Aviation disasters

Monday, January 31, 2011

“Sentiment and stock prices: The case of aviation disasters” by Guy Kaplanski and Haim Levy (Journal of Financial Economics, 2010, volume 95, pages 174-201).

Previous research shows that aggregate stock prices appear to be influenced by phenomena that affect the emotions and mood of a large part of the population concurrently.[i] One study using data from 26 international stock exchanges argues that good moods resulting from morning sunshine lead to higher stock returns.[ii] Other researchers report that stock markets fall when traders’ sleep patterns are disrupted due to clock changes with daylight savings time.[iii] A third recent study suggests that the outcomes of soccer games are strongly correlated with the mood of investors.[iv] After a loss in a World Cup elimination game, significant market declines are reported in the losing country’s market.

Guy Kaplanski and Haim Levy show that something similar seems to occur after major aviation disasters.  They argue that such unfortunate events lead to changes in sentiment.  When someone’s valuation of a security is influenced by misinformation (or irrelevant information) such a person is said to be a noise trader.  When many people are influenced in the same way at the same time (that is, noise is correlated among individuals) we say that sentiment is at work.  Sentiment can be driven by mass emotion or mood.

Economic losses vs. market outcomes

The authors of this study find that after American or European air disasters with 75 or more casualties, the U.S. stock market loses (on average) about $62 billion the day after.  Abstracting from the obvious human suffering, what is the economic cost such disasters?  The three main direct economic losers are insurance companies, aircraft manufacturers and airlines.  Since it is estimated that no more than $1 billion/event is lost by affected companies, the market impact seems way out of line, so what is going on?  In a simple valuation model, there are projected cashflows and a discount rate.  A component of the discount rate is the market risk premium (or equity premium) which comes from the amount of risk that is perceived to exist along with the aversion that investors in aggregate have to that risk.

Let’s consider the latter: do investors collectively become more averse to risk after such events?  The answer seems to be yes.  There are the three reasons to believe that fear and anxiety increase for a time after air crashes.  There is widespread media coverage after such events; the coverage is sufficiently emotional to arouse fear and anxiety; and when influenced by such emotions people to tend to become more pessimistic and shy away from risk.  A comprehensive examination of media coverage documents the first; and an assortment of psychological studies are cited to support the second and third.

Declines and reversals

It turns out that market movements in response to aviation disasters are mostly temporary.  As the press coverage wanes, the market gradually recovers.  As negative emotions dissipate and clear thinking returns, stock prices tend to reverse.  This reinforces the view that original price reaction was emotion-driven.

There is another way to gauge market fear, namely through imputing the implied forward-looking volatility that can be extracted from index option prices.  This leads to the VIX or (aptly termed) “Fear Index.”  Indeed we see dramatic rises in the VIX the day after disasters of this type.  Given that the level of the VIX tends to be fairly persistent, after the spike reversal tends to be slow.  The VIX is nothing other than the market’s prediction of future volatility.  It could be that after airline disasters there is a legitimate fear of political unrest (perhaps associated with terrorism).  Since actual volatility measured over time subsequent to the event does not rise appreciably, this lends credence to the interpretation that the market has got it wrong, that is, its fears were unfounded.

Consistency with limits to arbitrage

One reason that this research is quite compelling is that it seems to be consistent with a “limits to arbitrage” story.  The idea behind limits to arbitrage is that even when rational investors notice that sentiment has moved prices inappropriately, it may not be easy for them to capitalize on the apparent inefficiency.  Firms whose valuations are more subjective and which are harder to arbitrage logically should be, under the limits to arbitrage argument, more susceptible to price changes induced by aviation events.  A proxy for being hard to value is idiosyncratic volatility, and a proxy for being hard to arbitrage is firm size (where it is important to note that small-cap firms might not be easy to short-sell because of security non-availability).  In fact it is firms with high volatilities and low levels of market capitalization that are most impacted by such events.

It is becoming increasingly clear that there are both cognitive and visceral components to risk assessment.  This valuable contribution has reinforced the role of the second.


[i] See chapter 7 of Ackert, L., and R. Deaves, Behavioral Finance: Psychology, Decision-making and Markets, 2010, South-Western, Cengage Learning, for a discussion of emotions (and mood).

[ii] Hirshleifer, D., and T. Shumway, 2003, “Good day sunshine: Stock returns and the weather,” Journal of Finance 58(3), 1009-1032.

[iii] Kamstra, M. J., L. A. Kramer, and M. D. Levi, 2002, “Losing sleep at the market: The daylight saving anomaly,” American Economic Review 90(4), 1005-1011.

[iv] Edmans, A., D. Garcia, and O. Norli, 2007, “Sports sentiment and stock returns,” Journal of Finance 62(4), 1967-1998.

Share and Enjoy:
  • Print
  • Digg
  • StumbleUpon
  • del.icio.us
  • Facebook
  • Yahoo! Buzz
  • Twitter
  • Google Bookmarks