Deciding What Actually Needs Hedging

Every position carries more than one risk at once. Buy a semiconductor stock on a thesis about its next product cycle, and you've also bought exposure to the whole sector's cyclicality, to the broad market, to the dollar if the company's revenue is mostly overseas, and to interest rates if it's a growth name whose valuation leans on future cash flows. Hedging is not about eliminating all of that. It's about deciding which of those risks you're actually being paid to hold, and getting rid of the rest.

The dividing line: rewarded versus unrewarded

A risk is rewarded if you have a specific view that should make you money for bearing it. The product-cycle thesis is rewarded — you did research, you believe something the market doesn't yet, and if you're right you get paid for having taken that specific risk.

The sector beta that came along for the ride is, usually, unrewarded. You don't have a semiconductor-sector view; you have a single-company view. If the whole sector falls 8 percent on a macro scare unrelated to your company's product cycle, that loss buys you nothing — you weren't trying to bet on the sector, and losing money on a bet you never intended to make is pure waste, win or lose.

The test is simple to state and takes discipline to apply honestly: for each risk the position carries, ask "did I do work on this specific question, or did it just come along with the trade I actually wanted?" Hedge the second kind.

Why over-hedging is also a mistake

Hedging costs money — transaction costs, bid-ask spread, sometimes negative carry on the hedge instrument itself — and it dilutes conviction. A trader who hedges every risk a position could conceivably carry, including risks with genuine, if smaller, expected value, ends up with a position so neutralized that even being completely right about the original thesis barely moves the P&L. The goal is not zero risk. It's risk that matches your actual views, no more and no less.

Worked example

You're long $5m of a mid-cap industrial name on a thesis that a new contract win will re-rate the stock over the next two quarters. Breaking down what that $5m position actually carries:

• Company-specific catalyst risk (the contract, the re-rating): this is the trade. Not hedged — it's the whole point.
• Sector beta (industrials as a group): you have no industrials view. Hedge with a sector ETF or futures short sized to the position's estimated sector beta.
• Broad market beta: same logic, no market view here either. Hedge with an index future.
• Interest-rate sensitivity: modest for this name, since it's not long-duration growth. Checked, found small, left alone — hedging a risk that's already small just adds cost for no reduction that matters.
• FX: the company is domestic-revenue, so effectively none. Not applicable.

What's left after hedging sector and market beta is, as closely as position construction allows, a pure bet on the one thing you actually researched. That's the position doing its job — winning or losing on the question you meant to ask, not on questions you never had a view on.

Revisiting the decision, not just making it once

The rewarded/unrewarded split isn't fixed at entry. A risk that was unrewarded when you opened the position can become rewarded later, if you develop a genuine view on it — the sector hedge you put on for purely defensive reasons might come off if you later form a real thesis that the sector itself is about to re-rate, at which point removing that hedge is a deliberate new bet, not a lapse in discipline. The reverse also happens: a risk you were happy to carry unhedged at entry can turn unrewarded as the thesis plays out and the remaining upside is concentrated in a narrower part of the story than it was originally. Treating the hedge decision as something to revisit at each review, rather than a box ticked once on day one, is what keeps the position's actual risk matching your actual current view rather than your view from weeks ago.

Why this discipline compounds across a book

A single position with unhedged sector or market beta is a minor issue. A book of thirty positions, each carrying some unhedged, unrewarded beta because nobody separately checked, adds up to a portfolio whose largest single risk factor is one nobody has a view on at all — often larger than the sum of the individual stock-picking bets the book was actually built to express. Reviewing exposures this way, position by position, is what keeps a stock-picker's book from quietly turning into an accidental macro bet.