In my post entitled “Illustrating variability of returns of time” I simulated multiple dollar cost averaged portfolios over 15 and 20 year investment time horizons and found that the internal rate of return varied significantly depending on when an investment journey started and finished. This variability has significant implications for retirement planning (if you have not read that post please read it prior to reading this one).
Once I completed the simulation, I wondered what could be done if someone was unfortunate enough to have a portfolio that did not provide them with a high enough rate of return to meet their retirement needs. Does the person simply need to continue working and adding to their portfolio or can they scale back their work significantly?
To answer this question, I’ve taken the worst performing portfolios from the 15-year and 20-year simulations and extended them for up to 8 years to see what sort of returns could be obtained. In continuing the simulation, I chose not to add any additional funds to the portfolio in order to let the asset allocation drive 100% of the returns. Performing the simulation in this manner models the scenario where someone that has experienced insufficient returns (less than 6.50%) becomes semi-retired and scales back their work to a point where it covers their cost of living only.
15-year portfolios
For the 15-year portfolios $1000 was invested each month over the 15-year period for a total investment of $180,000.
MSCI Inc dividends
The worst performing portfolio invested 100% in the MSCI World Ex Australia index gave the investor a -2.43% return with a final amount of $150,408.
| Finish date | Final amount | IRR p.a. | Additional years |
| 31/8/2011 | $150,408 | -2.43% | 0 |
| 31/8/2012 | $168,899 | -0.75% | 1 |
| 31/8/2013 | $231,286 | 2.60% | 2 |
| 31/8/2014 | $266,771 | 3.68% | 3 |
| 31/8/2015 | $340,106 | 5.43% | 4 |
| 31/8/2016 | $341,679 | 5.05% | 5 |
| 31/8/2017 | $376,075 | 6.97% | 6 |
| 31/8/2018 | $467,426 | 6.51% | 7 |
| 31/8/2019 | $502,803 | 6.57% | 8 |
An investor with this asset allocation would have to leave their portfolio untouched for period of 4 additional years before the annualized return passed 5%, with the best result being 6 years where the growth rate reached 6.97%.
All Ords Inc dividends
During the sample period from January 1980 through to June 2020 the All Ordinaries index had a strong performance. The table below shows that the minimum annualized return was 6.00% (Note 1). Even holding for 1 additional year the rate of return jumps to 7.42% and rewards the investor with a portfolio of $347,829.
| Finish date | Final amount | IRR p.a. | Additional years |
| 30/6/2012 | $288,243 | 6.00% | 0 |
| 30/6/2013 | $347,829 | 7.42% | 1 |
| 30/6/2014 | $409,189 | 8.31% | 2 |
| 30/6/2015 | $432.397 | 8.09% | 3 |
| 30/6/2016 | $441,080 | 7.61% | 4 |
| 30/6/2017 | $498,946 | 8.00% | 5 |
| 30/6/2018 | $567,455 | 8.37% | 6 |
| 30/6/2019 | $630,095 | 8.53% | 7 |
MSCI+All Ords Inc dividends
The worst performing portfolio in this group commenced on 31/10/1996 and finished on the 30/9/2011. The $180,000 invested grew to $196,480 with an annualized rate of return of 1.15%.
| Finish date | Final amount | IRR p.a. | Additional years |
| 30/9/2011 | $196,480 | 1.15% | 0 |
| 30/9/2012 | $223,455 | 2.50% | 1 |
| 30/9/2013 | $290,754 | 4.92% | 2 |
| 30/9/2014 | $333,290 | 5.73% | 3 |
| 30/9/2015 | $370,058 | 6.68% | 4 |
| 30/9/2016 | $395,013 | 7.27% | 5 |
| 30/9/2017 | $445,109 | 8.35% | 6 |
| 30/9/2018 | $526,935 | 7.83% | 7 |
| 30/9/2019 | $581,703 | 9.44% | 8 |
We notice that after 4 years we see a satisfactory rate of return of 6.68% and after 6 years the rate of return climbs to 8.35% which is a value similar to the long-term return of the stock market.
20-year portfolios
For the 20-year portfolios $1000 was invested each month over the 20-year period for a total investment of $240,000.
MSCI Inc dividends
Having 100% of assets invested in the MSCI would have yielded a 0.77% annualized return over the 20-year period finishing in 31/8/2011.
| Finish date | Final amount | IRR p.a. | Additional years |
| 31/8/2011 | $259,576 | 0.77% | 0 |
| 31/8/2012 | $291,489 | 1.73% | 1 |
| 31/8/2013 | $399,155 | 4.09% | 2 |
| 31/8/2014 | $460,397 | 4.83% | 3 |
| 31/8/2015 | $586,595 | 6.13% | 4 |
| 31/8/2016 | $589,674 | 5.79% | 5 |
| 31/8/2017 | $649,035 | 6.02% | 6 |
| 31/8/2018 | $806,690 | 6.91% | 7 |
| 31/8/2019 | $867,744 | 6.94% | 8 |
In this case it would have taken an investor 4 years to increase their annualized return to 6.13%. Waiting a total of 8 years would provide a return of 6.94% and a portfolio value of $867,744.
All Ords Inc dividends
Dollar cost averaging the Australian all ordinaries index resulted in excellent performance. The worst performing periods are listed in the table below (Note 2).
| Start date | Finish date | Final Amount | IRR p.a. |
| 31/7/2000 | 30/6/2020 | $520,181 | 7.17% |
| 31/1/1999 | 31/12/2018 | $529,152 | 7.31% |
| 30/4/1996 | 31/3/2016 | $544,771 | 7.57% |
| 31/7/1992 | 30/6/2012 | $537,422 | 7.45% |
As you can see even the worst performing portfolios invested fully in the All Ordinaries resulted in a satisfactory performance so continuing the simulation to see how much better the returns could be are outside the scope of this post.
MSCI+All Ords Inc dividends
When combining the MSCI and All Ordinaries at a 60/40 split we see how diversification reduces portfolio risk. The worst returning portfolio had a finishing value of $355,979 with an annualized return of 3.77%. In this case, holding for an additional 3 years generates a return of 6.75% and the results follow an upward trend from there.
| Finish date | Final amount | IRR p.a. | Additional years |
| 30/9/2011 | $355,979 | 3.77% | 0 |
| 30/9/2012 | $404,852 | 5.69% | 1 |
| 30/9/2013 | $526,783 | 6.22% | 2 |
| 30/9/2014 | $603,850 | 6.75% | 3 |
| 30/9/2015 | $670,456 | 7.01% | 4 |
| 30/9/2016 | $715.678 | 7.00% | 5 |
| 30/9/2017 | $806,441 | 7.30% | 6 |
| 30/9/2018 | $954,692 | 7.85% | 7 |
| 30/9/2019 | $1,053,918 | 7.97% | 8 |
What the results indicate
Start and finish periods are important
The weakest performing portfolios in this dataset tend to be a result of the start date being in the 1990s and the finish date being after the Global Financial Crisis. This is due to the fact that the market experienced strong growth in the 1990s, but due to their small size at the time, the portfolios reaped limited benefit from this growth. They then experienced the Dot Comm bust which took a long time to recover and once the portfolios were near their finishing values, they experienced the negative returns of the GFC.
This illustrates how the sequence of returns is an important, but uncontrollable factor, in determining overall portfolio returns. In the ideal case an investor will be better off experiencing weak returns at the beginning of their accumulation journey and strong returns at the end. To put this into context a 50% loss to a $10,000 portfolio can be recovered in matter of months by continuing to contribute; whereas a 50% loss to a $1m portfolio will take years to recover from and is mostly dependent on market returns.
Diversification is important
As seen in the portfolios that were 100% invested in the MSCI Ex Australia index, negative returns are a very real prospect. The All Ordinaries performed better than the MSCI Ex Australia index over the sample period so adding it to the investment mix helped to increase returns.
It is possible to recover
For the poorest performing portfolios, simply holding the current 100% stock allocation for an additional 4 years produces satisfactory annualized rates of return in most cases.
How this impacts financial planning
Use an appropriate growth rate
Even though the median market growth rates are over 8%, there is no guarantee that your portfolio will achieve this rate of return. Investors should remember that half the values in the dataset will underperform the median. When doing your own financial modelling using a growth rate that is less than the median will increase your real-world chances of reaching your target within your modelled accumulation period. At the FFE house we use a 7% portfolio growth rate to model our portfolio trajectory.
Rebalancing your asset allocation
If you are approaching your retirement date and your returns have been satisfactory consider increasing your asset allocation towards bonds and other less risky assets. This will reduce portfolio downside risk as you approach retirement.
Maintain asset allocation if required returns are not met
If you find yourself close to or at your chosen retirement age and the market has fallen significantly, it’s likely that you will not have enough money to be financially independent or retire. This simulation indicates that maintaining a risky asset allocation (100% stocks) provides you a good chance of recovering, which is a testament to the resilience of stocks.
During this time, you should not withdraw any capital from your portfolio in order to give it the best chance of recovering. This means that you will need to continue working to meet your cost of living. However, this simulation shows that towards the end of the accumulation phase returns are almost entirely market driven, which means that contributing to your portfolio to help it recover is not vital to success.
Conclusion
Since we have limited control over when our investing journeys start and finish sequence risk needs to be given consideration by all investors. These simulations show how important it is to reduce portfolio risk as an investor nears retirement/draw-down phase. It also highlights the fact that shares, while volatile, are a good asset class to help investors recover from a down turn. Further to this it is important for everyone to be flexible with retirement dates, especially in the event that your portfolio underperforms.
Engineer your freedom
References
Dimensional, 2020, MSCI World ex Australia Index (net div. AUD), Dataset viewed 3/9/2020 <https://returnsweb.dimensional.com/>
Dimensional, 2020, S&P/ASX All Ordinaries (Total Return), Dataset viewed 2/9/2020 <https://returnsweb.dimensional.com/>
Notes
Note 1: The lowest performing 15-year All Ordinaries portfolio was actually from 30/4/2005 to 31/3/2020 with annualized return of 4.44%. The additional holding period was not modelled due to a lack of data
Note 2: The lowest performing 20-year All Ordinaries portfolio was actually from 30/4/2000 to 31/3/2020 with annualized return of 5.89%. The additional holding period not was modelled due to a lack of data