Here's a statement of the obvious: The opinions expressed here are those of the participants, not those of the Mutual Fund Observer. We cannot vouch for the accuracy or appropriateness of any of it, though we do encourage civility and good humor.
Support MFO
Donate through PayPal
Why A 100% Stock Portfolio Can Ruin Your Retirement
The writer makes a case for portfolio diversification during retirement. I don’t object to that conclusion, but I do think that the method used to justify that conclusion was heavy-handed and a little deceitful. He also did not report on the tradeoff that exists with portfolio survival probability and likely median end wealth. These are easily estimated using Monte Carlo codes.
In arriving at his conclusion, the author honestly admits that he selected an especially poor time period for the equity marketplace (like since 2000). That’s an especially weak approach if general conclusions are the goal of the study. His conclusions would be far more convincing if he had done multiple simulations using data from several timeframes. Again, this is easily accomplished using a Monte Carlo tool like from the PortfolioVisualizer website.
I did precisely that. I ran a half dozen reasonable simulations in about 6 minutes.
To get a respectable number of portfolio failures, I assumed a slightly high initial 4.5% drawdown rate, both adjusted for inflation and not so. I made runs using the entire available market data sets and similar runs using only the data starting from the year 2000.
Not unexpectedly, more portfolio failures were predicted for a 100% US Equity portfolio over a 60/40 US Equity/Total Bond portfolio for the more recent 2000 starting point. In a sense, the chosen data timeframe preordained the outcome.
Using the full market data sets, portfolio survival estimates were 80% for the all Equity portfolio and 87% for the 60/40 Mixed portfolio. Using market returns data starting in 2000, portfolio survival rates dropped to 50% and 67% for the all Equity and the Mixed portfolios, respectively.
What the author failed to report is that the end wealth for the surviving all Equities portfolio was always much higher than that for the Mixed portfolio. The differences were substantial. Therein rests the usual tradeoff: More risk; higher rewards.
I did a perturbation case that demonstrated portfolio survival odds would increase substantially if inflation drawdown increases were not needed,
The referenced piece would have been far more complete and more compelling if a simple analysis of the type I just described had been performed. The work required approaches zero. I’m astonished that many folks, including professionals, do not take advantage of this powerful tool when doing investment studies or making investment decisions.
As the writer said: “Sizable bond allocation cushions market volatility”. But there is a price for that cushion.
Comments
The writer makes a case for portfolio diversification during retirement. I don’t object to that conclusion, but I do think that the method used to justify that conclusion was heavy-handed and a little deceitful. He also did not report on the tradeoff that exists with portfolio survival probability and likely median end wealth. These are easily estimated using Monte Carlo codes.
In arriving at his conclusion, the author honestly admits that he selected an especially poor time period for the equity marketplace (like since 2000). That’s an especially weak approach if general conclusions are the goal of the study. His conclusions would be far more convincing if he had done multiple simulations using data from several timeframes. Again, this is easily accomplished using a Monte Carlo tool like from the PortfolioVisualizer website.
I did precisely that. I ran a half dozen reasonable simulations in about 6 minutes.
To get a respectable number of portfolio failures, I assumed a slightly high initial 4.5% drawdown rate, both adjusted for inflation and not so. I made runs using the entire available market data sets and similar runs using only the data starting from the year 2000.
Not unexpectedly, more portfolio failures were predicted for a 100% US Equity portfolio over a 60/40 US Equity/Total Bond portfolio for the more recent 2000 starting point. In a sense, the chosen data timeframe preordained the outcome.
Using the full market data sets, portfolio survival estimates were 80% for the all Equity portfolio and 87% for the 60/40 Mixed portfolio. Using market returns data starting in 2000, portfolio survival rates dropped to 50% and 67% for the all Equity and the Mixed portfolios, respectively.
What the author failed to report is that the end wealth for the surviving all Equities portfolio was always much higher than that for the Mixed portfolio. The differences were substantial. Therein rests the usual tradeoff: More risk; higher rewards.
I did a perturbation case that demonstrated portfolio survival odds would increase substantially if inflation drawdown increases were not needed,
The referenced piece would have been far more complete and more compelling if a simple analysis of the type I just described had been performed. The work required approaches zero. I’m astonished that many folks, including professionals, do not take advantage of this powerful tool when doing investment studies or making investment decisions.
As the writer said: “Sizable bond allocation cushions market volatility”. But there is a price for that cushion.
Best Wishes.