Business
People inside a situation use their thinking for two jobs at once: to understand what is happening and to change it. Because the same minds do both, what they believe feeds into what actually occurs. The book calls these the cognitive and the participating functions.
The author holds that market prices are always somewhat distorted, and that the distortion can change the underlying values it is supposed to reflect. A price is not a passive mirror of a company or a currency. It can push the real economy toward the price, as when exchange rates end up shaping the fundamentals instead of being shaped by them.
Where there are thinking participants, cause does not run straight from one fact to the next. It runs from fact to perception and from perception back to fact. Each turn adds connections that the facts alone don't show. This two-way loop is the core of what the author calls reflexivity.
In the author's view, financial history has two sets of actors: the competitors in the market and the regulators who oversee them. Each reacts to the other in a loop of its own. He argues that markets do not settle into balance on their own, because stock booms always ride on expanding credit and only some form of regulation can stop excessive instability.
Standard models assume markets only anticipate the future. The author argues that markets can also trigger events or prevent them from happening. A market's expectation is therefore one of the forces deciding what comes next, not just a forecast of it.
Markets resemble science in that both form hypotheses about the future and let events test them, keeping the survivors and discarding the failures. The difference is the standard of success: science aims at truth, while markets reward whatever works. Practical success can come without real understanding, which is why the author calls finance alchemy rather than science.
The author takes the opposite of the usual view and assumes markets are always wrong. In his account, the market's real merit is not that it allocates resources perfectly. It is that it gives participants a test by which they can spot their own misconceptions, and he credits markets with keeping his sense of reality intact.
The author describes his success as owing little to forecasting. By the time he explained a trend, it had often already changed, so he kept forming new hypotheses and often held two partly contradictory ones at once. What he relied on was a testable idea plus a ready plan for when events proved it wrong, not a confident prediction.
Because belief and reality reinforce each other, the author argues that self-reinforcing trends are common rather than rare. They build up for a while and then defeat themselves. Credit and collateral are a key example: more lending raises the value of what backs the loans, which justifies still more lending, until the loop runs in reverse.
The author says booms and busts are not mirror images. At the start of a boom, both the amount of credit and the value of the collateral behind it are at their lowest. At the start of a bust, both are at their peak. So the unwinding starts from maximum exposure and falls much harder than the rise.