How the market’s response mechanisms become drivers of the movements they respond to
The Channel
Rain falls on a hillside. At first, the water flows wherever the slope allows, spreading across the surface in no particular pattern. But as it flows, it erodes. Small channels form where the water concentrates. Once formed, the channels direct more water into themselves, which deepens the erosion, which directs still more water. The channel creates the conditions for its own reinforcement.
The original rainfall did not determine where the channels would form. The interaction between water and terrain did. And once the channels exist, they shape the path of every subsequent rainfall. The structure that emerged from flow now directs flow.
Markets exhibit the same circularity. Price movements trigger responses. Those responses alter structure. The altered structure shapes subsequent price movements. Cause and effect loop back on themselves until the distinction between them dissolves.
This is not merely a metaphor for market behaviour. It is a description of how markets actually work.
From Response to Cause
The earlier articles in this series described structure as something that shapes response: memory determines how the market reacts to shocks, attractors pull the system toward characteristic states, liquidity conditions determine whether disturbances are absorbed or amplified.
All of that remains true. But it is incomplete.
Structure does not merely respond to price. Structure moves price. The mechanisms that participants use to manage risk, hedge exposure, and target volatility are not passive observers of market dynamics. They are active participants whose responses feed back into the system they are responding to.
When a dealer hedges an options position, the hedge itself moves the market. When a volatility-targeting fund reduces exposure, the reduction itself increases volatility. When a risk parity strategy deleverages, the deleveraging itself creates the losses that triggered it.
The response is not separate from the cause. The response is part of the cause.
Dealer Hedging as Feedback
Options dealers do not hold directional views. They manage inventory. When they sell options to customers, they hedge their exposure by trading the underlying. The direction and intensity of their hedging depends on their aggregate position, specifically on their gamma.
When dealers are long gamma, they hedge by buying as prices fall and selling as prices rise. This is stabilising feedback. Their activity dampens movements, adding liquidity when the market moves and pulling prices back toward equilibrium.
When dealers are short gamma, the dynamics invert. They must sell as prices fall and buy as prices rise. This is destabilising feedback. Their activity amplifies movements, removing liquidity precisely when it is needed and pushing prices further in the direction they are already moving.
The market does not know or care why dealers are hedging. It only experiences the flow. When dealers are short gamma and prices begin to fall, their hedging adds selling pressure, which pushes prices lower, which requires more hedging, which adds more selling pressure. The feedback loop runs until something breaks the cycle: a change in positioning, exhaustion of the move, or intervention from other participants.
This is not a malfunction. It is the mechanical consequence of how options are hedged. The structure of dealer positioning becomes a cause of price movement, not merely a response to it.
Volatility Targeting as Feedback
Many institutional strategies target a constant level of portfolio volatility. When realised volatility rises, they reduce exposure to bring risk back to target. When realised volatility falls, they increase exposure.
In isolation, this is prudent risk management. In aggregate, it creates feedback.
When volatility spikes, volatility-targeting funds sell. Their selling increases volatility further, triggering more selling. The feedback loop amplifies the initial shock, not because any participant intended amplification, but because individual responses aggregate into a collective pattern. The same dynamic works in reverse: low volatility encourages buying, which suppresses volatility further, extending calm and building the conditions described earlier in the series.
Volatility targeting does not predict volatility. It participates in creating the volatility it responds to.
How Feedback Loops Couple
No single feedback mechanism dominates the market. Dealer hedging, volatility targeting, risk parity, trend following, and passive rebalancing all operate simultaneously. Their effects interact.
When these mechanisms align, feedback intensifies. A price decline triggers dealer selling, which increases volatility, which triggers volatility-targeting selling, which deepens the decline. The loops couple through shared sensitivity to price and volatility: the metronome effect operating through feedback.
When these mechanisms oppose, feedback dampens. Dealer long-gamma hedging may absorb selling from volatility targeters. Passive rebalancing may buy into declines that trend followers are selling. The market absorbs disturbances more smoothly.
The net effect depends on the configuration of all active feedback mechanisms at the moment of disturbance. This configuration is not directly observable, but its consequences are: the amplification zone from Article 4, the liquidity withdrawal from Article 5, the regime transitions from Article 6.
Feedback is the mechanism through which structure becomes cause.
Stability and Instability
Feedback is not inherently destabilising. It can work in either direction.
Negative feedback stabilises. A system with negative feedback resists displacement. Push it away from equilibrium and the feedback pushes back. Dealer long-gamma hedging is negative feedback. Mean-reversion strategies are negative feedback. These mechanisms absorb disturbances and return the system to its prior state.
Positive feedback destabilises. A system with positive feedback amplifies displacement. Push it away from equilibrium and the feedback pushes further. Dealer short-gamma hedging is positive feedback. Volatility targeting during stress is positive feedback. Margin calls that force liquidation are positive feedback. These mechanisms take small disturbances and make them large.
The balance between positive and negative feedback determines whether the market absorbs shocks or amplifies them. When negative feedback dominates, the market is resilient. When positive feedback dominates, the market is fragile.
The same market can shift from one state to the other depending on positioning, leverage, and the structural conditions that determine how feedback mechanisms interact.
Reading for Feedback Conditions
Feedback loops are not directly visible. But their preconditions are partially observable.
Dealer gamma positioning can be estimated from options open interest and market maker behaviour. When aggregate gamma is deeply negative, the conditions for amplifying feedback are present.
Volatility-targeting assets under management and their current exposure levels indicate how much mechanical selling a volatility spike might trigger.
Leverage in the system, visible through margin debt, fund leverage ratios, and risk parity allocations, indicates how much deleveraging a drawdown might force.
These estimates are imprecise. The data is incomplete, the models are approximate, and the thresholds that trigger feedback are not fixed. But imprecise orientation is better than none. You are mapping the channel, not predicting the rain.
None of this tells you when feedback will activate. It tells you what will happen if it does.
Implications for Architecture
Understanding feedback changes how you think about market risk.
Risk is not just about exposure to price movement. It is about exposure to the feedback mechanisms that will activate when prices move. A position that seems moderate in calm conditions may become dangerous if it sits in the path of amplifying feedback.
Timing matters structurally. The same price level can be safe or dangerous depending on whether the market is approaching it through positive or negative feedback. A gradual drift differs from a feedback-driven cascade, even if they end at the same price.
Independence from feedback flows is valuable. If your stops, rebalancing rules, or risk triggers align with the mechanical flows of large feedback mechanisms, you are not independent. You are part of the channel.
Survival requires designing for the feedback regime you are not currently in. Calm conditions do not test resilience to positive feedback. Stressed conditions do not test capacity to add exposure into negative feedback. Architecture must account for both.
The Channel and the Rain
The channel does not cause the rain. But once the rain falls, the channel determines where the water goes.
Feedback loops in markets work the same way. They do not cause the initial disturbance. But once the disturbance arrives, the feedback structure determines whether it is absorbed or amplified, contained or cascading.
Structure is not backdrop. It is mechanism.
The question is not just what will happen to prices. The question is what feedback will do to whatever happens.
This is the eighth article in a series exploring the deep structure of markets. Next: “The Regime Shift: Recognising Transition in Real Time”