Reflexivity in Markets

Beliefs can become part of the facts when they change funding, demand, collateral, or operating capacity—and the changed facts then reshape belief.

Reflexivity needs a transmission channel

A market price does not automatically change a company’s factories, customers, or cash. It can matter when the company can issue shares, use its stock to buy an asset, borrow against collateral, attract employees with equity, or gain credibility with suppliers and customers. Likewise, a bank’s reputation matters when depositors can withdraw funds and lenders can change terms. The belief becomes economically relevant through an action that changes the system.

George Soros used “reflexivity” for this two-way relationship between participants’ views and the situation they are trying to understand. The concept is broader than a price being wrong and narrower than the claim that all markets are self-fulfilling. The analyst should identify what perception changes, who acts on it, and which later observation shows the effect.

Positive and negative loops

A positive loop can begin with confidence. A company with a rising share price may issue equity on better terms, fund a useful investment, and attract people who improve the business. A lender that is viewed as safe may retain deposits and borrow cheaply, making its liquidity position stronger. The improvement is real, but it depends on the financing or demand channel remaining open and on the investment producing a return.

A negative loop works in the opposite direction. A rumour can cause customers to delay orders, employees to leave, lenders to shorten terms, or depositors to withdraw. The resulting cash pressure or service deterioration can make the rumour more credible. The loop can accelerate even when the original belief was exaggerated. This is not proof that the belief caused the entire failure; underlying weaknesses may have made the system sensitive to the shock.

Silicon Valley Bank shows the boundary

The Federal Reserve’s review of Silicon Valley Bank documents rapid deposit outflows, interest-rate risk, concentrated funding, and supervisory failures before the bank’s March 2023 collapse. The episode illustrates a reflexive channel: once depositors and counterparties lost confidence, withdrawals and funding pressure changed the bank’s liquidity conditions. It does not establish that sentiment alone caused the failure; the bank’s asset duration, uninsured deposits, risk management, and supervision were part of the underlying vulnerability.

The case is useful because it keeps perception and reality distinct. A withdrawal is an observable transaction. “The bank was doomed because confidence fell” is an interpretation. The analysis becomes stronger when it links the withdrawal, liquidity position, asset sales, communications, and regulatory action in sequence.

Where reflexive channels appear

• Funding. A higher share price or tighter credit spread can reduce the cost of capital, but only if the company can raise or refinance money on those terms.
• Collateral. Rising property or asset prices can support more borrowing, which can increase demand and prices; falling values can force deleveraging.
• Customers and suppliers. Confidence can improve bookings, payment terms, and willingness to commit; fear can delay orders or demand cash in advance.
• Labour and partners. A perceived winner may attract people and counterparties. A perceived failure may lose them before financial statements show the change.
• Market structure. Investor flows can expand or contract liquidity, affecting which prices are available and which firms can finance growth.

Each channel has a different duration and limit. A stock price cannot manufacture skilled workers immediately. Collateral cannot support unlimited borrowing. A customer may return after a temporary disruption, while a lost permit or supplier may take years to replace. Mapping the channel prevents “reflexivity” from becoming a story applied after every price move.

When the loop breaks

Loops weaken when cash runs out, lenders impose covenants, customers find alternatives, employees stop joining, regulation changes, or the physical capacity needed to deliver growth is exhausted. The loop can also reverse because the original belief was disproved by a product failure or financial report. Positive feedback does not remove constraints; it can move a system toward them faster.

There is no universal boom-bust timetable. A loop may persist for years, reverse abruptly, or be offset by a second channel. Nor is the downside necessarily faster than the upside in every market. The speed depends on liquidity, leverage, contract terms, information, and who can exit first.

What the concept cannot establish

• A rising price does not prove that the company gained operational value. The company must actually issue, invest, hire, or sell more.
• A falling price does not prove that a self-fulfilling collapse will occur. Cash, contracts, and physical capacity may remain intact.
• A feedback loop does not identify the original cause. It can amplify a weakness that existed before the belief changed.
• Market capitalization is not cash. A company cannot spend an unissued share price, and an issued share can dilute existing holders.
• Reflexivity is not a timing signal. Identifying a loop does not say when it will strengthen or reverse.

What investors can test

• Draw the loop in observable steps: belief, action, changed condition, and new belief.
• Measure the resource being changed—funding cost, deposits, orders, employees, collateral, liquidity, or customer retention.
• Separate direct evidence from interpretation. Price, withdrawal, spread, and cash are observations; “confidence caused the crisis” is a conclusion to test.
• Identify the constraint that can break the loop: cash, covenant, capacity, regulation, competition, or customer alternative.
• Check whether management has funding and authority to preserve the useful part of a positive loop without relying on continual market enthusiasm.

Reflexivity matters when expectations become an operating input. The investor’s task is to keep the feedback visible without confusing a powerful channel with an unlimited force.