This is sort of a followup to my last post; it’s just another thought that came to mind as I was typing up a response to someone in the comment section, and I thought I’d expand the thought into a quick post as I think it’s pretty important. (Blog readers take note: This is my Irving Fisher moment!)
Known Unknowns and Unknown Unknowns
Donald Rumsfeld said a long time ago:
There are known knowns. These are things we know that we know. There are known unknowns. That is to say, there are things that we know we don’t know. But there are also unknown unknowns. There are things we don’t know we don’t know.
And also Howard Marks said, it’s what you think you know that just ain’t so that is going to kill you. Maybe this would go into Rumsfeld’s “unknown, unknown”.
But OK. Enough of that. Where am I going with this?
I think a lot of people get caught up with the fact that the stock market is way overvalued. Buffett (and Tepper and some others) has said many times that the market is not bubbled up and is in a zone of reasonableness.
Many people put up Shiller’s CAPE chart to show how egregiously overvalued the stock market is. However, I have pointed out many times here that sometimes those macro, big picture charts can be very misleading. People look at charts like that and conclude that the market must go down. However, that is a total fallacy.
For example, for a very long time stock dividend yields were higher than bond yields because stocks were viewed as speculative.
Look at this chart:
(data from Shiller’s website: http://www.econ.yale.edu/~shiller/data.htm)
*for CAPE, I used the inverse so it can be compared to bond and dividend yields.
The data goes back to before 1900, but I couldn’t convert the dates from before 1900 in Excel (I used to be good at spreadsheets, but now seem to have forgotten a lot! Or they changed too much.), so we will just deal with post 1900 data. It shouldn’t really change anything. Oh, and I wasn’t able to put the entire time series in one chart so I divided it up into two sections; 1900 – 1970 and 1970 – 2014.
As you notice, throughout history, dividend yields were higher (and often much higher) than bond yields. But this flipped over in the 1950’s. This is around the time that Buffett started his partnership, and both his father and Benjamin Graham warned him against getting into the stock market.
I don’t know if this yield flip had anything to do with it, but you can imagine all the pundits warning people against investing in speculative stocks with dividend yields lower than bond yields and therefore not compensating investors for the additional risk taken. Since the 1800’s, dividend yields were always higher than bond yields.
Anyone who held that view would have been out of the market for the rest of the century, and only recently would have been able to get back in.
So, caution number one: The past can be a great guide in assessing the market, but clinging too much to one indicator, however effective and foolproof it looks historically, can be dangerous.
Dividend yield itself has been used for a long time as an indication of an overvalued market. I remember hearing as long ago as the early 1990’s that dividend yields of 3% (or below 4%) is just way too low. When it went under 2%, it was ridiculous. Some said it should be more like 4-5%. Well, if you needed 4-5% dividend yields, you would have been out of the market (or short) since the early to mid-1980’s.
This is the chart for 1970 – 2014:
Now look at this. We keep hearing how CAPE is way out of bounds and we are going to have a serious correction. OK. Maybe.
But look really carefully here at the chart back in the late 1980’s. One reason the stock market crashed in 1987 was because the stock and bond markets diverged. Look how the stock market earnings yield plunged lower as bond yields spiked. This discrepancy was corrected severely in a single day. Someone showed me an X-Y scatterplot of what happened (after the fact; I wasn’t in the business back then) and how neatly earnings yield tracked bond yields and how they diverged in 1986 and 1987, and then how it snapped back.
Today, if you look at the chart we are in the opposite situation. Bond yields continue to plunge lower, but it seems like CAPE is sort of stuck where it was for much of the 2000’s.
Of course, I would not go so far as to say that the market is undervalued and should actually be trading at a 50x p/e ratio.
But I don’t see a need for an imminent, harsh correction either. Looking at the above chart, it just looks like the market is floating around in a range more or less similar to where it has been since the late 1980’s.
Known Known
So, here’s a known, known. We know that the market is not cheap, and maybe expensive on a historical basis by certain measures. But the only thing we can conclude for sure from that is that future returns from a higher price will lead to lower returns.
A guarantee of lower return does not mean that the market has to go down.
For example, I still remember when bond yields went under 6%. That was nuts to people who started in the business in the 1970’s and 1980’s. The U.S. government was constantly devaluing the dollar and overspending. There was no way that the hockey stick (deficit) can be stopped, so there is no way that a 6% bond yield was going to be sustainable. When it went under 5%, it was a joke.
If you were a bond investor back then, maybe 6% bond yields were not enough for you. Then you could have gotten out and stayed in cash. Would you have done better? Or maybe you could have been convinced the bond market had to collapse. So maybe you would have shorted the bond market. Or maybe you would have bought tons of puts against your long bonds. Either way, you would probably have gotten a lower return than the 6% (actually, way more than 6% due to the bond rally since then) that was way too low for you.
You can make the same argument at 5%, then 4%, 3% etc…
People have been calling the top of the bond market for as long as I can remember. (Japanese bond market too!).
So, caution number two: Just because something is high or low doesn’t mean it has to reverse imminently. Some cycles are very, very long and can last for decades (bond bear market from 1940 to 1982, bull from 1982 and ongoing etc…)
Known Unknowns
Most of us have enough humility to admit that we really have no idea what will happen to the stock market, bond market, economy etc. But even then, many assume that due to the low interest rates (or high valuation), that bond prices must go down. Stock prices must go down.
Most admit that they don’t know when it will go down, but they are convinced it will go down soon enough to make their decision to be short or stay out a good one (this is key!).
Unknown Uknowns
But to me, that’s sort of the problem. People assume that just because bond yields are low, that they must necessarily go up. That stock prices are high, so they must necessarily go down. Yes, eventually they will. These things go up and down over time.
But if you really take a good, hard look at the charts above, those things can stay around this level for a long, long time.
Just as bond yields have been lower for a lot longer than anyone has ever expected, the CAPE can stay here for a long time too (as it does seem to follow bond yields to some extent).
If you are short the stock market (either by actually being short, or by having low exposure despite an equity mandate), you are simply betting on a rise in bond yields. So you are making a bet based on an economic forecast, basically.
What You Think You Know But Just Ain’t So
So if you think the market must go down because it’s expensive, maybe that just ain’t so! Again, I go back to bonds. The bond market doesn’t care if a bond investor is unhappy with a 3% yield or a 4% yield (Of course, there are some like Bill Gross that will call interest rates and make money).
Bears will argue that if you own stocks here, then you are basically betting that interest rates won’t go up. Well, not necessarily. Buffett has been saying rates are way too low and that the bond market is the biggest bubble ever, but has been buying stocks pretty aggressively in recent years anyway.
Conclusion
So anyway, I just wanted to point out that just because bond yields are too low for your taste (stock market CAPE is too high), it doesn’t necessarily follow that they must go UP (or CAPE go down). OK, so maybe they must go up eventually. But these cycles seem to be so long that it seems a bit silly to try to call the turn. (And no, that doesn’t make me a perma-bull. Call me a perma-agnostic! The market will keep going up and down as it always has; there will be plenty of brutal bear markets to come regardless of all of this stuff).
The only thing we know for sure when rates are at 4% rather than 5% is that the returns will (most likely) be lower. At 2%, we can know for sure that returns will probably not be high. But we can’t know for certain that rates will go up. And again, more specifically, we can’t know for certain that it will go up soon enough for timers to benefit from it.
Great post! Thanks for the super long range data. .. this is actually something I’ve been thinking about – is there any way to isolate and hedge equity risk premium ? (shorting S&P wont do b/c it involves not just risk premium but also growth prospects..). I know this is somewhat contrary to your discussion about hedging in your last post, but here’s what I mean-
The questions of rates and market premiums are not confined to macro investors, but also fundamental single name investors as well.. a typical stock investment thesis might look like this: "Stock trade at $15, fwd EPS = $1, so 15x PE. (insert various qualitative/fundamental reasons here) I think EPS will grow at 15% CAGR next 2 years, slap on the same 15x PE for $1.3 gets me $20 per share, so it’s a buy". Maybe you do a sensitivity table for your exit multiple..regardless, you're taking a big risk on that forward multiple.
But how do we know that in 3 years (or 5 years for that matter) ,the forward multiple will be 15x and not 10x? the forward multiple depend on 4 things:
1) the company’s growth outlook when you exit (so 2018-2022E earnings outlook if you’re looking to exit in 2017E)..
2) the company’s idiosyncratic risk premium (industry stability, market position..etc)..
3) rates
4) “equity risk premium”
The first 2 are the advantage of single name value investing, since you can be selective and only pick those stocks where fundamental/industry outlooks are reasonably visible. #3) rates you can theoretically hedge, not precisely but at least directionally. #4 (ERM) though, I’m not aware of anyway to isolate and hedge that – you can’t do it by simply shorting S&P b/c that’s dependent on market’s growth outlook as well (so that you’re not isolating ERM)
It seems to me if one of the bulge bracket banks can come out with a way of isolating /hedging ERM it would in fact do a big favor to value investors, by allowing them to focus on the single name fundamentals.. maybe it exists somewhere already and I just dont know..
sorry for the rambling but your post is very relevant to what i've been struggling with the past few days.. thanks
Hi,
That's an interesting question. I have never heard of any way to hedge equity risk premium, and there isn't even an accurate way to measure it either. People use earnings yield versus bond yield as a proxy, though.
As for the uncertainty of a future multiple, you just have to be conservative, I suppose. Some people use a 14x p/e multiple, for example, for a mature company that may continue to grow with inflation and gdp growth because that's sort of the average p/e ratio of the whole stock market over the past 100 years or so.
When you buy a stock, you sort of have to figure out what the company might be earning 3-5 years out and then put a reasonable multiple on it. Of course, you can't really know what the multiple will be exactly. But you can more or less make a reasonable estimate.
For example, for something like KO, you can assume a p/e ratio 20x. It may be higher or lower depending on many factors like the whole stock market value, interest rates, KO fundamentals etc.
But if you think KO is doing fine and will more or less chug along as it has, then it's reasonable to assume that in a "normal" environment, KO will be at 20x p/e. Of course, in a bear market, that might be 15x. Or in a severe one, maybe 10x. In a bubble, it may go up to 30x or 40x.
But what's important is, I think, the reasonable assumption that in the next five years, what KO would be trading at. It's not so important, I don't think, to depend on the environment being normal in exactly five years as the economy will cycle up and down. If things go down hard, they will eventually normalize.
So that's the way I would think about it.
If you try to hedge this or that out, I tend to think that the errors caused there would be far more damaging than the risk you can eliminate.
And besides, you are more likely to be wrong on the big decisions (pick wrong stock, sell out in a panic at the low etc.) rather than in the smaller factors (which arguable can be large over the short term).
Over time, errors in risk premium estimation would probably end up evening out so as not to be such a big factor.
That's my view, anyway.
But yes, I'm sure derivatives traders can make money if they can create an equity risk premium swap or hedge of some sort. My guess is that there wouldn't be that much interest and not a lot of liquidity (so therefore not too useful). Plus you would have to define it contractually (what exactly is the equity risk premium?).
Thanks for dropping by.
Thanks for posting! When comparing higher dividend yields in the past with todays below 2 percent, please consider that buybacks were much larger in the past 20+, years, today in fact maybe similar or above dividends, so perhaps it would be better to look at the total cash return, or just at earnings yield alone?
Yes, I know. Thanks for mentioning that. The point is that for various reasons these valuation metrics evolve over time. DY became less important due to share buybacks. Today's high-looking CAPE may largely be due to low interest rates etc…
But moving to what we can know something about, are you finding lots of cheap stocks these days? I'm sure not…
It strikes me as odd that you continue to put out articles pushing back at the minute portion of investors who are concerned that we are six years into a bull market and at some of the highest valuations (lowest expected returns) seen in the past 130 years!
It doesn't seem like this blog was around in 2008 – but how is this different than putting out articles in Dec 2008 saying "don't be too bullish people, I know the 10Y is at 2%, and CAPE expected returns are over 10%, but there where MANY times prior to 1950 when the spread was much bigger. This market could EASILY go lower from here."
For the people out there who claim "just look for good businesses at fair prices" – what's a fair IRR? If you can find a bunch of stocks where the multiple is fair and you think they'll grow EPS 10% a year – do you load up on those or do you save 20-50% of your capital for the chance that this state of cheery consensus may not last and you can buy 15-20% IRRs in a year or two? Of course, the broader market is far far worse than either of these scenarios.
In 2008 the 10Y was at 2% and most people weren't telling me to load up on stocks. Back then stocks were risky and bonds were safe so of course there should be a big spread! Now, basically everyone is telling me stocks should be priced in line with bonds (including you!). Just look to Japan – stocks are priced in line with bonds there…..well forget them Japanese ROEs suck…clearly in Germany stocks are priced in line with bonds….well forget them too its Europe. That basically leaves the U.S. The land of opportunity where rates are low, growth is great, the dollar is up, and we can all get rich together! Amazing what can happen in six years.
I'm not saying it's impossible for stocks to go to 50x. If you are at the blackjack table and you hit on 19 it will look good a small % of the time. But most of the time you'll lose….just like buying at these valuation levels IMO. Now is the time to be fearful, not greedy.
Fearful doesn't mean 0% in stocks (and by god not short the market), but it does mean underinvested for whatever normal is for each investor.
Is CFX a buy here? Outsiders for 18x and 15% annual EPS growth?
At the end of the day I am basically in the same camp as you. Although I think you can forecast the market LONG TERM, there will always be stocks within the market that are much better so we should just focus on those. But to the extent that you are going to comment on the market, why you put out these articles essentially agreeing with the consensus bullish view – at these valuation levels, six years into the bull – is a mystery to me.
Now you may say I'm distorting what you said – you are perma-agnostic. One may not be able to call a turn, but as Howard Marks has said one can see when the pendulum has swung pretty far in one direction and adjust positioning accordingly.
You may counter that Marks is currently saying "move forward but with caution." Not sure I understand this quote but I agree the S&P isn't as bad as it was in 2000, it may be as bad as 2007, but I think the median stock in the US is as bad as it has ever been. But even just looking at the S&P…why is not being the worst ever a reason to be optimistic? Even if it is the 3rd worst market in the past 40 years – isn't that a good reason to be underinvested?
I interpret Mark's comments as "things are bad, but if you find awesome bargains you should buy them." Thanks Howard.
Hi,
Good point and thanks for asking. I know I do seem to go out of my way with this stuff. And there is a reason. It's because when I talk to people, I seem to spend MOST of my time talking people out of worrying about this, that or the other thing.
And yes, I said the same thing I am saying now in 2007 and 2008. But I also did say that if a 50% down market is going to be a problem for you, then sell down to a level that a 50% decline isn't going to give you an ulcer. But do NOT try to lighten up now and then load up again after the storm passes. I warned many that this is not going to work even though it looks easy (it looks like a no-brainer to sell stocks at 28x p/e and then buy back at 8x p/e (you woulda sold in 1929, bought in 1932, sold in 1969, bought in 1982, sold in 1999, bought back in 2002 etc..).
But it is actually never as easy in real time going forward. And I have heard enough horror stories about people getting out at the wrong time and never getting back in etc…
So what I am saying now is basically no different than what I was saying in 2007/2008. The difference is that I did tell people to look at their holdings very carefully and make sure you own businesses that you will want to own even in a 50% bear market.
This is true now too. If you don't want to own in through a recession, then you shouldn't own it at all. If you want to own it in good times and want to get out before bad times, that's just a bad idea that doesn't work in practice.
What is a fair IRR for investors? I don't know. I tend to still look at 10% as a benchmark, but this will vary. Will the U.S. stock market return 10%/year going forward from this price level? Probably not. Does that mean you should get out of stocks? Not necessarily.
Should you stay out and wait for a chance to get in at higher IRR's? That's also a reasonable question.
If you had dinner with Munger or Buffett and asked that question, I bet they would tell you that it's not such a great idea to do that. You can't really know when that sort of opportunity will come.
As an extreme example, I have no MANY people who are convinced that a bear market can't end until p/e's go down to 7-8x. Why? Because that's how low they went in every other bear market in that past 100 years. It got there in 1932, 1972 and 1982.
So they were convinced that no bear market can end unless p/e got there. I read old newletters and reports from the 1980's after black monday and many missed the bottom because p/e didn't get down to 8x. They also missed the bottom in 1990, 1992, 1994, 1997 and 2002, I think. I heard the same arguments in 2009; that the bear won't end until the p/e gets to 7x.
It's a tough question. I know. If you want to wait for a better opportunity, maybe it's a good idea. Maybe not.
As for Buffett, you can see that he has really done well over the past 50 years and a lot of his major purchases have NOT been at major bear market bottoms or even at super-cheap prices. Many stocks were bought at reasonable to fair prices in not-so-cheap stock markets!
I do like CFX. But for companies like CFX and CBI (which BRK likes), the oil price collapse is sort of a monkey-wrench as capex will plunge in the energy sector. There's no getting around that. So you will have to look 'through the valley' and have conviction that management will manage through this and come out the other end kicking some ass. Or you can try to time in and out of it which some astute traders might be able to do. But for most investors, I think it's probably smarter just to sit through it.
Anyway, this is all very interesting to me. I spend time on it because I do think all the time what is wrong with my argument. I don't post here to change people's minds or to try to win arguments. I post precisely because I like to think about it, discuss it, and have my thoughts challenged, so thanks for commenting!
Sorry if my last post was a bit chippy. If anything it’s because you’ve set my expectations so high. I think this is one of the best blogs out there.
However, you have failed in your quest to prevent ulcers because I almost had one when I read the title ;)! Overvalued Market!?: YES, by almost every measure.
But as you astutely point out just because some other security is expensive (S&P) doesn’t mean anything for what you should do regarding your virtually unrelated securities.
However, I think the right tack isn’t to warn people (or yourself) that it’s tough to pick a bottom, and that just because it’s not 7x earnings doesn’t mean this couldn’t be the low (so many negatives….but all necessary!). Quite the opposite, I think the right message today is don’t be complacent just because US stocks aren’t the absolute most expensive ever and just because they don’t trade at parity with a risk-free security. I think it’s safe to say investors don’t have to worry about picking the bottom right now!
I absolutely agree that you should not wait to buy until after the storm passes. You should buy when the storm is bearing down upon you and about to swallow you whole! While Buffett’s buys may not be correlated with storms in the overall market – they are absolutely correlated with storms in the particular security he is buying.
And that is the message, consider selling securities you own where everything is pleasant and serene (CFX over $70?, POST over $50?) and moving that capital to places where the hurricane rages. Where are these places today? RUSSIA!? GOLD STOCKS!?
It’s laughable to even look at what I just wrote. In Russia companies are being stolen as we speak. No one is saying buy Russia right now – even Howard is too frightened. Gold Mining is the anti-buffett business, capital intensive based on a commodity that no one can predict and never gets used up. I’m not saying buy everything here, but it’s the place to look – maybe there is a security that has been tossed out with the bathwater. It never feels good to buy in the storm – which is exactly why people were selling in late 2008 – they just can’t stand the storm no matter how much they think they can when the sun shines.
Even if everything I just wrote is BS – which is likely – I feel 100% confident saying there is no storm at all in the S&P 500 or Russell 2000 right now.
PS: In CFX the storm is clearly coming. Heavily weighted to O&G capex AND heavy exposure to EM where currencies (and economies) are tanking. They seem destined to take an earnings hit. Market knows this and stock has traded down. Trailing earnings now appear to be a short term peak. No one knows how far earnings will fall, but is there really a big margin of safety at 18x peak?
There is certainly upside. If oil runs back up and margins expand and EPS is growing 15% on top of those bigger margins you are now at a price where you really profit in that environment. But that is a bullish environment! What about the environment where we get a real multi-year washout in capex and EMs and earnings go down 30% from here and grow at 15% from that lower base level? Right now you are paying 26x that number. BDT hasn’t been buying yet. The CEO finally bought a whopping $90k in October….why aren’t these dudes loading up right now? In any case, in the next two years I think there will be a very good opportunity to buy CFX. I’m not confident the price is “fair” yet.
Yes, I understand your point. And yes, you don't have to wait for 7x p/e, but you don't have to worry about holdings at the current level, either. That's sort of my point. When Buffett bought one of his biggest winners, it was in 1988, I think, when the market to many was still expensive, and KO was not cheap at all by conventional measures. It was a 'fair' price. In that sense, there are a lot of stocks in that 'fair' range of 15-20x. Yes, 15-20x p/e is not conventional value manager cheap, but still fair and reasonable, I think.
kk, what I love about you (among other things) is that your responses to comments as just about as interesting as your posts themselves.
Honestly, the piece missing from this article, is look at the past data for the US, and other comparable countries. (Are there any? I’d suggest Canada, and perhaps the UK for starters.) Now of the past data, how much represents a time like that in the US over the past few years where there is bona fide financial repression going on by the central bank? My guess is none. Food for thought.