William Hungiville III from FL posted over 10 years ago:

Very good article, as usual, from Dr. Israelsen.

Bryant Vann from VA posted over 10 years ago:

Nice analysis of historical data and presentation of the results! Would LOVE to see you repeat this with higher stock %s, some of the newer investment vehicles you mention, inflation/COLAs in the current ranges of say 8 to 10% as reported at chapwoodindex.com and shadowstats.com, and, perhaps lower payout ratios if that's necessary to last 35 years. Keep up the good work!

Jim Snide from OH posted over 10 years ago:

If the majority of one's retirement funds are in various IRAs, then one must at 70 1/2 start taking Minimum Required Distributions MRD's. The initial withdrawal starts at 3.65% and increases each year based on ones calculated remaining life and is 6.1% at age 83 and so forth. The rate of withdrawal is not constant as in your model, but increases every year. Perhaps the model, should consider taxable investments of the excess of the MRD beyond set percentages in your model and the effect of taxation on the withdrawals? I do appreciate your demonstration of the importance of higher allocation to stocks in a retirement portfolios - especially in the current interest environment for bonds.

Daniel O'Brien from IL posted over 10 years ago:

This article closing matches what I am doing with my retirement plan (assuming living to 100). However, my assumption for stock returns is lower then it has been for the pass 90 years. The dysfunction (debt levels, bad economic policies) by the federal government will certainly have a negative impact on returns for the foreseeable future.

John Rogers from CO posted over 10 years ago:

If I invest in a portfolio of dividend growth stocks and ETFs yielding 4% and growing at from 6 to 10%, I'll never run out of money.

John Blevens from CA posted over 10 years ago:

I plan on increasing the bond portion of my portfolio as I increase in age, not keep a set percentage until death. When I am 90 (if I am still around) I will have approximately 90 percent in bonds.

Herbert Schechter from MN posted over 10 years ago:

It is too shallow an approach to put your strategy on autopilot. Judgment must be exercised along the way to adjust for changing circumstances. Inflation, interest rates and dividends can go up or down, not necessarily together. Industries change. New companies emerge while others fail. The study is based upon average results during the historical period. After 30 years of declining interest rates, do you think the next period will be the same as the last? Interest rates have almost no place lower to go. This study in interesting history, but not the recipe for having a secure retirement.

Richard Cloutier from Georgia posted over 10 years ago:

A while back, I read an article which backtested withdrawal rates, not only on the basis of asset allocation, but also tested which asset categories from which to withdraw. For example, rather than withdrawing fixed amounts from each of the asset categories to make up a 3% annual withdrawal, the report looked at making withdrawals only from overvalued categories. This report looked at measuring stock valuations using CAPE and withdrew a majority (if not all) of the annual distribution from stocks, if they had a high CAPE rate. I thought it was an interesting perspective on the withdrawal debate.

Robert Greving from TN posted over 10 years ago:

Good article and analysis. I wish that analyses such as this took into consideration effective tax rates. I assume that the investor must pay taxes from the withdrawals they take. That can materially reduce the amount available for living expenses.

Victor Stankevich from NC posted over 10 years ago:

It would interesting to see additional portfolio allocations particularly in the more aggressive range; e., 80/20.

David Levine from NC posted over 10 years ago:

All interesting. I retired in 1998 and needed my portfolio to last at least 45 years as I was 58 and my wife was 52. My split was 35/65 equities/fixed income. When I retired most advisers thought 6% as a withdrawal rate was extremely conservative and 8% was more realistic - We were in a period of 15% / 20% market returns. I chose 3% as a withdrawal rate. The point being your guru/investment adviser is influenced by the most recent market returns but you can not reduce your withdrawal rate when the investment market turns against you. The other caveat I would make you aware of is when in 2008 you loss 38% in your equities did you sleep well or did you sell some. If you slept well then 2009/2010 markets did well by you. None of us know the future but you should know yourself and how well you can handle market downturns like 2008 or long term unmoving markets like the mid 60S through the early 80s. Remember it is a market out there and there is no magic. If you want to have more than market returns you must have more than market risks. Check your fees as that is just money out the window; unless your broker is delivering weekly cases of Veuve Clicquot to your doorstep. Please educate yourself and remember AAII has a weatlth of educational materials and studies; make use of them.

G. Michael Frey from OH posted over 10 years ago:

I am going to be 64 next month, and have generally embraced the "4% rule" as being a rational way to extract funds from retirement accounts. However, that being said: The market is NOT static...it fluctuates wildly, and the interest rate trends as of late (i.e. essentially zero rates) will not last forever. My gut feeling is that high quality dividend yielding stocks will do better than bonds, in the next several years (i.e. IMHO, reversion to the mean). Since interest rates are now more likely to rise than fall, I think owning stocks like XOM, PG, WMT, PFE, RTN, MSFT, AAPL, etc. are the way to go (some modicum of growth PLUS solid dividends over time). This upcoming presidential election may also throw in interesting "wrench into the gears"; there is not clarity or consensus on how the current presidential candidates will affect the market for equities.

Gary Jaffe from CA posted over 10 years ago:

The conclusion that a 65/35 equity/bond allocation will outlast a 25/75 equity/bond portfolio is useful -- but omits key asset classes. I have owned a portfolio of closed-end preferred stock funds for the past 24 years. My portfolio has beaten both the S&P 500 and the Barclays bond average with half the risk (Sharpe Ratio). My portfolio yields 8% and holds mostly investment grade securities. I withdraw 6% per year and leave 2% to cover inflation. I don't have to decide what assets to sell to generate income since my dividends arrive monthly. I have not had to sell any of my asset principal to generate income. I own no equities whatsoever and will not outlive my portfolio.

George Carter from WA posted over 10 years ago:

I would be interested in a comparison of 100% stock mutual fund, say the S&P500 index returns. This would eliminate the hassle to rebalance the stock/bond ratio which creates additional taxes when selling the higher flying stocks, too. As for the market corrections, one needs to focus on the rebound and taking 4%-6% out is not going to deplete the stock quantity which always recover value (e.g. returns since 2008 ~300%).

Dave Gilmer from WA posted over 10 years ago:

This was an interesting article and I especially liked the circle charts giving a good feel for which sequences ran out of money. However, the author bias the results to the downside (more years running out of money) by not using actual CPI data for the expense adjustment. You really shouldn't use average CPI data for the same reason you shouldn't use average return data. It misses the sequence risk, either positive or negative that exists from actual data. For instance in 1930 the CPI was -6.4%, in 1931 it was -9.3%, and in 1932 it was -10.3%. If fact using actual CPI data (furnished by Jim Otar's aftcast program) if the withdrawals started in 1929, it was not until 1944 before the investors actual CPI expenses got back above the original starting amount. In fact after 35 years starting in 1929 with a 65/35 AA and 4% CPI adjusted withdraw rate the investor has more money than he/she started with.

Dave Gilmer from WA posted over 10 years ago:

@ George, You can really eliminate the hassle of adjusting the stock/bond allocation by owning a balanced fund such as the Vanguard Wellington, or if you want to use indexes then use a Vanguard target date fund, with the "date" adjusted to your own risk tolerance. In general I have run many of these simulations using real stock data and CPI data and 100% stock positions do not fair well (some cases run out of money) when you approach a 4% and above withdrawal rate. Now you may say "well 90% of the time I will have plenty of money due to good stock market results." That is not how I prefer to design retirement scenarios -- if you run out of money "once" there is no second chance, so it is better to be conservative. What I have found is that as little as 20% bonds can start having a positive affect on your results. The sweet spot seems to be in the 30-60% bond range. There is also some recent work that indicates increasing your stock allocation after the first few years of retirement will improve your results. The reason being you are most subject to the dreaded sequence of return risk during early retirement.

Rick Haas from Michigan posted over 10 years ago:

At 71 years old I listened to all of the 3% to X% talk with great interest when I was younger. Then came reality last year. I reached 70 ½ and sat down with my investment companies to talk X%. Every time I said x% then answer came back “...that X% is up to you if want to deal with the IRS. All we can do for you is tell you the RMD that you owe the government. Just submit the request for the RMD so we can get the paper work done. Otherwise you will have to submit a separate special request for distribution and that could take awhile...’ When I protested to upper management I got the response “...all we are trained to do is submit RMD processing. If your investment advisor wants to submit a separate plan that we will keep on file for the IRS we can do that...” Since I am my own investment advisor I went with the RMD for now...

Bob Atwell from AZ posted over 10 years ago:

Did everyone not notice that absolutely every "failure" occurred in earlier years. Clearly something has changed over time and the underlying variable is NOT random. Imagine an experiment where somebody flips a coin 100 times and it comes up heads 50 of those times. That might seem reasonable until we then discover that it came up heads on the first 50 flips and tails on the next 50 flips. The probability of that happening with an unbiased coin is very nearly 0.