Portfolio , Retirement


We are living through the greatest test of the 4% safe withdrawal rate in history. Sounds like a big overarching statement but that’s what the data tells me.  A little over a year ago I questioned whether or not the year 2000 retiree would be the first to destroy the 4% rule. In other words, the year 2000 may turn out to be the worst time in history to retire. While the history of the year 2000 retiree is not complete we can take a look at his/her progress 12 years into retirement. It didn’t look good then and as of the end of 2011 it doesn’t look any better. In this post we’ll take a look at the concerning progress of the year 2000 retiree.

Lets look into the a person who retired on January 1, 2000. How has their retirement portfolio performed during the 12 years from 2000 through the end of 2011? Consider 3 retirement portfolios; a 100% US stock portfolio, the standard recommended 60/40 stock/bond portfolio and the IVY timing portfolio that I recommend in retirement. We use an SWR of 4%, a withdrawal rate that has worked back to the great depression and a starting value for the portfolio of $1M. The table below shows the annual return of the three portfolios from 2000 to 2011, the SWR of 4%, the value of the portfolio at the end of 2011, and the current withdrawal rate (CWR). The CWR is an early warning indicator of trouble for retirement portfolios. Now, on to the table.

Let me walk you through this table. The first thing to note is the investment returns from 2000 to 20011. Returns during these 12 years were lower than historical performance as one would expect due to the dot com and the financial crisis. Returns varied from a low annual return of 0.58% for a 100% stock portfolio to 11.76% for the timing portfolio. The next column simply shows the SWR used in this analysis and the last columns show the progress of this retiree’s portfolio in terms of the value of the portfolio at the end of 2011 and the CWR. For example, the 60/40 stocks/bonds portfolio with a 4% SWR would have left the retiree with a portfolio value of $642K (an almost 40% drawdown) and with a CWR of 8%. Basically this retiree has a high likely hood of running out of money in retirement, after all he/she has 18 years to go. This is on par or slightly worse than the previous worse case retiree, 1971, who after 10 years had a CWR of 8.5%. Of course, after 10 years, the 1971 retiree, benefited from the greatest bull market in history starting in 1982. Not a pretty picture for the year 2000 retiree. This is why I say that the year 2000 retiree may break the 4% SWR rule. Unless forward returns pick up the year 2000 retiree is in deep trouble.

In contrast, look at the retirement performance of the year 2000 retiree using the IVY timing model. With this investment portfolio the retiree’s portfolio is up to $1.9M and the CWR is 3% below the initial 4%. Looks like smooth sailing ahead. These are far from trivial differences. This difference is stunning. How would you feel after 12 years of retirement looking at a portfolio that is almost 40% less than when you started? This is the primary reason I think most retirees should follow the IVY timing model in retirement. Higher income and safer/higher probability outcomes in retirement. Isn’t that what we all look for in retirement? But wait. It gets better. As I’ve suggested , higher withdrawal rates are possible with the IVY timing model. Crunching the numbers, the SWR for the year 2000 retiree using the IVY timing model is 6.75%. Their portfolio value at the end of 2011 was $1.3M and their CWR was 7%. This is well within the safety range. That is the difference in beginning retirement with $67.5K/yr in income versus starting retirement with $40K/yr income with the same starting portfolio a more than 50% difference.

In summary, there is a high likelihood the 4% SWR will fail for the first time in history for the year 2000 retiree. The alternative that will be trumpeted by the financial industry is to use a lower SWR in retirement, maybe 2% or 3%. Ge, thanks for the help. How many retirements will be prevented or postponed? It doesn’t have to be that way. By using a different yet simple and automatic investment model, the IVY timing portfolio, the 4% SWR is extremely safe and in fact much higher SWRs, up to 6.75%, maybe used in retirement.

Tagged , , , , ,

28 thoughts on “ The greatest test of the 4% rule in history

  1. A great post that summarizes the last 12 years. Will the next 12 years be any better, or are we (the USA) now a mature economy that will no longer show any meaningful growth?

    1. Hi Terry. No one knows. Most likely markets continue to be range bound for a while. This is pretty normal after a major financial crisis. Economy will probably grow modestly but that doesn’t mean stocks will go up. That’s why I recommend the IVY model to most people. For more active investors a core portfolio of individual dividend paying stocks is the best way to go.


  2. Paul, implementing the IVY timing portfolio via ETFs is pretty straightforward for the US stocks, international stocks, US bonds, and US real estate portions; I like Vanguard and so would likely use VOO or VTI, VEU, BND, and VNQ for those portions of the portfolio (in fact, I am currently long all of those ETFs, with the exception of BND because of my perception of future investment-grade bond market returns. I do, however, own some high-yield bond closed-end funds). The commodities portion of the IVY timing portfolio is something I have struggled with, however. Commodities ETFs and ETNs that rely on futures contracts are too subject to exploitation by other futures market participants, not to mention contango and backwardation (things that I have a less than mature knowledge of). GLD (and I assume SLV) is a pain at tax time because you have monthly sales that cover expenses. Do you have any suggestions of commodity based ETFs/ETNs that might avoid those issues or perhaps a hand full of stocks (I currently hold XME, FCX, and NEM) for this portion of the IVY timing portfolio? Thank you. PS: It’s nice to have you back to bounce questions off of.

    1. Hi J. It’s good to be back. DBC is the best way to implement the commodity portion of IVY. That’s what all the historical results are based on.


      1. Paul, thanks for the reply. I owned DBC for a while which, at the time I owned it at least, was a futures based product. It seemed to me, admittedly without benefit of a rigorous analysis, that its price was trailing the prices of its various components; I will give it another look. An aside question, do you think mining stocks could, or should, be included in the commodities portion of the portfolio?

        1. J, no I don’t. I believe in keeping the passive and active portions of my portfolio completely separate. When I implement IVY, I implement IVY. I follow the rules to a tee. I don’t try to fine tune the model. The originator of the IVY portfolio has created an ETF, GTAA, that is a more active version of the IVY model. So far its results haven’t matched the simpler IVY model and you pay higher fees. My IRAs are 100% IVY, the rest of my portfolio I actively invest in individual investments. The IVY model is my benchmark. So, no I don’t think mining stocks belong in the commodities portion of the portfolio. And I wouldn’t touch a mining stock with a 10ft pole in my active portfolio.

          All direct commodity investments are futures based, that’s the only way to invest directly in commodities. What you’re referring to with the issues with a commodity ETF is roll yield. There is both positive and negative roll yield. The DBC fund is aware of this and attempts to mitigate this in this investment strategy. All the details are in the prospectus but overall it does a decent job. The biggest drawback to all commodity ETFs are the fees. If you really want to see the effect of roll yield look at the difference between two oil ETFs, USO an USL. USL is the much better product.

          Hope that helps.


      2. Paul, thank you very much for the in-depth and thoughtful reply. A lot to think about, and I will.

  3. Paul:
    I’ve always been a patient buy and hold investor in good times and bad. As I’ve approached retirement I’ve concentrated on income funds (AHITX, ABNDX, CAIBX, FCISX AND AMECX). My strategy is to follow the 4% SWR and reinvest the excess returns.

    Upon studying the IVY timing portfolio I’m intrigued but have a few questions. Do you follow this in a taxable account? Do you only execute trades (buy or sell) on the last trading day of the month? If not in a non-taxable account what do you do to minimize taxes..Income and capital gains.

    Thank you for your time.

    1. Hey Kevin, thanks for the comment. I personally implement the IVY in my IRAs, so non-taxable. Yes, you only trade on the day following the end of the month. You look at month end prices, make the buy/sell list and do it the next trading day. You can easily find the month end signals on-line either from world beta or dshort . I’ll do the month end updates here too.

      In a taxable account it can be just as simple. The extra returns from this model are way over and above the tax consequences. Also, the sell signals, the only ones that cause tax issues, don’t occur as often as you might expect. But you can do some extra work to minimize the tax consequences. For example, to book a loss you can sell an ETF and immediately buy a similar ETF to make sure you stay invested. For example, selling VTI and buying the SPY. May be Ill post on this in the near future.


  4. Paul:
    The more research I do the more questions I have. Today I started reading the IVY strategy. I hope you don’t mind but I’d like to ask if you invest in WASAX or do you primarily use the Vanguard ETF’s and why?

    Thank you for you time.

    K Daly

    1. Kevin, WASAX has nothing to do with the IVY timing portfolio. It just happens to have IVY in the name. The only fund or ETF that follows the IVY timing model is GTAA which was started by Mebane Faber the author and founder of the IVY timing portfolio. I highly recommend his web site, World Beta , and reading the white paper he wrote and the timing model, although it needs updating.

      I use the ETFs listed on Mebane’s site for the most part; VTI, VEU, IEF, VNQ, and DBC. Mebane’s ETF, GTAA is a more active version of the basic IVY timing model that hasn’t proven itself yet and more importantly the fees are high. Fees are hugely important to this strategy’s performance. At least at my broker, TD Ameritrade, all these are commission free ETFs as well.


  5. Paul:
    On Mebane’s site were he post the charts for VTI, VEU, IEF, VNQ, and DBC he list below the current closing price of each asset the 10 month SMA. It appears he is using a 300 day moving average to determine sell signals and rebalancing of the portfolio. In your posting you discuss 200 day averages. Is the difference because you are trading in your IRA and Mebane’s model is for a taxable account?

    K Daly

    1. Hi Kevin, the 10 month SMA is equivalent to the 200 day SMA. In markets only trading days are counted, so 200 days is equal to 10 months.


      1. Paul:
        With the posting of the months closing prices I’ve reviewed Mebane’s model and would like your input on strategy. As I see it, VNQ, VTI and IEF are still well above the SMA and are strong holds. DBC and VEU remain below the SMA, so following the strategy one should remain in cash for those positions.

        My question is in the case of DBC. It is only $ .23 a share (with a strong trend line) out of the buy range. Do you stretch a little and add the position or wait a month.

        VEU is also close ($ .55 a share) but the trend line isn’t as convincing.

        I’d appreciate your thoughts.

        K Daly

        1. Kevin, breaking the model rules is a slippery slope to get started on since it invalidates all the historical results.
          So, for DBC and VEU you stay in cash this month. It doesn’t matter how close they get. Don’t read any more into the model than its trying to sell you.
          Leave that for the active and more aggressive part of your portfolio.


  6. Hi Paul,

    I enjoy reading all of your posts and am happy to see you are active with them again. I started reading about the IVY portfolio when you wrote about it some time back. I moved most of my IRA into cash a few months ago and am ready to start. Would you do all the allocations at once? As of today, it looks like all ETFs will have allocations at month end, except for foreign. I am a little nervous that the entry points in REITs and US might be too high at this point. Would you recommend making all the allocations at once or perhaps do half the portfolio in IVY and leave the balance in cash, then increasing a percentage of the cash to IVY allocations over the next few months?

    By the way, we have a motorhome and love following you and your wife’s travels on Wheeling It!


    1. Hi Celia, thanks for the comment and I’m glad you’re enjoying our travel blog.

      You ask a great question. One that I’ve been waiting someone to ask for a long time. As you noted getting started with the IVY model can present some challenges. As you point out you can make all the allocations at one or do half or something like that. You can also wait until each ETF triggers the next buy signal. So, for example for VTI and VNQ (and IEF) you would wait until they trigger sell signals and wait again until they trigger buy again. That would be the safest approach but as you can probably imagine it could keep you out of these asset classes for a while. In other words, there is no one answer. It depends how you view and handle potential short term losses. In the long run the timing of your first buys won’t matter but its sure nice to get started with some positive results.

      I would do something like your 50/50 idea with a little twist. First, wait till the ETF (say VTI) pulls back. At least to the 50 day SMA, see if it holds, and if it does do your 50% buy of VTI. Or wait for a bigger pull back to the 200 day SMA like we had in June and if it holds then buy. Just an idea. This way you get in at a discount that would allow you to see some early positive results. Other wise your simple 50/50 approach would be great or even a 33/33/33 staged approach if you want to be a bit more conservative.

      Hope that helps.


  7. Thank you so much! That helps a lot Paul. I thought DBC would cross, but it didn’t. So will I do nothing with that ETF until it triggers a buy at some future month end.

    If I understand you correctly, I should watch the three etfs that are currently a buy. When they pull back to either the 50 or 200 SMA, then I can purchase my allocation at that time, regardless of the time of month. Two questions: Can you recommend a good site to follow those charts? Next, you indicated they should hold the specific lines. How long should they hold the line?

    I really like this simple method and appreciate you helping me get started with it without allocating everything at once.


    1. Celia, I hate the standard advice of just do the allocations, dont worry about it, it doesnt matter in the long run. Easy to say when its not your money.

      My favorite site to use for quick charts is stockcharts.com. They have easy to use and free charts with the basics you need all set up by default. The charts will automatically show you the 50 and 200 day SMAs. Also, you should learn to read bar charts – very helpful. You can find an explanation at stockcharts.com.

      Yes, you understood me correctly. You can purchase your allocation to the ETFs that are currently a buy when they pull back, hold those support lines, and reverse higher. Usually I like to see the ETF hold the line with a closing price above the line then have a follow through day where the price goes up. For example, look at the VTI chart from early June, here . See the prices on the first four trading days of June. The first day the price moved down quite a bit and closed right at the 200 day SMA. The second day it broke below the 200 day but came back and closed just below it. On day 3 it opens below the 200 day but has a very strong day and closes above the 200 day. This is when you’re all ears. On day 4 VTI opens up strong and continues up. This is the day I would have bought. Buying on day 3 would have been better in hindsight but the market could have easily reversed back down. Its helpful to see a confirmation of a price move even if you miss out a little. That’s an example.


  8. Paul, again thanks so much! You are a great instructor. I visited stockcharts.com and read through the explanations. I also found that I could set alerts with my online broker to notify me when each of the ETFs crossed either their 50 or 200 day Moving Average. I set alerts on all of them and will continue to check stockcharts.com as well. It feels good to have a plan.


    1. Great Celia. Glad it makes sense. Take your time and be patient. Markets these days will give you all kinds of opportunities.

      Now that you’re an expert… (just kidding), one more little subtlety that most people overlook. For your initial plan of looking to buy between month ends you can use the triggers from your brokerage to alert you to key events. However, the month end signals, like you’re looking for DBC and VEU are based on moving averages that are adjusted for dividends which your brokerage will not have. For the big month end signals make sure you use the charts at World Beta or Dshort’s monthly updates (or mine of course…) which take this into account. This is how Dshort describes it;

      Footnote on calculating monthly moving averages: If you’re making your own calculations of moving averages for dividend-paying stocks or ETFs, you will occasionally get different results if you don’t adjust for dividends. For example, VNQ triggered a buy signal in December based on adjusted monthly closes, but there was no signal if you ignored dividend adjustments. See the comparison here. If you use data from Yahoo Finance for dividend paying assets, you would use the right column of adjusted closes in calculating the monthly moving averages. Here is the link for VNQ. Because the data for earlier months will change when dividends are paid, you must update the data for all the months in the calculation if a dividend was paid since the previous monthly close. This will be the case for any dividend-paying stocks or funds.

      Clear as mud?


  9. Paul:
    I have a few questions for you regarding strategy using the Ivy model and the five ETFs that are tracked daily on World Beta. In the book, The Ivy Portfolio, they talk about a momentum strategy. Moving funds from a sector that’s posted a sell signal on the first of a given month and move those funds into a sector that’s still in a buy positon (above the 200 SMA). Do you feel this is a good strategy? Please explain.

    My second question involves using C shares in, for example, the Franklin Income fund. FCISX. A great income fund. I assume we can track track the 200 SMA of that fund. If so, would the same rules apply. Buying above the SMA and selling below the SMA. As long as you used C shares that had been invested for the minimum of one year (or A shares for that matter – but trying to keep fees at the lowest level) you could liquidate into their money market account until the next buy signal. If this is possible you might be able to take a basket of the Franklin Templeton Funds for additional diversity. Your thoughts would be appreciated.

    K Daly

    1. Kevin,

      I have only looked at the IVY timing model, not some of the other strategies that Mebane talks about in his book, like the relative strength strategy that you mention. So I can’t comment one way or the other. The IVY timing model is nice and simple which I think is one of its biggest strengths.

      FCISX is an open end mutual fund, not an ETF. The IVY model is designed for ETFs. I don’t know if what you suggest will work. From what I’ve found the moving average strategies tend to only work with bigger higher volume ETFs. For example, Ive shown that the 200 day SMA strategy works with some the dividend ETFs. An open end mutual fund is not subject to a lot of these trend effects IMO. But I don’t know. There is only one way to find out and that’s to look back at history and test the strategy. With the price history of the fund its not that hard to do in excel.


  10. Paul:
    I have a question for you regarding the Ivy Portfolio. Good income funds have provided a 5% dividend for the past two years while the stock market has languised. What has the overall return been for the IVY portfolio and what strategy would you recommend for a person looking for income.

    Kevin D

    1. Kevin, I’ve posted on the IVY portfolio returns several times before. See here and here for examples.

      I personally follow an income based strategy which I discussed at length in this post . And for the core strategy of dividend investing see the posts discussed here .


Comments are closed.