smitcat wrote: ↑
Sun May 17, 2020 8:08 am
geerhardusvos wrote: ↑
Sat May 16, 2020 11:55 am
MarkRoulo wrote: ↑
Sat May 16, 2020 10:36 am
willthrill81 wrote: ↑
Sat May 16, 2020 10:13 am
NoRegret wrote: ↑
Sat May 16, 2020 1:39 am
3. Lastly there aren't that many INDEPENDENT, NON-OVERLAPPING 30 year periods of good historical data, much less 40-60 years. If you have N observations of a random variable, the chance of the next observation falling out of the existing range is on the order of 1/N. It's a big problem if N is small.
3. This argument is used often and has some validity but is problematic. It assumes that overlapping periods lead to similar results, and the SWR research has robustly shown that to be completely false.
The other problem with this argument is that those who support it don't offer any real solutions. Some espouse Monte Carlo simulations, but these have their own problems that may result in them having limited value.
This isn’t a question of similar results or not. The fundamental problem is that with only three independent data points you simply can’t have 95% confidence.
There isn’t a “solution” either. We just can’t be as certain as we want to be.
Folks still need to make decisions, but we/they are fooling themselves if they are as confident as the 95% success claims imply.
Classic example of someone who wants absolute certainty and does not understand the data we have available to us.
All humans are striving for certainty. They need it. They want it. But it doesn’t exist in this world. So make a decision based off the analysis in the data available to us. My family has. That data overwhelmingly points to a 4% withdrawal rate working for almost any retirement length with a >50% equity portfolio.
"That data overwhelmingly points to a 4% withdrawal rate working for almost any retirement length with a >50% equity portfolio."
Actually there is a lot of work and data that suggests that the 4% rule of thumb would be best reduced for todays environment and also reduced for retirement horizons more than 30 years and for equity values of less than about 70%. Some of this type of work is readily available to review at Early Retirement Now (ERN) , here is a quote from one of their articles that combine SS for early retirees with SWR calculations:
"With today’s lofty equity valuations and measly bond yields, a 3.25% to 3.50% initial
withdrawal rate would be much more prudent."
And here is a link to that specific article of the series for further details:
https://earlyretirementnow.com/2017/07/ ... -security/
Big ERN is a homie. We exchange messages and I have read all his work. Both being German and economics majors, we are kindred spirits. When I leave the megacorp at age 40, I plan for an initial
wr of 3.5%-3.8% (for first 2-5 years) to avoid SORR based on his and other work. There’s overwhelming agreement in his community that while the future is unknown, it is likely to be similar to the past over long periods of time. Meaning that 4% WR or more is likely
what I will end up being able to withdraw on average over my retirement, but that will be variable and based on market performance
. There is a lot of luck with the specific year one retires. So it’s good to avoid SORR in the beginning, and then ramp up from there. A bond glidepath or other strategies can help with this as well.
Someone retiring in their 50s who has 25X and has worked up to that point, is financially independent free and clear especially given their SS income. For a 30 year retirement, 4% WR is supported by Big ERNs analysis of the data, and he’s said that for retirement lengths greater than 30 years a 4% wr often ends up being acceptable (with all the initial portfolio still in tact!!!). But we don’t know if we got lucky with when we decided to retire (will 2030 be a good year for me to retire? Hoping so!). So avoid SORR, and see where you’re at. If you are forced to give a blanket recommendation to early retirees, 3.5% WR is properly safe historically. Guess what, 4% WR is too in vast majority of cases, especially for 30-40 year retirements. CAPE levels, bond yields, market conditions, etc. are all things that change, and they don’t always reliably predict the future (Econ major here). In fact, I predict that there will be a very difficult economic time for the next 5 to 10 years (while I’m accumulating btw). But only God knows. What usually follows a difficult economic time? Growth.
If you don’t agree with the data or the approach, that’s fine, but you don’t have grounds to keep arguing and nitpicking without understanding the data or providing new relevant data or insights... There is cushion already
in 4% WR. It’s been stress tested over the worst economic periods in our history
. Those of us who are creative, can manage their money, will be taking a variable withdrawal, etc. will be fine thanks!