Irrational Exuberance — Robert Shiller on Stock Market Bubbles and Manias

AuthorAthenarium / Posted on

Irrational Exuberance - Robert Shiller

Understanding our market psychology

Robert Shiller published Irrational Exuberance in 2000 during the Dot-com bubble to explore the nature of investment mania and speculative bubbles. The book encourages people to see the stock market price not as a standalone entity, but as the aggregation of people’s choices. And that we shouldn’t underestimate how common it is for our thinking to mimic the herd.

Irrational Exuberance is timeless book on market psychology, and mandatory reading for investors and financial historians alike. This post will review the lessons that I took from Shiller’s work, including his views on bubble mechanisms, new era thinking and the anchors that shape market cycles.

Skip ahead

Natural Ponzi processes

Shiller says speculative bubbles are sometimes the result of a naturally occurring Ponzi processes. This is where increases in asset prices lift the confidence and expectations of investors, encouraging them to bid asset prices up even further. The cycle then repeats itself until some exogenous factor dispels the Ponzi process.

Changes in public attention and an aggregation in like-mindedness help to fuel the bubble. It’s a self-fulfilling prophecy in some ways, as people develop plausible stories to rationalise the feedback loop. The news media and some institutional investors too may reinforce the narrative. But like actual Ponzi schemes, stock market bubbles cannot grow indefinitely.

The Dotcom bubble

It’s untested investor enthusiasm that sustains most speculative bubbles. For example, articles like Mandel’s “The Triumph of the New Economy” justified the Dotcom era by pointing to increased globalisation, rising profits, growth in high-tech and low interest rates. Shiller provides a long list of reasons that contributed to the mania. This included the:

  • Rise of the internet during a period of strong earnings growth and optimistic forecasts from securities analysts;
  • Growth in media reporting of business news, and the frequency in which the public engage with stock news;
  • Confidence in Western finance, following China’s market reforms, Japan’s lost decade, and the 1997-98 Asian financial crisis;
  • Growth in employee stock option awards, which incentivised the veneer of corporate success to boost stock prices;
  • Investors that held onto their positions in expectation of further capital gains tax rate cuts;
  • Expectation that Baby Boomers would consume more and buy more stocks (encouraging others to preempt this trend);
  • Growth in defined contribution pension plans and mutual funds that invested heavily in equities; and
  • Growth in day trading, 24-hr trading and non-investment gambling opportunities;

This time is different

We often hear about the promise of new innovations, and how it’ll meet consumer demand, improve productivity and unlock firm profits. This is often true. Advances in technology and scientific understanding has been the cornerstone of long-run economic progress and development.

But there are times where our expectations and optimism far exceeds what is likely to happen. Shiller describes how new era economic thinking tends to come in waves. The narrative convinces the public and investors that the future is now brighter and less uncertain than before.

Economist Carmen Reinhart described this phenomenon as this-time-is-different syndrome. These investors neglect traditional models of valuation, or assume that they no longer apply. Shiller says such new era thinking has helped to propagate investor optimism during the 1900s, 1920s, 1950s, 1960s, 1990s and early 2000s.

Half truths and thoughts

Opinions in investing are likely to contain many half-truths and half-thought ideas. Many ideas are often mutually contradictory and unlikely to receive scrutiny within an analytical framework. Shiller warns against the use of broad generalisations, such as the baby boom or an ageing population, to rationalise valuations and expectations. Without further analysis, these stories are too simple to explain the complex demand and supply dynamics of a company, industry or economy. 

For example, while new technologies will impact economic activities, it does not always justify higher valuations for all companies. Innovation may improve a company’s productivity. But its relative efficiency might remain unchanged if every company can invest in it. The net impact on the bottomline is not immediately clear or always positive. It’s not uncommon for new technologies to destroy incumbent profits by way of disruption and newfound competition.

Mass market delusions

Aswath Damodaran referred to the above as the market delusion in Narrative and Numbers. That is, vivid stories that might appear plausible to us at the individual company level may not hold at the macro scale. Overconfidence is rife when every investor believes their stock or start-up will outperform the market and their competitors. This cannot happen by definition. Nonetheless, the imagery of technological progress is strong and can bring about new waves of enthusiasm.

Similarly, the expectation that stock portfolios will always recover from a precipitous falls, an assumption that many investors hold, can be dangerous. Japan’s Lost Decade is perhaps the strongest example. Similarly, inflation can disguise less than stellar historical records on a real-value basis. It’s possible that historical track record of the United State’s stock market is an anomaly and not the status quo for markets overseas or the future. We should take caution with assuming that stocks will outperform bonds in the long run with absolute certainty.

Compounding biases

Shiller noted that early evidence of speculative bubbles coincided with the introduction of mass newspaper distributions. Major market movements requires common thinking, which the news helps to create and reinforce. There are two features, with regards to the news media, that Shiller suggests we keep in mind: information load and information cascades.

Overloaded and over-explained

The news media enables people to spread, disseminate and reinforce ideas. On many occasions, these ideas are not supported by real evidence. The media today will write something (or anything) to rationalise the market outlook or day-to-day changes in stock prices, blurring noise with useful information. Furthermore, they will overuse superlatives and vivid colours (e.g. blood red for price declines) to generate emotional responses and more viewership. For the same reasons, they will also find creative ways to describe events as unprecedented records in financial history.

Information and attention cascades

Research suggests that big changes in stock prices don’t follow big news days as frequently as we would imagine. Human attention after all is temperamental. This means that the accumulation and ordering of news can give prominence to ideas that we previously ignored or deemed unimportant. So, tag-along news can influence stock prices in major ways over time.

This information cascade, coupled with word-of-mouth, can shift national attention (sometimes suddenly). Additionally, market news volumes tend to co-move with the size of the feedback loop, during both positive and negative bubbles. In these periods, we may also encounter price-insensitive buying or selling. Recognising these behaviours can help us to distinguish between valuations based on fundamentals or exuberance.

Psychological anchors

Market prices are not always driven by fundamentals. Instead, decisions are often biased by the closest available anchor. Shiller identified several anchors that can influence market expectations, attention and feedback loops. This includes quantitative anchors, moral anchors, overconfidence, heuristics and non-consequential reasoning.

Quantitative and moral anchors

For many investors, quantitative anchors are often the stock price or price-earnings ratio of recent memory. It could also be the nearest round-number milestone for a company or index. We also tend to compare company performance and stock prices to the country of its headquarters than to the the industry or geographies in which they compete.

Furthermore, periods of irrational exuberance often require compelling storytelling to justify existing price levels. Moral anchors tend to focus on the vividness, plausibility and consistency of qualitative factors, than on quantities or probabilities. We often construct simple reasons to make decisions, and require our decisions be justified in simple terms.

Similarly, professor Aswath Damodaran has also described how moral anchors are more commonly used for analysing emerging firms and industries that depend on charismatic founders and sustained faith in future potential. Where quantitative data is less accessible, we depend on narratives to rationalise our conclusions.

Overconfidence and heuristics

Individuals have a tendency to demonstrate overconfidence in their beliefs and reasoning. We often forget the probability for error at each stage of logical reasoning. This positions us to overestimate the likelihood of correctness in our conclusions. Philip Fernbach and Steven Sloman attributes some of these tendencies to our illusion of knowledge, highlighting our tendencies to ignore complexity when making decisions.

Furthermore, our tendency to seek patterns, to find best-fit explanations and to ignore probabilities can lead to overconfidence in our decision making (Annie Duke explores such processes in her book Thinking in Bets – A good read). Shiller notes that the strength of such heuristics and the speed at which people change their opinions may influence the pace and scale in which speculative feedback loops evolve.

Non-consequential reasoning

Psychologists Eldar Shafir and Amos Tversky termed the phrase non-consequential reasoning to describe people that are unable to form conclusions based on an assessment of hypothetical events that could occur in the future. That is, many people have trouble making decisions until after the events actually occur. It’s only after the occurrence of major events (e.g. stock price changes) that people realise their emotions and preferences, and act accordingly. Shiller suggests that such features in human reasoning can contribute to the fragility and unpredictability in psychological anchors.

Herds and epidemics

Many factors can contribute to and explain herd mentalities in investing. For example, people access and react to the same publicly available information at the same time. Additionally, social pressures, word-of-mouth and perceived authority are sources of information cascades that shape individual judgement. Research has found face-to-face and televised communications to shape emotional judgement more strongly than written mediums. Any onset of fear, anxiety and envy during such periods can encourage further conformity.

Social attention

The human brain is wired largely to focus on one or two major issues at a time, and is not well equipped to grapple the full complexities of complex systems such as financial markets. As an individual, we tend to have low awareness of how our attention shifts over time. Nonetheless, such aggregate shifts in attention will influence market valuations over time. For example, the public’s attention is very much attracted to bull markets and financial markets are likely to dominate our news and cultures during such periods.

Epidemic models

Shiller notes that the mathematics used in epidemiology to describe the spread of diseases is a powerful mental model for thinking about the spread of and decay in ideas among groups. In simple epidemic models, infection levels are influenced by the rate of infection and removal. If the removal rate is zero, infection levels will exhibit a logistics curve over time. However, if the removal rate is greater than zero and less than the infection rate, than the simple model predicts a bell curve in infection levels over time. Finally, if the removal rate is greater than the infection rate, then the infections will not occur.

The epidemic model is a simplified but useful framework for thinking about the spread of ideas and feedback mechanisms. It is helpful to think about the rate of infection and/or removal of ideas, and how such rates may evolve over time. These factors will shape the trajectory and evolution of ideas. Malcolm Gladwell shared similar ideas in his book The Tipping Point. To understand how some ideas and behaviours spread like wildfire, Gladwell recommends we consider the contagiousness of the messenger, stickiness of the message and the operating context (e.g. peer pressure, social proof, fear of missing out, etc.).

Rationalising irrational exuberance

Shiller highlights that we should not focus solely on factors that predominate the news to guide our valuations and expectations. History is a wonderful reminder of all the possible and unexpected events that can disrupt, support or destroy market value. For example, when positive or negative bubbles move too far in one direction, issues pertaining to reasonableness, fairness and resentment tend to surface. This may encourage some actors to reverse course (e.g. governments, unions, short-sellers, special interest groups, etc.).

Smart models, smart money?

Shiller suggests we take caution when rationalising current events within the confines of historical and esoteric academic models. For example, efficient markets theory would have many believe that all public information has been accurately reflected in financial prices, and that smart money would have driven asset prices towards their true value. We know from history that this is sometimes absurd.

However, the difficulties in predicting day-to-day changes does necessarily imply that predicting any change is impossible. Similarly, we should not assume that smart-money are always the only price setters, and that public information is always accurately incorporated into prevailing prices. Economists need to be more comfortable with the messier aspects of markets if academic theory is to be of greater utility for everyone. A willingness to learn and unlearn can go a very long way.

Further reading

References