Yesterday’s blog entry set forth the text of an e-mail that I sent to academic researcher Wade Pfau on December 18, 2010. Wade responded later the same day.
He said that he agreed with my comment re the Business Week article that safe withdrawal rates can rise a good bit higher than 4 percent as well as drop a good bit below 4 percent. However, he argued that the SWR was likely to remain low for some time and thus maintained that “perhaps it is not a major oversight on the author’s part.”
Wade noted that I had moved from my work on SWRS to examining the effect of valuations on allocations and expected returns and said that he expected to do research in that area within the next year. He observed that “I do like your term ‘Historical Surviving Withdrawal Rate’ as a more accurate description of what traditional studies like Trinity show.” He noted that he remembered me using that term in my posts at the Vanguard Diehards board.
Finally, Wade observed that the thread he started on safe withdrawal rates generated far fewer posts than threads he started on two earlier research papers. “It’s interesting to see the difference,” he said.
The text of my response e-mail follows:
Yes, all that is so. The initial SWR discussions were strange in that some reacted with great excitement and some with a great determination to shut down the discussions. My focus then turned to figuring out what was behind that. And that led to all sorts of explorations of all sorts of valuations-related topics.
I ultimately concluded that the confusion stems from the fact that Fama’s model (the Efficient Market Theory) starts from premises that are the opposite of those in Shiller’s model (Valuation-Informed Indexing — the title is mine but the core ideas are Shiller’s). I write a weekly column called “Valuation-Informed Indexing” at the ValueWalk.com site that aims to point out all the differences between the old model and the new model. The column is aimed at better-informed investors
(the type of people who practice Value Investing tend to be well informed, in my assessment) rather than at the typical middle-class investors to whom I direct most of my writings. I first wrote eight articles describing each of the four calculators that John and I developed together, and then added so far 20 weekly columns. Here is a link to the archives:
If I were going to pick out one column entry to illustrate the basic idea being explored, I think it would be this one (“Either Valuations Matter Not at All or They Matter A Great Deal Indeed”):
The best comment that has been made on my work was a comment by Columbia University Economist Rajiv Sethi, who said: “Rob Bennett makes the claim that market timing based on aggregate P/E ratios can be a far more effective strategy
than passive investing over long horizons (ten years or more). I am not in a position to evaluate this empirically but it is consistent with Shiller’s analysis and I can see how it could be true.”
That statement hits it on the head. It would be dogmatic for me to say that everyone must agree with me. I do NOT say that. What I say is that pretty much everything I say follows from what Shiller says in his book “Irrational Exuberance.” The subtitle of the book is: “The National Bestseller that REVOLUTIONIZED the Way We Think About the Market.” Shiller is making a declaration that he REJECTS the conventional model (the Efficient Market Theory, or Buy-and-Hold).
People have been trying to have it both ways now for 30 years. They advocate Buy-and-Hold and they also say that valuations matter. If valuations matter, Buy-and-Hold is dangerous. If the market is efficient, valuations don’t matter and Buy-and-Hold is the ideal strategy. It’s an either-or. I don’t mind people saying that they reject Shiller and thus they reject all that I say. That makes logical sense. But, if Shiller is right, there are all sorts of implications that follow from his revolutionary work. The strange thing is that, while Shiller’s work is well regarded, neither Shiller not anyone else has explored the strategic implications of his findings. Doing that has become my Life Project for the past eight years.
I’d like to illustrate how someone following the VII model would respond to your finding that the predicted withdrawal rate in 2008 was 1.48 percent. Drip Guy obviously saw this as a BAD thing, almost a catastrophically bad thing. The Valuation-Informed Indexer does not see it that way. It is just a data point that provides guidance on how best to invest. A low SWR is neither a good thing nor a bad thing. It is a neutral reality.
In 2000, the SWR for TIPS was 5.8 percent. That is an amazing SWR for a risk-free asset class. All that retirees had to do was to move their money into TIPS and they could retire with great safety. The problem is that this option was rarely recommended because under the Buy-and-Hold Model it is a logical impossibility. Buy-and-Hold posits that the higher returns associated with stocks are the result of the greater risk associated with stocks. So a risk-free asset class can never be expected to provide higher long-term returns than stocks. VII rejects this way of thinking about things. So Valuation-Informed Indexers are free to consider asset classes offering far better value propositions than stocks and thereby to solve the problem presented when the SWR for stocks is very low.
The SWR for TIPS is obviously lower today. But an argument can be made that TIPS offer an even better deal today than they did in 2000. Why? Because the P/E10 for stocks is on its way to 7 or 8 (we have gone to 7 or 8 in the wake of every trip to 25 or above — bull markets always cause enough economic destruction to bring valuations to half of fair value). When the P/E10 for stocks is 7 or 8, the most likely annualized 10-year return for stocks is 15 percent real. So money put in TIPS today will not earn only the amount provided directly through ownership of TIPS. It will earn that amount for a few years and then it will earn the far higher return available for stocks selling at a P/E10 of 7 or 8. The long-term return on that money will be a combination of the amounts earned on the TIPS and the amounts earned on the stocks once the stocks are priced well. Again, the low SWR for stocks today presents no problem. That is a temporary reality than can easily be avoided by moving assets to asset classes offering a more appealing long-term value proposition.
You say that it’s not too big a deal that the article doesn’t mention that the SWR will someday go up much higher than 4 percent. I of course am glad to see any article that points out that the SWR is sometimes a good bit lower than 4 percent. But I also think that the strategic point is a very important one. VII is a LONG-TERM investment strategy. All strategies are based on the idea that investors need to plan ahead. It is also an emotionally BALANCED strategy. It treats all price changes as having both positive and negative aspects. The same thing that causes low SWRs (high valuations) also causes crashes, which cause low valuations, which cause high SWRs. There is no such thing as a permanently low SWR. People don’t need to be concerned about low SWRs so long as they know how to respond strategically to take advantage of them in the long run.
The problem is that Buy-and-Hold does not permit consideration of the idea that super-safe asset classes can offer long-term returns far higher than those available from stocks. So long as an investor is working from that premise, all attractive options are closed to him at times of high valuations. Once we give up a belief in the EMT and in the Buy-and-Hold Model that follows from it, hundreds of doors open to us. My aim is to describe those opportunities to people.
Sorry for all the words. I hope that there is at least some useful stuff to be found in them. Please don’t feel any need to respond while you are away from the usual grind. If you have any questions or concerns or comments or added thoughts, I’d be thrilled to hear them at any time it is convenient for you to share them.