I’ve posted Entry #401 to my weekly Valuation-Informed Indexing column at the Value Walk site. It’s called The Stock Market Should Produce Gains of 6.5 Percent Real Every Year.
Juicy Excerpt: Economic booms and busts serve no good purpose. But are they not an inevitable part of our system? They certainly have been in the past. But isn’t it the purpose of economic research to teach us new things and thereby to make it possible for us to live better lives? Shiller published important economic research. He thereby made it possible for us to rein in the emotional impulses that have played a big role in causing booms (the creation of trillions in pretend money through the crediting of phony stock market gains on our portfolio statements). Why should we not believe that, if we learn the lessons his research points to, we can make a future very different and very much better than our past?
Why is it that, despite the huge bulls and bears of the past, the annual average return has always eventually reverted to 6.5 percent real? It’s because that’s the economic gain that our economy produces each year. Numbers on our portfolio statements indicating that the gain for the year was larger or smaller are temporary illusions. We can extinguish these illusions by thinking more clearly about the realities behind the numbers. When gains are higher than 6.5 percent real, we are borrowing from the future and, when gains are lower than 6.5 percent real, we are paying for borrowings made in earlier days. The real gain is always 6.5 percent real (at least that has always been the case this far into our history). The more we learn about how our market works, the more confidence we will have in that number and the less we will have in the numbers that appear on our portfolio statements and that are more the product of irrational exuberance or depression than of the underlying economic realities.
“It’s because that’s the economic gain that our economy produces each year.”
No it isn’t. Real GDP is the US has averaged between 2.5% and 3%.
The 6.5% real long term stock market return is a theoretical number. It is the return you would have received if you has reinvested all dividends throughout US stock market history.
A very small number of people may have reinvested every dividend they received in stocks but most did not. If all dividends had been reinvested in stocks that would have greatly increased the amount of money flowing in to stocks, driving up the price, lowering the dividend yield and hence lowering long term returns.
I agree that, if all dividends were reinvested in stocks, that would drive up the price of stocks and thereby lower the returns. The key is that we open up the internet to honest posting re the last 37 years of peer-reviewed research. So long as investors have access to good information, they will act in their self-interest and practice price discipline. So, when prices get too high and long-term returns drop to a point at which stocks no longer represent a sound long-term investing choice, people will sell stocks and thereby pull valuations down and long-term returns up.
If you don’t think that the long-term returns on stocks is 6.5 percent real, are you able to say what you think it is?
It seems to me that the choice of whether or not to reinvest dividends is up to the investor. Investors can earn 6.5 percent if they choose to and if they practice price discipline. Yes?
They also can make other choices and earn a lower return as a result. It’s up to them, you know? Good investment advisers should be encouraging people to reinvest dividends so long as stocks are available at reasonable prices (and not otherwise). I would go farther than that. If prices drop enough so that stocks are available at lower-than-fair-value prices, expert should advise people not only to reinvest their dividends but to increase their stock allocations until they reach an allocation at which their risk profile is where they determined it should be when they started out (stocks are obviously less risky when valuations are low and returns are high). Yes?
This is all pretty darn simple when you boil it down to the basics. Price discipline is the key to buying anything that can be purchased for money. Price discipline is what makes markets work. Stocks are a great asset so long as most investors are practicing price discipline. It is only when Buy-and-Hold strategies that disdain price discipline become popular that we get ourselves into trouble.
Price discipline is so important that, if it is part of your strategy, you can be unaware of most of what people talk about when they talk about stocks and still do very well. But if you fail to practice price discipline, you are virtually certain to suffer a wipe-out sooner or later even if you get everything else right.
I am a big believer in the importance of exercising price discipline when buying stocks, Evidence. I am a big believer in the importance of exercising price discipline when buying ANYTHING. But I feel particularly strong about the importance of exercising price discipline when buying stocks.
Does any of that help at all?
Price-Discipline-Advocate Rob
“If you don’t think that the long-term returns on stocks is 6.5 percent real, are you able to say what you think it is?”
I don’t think using the word “is” is appropriate.
The long term return on stocks (assuming fully re-invested dividends) “was” 6.5%. The actual return that investors achieved was somewhat lower (I don’t think I have ever seen such a figure published, it might be impossible to calculate)
What the future return on stocks “will be” is only guessable at this point.
I suspect that the real growth that investors achieved was around the same rate as GDP growth (minus costs). If stock market growth exceeded GDP growth then it would gradually make up a larger and larger percentage of the economy. I suspect that the long term future growth will be about the same (GDP growth minus costs)
Bill Bernstein addressed the subject of long term equity returns 20 years ago.
http://www.efficientfrontier.com/ef/198/returns.htm
Bill Bernstein addressed the subject of long term equity returns 20 years ago.
That’s a super article. Thanks much for posting the link.
Bernstein Lovin’ Rob
What the future return on stocks “will be” is only guessable at this point.
None of us has a crystal ball. That much is certainly fair to say.
We need to invest for our retirements, however. To do so responsibly, we need to form some sort of educated guess or estimate of the future return. We cannot achieve perfection in our execution of this task. But we have to give it our best shot. Otherwise, we cannot make intelligent choices re a very important aspect of all of our lives.
Say that the likely long-term average return on stocks really is 3 percent real. That pretty much tosses that 4 percent safe-withdrawal-rate concept into the trashcan, does it not? Stocks were priced at three times fair value at the top of the bubble. So you would have to divide the usual long-term return by three to identify the likely long-term return for people retiring at that time. Divide a three percent return by 3 and you a 1 percent likely long-term return. Is an asset class with a long-term return of 1 percent going to support a withdrawal rate of 4 percent for 30 years? It seems unlikely to me. Perhaps it could happen if the return each year were precisely 1 percent. But of course that’s not the way stock returns play out. If you were lucky and had better-than-average returns in the early years of the retirement, it might work. But if you had bad years up front, the thing would fail. Not safe.
And consider where things stood with Treasury Inflation-Protected Securities and IBonds at the top of the bubble. They were paying a risk-free 4 percent real at the time and that deal could be locked in for 30 years. So people had a choice between a risk-free asset class paying 4 percent real and a high-risk asset class paying 3 percent real if you don’t believe that valuations matter and 1 percent real if you believe that they do. Yowsa!
If stock market growth exceeded GDP growth then it would gradually make up a larger and larger percentage of the economy. I suspect that the long term future growth will be about the same (GDP growth minus costs)
This makes sense to me. I am not willing to sign on 100 percent to what you are saying, primarily because I have not seen this take explored by lots of other smart people and it is far enough removed from the conventional take (I am Mr. Conventional Take, as you know!) that I would need to see that before endorsing your position. But I think you have raised a super point. I would very much like to see it explored in greater depth.
If I had posting privileges at the Bogleheads Forum, I would put up a thread today asking people to share their thoughts re this one. I would like to know what Bogle thinks. And Shiller. And Swedroe. And Bernstein(providing a bit more detail that he supplied in the fine article that you linked to above). It’s a big deal. I am fairly confident that Bogle himself has referred to the 6.5 number. If that number is wrong, I would think he would want to know that.
I had a friend Brian who worked with me at Ernst & Young. He used a calculated that he found on the internet to tell him how big his retirement portfolio would be at age 65 monthly stock purchase numbers that he entered into it. The calculator assumed a 6.5 percent real return on stocks. I have always thought that that calculator was dangerous because it did not contain a valuation adjustment. The annual return for the past 18 years has been only 3.2 percent real. Brian’s portfolio is going to come up way short! (Actually, it won’t because he got a different job paying more money but that obviously won’t be the case for everyone who used that calculator.)
I wonder if it might be that the dividends portion of the stock return is reflecting the annual profits of the underlying companies while the increase in the value of the shares reflects an increase in the future profit-generating potential of the enterprise. My recollection is that that is how Bogle describes things. Bogle was the primary influence on my understanding of how stock investing works in the early days. Shiller became my primary influence only after Greaney put forward his first death threat on the evening of August 27, 2002, and 200 of my fellow community members endorsed it and I knew that the Buy-and-Hold thing was way too emotional a concept for me to follow or to recommend to others.
You are raising an interesting point. Your claims are pretty darn far out there, though, my Goon friend. I’d like to see what some smart people have to say about this idea before adding my endorsement. I would definitely like to see if you can get some good discussion going at the Bogleheads Forum. That would be super.
I certainly wish you the best of luck with it, in any event.
Highly Intrigued but Also Somewhat Skeptical Rob
“The annual return for the past 18 years has been only 3.2 percent real.”
That’s a shame. For the millions of people who put every last cent in the market at the peak of the internet bubble. Not one penny before, not one penny after. Yes, those people have it bad.
Let me know when you’re done cherry picking. I’d love a nice piece of pie.
18 years is not a small stretch of time, Anonymous. I picked a bad stretch of time. But when the stretch of time is 18 years in length, that’s not cherry picking as that term is generally understood.
And there’s good reason to believe that the next stretch of time will be a lot worse. We are today at the valuation levels that brought on the Great Depression. The Great Crash came after years of amazing returns. If a crash of equal size comes today (because today’s valuation levels are the same), it will be coming after 18 years of dramatically sub-par returns. That’s because the P/E10 level in 2000 was a lot higher than the P/E10 level that brought on the Great Depression.
There are all sorts of return scenarios that can play out. The lesson of history is that, so long as valuation levels are reasonable or better, all of the return sequences that pop up are positive and, when valuation levels get to crazy extreme highs, all of the return sequences that pop up are negative.
Overvaluation kills.
My sincere take.
Sweet As Cherry Pie Rob
“that’s not cherry picking as that term is generally understood.”
Granted there are many terms that are not generally understood the way you understand them.
So let’s check Wikipedia:
“Cherry picking, suppressing evidence, or the fallacy of incomplete evidence is the act of pointing to individual cases or data that seem to confirm a particular position while ignoring a significant portion of related cases or data that may contradict that position.”
Perfect example: Picking the worst possible time to invest and making an argument based on the ludicrous assumption that many people invested only at the one point in time.
18 years is not an individual case. It is a whole big bunch of cases strung together.
The entire historical record teaches us that stocks become a poor long-term investment class when prices rise to insanely high levels. Things have played out in the 18 years beginning in January 2000 just as you would expect if you had learned from the historical record how stock investing works in the real world.
If we see a 50 percent price crash in the next year or two or three, will it be another case of cherry picking if I point out that that will cause millions of failed retirements?
How far will you take this? If 150 years of historical data is not enough to make the point, what will be enough?
Wikipedia Checking Rob
“If we see a 50 percent price crash in the next year or two or three, will it be another case of cherry picking if I point out that that will cause millions of failed retirements?”
Yes, Cherry Picker Rob. You speak of that crash as if the world then stops, and that 50% loss is permanent. Anyone whose retirement is busted because of a 50% market crash is a victim of poor planning, not overvaluation.
A 50 percent loss in portfolio value is a big setback. That takes away many years of saving, in some cases it takes away decades of saving. If you have seen any of the charts that show how amounts saved compound over time, you know what a big deal it is to lose years or decades of saving effort.
It’s not that the world is going to stop. And it’s not that the 50 percent loss is permanent. It’s that avoiding 50 percent losses lets you achieve your goal of having enough to retire many years sooner and at greatly reduced risk. Given that Shiller provided us with the tool to know when those 50 percent losses are a likely event and when we don’t need to worry about them, why not put his powerful, Nobel-prize-winning insights to our advantage?
I personally do not see the downside, Anonymous. You should decide for you. It’s your money and so it is your call. But I think Bogle was right the first time. When he advised that we should use the peer-reviewed research as guidance on how to invest, I thought that made a lot of sense. So that’s what I do. And that’s what I advise others to do as well. I hope that’s okay by you.
True Boglehead Rob
“18 years is not an individual case. It is a whole big bunch of cases strung together.”
Not the way you use it. You are using an example of a single investment 18 years ago in the S&P500. Someone who invested every pay period in a diverse portfolio since 2000 would get a very different return than your example.
Some years better, some years worse. But the average annual return over the 18 years has been 3.2 percent.
Cherry-picking would be looking at people who invested prior to the 2008 crash and then sold at the bottom. That’s not a representative case. But showing the average annual return over an 18-year time-period covers a lot of cases.
You’re saying that the later years of the 18-year time-period are a bit better than the earlier years. That’s because the effect of the 2008 crash only hit those who invested in the earlier years. But the next crash will hit the people who invested after 2008 plenty hard. High valuations always exact a big price. You cannot know when the hit is coming. But you can know for sure that it is coming and that it will wipe out years or even decades of saving when it does.
I like having the probabilities on my side, Evidence. That’s the bottom line. When valuations change, the probabilities change. So I take valuations into consideration in every strategic question that I consider. I like to get good value for my stock buying dollar just as I do for my sweater buying dollar and my banana buying dollar and my haircut buying dollar. Sue me, you know?
Why do you think it makes you so angry that I don’t do what you do? It doesn’t make me angry that you do different than me. I don’t like you blocking people who have expressed a desire to hear what I have to say from being able to hear it. But it doesn’t bother me even a tiny bit that you walk to the bear of a different drummer. It’s your money. Why should it bother me how you go about investing it? But it sure does cause you a lot of concern coming the other way.
Value-Seeking Rob
“But showing the average annual return over an 18-year time-period covers a lot of cases.”
No it doesn’t. It covers a single investment made 18 years ago, a case that is almost certainly not representative of any real world investor.
Why didn’t you pick 20 or 25 years as your example?
Because it wouldn’t have produced the result you wanted.
Every year that you pick is going to produce a different number, Evidence.
If you want to avoid even a tiny hint of cherry picking, you would look at the entire 150-year time-period for which we have return data available to us. That’s what Shiller did. That’s what Pfau did. That’s what Russell did.
Guess what they found? That valuations affect long-term returns.
The same thing that is the case for 18 years is also the case for 150 years. Yes, you can find short time-periods when it is not the case. It was not the case from 1996 through 1999. If you want to go by that, you should go by that. Not this boy, you know. I am a long-term investor. In the long-term, valuations always make a big difference. So I take valuations into consideration when investing in stocks.
Long-Term Rob
“Guess what they found? That valuations affect long-term returns.”
I have known that for years.
Guess what else Wade Pfau found.
That trying to construct a market timing scheme based on that knowledge doesn’t work.
Here is my evidence for that claim
Here is the paper https://mpra.ub.uni-muenchen.de/29448/1/MPRA_paper_29448.pdf
9th page (numbered 8)
2nd row of data
Geometric Return for 100% S&P500 : 8.60
Geometric Return for MT rolling mean : 8.61
Geometric Return for MT rolling median : 8.59
rolling mean/median explanation
“More realistically, the “rolling mean” and “rolling median” values are calculated from 1881 to each subsequent year across the range of historical data.
8.61 and 8.59 as opposed to 8.60 for buy and hold S&P500, no significant difference for 130 years.
If you believe my analysis is incorrect please explain what I am doing wrong
Anyone who cares to know the realities can read the paper for himself or herself, Evidence. Wade offered a concise statement of what the entire historical record shows at the conclusion of the many months of research work that he put into the paper. He said”Yes, Virginia, Valuation-Informed Indexing works!” That tells the tale.
Another tale was told when you Goons (Buy-and-Holders one and all) responded by threatening to send defamatory e-mails to Wade’s employer in an effort to get him fired from his job if he continued doing honest work in this field.
Hey! Maybe Greaney really did include a valuations adjustment in his retirement study after all! You can never know for sure, right?
Felony Free (and Goon Enemy) Rob
You didn’t address the numbers I quoted.
Is my interpretation correct?
Buy and Hold 8.60%
Market Timing 8.59%/8.61%
If not, what are the correct numbers?
Anyone interested in knowing the realities should read the paper.
Then the person does not need to worry about whether your interpretation is correct or my interpretation is correct. He can form his own interpretation.
What a concept!
Co-Author Rob
What if that person asked the co-author of the paper a question, would the co-author answer the question?
I would answer a sincere question, Evidence.
Why do I have a funny feeling that your question will fail that test?
Sincere Rob
I have already asked a sincere question relating to the returns for the S&P500 versus Market timing contained in the paper.
Here it is again
“Is my interpretation correct?
Buy and Hold 8.60%
Market Timing 8.59%/8.61%
If not, what are the correct numbers?”
—
“Why do I have a funny feeling that your question will fail that test?”
Because you are unable to come up with an answer that supports your position and hence would rather not answer.
Okay.
Please take good care, in any event.
Stumped Rob
I apologize, I just realized my mistake.
I included numbers in my question.
As you have now retreated to classic hocomania I will stop contributing to this thread.
Fair enough.
Hang in there, man.
Manic Rob