August 1, 2009 Letters to the Editor
Updated: August 15, 2009.
Subject: The VII Investor Experiences Less Risk AND Enjoys Higher Returns
I received this letter from Rob Bennett.
These thoughts come in response to your article entitled "More Switching C Return Graphs."
There is a problem with making comparisons of Valuation-Informed Indexing (VII) and Passive Indexing that has been nagging at me for some time. I am going to try to explain it here and ask if it is possible to measure statistically the extent to which this factor causes confusion.
My sense is that VII always beats Passive in two important ways -- the VII investor experiences less risk and enjoys higher returns. In ordinary circumstances, that would make the case for the VII strategy very clear.
There is an oddity in the way that we measure stock performance, though, that makes things appear a little less clear than they really are, in my assessment. The oddity is that we generally ignore the risk factor when assessing performance. When you (or anyone else) tests how a Passive strategy does, you ASSUME that the investor will hold his stocks despite big price drops --just about every analysis I have ever seen does this. I see this practice as a big problem. One of the big edges provided by VII is that the investor avoids the big portfolio losses that cause stock sales and yet the effect of selling stocks at low prices (which obviously hurts the Passive Indexer far more than it hurts the VII investor) is assumed out of most statistical analyses.
I think that a better approach would be to assume that the investor will sell all his stocks and remain out of stocks for a time if he experiences a loss of a specified portfolio percentage that remains in place a specified amount of time. But I recognize that this adds a good bit of complexity to the analysis and that it is thus probably not a good idea to do this with all analyses (although I certainly think it is a good idea to do it occasionally to make the point).
In recent days I have been thinking more about a related problem -- the problem that stocks in general are the best asset class. The big benefit of VII is that it gets you to a lower stock allocation when a stock crash is coming. It is because of this benefit that VII beats Passive in the statistical analyses. However, I don't believe that VII is being shown to beat Passive by as much as it does in real life because the statistical analyses never show what happens with stock allocations in excess of 100 percent.
It is of course perfectly understandable that the analyses are set up the way they are. Stock investors should not be going with stock allocations of higher than 100 percent (to obtain stock allocations higher than 100 percent, investors would need to borrow to buy additional stocks, an obviously risky practice). In theoretical terms, though, there really is a problem with this practice. VII is a far less risky strategy than Passive Indexing. The benefit to the investor of taking on less risk can show up in these analyses only through a showing that he obtains higher returns (since the assumption that Passives will always hold their stocks rules out the possibility that it could how up through sales of stocks at the wrong time). But the VII investor is not able to translate his risk edge into higher returns (the only thing that counts in these analyses) because he cannot go to stock allocations of higher than 100 percent.
The reason why we don't look at stock allocations of higher than 100 percent is that we consider such stock allocations to be too risky. I of course agree with that assessment. However, the reality is that a stock allocation of 120 percent at a time of moderate prices is less risky than a stock allocation of 80 percent at a time of insanely high prices. If we were going to be theoretically consistent, we should either not count the benefits that a Passive Indexer gets from being at an 80 percent stock allocation at a time of insanely dangerous prices (because the risks here are so great that we simply refuse to look at the possibility, just as we refuse to look at the possibility of a VII investor being at a stock allocation of 120 percent at a time of moderate prices). OR we should look (at least sometimes) at the benefit that a VII investor gets from being able to go to stock allocations of higher than 100 percent at times of moderate or low prices.
The root problem is that VII is better in two different way. It is less risky AND it provides better returns. But we are measuring only one of the benefits. We are measuring the return edge. But by placing a limit on how high the VII investor's stock allocation can go, we are placing a ceiling on how much of a benefit we can see. The reality is that the reduced risk benefit in time TRANSLATES into an enhanced return benefit. The VII investor is (theoretically) able to go to a 120 percent stock allocation at times of moderate prices because he hasn't seen his portfolio wiped out in a price crash and because the risk of a 120 percent stock allocation at a time of moderate prices is not really all that great in a theoretical sense (I don't believe that there has ever been a time when a 120 percent stock allocation taken at a time of moderate prices would have brought terribly bad long-term results). But the artificial limit on the stock allocations that can be examined (the limit is proper in a real-world context, but artificial in a theoretical context) limits the extent of the edge we see for VII investors.
We are letting the Passive Investing mindset influence our analytical frameworks. I of course agree that no one should be going with a stock allocation of over 100 percent. People laugh if you even put forward the idea. But again, the reality is that a stock allocation of 120 percent at a time of moderate prices is less risky than a stock allocation of 80 percent at a time of insanely high prices (I am confident that this assertion would check out if tested statistically). So why do we permit an assumption that the Passive Indexer can use a stock allocation of 80 percent while the VII investor cannot use a stock allocation of 120 percent? If we did, the edge for VII shown in the statistical analyses would be greater.
This double standard re assumptions works to the benefit of the Passive Indexing model when it is tested statistically. Short-term timing does not work. So we cannot say how long it is going to be before a Passive Indexer is going to experience a stock crash. This means that the Passive Indexer gets the benefit of being at a high stock allocation for all the years when he is taking on insane risk. Yet the VII investor is not able to obtain the benefit of being at the more moderate levels of risk associated with being at a 120 percent stock allocation at a time of reasonable prices.
I believe that the implications of the anomaly being outlined here are big. I understand that the challenges of examining the problem through statistical analysis are large. Still, I believe that we will never be able to present statistical analyses showing the true horror of Passive Investing until we come up with statistical analytical techniques that at least make an effort to show these effects better than do most of the now-existing analytical tools. The now-available tools are showing that VII has been superior to Passive Indexing throughout the entire historical record. But they are not showing the full extent to which this has been so, in my assessment.
HERE IS MY RESPONSE
Thank you for another outstanding letter.
I have looked at investor reactions to a limited extent. Look at Current Research L: When Price Drops Occur. Scroll down to Reacting to Price Drops during DCA: 50%, Reacting to Price Drops: Middle Years and, especially, Locking In Failure.
What happens depends a lot on how deeply prices must fall before an investor reacts.
Current Research L: When Price Drops Occur
The appropriate analysis tool is the Scenario Surfer. It has full flexibility. The Investment Strategy Tester is excellent as well. It does not have the full flexibility of the Scenario Surfer. But it does an outstanding job when it can be used.
You can introduce leverage into the Scenario Surfer by increasing the initial balance. The key is that there is now a nonzero minimum balance that you must protect. If you start out with $120000 with $20000 in borrowed shares, you will be bankrupt if the total balance ever falls to $20000 or below. Your broker will sell your shares before this happens.
Thank you again for an outstanding letter.
Update: Margin requirements are in NOMINAL dollars. Our on site calculators are in REAL dollars. I don’t expect this to have a big effect. I consider it no more than a technical issue.
6% for the Early Retiree
I received this letter from Michael.
As always, great site and thought provoking. Thank you for your recent studies - very good.
I have asked variants of my question below before: 1) namely, how to think about 60-year retirement vs. 30-year and 2) taxes. Your response to the first was to think about 2 30-year runs but that was before the new Strategy Tester. And your response to taxes was, to paraphrase, it's difficult since it's so person and investment specific.
So, let me try to ask my question another way. Your recent Continuing 30-Year Withdrawal Rate analysis suggests somewhere between a withdrawal rate of 4.4% and 8% with some caveats (you may have less buying power in year 20 or you may only have 50% balance at year 30). And that got me to thinking. Let's assume I pose the following set of facts:
1) I need to plan for a 60 year retirement,
2) I want to have at the end of Year 60 100% of my original balance (inflation adjusted obviously),
3) Only 10% of my savings/investments is in tax deferred accounts (e.g., the bulk are in a taxable accounts),
4) I need a 6% withdrawal rate pre-tax, and
5) I am indifferent to strategy (VII, etc) and asset choices (annuity vs. dividend blend vs. income, etc) but to guarantee the goals above.
How would you go about assessing the probability that any I can safely withdrawal 6% per year given one (or more) of the strategies you've discussed/examined? Is it fair to take from your Stock Return section, the “Worst Possible” 60-year real stock return of 5.23% and add the VII Advantage of 4.2% per year from your July 23 post (I am using your Year 30 advantage as the same for my 60 Years. Is that appropriate or too aggressive?) to equal a 9.43% real return (“Worst Possible”) and therefore conclude that my 6% target seems very doable? Or am I being overly simplistic? How else would you handicap my likely success? Or perhaps more simply and generally, all else equal, would you think the 60 year continuing withdrawal rate should be higher or lower than the 30 year?
As always, thanks for the information and thoughts.
HERE IS MY RESPONSE
Thank you for a very interesting letter.
We have just had a market run up, which makes higher withdrawal rates difficult.
You need a continuing withdrawal rate of 6% per year (plus inflation). The dividend blend and income approaches come close, delivering about 5.5% today. A delayed purchase is likely to deliver the goods, but it comes with no guarantee.
My DVY+PFF “Practicing for Retirement” portfolio delivers a continuing withdrawal rate of 5.5% (plus inflation). It is not as well diversified as I would like and I think that you can do better. Because PFF is almost entirely in financial services, I would add some REITS and Master Limited Partnerships (MLP) for my high income component. One quality REIT, according to what I have read from others whom I trust, is Realty Income Corporation (symbol “O”). It has a current yield of 6.69%. One quality MLP is Kinder Morgan KMP. It yields 7.93%. PFF, by way of comparison, now yields 7.77%.
You should find MLPs especially attractive because of their favored tax treatment.
My dividend growth component DVY currently has a yield of 4.04%. I assign it a growth rate of 5.5% per year nominal. That is, I treat it as if it has a dividend growth rate equal to that of the S&P500 index. I think that it will do better. I am sure that you can do better than DVY by purchasing individual stocks.
The book, “Mergent’s Dividend Achievers,” is a good source for identifying companies with a commitment to dividend growth.
Can you wait a couple of years?
If so, you may be able to do a lot better.
I am showing the odds as 25%-25%-50% that the market will continue at today’s levels with P/E10=17, that the market will return to the P/E10=14 region or that the market will fall below P/E10=10, respectively.
Worst case: wait a couple of years and retire with a 5.5% per year continuing withdrawal rate (plus inflation).
Possible case: wait a couple of years and prices improve enough for you to retire at 6% (plus inflation).
Best case, about 50%-50%, and you will do exceptionally well. Even with a fixed allocation of 80% stocks, your continuing withdrawal rate would be above 8% (plus inflation).
Read also my response to Dennis in the June 15, 2009 Letters to the Editor, “Setting up a General Retirement Portfolio to use as a Guide.”
June 15, 2009 Letters to the Editor
It's Price Increases That Are "Lumps" for Most of Us
I received this letter from Rob Bennett.
I enjoyed your Note entitled "Take Your Lumps Early." It's not my intent to be picky in this comment. My intent is to point out how the language we all use re investing affects our psychology and thereby causes many of us to follow ineffective strategies.
I personally question whether drops in the price of stocks should even be referred to as "lumps". I believe that when we think that way (we all are guilty of this to different degrees), we are permitting our thinking to be influenced by the Passive Investing Mindset, which almost always encourages us to believe in the opposite of the true investing realities.
Most of us are both stock buyers and stock sellers.
To stock sellers, price drops are indeed "lumps." To stock buyers, the opposite is so. To stock buyers, price drops are the best thing that could happen.
I think we need a change in attitude more than anything else. The Passive Investing model encourages us to think of price increases as good news and to think of price drops as bad news. For those who expect to be buying stocks for some years to come, the realities are precisely the opposite (not for the first time!) of what is taught under the Passive Investing model. It is generally price increases that are "lumps," not price drops. A price drop causes the value of the shares we already own to diminish (the lump side of the equation) but it also makes all the shares we have yet to purchase more appealingly priced (the more important side of the equation for most).
Let's all begin rooting for another stock crash!
(That's a joke -- that would actually be taking it too far as another crash would worsen the economic crisis and thereby do more harm than good.)
HERE IS MY RESPONSE
Thank you once again. You have written an interesting letter.
Lower prices are always a boon to accumulators.
Lower prices cause a problem for retirees who sell shares for income.
We normally think of dividend and income investors as being immune, but that it not always true. They may need to make a large purchase, possibly to cover an unexpected expense. Most of the time, however, they are reinvesting some of their income to grow it as fast as or faster than inflation. When they do, they are better off with low prices.
Yes, it is mixed.
In terms of another crash: I think it is coming. I don’t see how we can avoid it since we are facing stiff tax increases. I would prefer to get the bad news over with sooner rather than later.
Letters to the Editor in 2009
Letters to the Editor in 2009
Letters to the Editor in 2008
Letters to the Editor in 2008
Letters to the Editor in 2007
Letters to the Editor in 2007
Letters to the Editor in 2006
Letters to the Editor in 2006
Letters to the Editor in 2005
Letters to the Editor in 2005