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.
Peter Lynch recommended a 100% approach in one of his books. He claimed he had his analysts calculate what would happen if a retired person placed all assets in the market and concluded that a 7% withdrawal rate would work even if the market crashed during that person's retirement. I wonder if Lunch still believes that.
I thought for most humans with risk tolerance 80-20 worked best as it showed little gain given up compared to 100% equities and was much smoother. To me the real interesting asset allocation issue is whether to invest in international and whether to hedge currency. Obviously you should in situations like Japan in the last few years where it was clear the govt. intended to weaken the currency
Peter Lynch made a rookie blunder when he recommended a survivable 7% withdrawal rate for a retirement portfolio. He based the faulty endorsement assuming historical average portfolio returns with zero volatility, zero standard deviation in those returns.
During retirement, portfolio returns variability is a killer.
One early critic of the erroneous Lynch analysis was the team from Trinity University. Their work became known as the Trinity Study. The professors, Philip L. Cooley, Carl M. Hubbard, and Daniel T. Walz, published a paper titled “Retirement Spending: Choosing a Sustainable Withdrawal Rate”.
They basically concluded that something like a 4% annual drawdown was more realistic in terms of portfolio survival. Here is a Link to a recent Forbes article that reviews the work:
Much work has been done to update these findings. The Monte Carlo simulation codes are terrific tools to assess permissible withdrawal rates. They are now accessible on the Internet. These codes permit the user to explore countless scenarios in a quick and convenient way. The basic output is a portfolio survival estimate. What-if scenarios can be used to test the robustness of various portfolio construction options.
Here is a Link to the easiest code to input. It is on the MoneyChimp website. It is not the most sophisticated code, but it demonstrates the power of this tool:
Note that since these simulations are based on random return selections that are coupled to the statistical input, results will vary somewhat even for identical inputs. That’s the nature of market uncertainty that is captured by Monte Carlo methods.
Comments
Peter Lynch made a rookie blunder when he recommended a survivable 7% withdrawal rate for a retirement portfolio. He based the faulty endorsement assuming historical average portfolio returns with zero volatility, zero standard deviation in those returns.
During retirement, portfolio returns variability is a killer.
One early critic of the erroneous Lynch analysis was the team from Trinity University. Their work became known as the Trinity Study. The professors, Philip L. Cooley, Carl M. Hubbard, and Daniel T. Walz, published a paper titled “Retirement Spending: Choosing a Sustainable Withdrawal Rate”.
They basically concluded that something like a 4% annual drawdown was more realistic in terms of portfolio survival. Here is a Link to a recent Forbes article that reviews the work:
http://www.forbes.com/sites/wadepfau/2015/06/10/safe-withdrawal-rates-for-retirement-and-the-trinity-study/
Much work has been done to update these findings. The Monte Carlo simulation codes are terrific tools to assess permissible withdrawal rates. They are now accessible on the Internet. These codes permit the user to explore countless scenarios in a quick and convenient way. The basic output is a portfolio survival estimate. What-if scenarios can be used to test the robustness of various portfolio construction options.
Here is a Link to the easiest code to input. It is on the MoneyChimp website. It is not the most sophisticated code, but it demonstrates the power of this tool:
http://www.moneychimp.com/articles/volatility/montecarlo.htm
Note that since these simulations are based on random return selections that are coupled to the statistical input, results will vary somewhat even for identical inputs. That’s the nature of market uncertainty that is captured by Monte Carlo methods.
I hope you find the references useful.
Best Regards.