Short But Sour
A slow Monday, and the S&P could barely manage a +0.3% rally. That’s tantamount to a selloff, these days, as the all-time nominal highs are a mere 20 points away. I’m not joking: with stocks up 9.1% on the year, the S&P is averaging +0.19% per day, which means the all-time highs ought to be reached by next Wednesday.
Meanwhile, our measure of valuation for equities has reached levels not seen since July of 2011. The expected compounded after-inflation return for the S&P 500, inclusive of dividends, is just 2.00% (it got to 1.81% in July 2011 – and, for the record, it stood at 0.83% at the end of 2006).
The VIX tumbled today to the lowest level since April of 2007 (see chart, source Bloomberg), two weeks after Fed Chairman Bernanke told Congress that the “subprime crisis” was likely to stay “contained” (which it did, in roughly the same sense that the universe itself has a boundary).
Now, I don’t want to follow the usual course and list all of the things we could be worrying about (Italy, Cyprus, France, Iran, North Korea…) to somehow argue that prices are too high. After all, there’s always something to worry about. No, that’s not my argument at all. My argument is that prices are too high regardless of what the news is.
Over the next ten years, compounded real returns after inflation will likely be in the neighborhood of 2% per annum. They could reach 5% per annum, but they could be -3% per annum with equal probability. Note that these are real returns I am speaking of, so there is no reason stocks can’t continue to reach new nominal highs especially if consumer prices continue to accelerate.
(And here’s an odd fact: while equity market volumes over the last few years have been shrinking persistently, the gap between 2012 and 2013 has been narrowing over the last month and a half. That is, volumes are still running about 78mm shares/day below the year-to-date pace in 2012, but at the end of January that figure was 113mm shares/day. So volume is still shrinking, but no longer monotonically.)
So I’m not sure when we are going to get a significant correction on the order of late 2011 (~20%), but we are overdue. Frankly, if I thought the correction was likely to be no more than 20%, I probably wouldn’t even be particularly concerned, because 10% and 20% corrections happen in healthy markets. But I can’t discount the possibility of a 2000-2002 or 2007-2008 sort of decline. The conditions are “right” for just such an occurrence, unfortunately.
Leave a Reply to Michael AshtonCancel reply
>>Bernanke told Congress that the “subprime crisis” was likely to stay “contained” (which it did, in roughly the same sense that the universe itself has a boundary)…<>…I can’t discount the possibility of a 2000-2002 or 2007-2008 sort of decline. The conditions are “right” for just such an occurrence, unfortunately.<<
SPX earnings are on roughly a $100 run-rate. In order to get that sort of decline you're going to need to see expectations for a huge slide in earnings or a massive spike in interest rates or some combination of those two things. What makes you think that conditions are "right" for this now? Or do you think it's something else that will crash the market?
Somehow your comment section butchered my comment. I meant to say that the “Bernanke/expanding universe” line was hilarious. I then posed the stock market question in response to the second quote from the blog entry.
Ha, I figured out what you meant! Thanks, Mark!
I think that earnings are unsustainable at this share of GDP, so even if we avoid a recession while Europe is having one earnings will probably still decline. But even if they don’t decline, the market is currently priced at almost 23x ten-year earnings, while 16 is the long-term average. So even with no correction in margins, you could easily lose a third of the market’s value and not have stock prices look cheap according to traditional metrics.
…but I also mean that conditions are right in the sense that implied volatilities are extremely low, bullishness is off the charts, retail money is flooding into stocks…all of these things are not typically associated with the early stages of extensive bull markets. Although it could be different this time.
p.p.s…to be clear, I’m not CALLING for a crash. I’m calling for weak returns for 10 years. I’m just observing that there are lots of ways to get to 2% after-inflation for 10 years. One of them is the market goes up 200% but prices go up 180% (ignore compounding)…which isn’t a crash…or you could have a major discontinuity tomorrow, followed by a nice 10-year rally. But the breakage risk is higher now than it has been in several years.
Yes, understood. I just perceived that you were seeing something that merited a crash “right now” and i was wondering what it was. That Shiller “10-year indicator” is at best a long-term thing and definitely not a timing mechanism. As for sentiment, the last AAII poll i saw had shown a sudden swing towards bearishness following just two down days– exactly the kind of classic “wall of worry” that makes for a climbing market. And while the Investors Intelligence poll is in the mid-50s, it doesn’t get extreme until around 60% or so. Meanwhile, the SPX is only around 4% over its 50-day SMA which doesn’t even mean that the market is terribly short-term overextended. As for “earnings being unsustainable at the level of GDP,” well, even a drop from $100 to, say, $90 would be unlikely to lead to an outright crash unless we get a spike in interest rates, as 14x 90 is still 1260 and while that’s a 24% correction it’s still nowhere near 2001/2008 territory.
The thing that i think could crash this market is a big spike in interest rates (especially one caused by something other than a booming economy) and I just don’t see a near-term catalyst for that.
P.S. Lest I sound like some kind of crazed bull, I currently have a very large cash position in my fund because I don’t particularly like the prices out there, but there’s a big difference between “not cheap enough for a value guy like me” and “nosebleed territory.”
This is sad. You see a 50% drop materializing out of thin air. Good luck to you moron, I bet you have been bearish since Dow 7000.
I’m approving this because I think even people who engage in ad hominem attacks with no reasoning ought to have a voice.
I AM curious whether the poster means Dow 7000 in 1997 or in 2009. Whoever has been out of stocks since 1997 is kicking this guy’s butt. Inflation-linked bonds are up 175% since then, with dramatically less volatility. But as for me, I can only say that I’ve been under-invested in stocks since pretty much constantly since the end of 1998. Since then, the S&P total return is up 59.8%, compared to 113.8% for commodities, 164.8% for ILBs…but only 42% for T-Bills.
However, I suspect he means that he caught every turn since 1997, and is so unbelievably wealthy that he has nothing better to do than to troll around the internet and trumpet his investing prowess! Clearly, no one who was embittered by his own failure would do that. Surely?
“In my opinion, you have to be wildly optimistic to believe that corporate profits as a percent of GDP can, for any sustained period, hold much above 6%… Maybe you’d like to argue a different case. Fair enough. But give me your assumptions. The Tinker Bell approach – clap if you believe – just won’t cut it.” Warren Buffet, 1999,
We are about 10% of gdp right now. So, a return to normal would take us to $60, not $90. Multiply that by 14, and you get 1080. Add in a little bit of overshooting, and you could easily get a 2007/2001 style decline. Let Mike, I’m not predicting this by any stretch, but just sayin’
I’m a little bit tempted by something I wouldn’t ordinarily touch right now: VXX. Any thoughts? (Just for informational purposes, of course.)
Eric,
You make a fair point re. historical ratios of profits to GDP, but let me ask you this: in Q4 U.S. GDP growth was 0.1% and Europe had awful contraction and the SPX was STILL on a $93 earnings run-rate while the current quarter– which is almost over– is supposed to be back over $100 annualized. So if companies are able to crank out this kind of earnings with this kind of lousy economy and yet the fear of “economic lousiness” is discouraging the kind of competitive capital investment that could drive those margins back down, then what do you think WILL kill the the margins? I suppose one could argue that after several more years of this “new normal” capital investment WILL come in to steal away some of those excess profits, but that’s nowhere close to being a near-term catalyst. Alternatively, one could argue that inflation will kill the margins but that still might raise (or at least put a floor under) nominal stock prices and “crash” is generally a very rapid “nominal” event.
MArgins will be killed by growth. Unemployment helps explain why the division between labor and capital is where it is (and that’s the flip side – if margins are high, wages are low and v.v….with GDP, it really is close to a zero-sum game between the two of them). So anything which raises the cost (and benefit) of employment or swings power to employees will do it. Like, maybe, Obamacare.
P.S. i had the same thought you did today re. the various VIX ETNs. The problem is that there have been multi-month periods when the VIX has averaged in the 10s and that’s enough time for the degradation of those products to kill you. If you’ve got a non-timing-critical way to play this (on, say, a move below 11) I’m all ears.
How about politically driven austerity? (read the news lately?) Or how about the fact that the 2007 debt crisis has never been addressed and we might get a re-echo of it, where people start paying back down their debt, or defaulting, or increasing their savings.
I can’t predict: but the surplus of corporate profits comes from a deficit of government and personal savings/debt. that’s practically an accounting identity. the pendulum swings back and forth.
as for what will be the catalyst: i think its unlikely that the earnings will fall first, and then be followed by the market. that’s not how things usually go.
Actually, the ratio of household debt to income is down significantly from where it was a few years ago (while still far higher than historical norms, and of course if you assign a pro-rata share of the Federal debt to each household…). However, if Bernanke continues to blow his “echo bubble” in the housing market that ratio could decline further (at least for a while). And sure, the market is likely to smell out a broad earnings decline around three months or so ahead of time (it used to be far longer, I think), but my question was: What is the near-term catalyst to tank those earnings (or spike rates and compress PEs)? Those things you mentioned could (and probably will) happen at some time in the next five years… I just don’t see it as a high-probability bet that they happen NOW as opposed to, say, a year or two from now.
I think government austerity and foreign sales are bigger near-term threats. but domestic personal consumption will threaten too. and if mike is right about commodity prices, they can be hard to pass on to consumers in a stagnant economy.
Yes, that’s all possible, but then you really need to be calling for outright U.S. GDP contraction because, again, with 0.1% growth and European contraction in Q4, earnings weren’t all that bad. And with that idiot/maniac Bernanke printing $80 billion/month…
As an aside, let me give you a legitimate bull case: the budget deadlock in Washington gets broken by mutual agreement to heavily means-test Medicare and Social Security and this gets combined with a tax overhaul that meaningfully lowers rates and eliminates (or caps) a bunch of deductions. What do you think the market would do if this were to happen? (I think there’s a non-trivial chance that it will.)
I was just looking at this today in the latest Z.1. Household debt has plummeted from 14T to…wait for it…12.8T since before the crisis. The ratio relative to income is down a bit more because incomes are up a little. The debt SERVICE ratio is down a lot because of low interest rates, though.
Markets don’t need catalysts to tank, or at least it’s not necessary that we know what the catalyst is at the time. That’s the message of 1987 (and Shiller’s surveys around that time, which showed no one even mentioned program trading at the time) and other similar studies. Heck, we were already crashing in 2007-8 long before most people knew there was a generalized credit crisis.
You can’t call crashes, but you can identify when the market is in critical states where crashes are more likely (I recommend “why stock markets crash” by didier sornette, by the way!
“Markets don’t need catalysts to tank, or at least it’s not necessary that we know what the catalyst is at the time. That’s the message of 1987 (and Shiller’s surveys around that time, which showed no one even mentioned program trading at the time)…”
But stocks were a lot more expensive in October of ’87 than they are now (at least without margin-adjusting current profits). As for that crash itself, a lot of people understood immediately that its massive size was caused by “portfolio insurance” and nothing truly fundamental unlike, say, stocks in 1999-2000 or housing prices in 2006 which were “fundamental accidents waiting to happen)
Gosh, Mark, I love you but the first thing you said there is false. In September 1987 the Shiller P/E was 17.7 and it’s nearly 23 now.
and no, as far as I know, there is no way to play VIX without paying for cantango. (other than, of course, just buying the options, which is also somewhat tempting.)
Shortest article in a while, and by far the largest list of comments! Thanks!!
Yes, but again you’re talking “Shiller.” The actual 12-month trailing PE then was around 22 while right now it’s around 16 which, again, is definitely not “cheap” but also not “accident waiting to happen” territory.
Mark, you have to either look at Shiller or else look at tralining PE and consider earnings to GDP. You can’t say “yes I know earnings to GDP are high but stocks aren’t that expensive on a 12-month trailing PE basis.”
Eric: This is a fair point re. adjusting the PE for margin compression but it brings us full-circle back to looking for the margin compression catalyst. Maybe Michael is right that it will be Obamacare or higher commodity prices. Or what if corporate tax reform encourages companies to repatriate their overseas earnings and to no longer park as much overseas– in that case, earnings would decline as those companies start paying taxes that they’re currently avoiding. I just don’t see any of this stuff (with the possible exception of Obamacare) happening immediately as opposed to “they could just as easily happen several years from now.”
Michael: have you looked at “Shiller” as a timing mechanism? My recollection is that it’s lousy as it can stay high (or low) for very long periods of time.
I don’t believe in market timing. I’ve never seen any research that showed it was possible, and never met anyone who was consistently good at it. Shiller is certainly not a timing tool. But I don’t mind missing 10% rallies if it means I miss 20% declines too.
But you can’t just take the most recent 12 months unless you expect all of the future 12 months to look the same. You’ll always be overinvested at highs, when earnings haven’t fallen yet but the stock market has. By the way, most investors today wouldn’t think a 22 trailing multiple is high. But investors were much more conservative then.
The point of the Shiller PE is to measure earnings versus a whole cycle or two, because you’re not getting a 12 months in the future that’s guaranteed to be like the ones just past. However, it’s much more likely that the average of the next 10 years will be pretty much like the average of the last 10.
‘there are lots of ways to get to 2% after-inflation for 10 years. One of them is the market goes up 200% but prices go up 180% (ignore compounding)…’;
With Bernanke now floating trial balloons that maybe QE4evah never will be withdrawn at all (as I have always maintained would happen), your 200% nominal gain scenario is not so outlandish at all.
Probably the flood of comments is because we all missed you during your self-imposed mourning period for Chávez. *wink*
Jim – bad incentive! I should go away more often! 🙂
Mike
‘I don’t believe in market timing. I’ve never seen any research that showed it was possible, and never met anyone who was consistently good at it.’
Eh, I’m surprised by your statement. Don’t know whether you’ve seen this paper by Baltas and Kosowski, “Improving Time Series Momentum Strategies,” but if not I would call it to your attention:
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2140091
Their 10-year test period using intraday commodities data perhaps could be criticized as too short, though it constitutes a mountain of data (1.25 million observations) given the 30-minute frequency. But the mathematics and statistical testing seem quite thorough. They conclude that ‘time-series momentum profitability implies strong autocorrelation in the individual return series of the contracts.’
To me, this paper is a logical companion and follow-up to Gorton and Rouwenhorst’s “Facts and Fantasies” paper, amplifying their equal-weighted rebalancing return with a long-short switching strategy derived from momentum indicators.
I should have been more precise, knowing Jim was reading!
A momentum strategy is not strictly a timing strategy – you’re going with the flow (this works great on currencies too, by the way!). Timing strategies involve reversing “at the top” or “at the bottom” based, generally, on some sort of oscillator if it’s technically-based, and that’s what I meant I’ve never seen anything that works consistently, in advance.
Put another way: a momentum strategy will always be long at the top and short and the bottom. It is hard to calibrate trailing stops so that you get out of disasters (1987) but don’t get out of knee-jerk corrections that keep you from missing the big trend. And if you have a momentum system that’s always in the market, it’s really, really hard to avoid getting chopped up in sideways markets.
I spent the first few years of my career as a quantitative analyst for a technical analysis firm, and spent a lot of time trying to design models like this. Today’s quants are much quantier, though, and might have found some way to tease more signal out of the noise. But that’s the fundamental limitation – the noise just swamps the signal.
This is why I invest based on probabilities for long-term value. I tend to get the stories roughly right, but my timing is abysmal. Which is why I want to be very clear that I’m not calling for a crash!!! I’m just saying that the conditions are right, making it POSSIBLE.
p.s. I haven’t read that paper, but I will! Thanks for posting it!
Appreciate your clarification/confession (that you were a techie). 😉
In looking up the Baltas/Kosowski paper, I came across another paper by the same authors which is more accessible and more comprehensive, starting with data from 1974:
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1968996
I’m no statistical jock, but their work looks thorough to me, and the Sharpe ratios they report are impressive.
Apparently, being diversified among a sufficient number of contracts, using several different lookback periods, somewhat ameliorates the disadvantage of being ‘long at the top and short at the bottom,’ which of course affects all of these strategies.
Well, it does until it doesn’t. I wonder if that’s not just another kind of optimization? Do they pick their periods with a priori reasoning or just by crunching it? Seems like they’re “diversifying” across different time periods etc, which as we know works until the correlations (in this case, across lags) approaches 1…still, I guess my more substantive objection will be “why aren’t these guys billionaires then?” 🙂