Skip to content
WealthTicker
Learn · Module 5 — Advanced metrics, taxation and wealth architecting

The psychology of a market crash: behavioural finance that survives contact

Loss aversion, herding and action bias are not character flaws — they are the default settings. The pre-commitments that work, because judgement in the moment does not.

Last reviewed 21 Apr 2026

Every investor knows they should buy low and sell high. In a crash almost nobody does, and the ones who fail are not stupid — they are running perfectly normal human software on a problem it was not built for.

Understanding the specific failure modes is the only defence that works, because the one thing you cannot do in the moment is reason your way out.

Why a fall feels worse than a rise feels good

Loss aversion is the foundational finding: a loss registers roughly twice as strongly as an equivalent gain. A portfolio down 20% does not feel like the mirror image of one up 20%. It feels about twice as urgent.

That asymmetry is why “just hold on” is such weak advice. It asks you to override a response that is doing exactly what it evolved to do.

Layered on top:

  • Recency bias. Whatever is happening now feels permanent. In March 2020 the fall felt endless; by late 2021 the rise did.
  • Herding. Everyone selling is powerful evidence that selling is correct — and it is the single least reliable signal available, because it peaks at the bottom.
  • Availability. Vivid, repeated headlines make catastrophe feel probable. The coverage is loudest precisely when the risk has already been priced in.
  • Action bias. Doing nothing feels negligent. In investing it is usually the highest-value action available, and it is the hardest to take.

The two mistakes that do the damage

Mistake one: selling into the fall. This converts a temporary decline into a permanent loss. The units are gone and do not participate in the recovery.

Mistake two, which is worse: not coming back. Having sold, the investor waits for “clarity”. Clarity arrives well after the recovery, so they re-enter higher than they left. This is buying high and selling low with extra steps, and it is the largest documented source of the gap between what funds return and what investors actually earn.

There is a third, quieter one: stopping the SIP. A SIP’s entire mathematical benefit comes from buying more units when prices are low. Pausing during a fall means the mechanism is switched off in the only period it was going to help — see rupee-cost averaging.

The defences that actually hold

Willpower is not one of them. What works is deciding in advance, while calm:

  1. An allocation you can survive. If a 40% equity fall would force or panic you into selling, your equity share is too high — regardless of what a risk questionnaire said. This is the real job of asset allocation.
  2. An emergency fund that is not equity. Most forced selling is not sentiment; it is a job loss meeting a portfolio. Six months of expenses in a liquid or short-duration fund removes the mechanism.
  3. Automation. A SIP that continues by default requires an active decision to stop. Make the good behaviour the path of least resistance.
  4. A written investment policy. One page, written in a calm month: what you own, why, and what would legitimately make you sell — the genuine reasons, not “it fell”. Read it during a crash instead of the news.
  5. Rebalancing as a rule, not a judgement. Pre-committed bands force you to buy the asset that fell without anyone having to feel brave about it.
  6. Look less. Loss aversion is triggered by observation. Checking a portfolio daily during a fall guarantees you experience every decline and none of the long-run compounding.

Pitfalls to avoid

  • Do not confuse “I have a high risk appetite” with capacity. Appetite is measured on a form in a bull market; capacity is measured by cash flows and horizon.
  • Do not wait for the bottom to re-enter. It is only identifiable afterwards. If you must move, move on a schedule rather than a view.
  • Do not treat a crash as evidence your fund is bad. Everything in the category fell. Compare against the peer group, not against zero.
  • Do not go looking for confirmation. In a fall, every forecast you seek out will agree with your fear, because that is what gets published.

Key takeaway

A crash is not primarily a test of analysis; it is a test of pre-commitment. Every decision that protects you must be made before the fall, because during it your judgement is measurably impaired and you will not notice. Set an allocation you can hold through a 40% decline, keep a non-equity emergency buffer, automate the contributions, write down what would genuinely justify selling — and then, when it happens, do considerably less than feels right.

More in Module 5 — Advanced metrics, taxation and wealth architecting