Do Stock Markets Exhibit Price Bubbles?
Review of “Bubbles for Fama” by Robin Greenwood, Andrei Shleifer, and Yang You
“Bubbles for Fama” by Robin Greenwood, Andrei Shleifer, and Yang You challenges Eugene F. Fama’s assertion that stock prices do not exhibit price bubbles.
By analyzing US industry returns from 1926 to 2014 and international sector returns from 1985 to 2014, the authors present four main findings. The abstract effectively outlines these findings: Fama’s claim is partly correct in that sharp price increases do not predict unusually low future returns; however, such increases do predict a heightened probability of a crash, and certain attributes of price run-ups can help forecast both crashes and future returns. This sets the stage for a nuanced exploration of the predictability of stock market behavior.
What Are The Main Conclusions?
- Sharp Price Increases and Future Returns:
- The paper confirms Fama’s view that sharp price increases in industry portfolios do not, on average, predict unusually low returns. The average returns following a price run-up are unremarkable, matching those of the broader market. This finding challenges the conventional wisdom that price bubbles necessarily lead to poor future performance.
- Probability of a Crash:
- Sharp price increases are linked to a substantially heightened probability of a crash. The authors define a crash as a 40% drawdown within two years and demonstrate that the probability of such a crash increases significantly with higher past returns. This insight adds a critical dimension to Fama’s conclusions, suggesting that while returns may not be predictably low, the risk of a significant downturn is elevated.
- Attributes of Price Run-Ups:
- The paper identifies several attributes that help forecast an eventual crash, including increased volatility, turnover, stock issuance, and specific price patterns during the run-up. These attributes also have predictive value for future returns, supporting the idea that more information beyond price increases can improve predictions.
- Predictive Characteristics for Future Returns:
- Some characteristics of sharp price rise episodes, such as volatility, issuance, and the relative performance of new versus old firms, help predict future returns. This finding aligns with Fama’s broad thrust but suggests that additional factors can enhance predictability.
The authors employ a robust methodology, analyzing episodes of significant price increases in US industry and international sector portfolios.
Identifying 40 episodes in the US market since 1928 and conduct out-of-sample tests using international data from 1985 to 2014. This comprehensive approach ensures the generalizability of their findings.
Authors conclude Fama’s claim is only partially correct!
While a sharp price increase does not predict low future returns, it does signal a higher probability of a crash.
Furthermore, other attributes of price run-ups can help forecast future returns and crashes. The authors argue that understanding these attributes provides a more comprehensive view of price bubbles, aligning with historical perspectives on bubbles involving innovation, displacement, and paradigm shifts.
Strengths and Limitations
Strengths:
- Comprehensive Analysis: The study’s extensive data set and rigorous methodology lend credibility to its findings.
- Nuanced View: By distinguishing between future returns and crash probabilities, the paper offers a more detailed understanding of stock market behavior.
- Broad Implications: The findings have significant implications for investors, policymakers, and researchers, challenging traditional views on market efficiency.
Limitations:
- Statistical Power: The rarity of large price run-ups limits the statistical power of the analysis. The authors address this concern by using international data and adjusting for multiple comparisons.
- Predictive Value: While the paper identifies predictive characteristics, the inherent difficulty in calling the peak of a bubble remains a challenge.
In conclusion, as a result of providing empirical evidence that sharp price increases predict higher crash probabilities and identifying attributes helping forecast future returns, authors offer a more nuanced view of market behavior.
Read the full paper:
Bubbles for Fama – ScienceDirect
