Portfolio, Retirement, Uncategorized

Higher withdrawal rates via diversification & timing

In an earlier post I introduced a method to time the market, based on the 200-day moving average, that produced higher investment returns, lower volatility, and lower drawdowns than a buy and hold portfolio. As I’ve stated here many times lower drawdowns (max principal loss in any period) is critical for retirement portfolios. The fact the retirees need to make withdrawals every year, no matter what the market is doing, makes principal protection just as important as investment returns.

For this post I wanted to know, can you have your cake and eat it too? Can you have higher investment returns and higher safe withdrawal rates by using proper diversification and market timing? I ran the numbers for portfolios from 1973-2008. Unfortunately, most of the ETFs used for the diversified portfolio were not available before 1973. I used 4 different portfolios; 100% US stocks (S&P500), a traditional 60/40 Stock-Bond retiree portfolio, a buy and hold 5 asset class diversified portfolio, and timing the 5 asset class diversified portfolio. The 5 asset class portfolio is evenly split between US stocks, International stocks, US bonds, US real estate, and commodities. Lets look at the portfolio statistics first:

As the chart shows, as far as investment returns go, you CAN have your cake and eat it too. Not only did the timing portfolio have the highest return but it also had the lowest volatility (risk) and the lowest drawdown (least worst year). By the way, the worst year for all these portfolios during this time period was 2008. I’d say these are pretty impressive results – they not only show the power of diversification, but also of timing. But do these results translate to higher withdrawal rates (SWRs) for retirees? Lets look at the results:

The SWR for the traditional retiree portfolio during this period (1973-2008) was 5.33%. This is higher than the SWR for longer time periods due to avoidance of the awful years of the great depression and the mid 60s combination of high inflation and poor stock market returns. The chart also shows the powerful effect of diversification. By diversifying among the 5 asset classed mentioned the SWR increases to 6.19% or a 16% increase. And adding timing to the diversification increases the SWR further to 6.78% or a further 10% increase. So, diversification and timing could increase retirement withdrawals by 26%. Such an increase would provide many retirees a much more comfortable retirement or similarly allow more people to retire earlier with less assets.

Now, there are some caveats to this analysis. The data set is limited. The data only goes back to 1973. The biggest effect of this is that it limits the number of 30 year retirees portfolios to run the data against. This makes me take the absolute SWRs with a grain of salt. I think its more prudent to use the percent increase in SWRs for the impact proper diversification and timing can have. In this case I would take the 4% SWR that has been tested back to 1900 and apply the 26% increase in SWR due to diversification and timing. That would take the SWR to 5.04%.

There is a lot of work that has not been done on retirement withdrawal rates. Maybe as more boomers start to enter retirement more emphasis will be put in this area. The two biggest issues with withdrawal rate studies are that they do not take into account proper diversification and they don’t take into account the impact of negative returns on retirees and options to avoid that impact. I hope this and other post of this topic show that a retiree has many options to increase their withdrawal rates in retirement.

In conclusion, stay diversified and watch those negative returns! You’ll be better off no mater how you structure your retirement.

Full Disclaimer - Nothing on this site should ever be considered advice, research or the invitation to buy or sell securities. These are my personal opinions only.

Tagged , , ,

About paul.novell@gmail.com

16 thoughts on “Higher withdrawal rates via diversification & timing

  1. Paul:
    What is the best place to view the 200 moving avarages of one’s positions? How far off the curve’s edge is the most judicious place to buy or sell a position. Should these buy/sell moves the staggered or total? Does the dividend payout of a position influence a trade moving average change trade if one is still within one’s game plan with a position in spite of it moving lower? Your “tips” so far have placed me in the 2K a month dividend income. Just looking to continue to remodel. Thank you

    1. Hey Rick. I usually use the Yahoo finance chart to see the 200 day moving averages. It doesn’t show up on the standard charts but you can add it to the charts.
      Most charting sites give you the option to see the 200 day moving averages. The buy/sell decisions in the timing system are made only once a month, at the end of the month. If the price is above the 200 day you stay in or buy, if the price is below the 200 day you sell or stay in cash. Dividend payouts do not affect the system – its purely based on price. Dividend yields in general aren’t high enough to affect the 200 day when the stock goes ex-dividend for example.


  2. Hey Paul. Thanks for this article. Very good. I wanted to tell you I got a Kindle for Christmas!! I was getting frustrated trying to read charts in finance books in the IPhone. At 51, it was either that or get new glasses. Anyway, was reading The New York Times (how cool us it to be able to subscribe to that and WSJ and Time and Newsweek on the Kindle. I know I am long -winded but there is a point to my story – I am just so excited about my Kindle. There was this story in the NYT about a man who is a master in the investment world who was diagnosed with cancer. He stopped further treatment and instead wrote an amazing book. It is already in the Top 10 on Kindle Bestsellers. Thought you would like it. He knew he would probably not live past Mar 2011. Wanted to leave a legacy and teach people an investment method that is his life’s secret. The Investment Answer by Daniel Goldie and Gordon S. Murray. Here’s another one your readers might like, too: Early Retirement Extreme by Jacob Lund Fisk (also on Kindle). Have fun in beautiful places. I’ll be seein’ ya on the Net.
    Warmest Regards,

    P.S. I have a hand-written note from the man – Warren Buffet himself- after I wrote him a thank you note for his inspiring character. How cool is that? 🙂

    1. Hey Lynne, I love my Kindle as well. Thanks for the book recommendations. I’m always looking for a good read.

  3. Just stumbled across this blog while reading your RV blog and it is nice too see others that RV and hope to generate income while doing so… now on to the question…

    You say the buy/sell decisions in the timing system are made only once a month, at the end of the month… so have you done any back testing to see what the results would be if done every week say on a Thursday close (since in recent times it seems as if Fridays have been very volatile)?

    1. Mebane Faber, the guy who created this system, did testing on different time frames for moving averages and such and found no big differences. Also, end of month falls on whatever day it falls, not just Fridays. For example, end of month this Nov fell on a Tuesday. So. while he hasn’t tested the exact method you mention based on his results I’d say it wouldn’t make much difference to the system’s results. Check out Mebane’s paper on the system. I link to it in this post: https://investingforaliving.us/2010/11/10/reducing-risk-and-enhancing-returns-through-market-timing/


  4. Paul:

    It makes sense. Even if one can avoid sliding one down trend into wipeout territory, as I have done, is worth this effort. Now let’s see if I can set up my cruise control. 🙂

    I use Big Charts.com There are interactive charts available. There are many choices and filters in the upper and lower indicator windows. It looks like I should set the upper indicator daily prices chart with SMA(9)in a YR or 6 month range. How would you set the lower indicators?… Momentum, MACD and DMI seem to be the simplest. It looks like if momentum rises with a change in trend direction and a +MACD line crossing it might be a good “buy” trading time. The same would apply in the reverse situation as a good “sell” time. It also looks like looking at these grafts should become part of a daily routine.

    Am I even in the ball park or am I reading too much into this?

    Happy Holly Days for you Nina and your family


    1. Rick. First, happy holidays to you too.

      Big Charts is a good one to use. You only need to track one indicator, the simple 200 day moving average. That is the only indicator Mebane’s system uses. If at the end of the month the closing price for the asset class is above the 200 day moving avg you stay invested or buy into the asset class if you were out the previous month. And vice versa.

      No need to get fancy but if you want to use another indicator to give you a potential heads up then use a shorter simple moving avg like the 50 day. I use that one to give me a better feeling of the the short term momentum of the asset class but I don’t make decisions off it.

      There are litterally thousand of other technical indicators. Unless you are trading, not investing, they are not worth it in my opinion.

  5. How do you first enter the asset class using this method? Do you just buy whenever, or do you wait for the stock to fall below the 200 moving average and just start to rebound for the first time?

    1. A couple of options. Wait until it crosses below then back above the 200 day for a buy. Or buy whenever you want to invest and the its above the 200 day SMA. Over the long term it doesn’t matter but the short term results of investing right away may suffer. When I did it, I waited. The author Mebane Faber says just to invest in the strategy and ignore short term results.


Comments are closed.