in my previous article there was only a 1.63% probability of not running out of money if the worst 10 years of historical returns happened early in retirement.
Given this risk, I’ve given the 4% rule a score of 3 out of 10 in addressing sequencing risk.
Levels of lifetime spending – 3 out of 10
The goal of retirement is to maximise lifetime spending while not running out of money. Other goals like leaving a bequest may be a factor in decision making. This is a difficult balancing act.
In a scenario with low inflation, strong returns early in retirement, and a normal lifespan the 4% rule dictates a level of spending far below what a portfolio could otherwise support.
This is by design as the 4% rule is intended to protect against worst case scenarios. If no adjustments are made the rule results in a surplus of funds in most scenarios.
As a result, I’ve given the 4% rule a score of 3 out of 10.
Just spend income
This concept is straightforward. A retiree amasses a pool of assets that generates income which is used to pay for life. Spending fluctuates based on the level of income.
The advantage of this approach is that a retiree won’t ever run out of money which accomplishes the baseline goal of retirement.
However, this comes with challenges. More assets are generally required to support a similar level of spending as a set withdrawal strategy. This is somewhat mitigated in Australia given historically high dividends and franking credits. The overall level of retirement savings needed will depend on asset allocation.
Another challenge is keeping up with inflation. If this doesn’t happen a retiree’s real – or inflation adjusted – spending will drop. Historical data illustrates this problem.
Between the start of 2016 and the end of 2025, the Vanguard Australian Shares ETF (ASX: VAS) delivered total income growth of 10.72%. This growth meaningfully trailed cumulative inflation of 36% meaning real spending for a retiree would fall by just under 23%.
A final challenge is dealing with fluctuations in annual income. Historically the volatility of dividends is significantly less than share price volatility. But there have been instances like the pandemic when dividends dropped significantly.
Longevity protection – 10 out of 10
If a retiree only spends income there is no possibility of running out of money. You can’t get much better than that for longevity protection – hence, the score of 10 out of 10.
Sequencing risk protection – 10 out of 10
The order of returns is irrelevant for an income strategy as a retiree simply spends available income even if it drops significantly in an event like the pandemic. Once again, a score of 10 out of 10.
Levels of lifetime spending – 1 out of 10
If a retiree only spends income their portfolio would continue to grow throughout retirement. However, growth would be at a slower rate than published historical levels as income makes up a meaningful portion of total returns.
Dividends made up approximately 55% of ASX 300 returns over the last decade and about 12% of S&P 500 returns. For bonds held to maturity and cash, returns are entirely made up of income.
Dying with significant assets may fulfill other goals such as leaving a bequest but levels of lifetime spending are lower than possible. As a result, I’ve given a score of 1 out of 10.
Bucket strategy
The bucket strategy divides assets into different buckets and allows the retiree to pick what to sell based on market conditions. A bucket strategy differs significantly from the first two rules-based approaches as a retiree will have to make a series of decisions over the course of retirement.
The bucket strategy relies on the judgement and decision making of the retiree. More skill is needed to set-up and execute this approach.
The typical bucket strategy will involve three sets of assets – short-term assets primarily consisting of cash, medium term assets consisting of bonds and income producing shares, and a long-term bucket consisting of growth shares.
The short-term bucket is used to support spending and is re-filled with income and asset sales from the other buckets. In times of extreme market stress the retiree may forgo or limit asset sales and spend down cash which can be replenished later.
The bucket strategy is not a spending rule and instead is a portfolio construction and management approach. I’m going to assume that flexibility is also used with spending and a retiree can start out with a spending level of 5% to 5.50% (more on this later).
Longevity protection – 7 out of 10
Longevity protection and sequencing risk are related. Given lower sequencing risk and the flexibility of spending the bucket strategy provides more longevity protection than a set withdrawal approach.
As a result, I’ve scored the bucket strategy 7 out of 10.
Sequencing risk – 7 out of 10
The bucket strategy is designed to protect against sequencing risk. If a retiree has the misfortune of facing a bear market early in retirement than cash can be used to support withdrawals while markets recover.
This eliminates the need to sell shares in a bear market which reduces the impact of sequencing risk.
During the average bear market it takes 27 months for prices to reach the same level as prior to the crash. In particularly harsh bear markets recovery has taken up to 5 years.
A judgement call needs to be made on how many years of living expenses to hold in cash. The more cash held, the lower the sequencing risk. But there is a trade-off as more cash results in lower long-term returns. Lower returns reduce longevity protection.
No matter how much cash is held, the bucket strategy is more effective in dealing with sequencing risk than a set withdrawal method so I’ve scored it a 7 out of 10.
Levels of lifetime spending – 7 out of 10
Modeling shows the bucket strategy can support a higher level of initial withdrawals given the elimination or reduction of sequencing risk. Like all modeling there are several assumptions involved.
Unlike the rules-based approaches many of those assumptions are reliant on a retiree making good decisions. Given the higher initial spending I’ve given a score of 7 out of 10.
Final thoughts
This exercise has demonstrated the need for nuance in a retirement strategy. Any strategy should be based on the personal circumstances of each retiree.
Owning your own home outright, the percentage of spending dedicated to wants vs needs, and other sources of retirement income like the age pension will all factor into the approach each person should take.
Rules based strategies and rules of thumb are unlikely to suit the needs of any individual retiree. Hybrid strategies taking elements of each of the three approaches I’ve outlined are likely to result in better outcomes.
Like any approach to personal finances a strategy should suit your individual circumstances and temperament. As always, knowledge is the key to successfully navigating evolving market conditions over the course of a long retirement.
Mark LaMonica, CFA, is Director of Personal Finance at Morningstar Australia.