Risk Revisited

[RobBennett]

Kevin M raised a point about risk last week in the thread titled “Is Buy-and-Hold Just a Marketing Pitch?” He said: “I'm not particularly enamored with any of the funds in my plan and often thought parking it in a bond or money market option was not risking enough to get a good reward. Obviously through the last few years I'm rethinking that.” In this thread-starter, I am going to try to show how those following the Buy-and-Hold Model (which is of course the dominant model today) look at risk very differently from those following the Valuation-Informed Indexing Model (which I favor).
I view the attitude toward risk that Kevin is describing (he is accurately describing the Buy-and-Hold approach) as exceedingly dangerous. In all other areas of life endeavor, we think of risk as something to be avoided. When it comes to investing, we think of risk as something to be sought out. I believe that this is why we are in an economic crisis today. We have taught millions of middle-class people to seek out risk and the result is that we have collectively taken on so much excessive risk that the losses have become big enough to crater the entire economy. We need to rethink this risk question!
The idea that you sometimes need to be willing to take on risk is of course correct. If you are not willing to take on any risk at all, you will not obtain returns big enough for you to achieve your investment goals. The proper approach, though (in my view!) is not to take on risk just for the sake of taking on risk. The proper approach is to take on risk only when sufficient compensation is provided for taking on the risk to justify the losses that may result. This is the critical step in the analytical process that the Buy-and-Holder fail to take.
The root problem is that Buy-and-Holders treat compensation for taking on stock risk as a static thing. Buy-and-Holders reject the idea that we can know in advance when stocks will provide good returns and when they will provide poor returns. So they use historical averages. Stocks on average provide a return of 6.5 percent real. So on average they provide sufficient compensation to justify taking on considerable risk. If the amount of risk associated with investing in stocks AT A PARTICULAR TIME really could not be known in advance, the entire Buy-and-Hold Model really would make sense. If the best we can know is the compensation for risk provided on average, that’s what we must use as our guide to making allocation decisions and the average compensation for risk provided by stocks is great enough for stocks to come off looking better than any other asset class. So it makes sense if this is so to always go with a high stock allocation.
But why do we believe that it is not possible to know the compensation for risk provided by stocks in particular circumstances? Why can’t we determine each year whether the risk associated with stocks is worth taking on at that particular time?
It turns out that there is no solid reason for believing this. Lots of smart people came to believe it for a time because they came to believe in the Efficient Market Theory. But there was never any hard evidence supporting the Efficient Market Theory. it was an HYPOTHESIS, nothing more. When it was tested in 1981 (by Yale Professor Robert Shiller), it failed the test. We are not bound for all time by any principles that follow from a belief in the Efficient Market Theory. We are free to consider whether the compensation for taking on stock risk is a variable thing, whether there are some times at which the likely return on stocks is great enough to justify the risk involved in investing in them and other times when it is not.
There is only one thing you need to know to perform the assessment -- the extent to which stocks are overvalued or undervalued at the time you are thinking of buying them (the approach I am describing only works with purchases of index funds, the prices of individual stocks are too influenced by the fortunes of the particular companies for valuation assessments to be highly meaningful). A regression analysis of the historical data tells you the extent to which valuations have always affected long-term returns (this does not work in the short-term because investor emotion is the dominant influence on stock prices in the short term and investor emotion is highly unpredictable) in the past. Knowing that, you can form a good assessment of what your return will be in 10 years. Once you know the return you will receive from stocks in 10 years, you can compare it to the return you will receive from super-safe asset classes and determine whether the risk of investing in stocks is at this particular time worth taking on and to what extent it is worth taking on (what your stock allocation should be).
The most dramatic example of the benefit is of course supplied by looking at the time when stocks were the most overpriced they have ever been. In January 2000, the range of possible annualized 10-year stock returns extended from a negative 7 percent to a positive 5 percent (the annualized return is the average return you would receive for each of 10 years running). Treasury Inflation-Protected Securities (TIPS), the safest asset class imaginable, were paying 4 percent real. You had to get an annualized return of considerably more than 4 percent real from stocks to justify taking on the risk of investing in them. But there was only a one in ten chance that stocks would pay more than TIPS over the next 10 years. Stocks were indeed more risky than TIPS. But there was no compensation being paid to investors for taking on this added risk (in fact, there was an equity risk penalty in place at the time). This is a case in which taking on the added risk was a bad idea.
The only objection that I know of that can be raised to the idea of using the historical data to know when the risk of investing in stocks is worth taking on is that we do not have enough data to make definitive pronouncements as to how much compensation is being paid to take on the risk of owning stocks at particular times. There is some weight to this objection. We do not have enough data to answer every possible question, and, even if we did, we do not have a strong enough understanding of how stock investing works to possess perfect knowledge as to how to make use of the data. We need to proceed with caution.
It must be kept in mind, however, that the same objection applies to all strategies recommended under the Buy-and-Hold approach. There are millions of investors investing in stocks today without taking into consideration the compensation for stock risk being paid in particular circumstances. The strategies developed under the Buy-and-Hold Model were developed with use of the same limited data set that was used to develop the Valuation-Informed Indexing Model. The same lack of perfect knowledge applies in both cases. We need to proceed with caution when applying Buy-and-Hold strategies too.
My belief is that we need to open a national debate on these questions. The main point here is that the Buy-and-Hold approach to investing is not the only viable approach. There was once a time when smart people thought that it was not possible to know how much the compensation for taking on stock risk changed from time to time. Today there are a lot of smart people who think it IS possible (this group is still a minority, to be sure). A vigorous debate would help both Buy-and-Holders and Valuation-Informed Indexers better understand their own positions and the positions of those following the other strategy.
I personally believe that Kevin is making a mistake in thinking that taking on more risk always leads to better returns. It could be that I am right, it could be that I am wrong. The only way for us as a society to find out is for us to hold lots of discussions of questions that many had once thought had been settled once and for all during the Buy-and-Hold Era.
If it is really possible to know in advance the compensation you are likely to obtain for taking on stock risk, it is possible for all of us to invest in the future far more effectively than we have ever invested in the past. We can obtain much higher returns at greatly reduced risk. I am excited about the possibility of participating in a national debate aimed at finding out one way or another for sure.
Rob

[Robert Muir]

The truth is out there, eh?
One thing that bothered me about Bernstein's "The Investor's Manifesto" was his discussion of the risk premium.
Bernstein starts the book talking about a discussion he had with a friend on whether investors at the time (during the tech bubble) were smart or dumb. If the market was overvalued, then they were dumb because they were not insisting on being compensated with a premium for the risk they were taking for investing in equities. If the market was trading at a new, higher, P/E level, then they were smart to invest and earn the dividends forthcoming.
He then goes into a discussion of what his friend meant about a risk premium and how it turned out the investors were dumb as the market soon after took a huge dive.
But then Bernstein spends the rest of the book showing how it was impossible to win by purchasing individual stocks or to time the market and that the best way to invest is to buy and hold in specific asset allocations. So, in other words, even as he said the market appeared to be priced way too high, still the best thing to do was to buy into it. He never states it, but I'm guessing that he believes that it is possible for the market to start trading at a new plateau.
So Rob, my understanding is that you've been out of the market since '96 and you don't plan on buying back in until the P/E is below 10 or so. I can understand that thinking. If historically, the market P/E is around 15-16, it doesn't seem logical to buy into it when it's trading over 20 since reversion to the mean may be imminent and it's better to buy in when it's lower.
However, what if all the money poured into 401k and IRA funds since the mid 90's has pushed the market up to a new mean? It could be another 20 years before we see a P/E of 10 or lower for the overall market, when Boomers start withdrawing their kit with more gusto.

[RobBennett]

"what if all the money poured into 401k and IRA funds since the mid 90's has pushed the market up to a new mean?"
Thanks for your question, Robert.
If you believe that that is what has happened, obviously you should invest pursuant to that belief. I am a flawed human like all the others so it is of course possible that my take is mistaken.
The problem that I have with the Buy-and-Hold approach is that the assumptions being used are unstated. Did the people who stayed with the same stock allocation when valuations got high really think it through logically and come to the conclusion you are suggesting here? Or do they think of staying at the same stock allocation as a neutral and safe position? I think that in many cases it is the latter explanation that applies and I see that as dangerous. That means that many people were going with high stock allocations not because of a personal belief that they would work out but just because they were lulled into a false sense of security that high stock allocations are okay at all times.
My recommendation is that people think this stuff through. If you really believe that a high stock allocation makes sense regardless of what valuations are, you will have the confidence to stick with your plan through bad times. If you were just lulled into thinking that there is no need to pay attention to valuations because sticking with the same allocation seems on first impression to be a neutral choice, you will abandon that position when times get hard. And no strategy will work unless you have enough confidence in it to stick with it through hard times.
Having confidence in your strategies is important. To have confidence, you need to think through this stuff. The most important decision an investor makes is his choice of stock allocation. Buy-and-Hold turns that critically important decision into an automatic one, one made without thought. That alarms me.
Rob

[jacob]

Academics do cause all sorts of damage don't they?
Main main problem with risk is when it's defined as volatility rather than uncertainty. Volatility is mathematically tractable but it is to risk what price is to value: not necessarily the same.
The risk-reward thinking has led to all this "in the long run" hoping. The problem with the long run is that it can be a really long time---certainly on the order of 30 years. I think if we hadn't just had a 20 year mega bull run, people's attitudes to stocks would be much different.

[RobBennett]

"what if all the money poured into 401k and IRA funds since the mid 90's has pushed the market up to a new mean?"
I'd like to try another, more tricky, response to this one.
There's a suggestion here that, when more people buy stocks, prices go up. There is a widespread belief that this is so. If it is, then demographics affects stock prices. As Robert is suggesting, it could be that we had the huge bull because the Baby Boomers were in their high-earning years and that we will have a horrible bear when they all retire.
I certainly don't reject that idea. But my primary belief is that our knowledge of how investing works is more primitive than most of us realize. So I think it is a good idea to question even our most fundamental beliefs about how this stuff works. I am not sure that it is so that having more people own stocks necessarily translates into higher prices.
The price of a stock is set by the people who bid for it. nothing else. This is one of the weird things about stock investing. We think of prices as reflecting realities. But in the case of stock prices, there is no requirement that any reality at all be reflected.
Say that as a people we decided that it was too hard to save enough for retirement. We could as a group bid stock prices up to two times their current value. There is not a thing in the world to stop us. Our portfolio statements would then indicate that our retirement plans were coming along just fine. We would of course be fooling ourselves. The newly created wealth would be cotton-candy nothingness fated to be blown away in the wind in a few years. But temporarily at least, we really can set prices at whatever level we want them to be.
The stock market is different from every other market in this regard. In all other markets, there is a reality to the price set. For example, in the grocery market, there is a reality to the price assigned to bananas because if a silly price is set, the owner of the grocery store is penalized by having to honor the silly price. But in the stock market there is no immediate penalty for setting a silly price. So I think we need to think about stock prices differently than we think about the prices set in any other type of market.
Now -- does the fact that there are more people at an age at which people like to buy stocks necessarily mean that stock prices must go up? I don't think so. There is not a limited number of shares that everyone must fight to own. Stock shares are just pieces of paper. The only thing that matters is the value assigned to those pieces of paper, not how many of them exist.
Say that the demand for stocks increased by 10 times. Does it follow that stock prices would go up dramatically? I don't believe it necessarily follows (I don't deny that we might see this effect, only that we must see it). If for some reason all the people owning stocks had decided at the time that low prices were appropriate, low prices would prevail, regardless of how many people were participating in the setting of the price.
What I am saying is that it is investor emotion that is the dominant effect on setting stock prices in the short term, not any of the economic factors (such as demographics) that we usually look to to explain stock prices. I think that we like to find economic justifications for stock prices because it scares us to think that prices are just "made up." It is indeed sort of scary to think this way! So I understand the concern. Still, I think it is best to face realities and it really does seem to me that investors possess the power to temporarily set stock prices wherever they please in defiance of all economic realities. If we shy away from this truth (presuming that I am not mistaken and that it really is a truth), we hurt ourselves by doing so.
Of course it is also possible that demographic changes might affect investor emotions and thereby might affect stock prices even though there is no necessary connection between demographics and stock prices.
I hope that no one gets the idea that I am offering definitive takes here. My aim is just to throw out ideas that perhaps have not been considered too often in the past. I certainly am no investing expert. I just question whether a lot of the people who we think of as investing experts are all that much more expert than us common folk. Robert's story about Bill Bernstein is a great illustration of what I am getting at. Bernstein is very smart; I have learned a lot from him. But I catch him in logical inconsistencies ALL THE TIME. I very much wish that he would make an effort to be more humble in the way he states his views.
Rob

[RobBennett]

"The problem with the long run is that it can be a really long time---certainly on the order of 30 years."
Yes. For what it's worth, I believe that Buy-and-Hold works well if you go 30 years out. If you look at my calculator (The Stock-Return Predictor), you will see that it gives good 30-year results for those going with high stock allocations REGARDLESS OF THE VALUATION LEVELS THAT APPLY WHEN THE SHARES ARE PURCHASED.
The problem with going with 30-year numbers is -- What percentage of the population following the strategy will be able to hold for 30 years? The P/E10 level went to 33 in the late 1920s and we saw a price drop on 90 percent nominal/80 percent real in the years that followed. We left a P/E10 value of 33 in the dust this time, stopping only at a P/E10 level of 44. So we should be expecting something worse this go-around. How many middle-class people can take a loss of 80 percent of their accumulated wealth of a lifetime and not sell any shares?
Buy-and-Hold works in a hyper-technical theoretical sense. If you really did hold for 30 years, you really would get the results advertised. But has anyone every identified one living soul who actually pulled this off through a complete secular bull/secular bear cycle? I have never seen a single name put forward. So I am highly skeptical. If you sell even a percentage of your holdings in the down cycle, none of the numbers used in Buy-and-Hold analyses apply. I think it is entirely possible that most Buy-and-Holders will end up seeing a real-world long-term return of only about 3 percent real rather than the 6 percent real that they are anticipating.
My view is that the best model is the one that does the most to limit price swings. The ideal would be if stocks provided that 6 percent real return each year, rather than sometimes providing 30 percent plus and other times providing 25 percent negative. I believe that the key to getting to that ideal state is taking the emotion out of stock investing (to the extent possible). This is why I favor telling people the valuation-adjusted numbers. If we had told people in 2000 that their portfolios had a real, lasting value of only one-third of the numbers on their portfolio statements, they would have been far less excited about stocks at the time and far less disappointed in them today.
It is my view that Buy-and-Hold caused the bull market and that the bull market caused the bear market and that the bear market caused the economic crisis. I believe that if we followed a practice of always providing valuation-adjusted numbers as well as nominal numbers, we could make stock prices far more stable than they have been in the past and thereby make the entire economy far more stable too.
Rob

[Kevin M]

Lookie here, I spawned another thread.
"I personally believe that Kevin is making a mistake in thinking that taking on more risk always leads to better returns."
I agree - hence my questioning and reflection of what I should be investing in right now. In the past listened too much to the "experts" that told me I was young enough and should be invested in stocks. I could make up the losses over the long-term.
I think what Jacob said about volatility is how most people view risk. That you have to live through the volatility of stocks to get a good return.
Not to say I think stocks are a bad investment right now, jus that I'm not sure I want all my 401(k) money tied up there. The problem is the lack of other options in our plan - a few bond funds and a money market fund.

[RobBennett]

"In the past listened too much to the "experts" that told me I was young enough and should be invested in stocks. I could make up the losses over the long-term."
Kevin, I hope you didn't take anything I said as being any sort of personal criticism. I certainly did not mean it that way. I don't believe that any of us fully understand investing today. I definitely include Rob Bennett in that group. I include John Bogle in that group. I even include my hero Robert Shiller in that group.
I'm not just saying that either. I probably would say that Shiller is the best informed person out there today, but I indeed could point to things that I think he has gotten wrong if put to the challenge. And I have included the words "I Was Wrong" in several of my podcast titles when I discovered something that I had said at an earlier time and had over time come to question.
The one good thing you can say about us all being at a primitive level of understanding is that we can make up a lot of ground fast if we put our minds to it!
The stuff that they tell young people really bugs me, Kevin. I am of the view that we are all being hurt but that young people are being hurt the most. I recently wrote an article making this precise point titled "Young Investors Taking Hardest Hit":
http://deathby1000papercuts.com/2010/08 ... rdest-hit/
Juicy Excerpt: Older investors at least saw some benefit from the insane bull market while it was going on. The younger investor is experiencing the downside of that insanity without having had the opportunity to enjoy the upside. Truth be told, this young investor’s chances of enjoying compounding returns starting at age 35 are not at all good. We are still at high valuation levels. We have fallen from insanely high levels of overvaluation down to merely dangerously high levels of overvaluation. But it could take another five or ten years (depending on how soon we face up to the realities) for us to pay back the $12 trillion investing debt we incurred in the late 1990s. A woman who did the right and rare thing by starting to save at age 25 may not see any significant compounding magic until she turns 45!
I point out at the end of the article that about the only thing every age-35 investor has going for him is that every age-55 investor would trade places with him in 10 seconds if he could!
Rob

[Kevin M]

"Kevin, I hope you didn't take anything I said as being any sort of personal criticism."
Not at all. I'm happy to talk about my experiences if other people can learn something from them. I'm 35, and shudder to look back at how my 401(k) has performed since I entered the workforce 10+ years ago. My best guess - the only growth I've seen is in my monthly contributions. I'm not blaming anyone for this, and like you said the 55 year old would gladly change places, just pointing it out so people question what they are told.

[Robert Muir]

"I am not sure that it is so that having more people own stocks necessarily translates into higher prices."
That's the thing, *more* people don't own stocks. New issues and bankruptcies notwithstanding, the number of stocks is a finite number and the more people *wanting* the stocks is what drives the price higher.
With company pensions disappearing and workers sent to 401k and IRA plans for retirement planning, the upward pressure on stock prices has predictably grown stronger. The fear is that when the boomers start selling, it will either cause the longest bear market in history or (as they freak out and start dumping their investments) the greatest crash.

[RobBennett]

"Not at all. "
That's great. I only referred to you in the thread-starter because I thought your comment in the earlier thread was so on point.
There are millions of people asking the sorts of questions that you are asking today. I wanted to point to something real to get things going.
"I'm not blaming anyone for this"
My personal thought is that it would not be such a bad thing if some of us started blaming some people to a limited extent. I certainly don't think it is a good idea to become mean-spirited or uncharitable. But I also think that there are responsibilities that come with putting one's self forward as an investing "expert" and I question whether some have entirely met those responsibilities.
I'll give a concrete example. Jonathan Clements was for many years the personal finance columnist for the Wall Street Journal. He wrote a column after the crash saying that he didn't think that he has anything to apologize re the investing advice he had been putting forward for years. Some readers had obviously been suggesting that, given what had happened to stocks, he should be apologizing.
I liked Jonathan's column. I exchanged a few e-mails with him one time and he struck me as a perfectly nice guy. So I certainly do not want to screw him to the wall or anything like that. But I think it would be WONDERFUL if he would apologize. Perhaps he cannot say that he got things wrong. I don't think he really believes that at this time. But he could say that he apologizes for coming across as too sure of himself re things that none of us can really be entirely sure of at this point.
That would be a big deal. Once a few big names do that, I believe that the floodgates are going to open and we are going to hear all sorts of new ideas on all sorts of investing questions that a lot of people have been holding back sharing on for a long time now. People have been holding back because they don't want to offend those giving the conventional take. But to advance together we need to get some new ideas on the table.
I think that this is what this economic crisis is all about. Stock prices can return to fair value in two ways: (1) by investors learning that the value proposition is no longer there and selling; or (2) through price crashes. We forced Option Two and that was a huge mistake. Now we need to figure out together how to swear off Option Two and get the job done through the use of Option One in the future.
There have to be some public expressions of humility to get us there. So to some extent we need to start "blaming" those who have offered bad advice. Again, not with the purpose of making them feel bad. The purpose is to get them to a point where they can offer much better advice in the future, which in the long run helps them as well as all the people who listen to their advice.
I believe that we are working our way through an elaborate social dance. How do we get to the point where it becomes possible to say openly that we all goofed up? I believe that once we get that part of the job out of the way, the rest is going to come easy.
Rob

[Robert Muir]

@Rob "Stock prices can return to fair value in two ways"
Not to pick nits, but there is really only one way for stocks to return to fair value and that's for enough people to believe they are currently overvalued and sell, with no takers until they are priced at what *you* believe is a fair value, at which point you could then buy. Of course, given the nature of herd behavior, to reach that level from the current level would be the definition of a crash.