What the book covers

The book discusses credit cycles and abrupt market changes at six levels:

1. Debt and output. Minsky’s financial instability hypothesis and the differential-equation models of Goodwin and Keen: under what conditions a boom pushes debt beyond what can be serviced.
2. Self-organised criticality. How the sandpile model of Bak, Tang and Wiesenfeld produces “avalanches” of every size without anyone tuning a parameter; how to estimate a power-law exponent correctly; and how much evidence there is for “bigger crises after long calm” as a hypothesis.
3. Liquidity spirals. How margin requirements and falling prices amplify each other; and how much of “correlations go to one in a crisis” is real and how much a statistical artefact.
4. Volatility clustering. Why large moves tend to follow large moves, and where a constant-volatility normal model of risk goes wrong.
5. Implied volatility and skew. Recovering implied volatility with Newton’s method; how a price process with jumps makes out-of-the-money puts more expensive (the “skew”); and mean reversion of a volatility index.
6. The early-warning controversy. What critical slowing down and log-periodic power laws can and cannot do on simulated data.

Chapter 7 reads this modern research beside the Book of Changes’ hexagram Tai (“there is no level without a slope, no going without a return”; “the city wall falls back into the moat”), the Daodejing’s “reversal is the movement of the Way” and “things that grow strong grow old”, and the Discourses on Salt and Iron’s debate on price stabilisation. Chapter 8 offers a four-step checklist for observing the state of the market and the fragility of one’s own portfolio, and a pre-registered randomised controlled experiment.

Claims corrected on checking

Several claims behind this book needed correction once they were checked. The periodic clearing of credit cycles is not an “arithmetic necessity”: the Goodwin–Keen model shows one mechanism, and from nearby starting points the same model can converge to a steady state with finite debt or move towards exploding debt, depending on parameters and initial conditions. The sandpile model is not a model of markets, the fat tails of financial returns have other explanations, and “markets are in a self-organised critical state” is only an analogy and a hypothesis. “Long suppression of small volatility builds up bigger structural tension” has some historical support but should still be treated as a hypothesis. Part of the rise in correlation during crises is a statistical artefact of selecting volatile periods. The implied-volatility skew of options reflects risk-neutral prices, which include a risk premium, and cannot be read directly as “the market’s probability of a crash”. Log-periodic power laws are easy to fit after the event, and accurate timing in advance is statistically very unreliable. Each chapter says so where it arises, and the Editor’s Notes published with the book files collect these points.

How to read the boxes

As in other 5D Library books, different kinds of statement are marked separately.

About this edition

This is version 1.0.0, the first complete edition: a preface and eight chapters, in Chinese and English. Corrections and missing references are welcome through the 5D Library website.