« Zero-cost » hedging: what a collar, a participating forward and a barrier really cost
Three structures presented as free, three different ways of hiding remuneration inside the guaranteed rate. How to compare them with each other and with the plain forward, which remains the yardstick.
« This solution costs you nothing. » The sentence is literally accurate — no premium will be debited — and economically misleading. A bank does not provide protection for free; if nothing changes hands at inception, the remuneration is elsewhere.
It sits in the level of the guaranteed rates, and it gets there in three different ways depending on the structure.
The three structures and how they work
| Structure | What you get | What you give up | Where the remuneration sits |
|---|---|---|---|
| Collar | A guaranteed floor rate | All upside beyond a cap | In the gap between the floor and the cap |
| Participating forward | A floor rate, plus partial participation in the upside | A share of the upside, fixed in advance | In how far the floor sits below the plain forward |
| Knock-in / knock-out forward | A rate markedly better than the plain forward | The protection, if the barrier is touched | In the probability that the barrier is touched |
The plain forward is the yardstick. Every structure is measured against it: what rate does it guarantee you, compared with the forward rate you would have obtained without it? The gap between the two is the price of what the structure adds. If it is wide, the structure had better add a lot.
The collar, the most readable structure
You buy protection at a floor level and fund that purchase by selling the upside above a cap. The two options offset in premium.
Its readability comes from both bounds being explicit: you know exactly what range you are in. Checking it means repricing the two legs separately and seeing whether they really offset, or whether the leg you are selling is worth materially more than the one you are buying.
The participating forward, harder to read
You get a guaranteed floor rate, less favourable than the plain forward, and in exchange you participate in a fraction of the upside — often half.
The difficulty is that two variables move together: the level of the floor and the participation rate. A proposal can show a generous participation and a heavily shifted floor, or the reverse. Comparing two participating forwards without repricing them means comparing two numbers that do not measure the same thing.
- 1Note the plain forward rate
On the same notional and the same maturity. That is your reference.
- 2Measure how far the floor is shifted
By how much is the guaranteed rate less favourable than the plain forward? That is what you are paying.
- 3Set that against the participation
What share of the upside do you get back, and from what level? That is what you are buying.
- 4Compare the two
A 50 % participation in an upside that has only a small probability of occurring is not worth the floor shift you are being asked to accept. The calculation can be done; it should not be estimated.
Barriers, in a category of their own
Knock-ins and knock-outs do not belong to the same family. The two previous structures trade a share of the upside for protection; these trade the reliability of the protection for a better headline rate.
A knock-out forward protects you as long as the market has not reached a certain level. But that level is, by construction, reached in large market moves — precisely the ones you wanted protection against. The protection disappears at the moment it would matter most.
That does not make them illegitimate: a barrier can suit a company with a documented view on a range of rates that explicitly accepts the scenario where the protection falls away. It does not suit a company trying to protect a margin, which is the situation of the great majority of mid-market businesses.
Two details change their value a great deal, and can be read on the proposal: how the barrier is observed — continuously or on set dates — and whether it is American or European. A continuously observed barrier is far more likely to be touched, and is therefore worth far less to the bank than the improved rate suggests.
The decision grid
| Your situation | Suitable structure |
|---|---|
| Certain flow, thin margin, safety first | Plain forward. The cheapest and most readable structure. |
| Certain flow, wanting to capture some upside | Collar, provided the gap between the bounds is checked |
| Certain flow, upside seen as likely but uncertain | Participating forward, after quantifying the floor shift |
| Conditional flow, not yet won | Plain option. Maximum loss bounded by the premium. |
| Documented view on a range, risk explicitly accepted | Barrier, with written sign-off from the managing director |
| Need to protect a commercial margin | Never a barrier |
The question that settles the discussion
Faced with any structure, one question is enough to bring the conversation back onto verifiable ground: what plain forward rate would I get on the same notional and the same maturity?
That rate always exists, it can be computed, and it measures what the structure really costs you. A proposal that cannot be compared with that yardstick has not been understood — not by you, and sometimes not by the person presenting it.
Reprice your structure leg by leg
Collar, participating forward, barrier: the pricer values each component and shows the gap to the equivalent plain forward.
Analyse my structureNot ready to open the pricer yet?
Get the one-page FX policy template — hedge ratios by horizon, tolerance band, delegation table. The one your auditor will ask for.
- One field only
- One-click unsubscribe
- No data resold
Frequently asked questions
The questions finance departments ask most often on this topic
Is a zero-cost hedge really free?
No. The absence of a premium means the optional legs bought and sold have been calibrated to offset, not that the bank is working for nothing. Its remuneration sits in the level of the guaranteed rates, where it can only be read by repricing each leg separately. These are the structures where the margin can become very significant and exceed 2 % of notional.
What is the difference between a collar and a participating forward?
A collar sets two explicit bounds: you are protected below a floor and give up all upside above a cap. A participating forward has no cap but a floor less favourable than the plain forward, in exchange for participation in a fraction of the upside. The collar is more readable; the participating forward requires comparing two variables that move together.
Why advise against barrier products for protecting a margin?
Because their protection is conditional. A knock-out stops protecting as soon as the market reaches a certain level — that is, during large moves, the very ones the hedge existed for. Protecting a commercial margin requires protection that holds in the adverse scenario, which a barrier does not guarantee.
How do you compare two proposals with different structures?
By reducing both to the same yardstick: the plain forward rate on the same notional and maturity. The gap between each structure's guaranteed rate and that plain forward quantifies what it costs; what it delivers in exchange is then compared. Without that common yardstick, two structures are not comparable with each other.
What is an accumulator and why is it in a category of its own?
An accumulator, or TARF, spreads transactions over many fixing dates, often with leverage on the adverse leg. It is no longer a hedge but a series of conditional commitments whose loss is not bounded in adverse configurations. Its headline rate is attractive precisely because that risk is not visible at inception.
Read next
Pricing an FX option: how to check the premium you are charged
9 min read
Reading an FX term sheet: the eight lines that set the price
10 min read
Calculating a forward rate: the formula, a full worked example, and the mistake to avoid
11 min read
Currency risk for mid-market companies: the complete guide for a finance director
14 min read
Hedge accounting under French GAAP: ANC Regulation 2015-05 explained for a mid-sized company
6 min read