A hedge is a purchase with three prices, and the discipline is to compute all three before deciding.
Premium drag. A standing protective-put program at Worked example 1's parameters (28% volatility, 6.5% rates, quarterly rolls) costs 2.82% of NAV per quarter for a 5%-out-of-the-money put, about 11.3% per year; a 10%-out-of-the-money put costs 1.44% per quarter, 5.8% per year. Those are not rounding errors on an equity portfolio's expected return; they are most of it. Permanent insurance on an appreciating asset is almost always a losing proposition, which is why professional put use is episodic and event-driven and why collars dominate in practice.
Opportunity cost. A covered call's cost is invisible until the stock runs. In practice problem 15(a), a ₹3,200 stock with a ₹3,400 call sold for ₹86 that finishes at ₹3,750 earns ₹286 instead of ₹550: ₹264 per share forgone, 8.25% of the position, and the premium looked like free money right up until it wasn't.
Basis and residual. The mid-cap hedge in the basis risk section removes 51.84% of variance; if it costs anything in fees, rolls or margin you are paying that for half a hedge, and the right comparison is against simply owning less.
Advertisement
When is hedging worth the cost?
When hedging is genuinely right: and this program's answer is narrower than the industry's: when the exposure is involuntary (a concentrated legacy holding you cannot sell without a tax event; a contracted receivable; a floating-rate liability), when the horizon is defined, and when the residual after hedging is smaller than before by an amount worth its price. Hedging a diversified portfolio you chose to own, permanently, is usually a confession that the position is too big, and the cheaper fix is M9.01's: own less.
{
"skill": "Building and checking an option payoff / breakeven table for a multi-leg structure",
"use": "Have an assistant lay out the leg-by-leg payoff table and the closed-form breakeven, maximum-profit and maximum-loss expressions for a structure you specify, so you have a second opinion to check your own by-hand table against — never a first opinion to copy.",
"tools": ["Chat assistants — Claude, ChatGPT, Gemini", "Code-execution tools — an assistant that writes and RUNS Python (numpy/pandas), so the arithmetic is executed rather than predicted", "Spreadsheet AI — Excel Copilot or Google Sheets AI, over a grid you built"],
"prompt": "For this position: <LEGS with quantities, strikes, premiums, multiplier>. Build a table with one row per underlying price from <LOW> to <HIGH> and one column per leg, showing each leg's payoff and the net profit. Then give the breakeven(s), maximum profit and maximum loss as algebraic expressions in the strikes and premiums, and evaluate them. Do NOT round intermediate values. State every sign convention you used.",
"verify": "Recompute at least three rows by hand, including one in every linear region and one at each kink; check the degenerate case (set both strikes equal — a collar must collapse to the risk-free forward, a spread to zero width); confirm the contract multiplier was applied exactly once; and check the structure against put-call parity. Any market data (spot, premiums, lot sizes, contract specs) must be traced to a named primary source — the exchange's own contract-specification page or an official quote — not to the model (Guardrail 1-5, especially #1 Trace and #3 Recompute).",
"diy": "You must be able to write the payoff function and solve for the breakeven from Rule 1 and Rule 2 without help. This is a gated skill: the model's table is a cross-check on yours, and if you have no table of your own there is nothing to cross-check.",
"market": "IN"
}