Index option prices since 1988 embed an irrational crash premium beyond any plausible expectation

Evidence

The volatility skew described in the previous section is qualitatively consistent with standard models of rational learning, but quantitatively the options market has overlearned. For a prolonged period, excess riskadjusted profits have been available from selling index puts at the market prices that have prevailed since 1988. Remarkably, this is the case even when artificial crashes of 20% in a month are introduced into the computation with probabilities as high as one every four years (Jackwerth 2000). … To explain observed index option prices since 1988 requires artificial addition of crashes of almost 1987 severity more frequent than one every four years, when “even a 20% [one-month] crash every 8 years seems to be a rather pessimistic outlook” (Jackwerth 2000, p. 447). “The most likely explanation,” Jackwerth concludes (p. 450), “is mispricing of options in the market.” — Donald MacKenzie & Yuval Millo, Constructing a Market, Performing Theory, 154

Categorization

  • XREF: This is the empirical core of the performativity argument that will follow—the market prices in the trauma. Connects to behavioral finance’s equity premium puzzle and Shiller’s irrational exuberance literature. — connects to: Explaining post-1987 option prices requires crash frequencies roughly four times historical rates — Donald MacKenzie, An Engine, Not a Camera: How Financial…, 913; Index option prices clustered around Black-Scholes values despite greater model complexity — Donald MacKenzie, An Engine, Not a Camera: How Financial…, 786; Index option prices embed a skew that actual price movements do not justify — Donald MacKenzie, An Engine, Not a Camera: How Financial…, 911 (+2 more)
  • QUESTION: If the premium is truly ‘excess’ by rational models, what does the persistence of profits from betting against the skew imply about the efficient-market floor? Worth chasing Jackwerth’s calibration.

Broader Connections