Collar

Definition

A collar is long stock plus a protective put below the market plus a short call above it: the call premium finances some or all of the put’s cost, bracketing the position between a floor and a cap through expiration. A zero-cost collar sets the strikes so the two premiums offset. It converts open-ended equity exposure into a defined band — protection paid for with upside.

How it works / structure

• Legs: +100 shares, +1 put K_p (below spot), −1 call K_c (above spot), same expiry.
• Payoff at expiry: floored at K_p − net cost, capped at K_c + net credit (both relative to basis); between the strikes, stock P&L adjusted by the net premium.
• Parameters (engine-executable): put strike (floor distance — % OTM or delta), call strike (cap distance), expiry, net-cost target (zero-cost vs debit-tolerant), roll policy at expiry (mgmt-rolling — a standing collar program re-strikes each cycle), dividend calendar check (short call assignment risk — event-dividends-ex-dates).
• Skew economics: equity smirk means the put bought is IV-richer than the call sold (opt-volatility-skew) — zero-cost collars on equities cap more upside than the downside they floor, an asymmetry that IS the skew premium being paid.

When it applies

Concentrated positions that cannot be sold (restrictions, tax, conviction) but need drawdown bounds (risk-max-drawdown-budget), pre-event protection on holdings, and portfolio-level drawdown control implemented per position. The band’s cost is upside — the structure suits holders whose priority is the floor.

Risk profile & failure modes

• Cap regret dominates realized outcomes: equities spend more time rising than crashing; the modal collar outcome is upside surrendered, which quietly compounds against the position over repeated cycles.
• Band-gap losses persist: between strikes the position is still equity — a grind down to just above the put strike is a real loss the floor never triggers on.
• Skew drag: perpetual zero-cost collaring systematically sells cheap IV and buys rich IV; the Szado-Schneeweis QQQ study found implementation details (strike width, tenor) drive results materially.
• Early assignment on the call near ex-dates unwinds the structure at the cap prematurely (ms-expiration-exercise-assignment).

Evidence & limits

Mechanics are contract arithmetic (OCC). Szado-Schneeweis (2010) documented that collar performance varies widely with implementation and period — protective in drawdown-heavy samples, a drag in appreciating ones; there is no evidence collars add return, only that they reshape the distribution. “Free insurance” framing of zero-cost collars is folklore — the cost is the cap and the skew spread.

Falsifiable-thesis examples

Illustrations only, not signals:

• “X, collared at K_p/K_c for the next quarter, will finish inside the band (neither leg in the money at expiry)” — falsified by an outside finish.
• “A rolling quarterly zero-cost collar on Y will show a smaller maximum drawdown but lower total return than unhedged Y over the next year in replay” — falsified by either half of the pair.

Cross-references

• Components: strategy-covered-call (the cap side), strategy-bear-put-spread (spread-financed floor alternative)
• Pricing asymmetry: opt-volatility-skew
• Purpose framing: risk-max-drawdown-budget, risk-scenario-analysis
• Management: mgmt-rolling, mgmt-assignment-handling