Protective Puts and Collars

A protective put is not a hedge, it is insurance

A protective put pairs shares with a long put at a strike below the current price, so the position has a floor: below the strike, the put gains what the shares lose, and the loss is capped at the distance from the entry to the floor plus the premium. What it does not do is make the position safe, and the distinction matters because the two are conflated constantly. The put does not remove the risk of the shares falling to the strike — it removes the risk of falling *below* it. Between the current price and the strike the holder is unhedged, and the premium is paid either way, so a stock that drifts down 6% into the floor loses the same 6% plus the cost of the protection. The pricing is therefore best judged the way any insurance is judged: as a recurring cost against a loss that may not arrive. The worked case is two hundred shares worth $13,600 with a $2.10 put, which costs $420 and covers two contracts — 3.09% of the position for the period. The relevant comparison is not the loss it might prevent but the drag it certainly creates: if the put is rolled every quarter at a similar price, the position pays a little over 12% a year for a floor two dollars below the market, and it must earn that before it earns anything for the holder. That is not an argument against protection; it is an argument that protection has to be either cheap (a wide strike, a short period, a credit from a sold call) or paid for by income the position generates (dividends, covered calls, or the put-writing that the wheel rotates into). The time horizon matters more than the strike, and this is where most protective-put decisions actually go wrong. A put bought to cover an event — an earnings report, a court date, a policy meeting — is a cheap, dated hedge whose cost is comparable to the uncertainty it covers. A put bought to cover “the next year” is a rolling position whose accumulated cost is comparable to the drawdown it protects against, and it will be renewed or abandoned precisely at the moment it is most expensive, because implied volatility is highest when the market has already fallen. That is the structural weakness of long-dated protection: it is cheapest to buy when it is least needed and most expensive when it is most wanted, and a plan that commits to a strike, a period and a budget in advance is the only way to avoid buying high and abandoning low. The floor, priced — Position: 200 shares at $68.00 = $13,600 · $65 put at $2.10: $420, or 3.09% of the position for the period ← · Worst case with the put: A $13,000 floor less the $420 premium — about 4.5% below the current price · Annualised if rolled quarterly: A little over 12% a year in protection cost Read the second row again: the floor sits below today’s price, so the position can still fall to the strike unhedged. The put caps the tail rather than the loss, and the premium is paid whether or not the tail arrives.

The collar: selling the ceiling to buy the floor

A collar pairs the protective put with a short call above the market, so the credit from the call reduces or eliminates the cost of the floor. In the worked case the put costs $2.10 and the $75 call pays $1.35, so the net cost is $0.75 a share, or $150 on two hundred shares — a floor two dollars below the market for a seventh of the cost. The economics are not magic, and stating them plainly is the whole value of the lesson: the trader has given up everything above $75 to pay for protection below $65, so the structure is a bet that the stock stays between the two, or at least that the floor is worth more than the ceiling being sold. At $80 the position has forfeited (80 − 75) × 200 = $1,000 of gain — more than six times the cost of the floor it financed — which is the number that makes the trade-off real rather than theoretical. That comparison gives the two conditions for a collar to be sensible. The first is the covered-call test from O13, applied to the sold strike: the trader must be genuinely content to have the shares called away at $75, because the collar will deliver exactly that in a rally. The second is that the floor must be somewhere the position is willing to be protected — a strike below the entry makes the collar a loss-limiting structure, while a strike above the entry locks in a gain and turns the collar into a way of monetising a tax or a valuation decision. Which of the two the trader wants changes the strikes, so writing the intention down first is not bookkeeping; it determines the structure. Two practical notes complete the picture. A collar whose strikes are close around the current price is called a fence or a risk reversal depending on the construction, and the tighter the collar the cheaper the floor and the smaller the range the position can profit in — the same trade-off, dialled. And a collar has a real cost beyond the payoff: it caps participation, which over long bull markets is a systematic drag, and it is usually executed as a multi-leg order whose spreads have to be paid on entry and exit. The honest summary is that a collar is a decision to convert a stock position into a bounded one for a period, which is exactly right for a position the holder wants to keep but cannot afford to see collapse, and exactly wrong for one they believe will compound. • Collar = long put + short call: the ceiling pays for the floor. • Look at the forfeited upside in dollars, not in percentages, because that is what pays for the protection. • Only sell the call at a strike at which being called away would be acceptable. • A below-entry floor limits loss; an above-entry floor locks in gain — the intention decides the strikes. • Caps are a systematic drag in a bull market, and multi-leg orders pay the spread twice. The error to avoid is treating a collar as a free hedge because its net cost is small. The premium is being paid in upside rather than in cash, and the size of that payment scales with the rally — which is the scenario in which the holder most wanted to own the position. Before entering, compute the forfeited gain at a 20% rally and compare it with the protection the floor actually provides at a 20% decline. If the two are not comparable in the holder’s own terms, the collar is being sold on the basis of its cash cost rather than its economics.

What protection actually costs per year

A protective put’s premium is quoted as a percentage of the position, and the number that decides whether the programme is worth running is the same premium expressed as a rate — what the protection costs per year, against the downside it removes. That conversion turns an appealing-sounding quarterly cost into a hurdle the portfolio has to clear every year simply to get back to where it started. The arithmetic is short and the result is usually higher than expected. A put costing two percent of the position for three months is roughly eight percent a year if rolled four times, before any compounding of the drag — and the position has to earn more than that in an ordinary year to break even relative to holding the stock uninsured, which is a demanding requirement for an equity portfolio with a long-run return in the high single digits. The cost whittles down in the money as the strike falls, since a lower floor is cheaper, and it rises sharply as the strike approaches the spot. The reason it feels cheaper is a specific feature of how the option market prices it. Downside puts usually carry higher implied volatility than calls at the same distance from the spot, and that **skew** is the market charging a premium for crash protection. The consequence is that the cost of protecting against the event a household actually fears is systematically higher than a symmetric measure of volatility would suggest — the insurance is expensive exactly because everyone else also regards the tail as the risk worth insuring. The seller on the other side of that trade is collecting a premium for bearing the tail, which is the mirror of the carry and short-volatility observations elsewhere in the curriculum. That is why the honest comparison is against the alternatives rather than against nothing. A **stop order** costs nothing in premium and fails exactly in the case a put is for, because it cannot fill through a gap. A **collar** sells the upside to fund the floor, which makes it nearly free and converts the position into a range — a real choice, and one that should be made explicitly rather than discovered. A **cash allocation** funded by selling part of the position produces a floor without a premium, at the cost of permanently lower exposure. And **reducing the position size** achieves a better risk outcome than hedging at almost any horizon, because it costs nothing and works in every state of the world, which is why sizing is upstream of hedging in this curriculum. The put is the right instrument when the exposure must be kept and the tail must be bounded over a specific window; it is the wrong instrument when the honest answer is that the position is too large. The practical form of the decision is a written sentence: over what period, against what level, for what cost as a percentage per year, and what alternative that money could fund. A protection programme that cannot be described that way is usually an expense being paid for reassurance, and reassurance is available for less. • Convert the premium into an annual rate: three months at two percent is roughly eight a year. • Skew makes the tail the expensive thing to insure, precisely because it is the tail. • A collar is nearly free because the ceiling is sold, which is a trade rather than a saving. • Resizing the position beats hedging it for almost any horizon, and costs nothing. A useful check before rolling protection for another quarter: add the premiums paid over the last two years and compare the total with what the position is worth today. A programme that has cost a tenth of the account and has not been used is a decision, and it is worth making deliberately at least once a year.

Sources

• Protective puts, collars and the cost of insuranceMcMillan, “Options as a Strategic Investment”
• Hedging costs and the drag on long-run returnsIsraelov & Nielsen, “Covered Calls Uncovered” (Financial Analysts Journal)
• Pre-commitment and the price of tail protectionTaleb, “Dynamic Hedging” (practical hedging sections)