The log-normal assumption may have been self-undermining through portfolio insurance.

Evidence

There is a tempting argument that the assumption of log-normality was self-undermining, via its incorporation in the techniques of portfolio insurance. Unfortunately from the viewpoint of analytical neatness, however, the extent of the contribution of portfolio insurance to the 1987 crash is unclear. — Donald MacKenzie & Yuval Millo, Constructing a Market, Performing Theory, 191

Categorization

• SEED: Angle: a performativity story where a theory’s own adoption destabilizes the very distribution it assumes — the model becomes its own failure mode, though the empirical evidence is murky.
• QUESTION: What does the empirical literature actually say about portfolio insurance’s role in the 1987 crash? The author defers on the size of the contribution.
• XREF: connects to: Portfolio insurance’s real risk was internal self-impact, not external shocks. — Donald MacKenzie, An Engine, Not a Camera: How Financial…, 807; Portfolio insurance designed to limit losses can amplify market volatility. — Shapiro, Susan P., The Social Control of Impersonal Trust, 144; Market conditions moved toward the model’s assumptions partly because of the model itself — Donald MacKenzie, An Engine, Not a Camera: How Financial…, 1155 (+2 more)

Related

• Portfolio insurance’s real risk was internal self-impact, not external shocks. (similarity 0.87)
• Portfolio insurance forced clients to miss the entire market rebound after the crash (similarity 0.85)
• Portfolio insurance’s inventors concede it helped cause the 1987 crash (similarity 0.85)
• Portfolio insurance sells more stock as price falls, creating mechanical downward supply. (similarity 0.84)
• Price crashes in the model depend on whether traders know insurers’ selling rule. (similarity 0.84)
• Portfolio insurance collapsed after 1987 partly because it became socially tainted (similarity 0.83)
• Portfolio insurance trades could double NYSE volume on a 3 percent drop (similarity 0.82)
• Portfolio insurance largely protected client floors even during the 1987 crash (similarity 0.82)
• Portfolio insurance designed to limit losses can amplify market volatility. (similarity 0.81)
• The 1987 crash stemmed from ignorance of mechanical selling, not portfolio insurance itself. (similarity 0.81)
• Widely adopted portfolio insurance would amplify market price movements (similarity 0.81)
• The 1987 crash was a 27-sigma event under the log-normal model. (similarity 0.81)
• Observed portfolio-hedging sales would have caused under one percent of the 1987 crash. (similarity 0.80)
• Portfolio insurance creators accepted increased volatility as unproblematic (similarity 0.80)
• The 1987 crash made its own portfolio-insurance inventor fear he caused a world crisis (similarity 0.79)

Broader Connections

• Portfolio insurance’s real risk was internal self-impact, not external shocks. #expands (0.85) — Donald MacKenzie, An Engine, Not a Camera, 807
• Market conditions moved toward the model’s assumptions partly because of the model itself #corroborates (0.82) — Donald MacKenzie, An Engine, Not a Camera, 1155
• Portfolio insurance sells more stock as price falls, creating mechanical downward supply. #expands (0.78) — Donald MacKenzie, An Engine, Not a Camera, 867
• Price crashes in the model depend on whether traders know insurers’ selling rule. #expands (0.75) — Donald MacKenzie, An Engine, Not a Camera, 868
• Portfolio insurance designed to limit losses can amplify market volatility. #complicates (0.72) — Shapiro, Susan P., The Social Control of Impersonal Trust, 144