“The term bubble should indicate a price that no reasonable future outcome can justify.”” – Cliff Asness
“[A bubble is] a period of enthusiastic bidding up of prices by a growing group of enthusiastic investors that goes on too long and is carried away by its own momentum.” – Robert Shiller, June 19, 2015
Asset bubbles came into the mainstream vocabulary after the DotCom Bubble. After the Housing Bubble they became an obsession. The term is used so often that it has ceased to have much meaning. This is unfortunate because it obscures one of the most critical risks to our global economy. For this reason, I reviewed what leading minds in academia and Wall Street have to say about bubbles. In this observation I summarize four distinct theories on asset bubbles by Eugene Fama, Robert Shiller, George Soros, and Cliff Asness. I then weave these theories into a comprehensive definition/recipe for identifying an asset bubble.
Eugene Fama vs Robert Shiller
Eugene Fama and Robert Shiller both won the 2013 Nobel Prize for their work in asset pricing despite having diametrically opposing views on asset bubbles. Robert believes bubbles exist while Eugene does not. After receiving the prize, Eugene challenged Robert to the academic equivalent of a duel. Eugene would accept Robert’s theory if he could predict the next 10 bubbles. Robert accepted the challenge stating that he could, assuming he doesn’t die first. Unfortunately for Robert, and popular opinion, it turns out bubbles are harder to predict than he thought. The problem is that what we presume to be a “divergence from fundamentals” may actually be “regime change”.
As arguably the most prominent supporter of the efficient market hypothesis (EMH), Fama doesn’t believe asset bubbles exist. When asked if he thought markets were efficient during the recent Financial Crisis, Fama said, “I think [markets] did quite well in this episode. Prices started to decline in advance of when people recognized that it was a recession and then continued to decline. There was nothing unusual about that. That was exactly what you would expect if markets were efficient.” But how can this be? Of course there are bubbles…right?
In the aftermath of the Housing Crisis most academics and investors have gravitated toward the view that bubbles do exist, but debating the existence of bubbles is not really interesting. What matters is whether bubbles can be identified before they pop, or if they are only visible in hindsight. Eugene’s view is that bubbles can only be identified in retrospect. If investors could consistently forecast bubbles then they would cease to exist because investors would bid down their prices before they got too high. Therefore, even if bubbles exist, the term is not useful because a process for identifying bubbles can’t, in his view, be used to beat the market.
Robert Shiller is widely regarded as an expert (if not the expert) on identifying bubbles. He is one of the only economists who predicted, for the correct reasons, the Dot-Com and Housing bubbles as documented by the publications “Irrational Exuberance“. The first edition, which featured the now famous CAPE ratio, was released the same week the NASDAQ Composite began its 78% decline peak-to-trough. The second edition came out 14 months before the peak of the housing bubble and subsequent drop of 38%.
Shiller essentially defines a bubble as having two components: a run up in prices and a divergence from fundamentals. Recognizing a run up in prices is easy. The hard part is defining and measuring “fundamentals”. In 2000, stock prices were high relative to “fundamentals” which Shiller defined as the 10 year average of corporate earnings (i.e. cyclically adjusted price-to-earnings or CAPE). In 2006, home prices were high relative to “fundamentals” which Shiller defined as household income and rental rates. With the benefit of hindsight it is easy to see that these assets were overpriced.
However, Shiller’s more recent predictions from June 15, 2015 suggest that these two components, while necessary, may not be sufficient.
- I think that it’s common sense to lean away from current high-CAPE countries like the United States and lean toward low-CAPE regions like Europe
- Right now, oil prices are in a negative bubble. They’re extremely low.
Since then oil prices and European stocks relative to the USA tanked. What went wrong? Shiller may still turn out to be right, but these huge losses beg an explanation.
One explanation is that it is notoriously difficult to time when a bubble will reverse. Alan Greenspan once said that, “The nature of a bubble as I would define it makes it impossible to determine when it will deflate.” However, at some point we have to accept that a bubble call is simply wrong or else the meaning of the word “bubble” losses significance.
A second explanation is that it is very difficult to distinguish between a “divergence from fundamentals” and a “regime change”. A divergence from fundamentals implies that the prices have moved away their “intrinsic value” which is generally calculated using “fundamental” drivers or characteristics. A “regime change” refers to any change in the market that redefines what the fundamentals are and/or their relation to an asset’s “intrinsic value”.
Robert Shiller’s predictions of rising oil prices and a rise in European vs US stocks assumed a reversion of prices to their long run historical mean. Oil prices at $60 looked attractive relative to their historical average of $81 during the previous 10 years. CAPE ratios for the S&P 500 were 26.5 compared to its 120 year average of 16.69, and far higher than European stocks at around 13. So what went wrong? To find out, let’s review the leading explanations for why oil prices fell and USA stocks outperformed Europe:
- The drop in oil price is largely attributed to the explosion in US fracking which reduced OPEC’s market share. This reduction in market share, along with the removal of Iranian oil sanctions and lack of trust between OPEC country leaders, has led OPEC leaders to end years of collusion leading to record low levels of excess capacity. The result was a flood of oil and cheaper prices. This is why OPEC failed to reach agreement three weeks ago, and why any agreement they reach will probably be either immaterial to oil prices or short lived.
- The out-performance of USA vs European stocks is largely attributed to the failed Euro experiment. Mervyn King, the former Governor of the BOE, said as much in his excellent new book, “The End of Alchemy“. King is joined by George Soros, Milton Friedman, and dozens of other leading minds. Of course, not everyone believes the Euro is doomed, but just about everyone does agree that it will be many years, if not decades, before the Euro could gain back its former glory.
Do these explanations point to temporary conditions or long term shifts that are unlikely to reverse? The answer is critical. Each is entitled to their own view, but the point is one must distinguish between temporary conditions we can chalk up to a “divergence from fundamentals” and long term factors more appropriately filed under “regime change”.
So what can we learn from Eugene Fama and Robert Shiller? Mean reversion metrics only work in the absence of “regime change”. What this means is that we cannot conclude that a financial asset is in a bubble merely from observing a run up in prices and a divergence from fundamentals. We first need to review the market for any “regime changes” that could change the fundamentals of the market and how asset prices relate to these fundamentals. We learned that Robert Shiller’s definition of bubble provides too many false positives.
George Soros and Reflexivity Theory
George Soros, the $24.9 Billion investor, is another widely regarded expert in predicting asset bubbles including the 1992 Black Wednesday UK currency drop and 2008 Financial Crisis. His Reflexivity Theory provides a useful complement to Robert Shiller’s model as it focuses on the psychological aspects of bubbles. Shiller argues that we can’t have a bubble without having two components: a run up in prices and a divergence from fundamentals. Soros argues that a bubble must also have a self-reinforcing “narrative” to justify this divergence in fundamentals.
Reflexivity Theory, when applied to financial markets, essentially states that financial asset prices diverge from fundamentals when driven by a self-reinforcing “narratives” which can propel investors to buy long after any rational justification can be found. The reason is that asset prices, unlike the weather or physical laws, are influenced by market participants. If all market participants wake up tomorrow believing that the sky is green, they will all immediately realize their mistake. However, if all market participants wake up tomorrow believing that interest rates will stay lower for longer, then all financial asset prices will rise and market participants will be emboldened with new confidence in their investing prowess, and double down.
Reflexivity Theory provides a necessary step in our formula for identifying asset bubbles. The self-reinforcing narrative is the cause of the bubble. It justifies an observation that prices have “diverged from fundamentals”. If we fail to identify this narrative in the media then we may be confusing a “divergence from fundamentals” with a “regime change”. In 2000, the “narrative” was that internet companies without a plan for turning a profit could continue to grow their share price. In 2006, the “narrative” was that home prices could not fall. In both cases the narratives were self-reinforcing because the more people believed that somehow “This Time is Different“, the more prices rose, confirming their belief in the narrative.
The psychology behind the “self-reinforcement” mechanism is what overcomes often, but not always, efficient markets. The specific heuristics involved are well known, but often ignored by traditional economists and other advocates of the efficient market hypothesis. Charlie Munger provided a near comprehensive list of these heuristics in his June 1995 speech, “The Psychology of Human Misjudgment“. Heuristics most relevant to the formation of asset bubbles include:
- Self-confirmation bias is the act of seeking out information that confirms our prior beliefs.
- Pavlovian association is the act of misconstruing past correlation as a reliable basis for decision-making
- Reciprocation tendency is the act of doing more of something that has recently made you feel good (like making money on an investment)
- Over-influence by social proof is the act of placing too much weight on the conclusions of others, particularly under conditions of natural uncertainty and stress
These heuristics compound on each other in the formation of a bubble. The “narrative” starts gathering steam with a fast run-up in prices (such as the 52% gains in long maturity UK gilts YTD). This run up gives early investors a heavy dose of self-confirmation bias. The good feeling of MtM gains leads to a reciprocation tendency exhibited by buying into the trend. Others see the move and are compelled to join due to over-influence by social proof. After a while the act of buying into the trend takes on a Pavlovian like response where investors react to rising prices like dogs to the dinner bell.
So what can we learn from George Soros? A bubble cannot form without a self-reinforcing narrative. It’s this narrative that provides a potentially credible explanation for how a price could diverge from its intrinsic value.
Clifford Asness and his “Bubble Logic”
Clifford Asness, a former Fama PhD student and hedge fund manager, says that one of his “pet peeves” is that “an asset or a security is often declared to be in a bubble when it is more accurate to describe it as ‘expensive’ or possessing a ‘lower than normal expected return…the term bubble should indicate a price that no reasonable future outcome can justify.”
Combining Cliff’s definition of bubble with Robert’s is useful because Robert’s allows for too many false positives. For example, Google and Amazon both experienced a run up in price that can’t be justified by fundamentals like P/E ratios. Buyers of these stocks are also often very enthusiastic (read ‘emotional’) buyers so one might consider them “irrational” to some extent. But these stocks can’t be in a bubble because their prices have been high for several years!
So what can we learn from Cliff Asness? The term bubble should indicate a price that no reasonable future outcome can justify. Robert Shiller’s definition provides necessary, but not sufficient conditions.
The following is my definition/recipe for identifying an asset bubble:
- Confirm a recent continuous run up in prices
- Review changes in the market for evidence of “regime change”
- Define “fundamentals” in a manner consistent with any “regime change”
- Determine if there has been a “divergence from fundamentals”
- Assess the plausibility of this divergence by evaluating the power of the self-reinforcing narrative
- Confirm that no reasonable future outcome can justify the current price
Feel free to share your own thoughts on asset bubbles and whether there are any present today. There is no better compliment you can give us than your thoughtful criticism. You can reach us at email@example.com, or follow us on Twitter @intuitecon
Disclaimer: These are our personal views. This article is not, and should not be regarded as investment advice or as a recommendation regarding any particular security or course of action. Our hope is that these observations will merely help you to more critically examine your own beliefs about finance and stimulate dialogue.