Double, double toil and trouble
Fire burn, and cauldron bubble.
[dc]B[/dc]ubbles seem to be the theme of the day. Headlines like “Why Stocks Are Undoubtedly Experiencing a Massive Bubble”, “Nobel Prize economist warns of U.S. stock market bubble”, and “Shiller Worried About ‘Boom In US Stocks… Bubbles Look Like This'” raise dire concerns. At the very least, they do what they are intended to do, which is to strike fear in the hearts of readers and to get attention for the articles that follow. There is much debate, with good logic on both sides of the argument. Economists build and revise models, researchers study data, reporters write articles, and bloggers can write anything they want, but I propose there is only one question that matters—how should we trade in this market? (Trade, in this sense, includes long-term investors and managers as well, not just short-term traders.)
Even at the pinnacle of academia, there is disagreement. Robert Shiller, the Yale economist and recent Nobel Prize winner, recently proposed a checklist could be used to diagnose bubbles: a sharp increase in prices, great public excitement about those increases, an accompanying media frenzy, stories of fortunes being made in the asset, growing interest in the asset among the public, new theories to justify these price increases, and a decline in lending standards. Eugene Fama, who shared the Nobel Prize with Shiller this year, said recently “I don’t even know what a bubble means. Those words have become popular. I don’t think they have any meaning”—soundly denying that bubbles exist. If two of the leading minds in the field cannot even agree whether bubbles exist or not, is it any wonder that investors are confused? (Indeed, if markets were truly efficient, in the academic sense of the word, bubbles could not exist. The existence of bubbles is one potentially effective challenge to the Efficient Markets Hypothesis.)
One thing that is interesting about Professor Shiller’s list is the qualitative element. I have always held that market analysis is not a pure science, that the market is not a pure math problem. (I suppose this should be obvious from the title of my book.) Much of research is about quantifying everything that can be quantified, and trying to understand the relationships between data points. This is a noble pursuit, but, in the end, there may be some predictive elements that simply cannot be quantified—that must be interpreted with an element of Art.
This is not an exhaustive or authoritative list, but let me share some thoughts about bubbles in financial markets:
- The Bard’s witches got it right: double, double toil and trouble. One of the characteristics of bubbles is that the rate of trend increases dramatically, often in several steps. (If “trend” is the average rate of return over a time period, we are looking at the first derivative of those returns.) The rate of trend can only double so many times before that trend becomes unsustainable and vulnerable to collapsing into the vacuum on the other side.
- We need to think about our definitions carefully. Bubble should probably apply to a move over a significant (perhaps greater than three year) timeframe, but remember the patterns of the market are fractal. A “bubble” appears nearly every day, intraday, in some asset.
- Professor Shiller’s points about public interest becoming mania are certainly valid. I remember a preschool teacher telling me that she was leaving her job teaching and becoming a daytrader since she’d turned $5,000 into nearly $200,000 in six months. This was, of course, in late 1999. I remember a taxi driver telling me about his profits in Crude Oil futures in 2008. By the time a market becomes a part of the (very short-lived) zeitgeist, the move in that market is probably nearly over.
- Valuations are problematic. Most people who point to valuations as a reason that a market move is “wrong” must not have looked at the long history of those valuations. Very high valuations are commonplace in several stages of extended bull markets, not just at the end. Valuation matters, but the interaction of prices and valuations is much more subtle and complex than is commonly assumed. Perhaps high valuations do indicate a market is overvalued, but perhaps high valuations will lead to much higher valuations.
- This is a more technical element: in normal, healthy trends there is an alternation of buying and selling pressure. Bubbles always show a condition called a “buying climax” in which that alternation fails, and buying leads to more buying in an ever-increasing feedback loop. The normal fluctuations of the market become muted and “one-sided” trading sets in. We say that the market “goes in a straight line”, but this is as much a reflection of the psychological backdrop as of actual price movements.
- Identifying a bubble in real time is much more problematic. I’ve said elsewhere that the most important problem in technical analysis is separating “overbought” conditions from true strength. I can show you a reliable bubble indicator that has flagged every major bubble in every asset, but it would’ve taken you out of stocks in 1992, 1995, and many other times as the Nasdaq rocketed to a 1,000% gain. Here, again, the importance of a holistic approach to market analysis is important: blind adherence to overbought/oversold indicators will always take you out of trends early. Perhaps the naysayers are right and stocks are in a bubble right now. Perhaps that bubble will last five years and stocks will quadruple in value. I don’t know, and neither does anyone else.
- One of the key elements of a bubble is what happens afterwards. If the conditions were truly “bubbly” then overwhelming selling pressure carries the market into a sharp selloff and it is many years before prices return to those highs. Consider stocks after 1929, the Nasdaq after 2000, Crude Oil after 2008, Wheat in early 2008, Real Estate in 2007: these are typical “post-bubble” movements.
My sense of the current bubble debate is this: There is scant technical evidence for stocks currently being in a bubble. Most major indexes show few signs of “overheat” or overextension, but, rather, we see evidence of a strong, sustainable trend. (If you’re looking at long term charts and thinking otherwise, make sure those long-term charts use log-scaled axes which accurately reflect the percentage growth rates over long time periods.) We are virtually obligated, over the intermediate to long term, to focus on being long stocks. Expect “experts” to tell us why the market’s move is wrong—this probably just normal dissension which is essential for fueling a further trend. The market does not work like most people think it does; markets work on expectations and expectations of expectations, and this will always frustrate people who expect markets to respond to business school-style logic.
Of course, there is always the chance the naysayers are right. The bull market could end dramatically today. Markets turn, sometimes seemingly on a dime, and it does not take a bubble to do serious damage to unprepared, emotional investors. The bubble debate may be focusing on things that are not actually important. This is why we use a disciplined technical process: it allows us to focus on what is important and to manage the risk at points like this without relying on speculation or undue emotion. Bubble or no bubble? No one knows, and it really doesn’t matter.