By failing to account for volatility of investment returns, most financial plans and retirement calculators are selling retirees
Imagine this: you are planning on packing up and moving to Melbourne for 1 year, you ask someone what clothes you should pack and they tell you the following:
“The historical average temperature of Melbourne is 15.6°C, at this temperature you’ll be comfortable wearing jeans and a thin cotton jumper. Assuming that you wash your clothes once per week; packing 7 pairs of Jeans and 7 thin cotton jumpers will provide you with adequate clothing for the year.”
Would you question the quality of their advice?
If you are reading this from Melbourne you already know that this is highly flawed advice; during Summer days of 36°C + jeans and a cotton jumper will have you suffering, and on Winter mornings when it gets down to 2°C you will be just as uncomfortable.
If you made plans based on average temperate, you will spend a most of the year living in significantly less comfort than the people who have taken the variance of temperature into account when planning their wardrobe.
In reality no one would plan their wardrobe around averages, however this is exactly how the majority of people plan for their retirement, it is exactly how the VAST majority of retirement calculators work, and it is exactly how most financial plans work.
They all assume that your chosen investment portfolio with earn average returns year in year out, but just like Melbourne weather won’t be 15.6°C all day every day, in reality investment portfolios won’t provide average returns year in year out.
Volatility of returns becomes a problem in retirement because retirees need to eat (and use electricity, petrol, spoil grandchildren) regardless of what investment markets are doing.
When consistent withdrawals are made from a portfolio sequence of returns risk becomes a problem.
Investment returns at the start of retirement are more impactful than returns later in retirement, this is because retirement portfolios are generally at their biggest at the start of retirement (you have been saving the longest and have not yet started spending) a bigger percentage change in this period will have a bigger impact on absolute wealth.
Poor returns at the start of retirement can have large impact on your financial situation.
The risk of getting poor returns at the start of your retirement is known as sequence of returns risk.
Below are three sets of possible investment portfolio returns over a 20 year period.
Each set has an Average (Geometric Mean) of 5.8% per annum.
For return sets A &C I have introduced volatility. I have generated a set of returns with a geometric mean of 5.8% and a standard deviation of 7.38% , these are not unrealistic figures for a diversified retirement portfolio.
To return set A I have taken the generated returns and shuffled the good years to the front.
To create return set C I have taken the same 20 returns and manually shuffled the bad years to the front.
It is important to remember that return set A & C contain the same returns just in a different order.
We can now compare the performance of the 3 portfolio’s.
Firstly let’s look at how they compare for an amount invested for 20 years and left alone with no additions and no withdrawals.
As we can see, because we are not making withdrawals, the sequence of the returns does not matter. Poor returns at the start will be offset by good returns at the end and vice versa.
We know look at the performance of the same 3 sets of returns in a simulated retirement situation; we start with $374,000 portfolio and withdraw $25,000 (indexed) every year to fund expenses.
We can now see sequence of returns risk in action, return set C is exhausted before year 13, portfolio B shows the portfolio lasting 20 years (but will not make it to 21) and Portfolio A shows an impressive $322,700 left at the end of 20 years; three drastically different outcomes, from 3 portfolios all with an average return of 5.8%.
Admittedly, I have manually shuffled the returns of portfolios A and C to highlight the potency of sequence of return risk, these figures are exaggerated, below I have randomly generated 20 more returns sets with a geometric mean of 5.8% and a standard deviation of 7.38% and compared them with portfolios A, B and C.
We can see that if withdrawals are contributions are not made, the sequence of returns still does not matter.
We can also see that the sequence of returns continues have a significant impact of outcomes; the best set of generated returns still has $173,989 left after 20 years while the worst outcome using non manually altered return sets has the portfolio running out soon after 14 years.
Assuming a year in year out return of 5.8% (as most calculators and financial plans do) this scenario would be considered a success; however once volatility is added to try and better represent reality 9 out of 20 possible return sets failed to last the desired 20 years.
In fact I generated 10,000 possible return sets (a MoneyGeeks standard) for this scenario and 45% of return sets failed to last the 20 years.
Over the years of providing financial advice, one of the most common questions we have been asked is:
“Do I have enough to retire?”
Across Australia and on the internet this question is being answered inadequately.
Consider this example:
Mr & Mrs Smith, are about to retire, they are both 66 and have $374,000 built up in superannuation. They estimate the age pension will meet their day to day expenses. However Mr. & Mrs Smith have two main financial goals that they consider important: they wish to take their Daughter and her family out to dinner once per week, they also wish to the fly to the UK every year for the next 20 years to visit their ex-pat son (after 20 years they estimate health will preclude them from international travel they also hope that their son will be in a financial position to visit them regularly & that their daughter can start buying them dinner) they estimate the cost of these extra activates to be $25,000 per annum.
Mrs Smith is retiring from a job she enjoys; she likes the people and has been offered a job of 2 days per week to help train her eventual replacement, this role appeals to Mrs Smith and she has confirmed she will be given time off every year to visit her son.
The income from this two day per week position would be enough to pay for their extra expenses (even with a reduction in the age pension).
Mr and Mrs Smith agree that if they can afford to retire and live out their 20 year plan Mrs Smith will turn down the position. They consult a financial planner who recommends a diversified portfolio that is expected to return 5.8% per annum, the planner shows them projections confirming that their money will last 20 years and exclaims “Great news, you can retire”.
Mrs Smith informs her employer she is retiring and will not accept their offer of 2 days work per week.
In this example Mr & Mrs Smith have just signed up for a strategy with a chance of success of 55% (as seen above).
Incidentally, if Mrs. Smith had continued to work and their nest egg has three more years to grow with only 17 year to pay out the chance of success would grow to 90%.
The vast majority of retirement calculators and financial plans fail to take volatility into consideration; this can lead to poor decisions.
So where can you fund adequate information to base your decisions on?
The process of generating and testing these possible return sets is called a Monte Carlo Simulation; by searching “Monte Carlo Retirement Calculator” you will find multiple versions. However these are all US based, the US greatly lags Australia in another important aspect of retirement; social security. Because the Age Pension is means tested, it may increase in years of poor returns and offset some of your losses; this means US based calculators will tend to underestimate the chance of success for Australians.
At MoneyGeeks we have developed a calculator that takes volatility and the age pension into consideration; we utilize this calculator in our GeekPeeks.
A GeekPeek will provide you with two chances of success to consider.
Check out a GeekPeek here