Risk management has a public-relations problem. In a drawdown, everyone wants it. In a calm market, it looks like friction—limits that cap upside, checklists that slow trades, reports that nobody reads. I have come to believe that the quiet periods are exactly when risk work either compounds or decays.
When volatility is low, three illusions grow. The first is that leverage is free. Tight spreads and easy financing invite larger notionals for the same dollar of day-to-day noise. The second is that concentration is insight. A few positions that “always work” begin to dominate P&L, and the book starts to resemble a personality more than a portfolio. The third is that operations can wait. Documentation, reconciliation, key-person backups, and scenario drills feel optional until they are not.
Good risk practice in calm weather is mostly unglamorous. It is refreshing limit frameworks while people still listen. It is measuring gross and net exposures, not only mark-to-market volatility. It is asking whether liquidity assumptions still match position size. It is testing what happens if a prime broker, exchange, custodian, or data vendor fails for a day. None of those exercises predict the next crisis. They reduce the chance that the next crisis is also an internal organisational failure.
I like to separate market risk from process risk. Market risk is the unavoidable consequence of seeking return. Process risk is optional: unclear mandates, inconsistent valuation, unreviewed models, and decision rights that shift with whoever speaks loudest. Process risk is what turns a bad month into a franchise event. In calm markets, process risk is cheap to reduce and easy to ignore.
Position sizing deserves special attention when realised volatility is suppressed. Scaling into “comfortable” risk using recent noise can produce positions that are violently uncomfortable when noise normalises. A simple discipline helps: size as if tomorrow’s volatility could jump toward a longer-history percentile you actually respect. If that size feels too small to matter, the strategy may not be robust enough to deserve capital.
Hedging also changes character in calm regimes. Cheap hedges can be worth buying when complacency is high; expensive hedges can be worth skipping when fear is already priced. That sounds obvious until you watch institutions buy protection after the move and call it prudence. Timing hedges is hard. Having a standing policy for when protection is a budgeted cost of doing business is easier than improvising under adrenaline.
Communication is part of risk management. If the investment committee only hears from risk after losses, risk becomes a courtroom. If risk joins idea generation early—challenging liquidity, crowding, and tail assumptions—it becomes a design partner. Calm markets are when that cultural habit can be built without blame.
Digital-asset sleeves make the calm-period problem sharper. Continuous trading, venue fragmentation, and operational key management create failure modes that traditional desks sometimes underweight. A risk framework that only watches price beta is incomplete. Wallet controls, withdrawal policies, oracle dependence, and counterparty concentration belong on the same dashboard as market risk. Again: boring until it is not.
My practical advice is not to romanticise stress tests as prophecy. Use them as rehearsal. Write down who can cut risk, under what authority, with what tools, and how quickly capital can be raised or returned. Then run a short drill twice a year while people are still polite. The goal is muscle memory, not theatre.
When markets are calm, the temptation is to declare victory and loosen the bolts. The better instinct is to treat calm as inventory time: check the bolts, label the tools, and make sure the people who will need them still know where they are.
Governance is risk management’s skeleton. Who can approve exceptions? How are they recorded? How long can an exception live before it becomes the new normal? Calm markets are when exceptions proliferate because “nothing bad happened.” That is how limits die by a thousand waivers.
Model risk deserves a calendar slot too. Pricing models, VaR engines, and now machine-learning signals all embed assumptions. Schedule a review of those assumptions while P&L is dull. Waiting until the model breaks in a crisis guarantees you will rebuild under panic.
Vendor and data dependencies are easy to undercount. A single delayed feed can freeze a desk that believes itself sophisticated. Map critical paths. Keep a manual fallback that is ugly but workable. Resilience is not elegance.
Above all, keep teaching. New joiners inherit culture faster than documentation. If seniors treat risk as optional bureaucracy in bull markets, juniors will treat it as optional forever.