Layered hedging: a 12-month framework for importers

The problem with "all or nothing"

Australian importers in the $5-50M annual programme range tend to fall into one of two camps. The first hedges nothing, runs the FX exposure naked, and explains the resulting margin variance to the board four times a year. The second hedges 100% of the year's expected payments in a single pass, usually on the day the budget is approved, and prays they didn't pick the worst possible spot.

Both approaches are answers to the same problem (currency volatility in the cost of goods), and both are bad. The first leaves you fully exposed to the very thing FX risk management is supposed to remove. The second concentrates timing risk in a single decision date, which is its own kind of speculation.

A layered hedging programme is the compromise that the corporate treasury world settled on decades ago. It works by spreading hedge execution across time, building a portfolio of staggered forwards (and occasionally options) at different tenors, so that no single market level dominates your average effective rate. The cost is some operational complexity. The benefit is that your hedging decision stops looking like a directional bet on AUD/USD.

This article sets out a practical 12-month layered hedging framework for an Australian business running an annual import programme between $5M and $50M equivalent. The numbers and ratios are calibrated for that range. The principles scale up or down.

Why layer at all

Three concrete things change when you move from one-shot to layered hedging.

First, you stop trying to predict the future. A single 100% hedge at the start of the year is an implicit forecast that today's rate is the right rate. You have no actual basis for that view, and neither does anybody else. Layering accepts the forecast problem is hard and replaces it with averaging.

Second, you decouple FX risk from forecast risk. Importers don't actually know with certainty what they'll buy in nine months. Order books drift, demand changes, supplier mixes shift. If you've hedged 100% of a budget that turns out to be 70% of actual, the over-hedge is now a speculative position. Layering matches hedge intensity to forecast confidence: heavy in the near term where exposure is committed, light in the back end where it's still budgetary.

Third, layering creates a routine. Treasury becomes a process rather than a series of judgment calls. The decision becomes "have we executed this month's layer" rather than "should we hedge today?" That shift, more than any specific ratio, is what separates teams that hedge well from teams that hedge sporadically.

The core framework

The framework presented here covers a rolling 12-month exposure, divided into four 3-month buckets, with declining hedge intensity as tenor extends.

• Bucket 1 (0-3 months): 75% hedge ratio
• Bucket 2 (3-6 months): 50% hedge ratio
• Bucket 3 (6-9 months): 30% hedge ratio
• Bucket 4 (9-12 months): 15% hedge ratio

These ratios apply to forecast import volumes in each bucket. The intuition is straightforward. Near-term volumes are confirmed by purchase orders or supply agreements. Hedge them aggressively because the underlying is real. Far-term volumes are still partly forecast. Hedge them lightly because over-hedging a forecast that doesn't materialise turns the hedge into a directional position.

The numbers are not gospel. A business with very stable demand and long supply lead times might run higher ratios in the back end (say 40% / 25%). A business with volatile demand or high cancellation risk might pull them down (say 25% / 10%). The 75/50/30/15 cascade is a reasonable starting point for a typical distribution business. Calibrate from there.

The key discipline is that ratios are set by policy and applied mechanically. They are not adjusted in response to market levels. The moment you start saying "AUD looks cheap, let's go to 90% in the front bucket this quarter," you're back to running directional views, just dressed up in policy language.

Rolling the book

The book is rolled monthly. Each month, three things happen.

Bucket 1 has the next month's payments settle and drop out. The remaining months in the bucket move closer to settlement. Any new payments that have firmed up are added to the bucket and topped up to the 75% ratio.

Bucket 2 contributes one month into Bucket 1, and gets a new month added at its back end (now 6 months out). The hedge ratio is rebalanced to 50% of the updated bucket forecast. Buckets 3 and 4 work the same way, each absorbing one month from the bucket behind and contributing one month to the bucket in front.

A new month enters at the 12-month tenor with a 15% hedge ratio applied.

In steady state, the team executes a small number of trades each month: top-ups to maintain ratios, new layers at the back end, and any rebalancing required by changes in the underlying forecast. After the first 12 months of running the programme, the book is fully populated and the monthly activity becomes routine.

Forecasting the underlying

The framework only works if the underlying forecast is reasonable. Garbage in, garbage out, with the additional hazard that hedging a bad forecast manufactures real P&L consequences.

For a $5-50M programme, three forecast tiers are usually enough.

Confirmed exposures are payments backed by signed purchase orders, supply contracts, or comparable evidence. These are the hedge-eligible base. They almost always live in Buckets 1 and 2 and frequently extend into Bucket 3.

Highly probable exposures are forecast on the basis of historical run-rates, customer pipelines, or production plans, but not yet contractually committed. These typically populate Buckets 3 and 4. Under AASB 9, hedge accounting requires the hedged item to be highly probable, which sets a real evidentiary threshold. If you can't articulate why the volume is highly probable, you can't designate the hedge.

Speculative exposures are forecast volumes that are aspirational or scenario-dependent. These should not be hedged. If you find yourself with hedges against speculative volume, you've stopped hedging and started trading.

Each month, the treasury team reconciles confirmed against forecast. If the confirmed pipeline is significantly lighter than the forecast, the hedge ratios in the affected buckets need to come down before new layers are added.

Choosing the instrument

For most of the book, plain vanilla forwards are the right instrument. They are simple, cheap, and produce deterministic outcomes. Australian banks and AFSL-licensed non-bank specialists offer them through standard documentation, and most SME-scale programmes can be supported with a forward dealing line on the back of audited financials.

Window forwards are useful in the front buckets where supplier payment timing slips by a week or two. The cost is small (a few additional pips depending on the window length) and the operational benefit is real: you avoid forced rolls when a payment date moves.

Options have a role at the long end of the book, particularly in Bucket 4, where forecast confidence is lower and the benefit of a non-binding instrument is highest. A bought call (right to buy USD at a strike) gives downside protection without locking in a rate, which suits a position where the underlying might not materialise. The cost is the premium, which can run 1-3% of notional for at-the-money strikes at 9-12 month tenors. For a $5M back-end exposure that translates to $50,000-$150,000 of premium, which has to be justified against the cancellation risk it covers.

Participating forwards (also called "leveraged" or "participating" structures) sit between forwards and options. They protect you from adverse moves while letting you participate in some of any favourable move, with no upfront premium. The catch is that they typically involve a leverage clause if the rate moves a long way in your favour, meaning you may end up obliged to buy more notional than you originally needed. They can be appropriate in specific circumstances, but read the term sheet carefully. Many treasury teams have learned about leverage clauses the hard way.

For a typical $5-50M programme, the instrument mix often looks like vanilla forwards across Buckets 1-3 with a small allocation to window forwards in Bucket 1, and a split in Bucket 4 between vanilla forwards on the high-conviction portion (say half of the 15% hedge ratio) and options on the rest. This keeps complexity manageable. Treasury teams that go further down the structured products path tend to find the operational and audit overhead exceeds the benefit at this programme size.

Triggers vs. calendar execution

Two execution philosophies dominate.

Calendar-based execution runs the layering programme on a fixed schedule. The first business day of each month, the team executes the month's required trades regardless of where the market is. The discipline is mechanical and therefore robust. The downside is occasional execution into adverse market levels.

Trigger-based execution sets price levels at which layers will be added. For example: "execute the 12-month back-end layer when AUD/USD trades above 0.6650, or by month-end at the latest, whichever comes first." Triggers introduce some opportunism without abandoning discipline.

For most $5-50M programmes, a hybrid works best. Calendar execution as the default, with a small trigger overlay (say 10-20% of monthly required notional held back for opportunistic execution within a defined band). Anything more aggressive starts to drift back into market timing, which the framework was designed to avoid.

Governance

A layered hedging programme needs governance commensurate with its size. For a $5-50M programme, that usually means the following.

A board-approved hedging policy, refreshed annually, that sets out objectives, authorised instruments, ratio bands, counterparty limits, and authority levels. The policy is short, typically 4-8 pages. It is also non-negotiable in execution. The point of having a policy is to remove judgement calls from individual transactions.

A delegated execution authority, usually held by the CFO or treasurer, with named alternates. The authority defines who can execute trades up to what notional, and what requires escalation.

A monthly hedge report to the audit committee or finance committee. The report shows current hedge ratios by bucket, the effective rate of the existing book, mark-to-market position, and any policy exceptions. Five pages is plenty.

A dealing list of approved counterparties, with credit limits per counterparty. For a programme of this size, two or three counterparties is normal. Concentrating with one creates dependency. Spreading across more than four or five creates operational drag.

Annual independent review. Either an internal audit function or an external advisor reviews the programme annually for policy compliance, valuation accuracy, and effectiveness against stated objectives.

None of this is heavy. A $20M programme can be governed adequately by a CFO, a finance manager, and a quarterly slot in the audit committee agenda.

A worked example

Consider an Australian distributor importing $24M of inventory annually from US suppliers, paid USD-denominated, broadly evenly across the year ($2M per month). Today is January.

The book at steady state, applying the 75/50/30/15 framework:

• Bucket 1 (Feb-Apr, $6M forecast): 75% hedged = $4.5M of forwards outstanding
• Bucket 2 (May-Jul, $6M forecast): 50% hedged = $3.0M
• Bucket 3 (Aug-Oct, $6M forecast): 30% hedged = $1.8M
• Bucket 4 (Nov-Jan, $6M forecast): 15% hedged = $0.9M

Total outstanding hedge book: $10.2M USD notional. At an average AUD/USD around 0.6500, that's about AUD 15.7M of notional commitment.

In a typical month, treasury executes a top-up trade in Bucket 1 to absorb the new month entering at 75%, rebalancing trades in Buckets 2 and 3 to maintain ratios as months roll forward, and a new layer at the back end (now month 13) at 15%. Total monthly activity: typically 3-5 trades, often a single counterparty visit. Annual trade count: 40-60.

If AUD/USD drops 5% over six months, the existing book continues to settle at locked rates, providing the operational stability the framework was designed for. New layers added in the lower-rate environment will eventually pull the average effective rate down, but slowly. The smoothing effect is the entire point.

The opposite move, AUD strengthening 5%, produces opportunity cost on existing hedges (the locked rates are now worse than spot), but again, this is what the framework signed up for. New layers added at the higher rate eventually lift the average back up.

The risk that the framework specifically protects against is the scenario where AUD drops sharply for an extended period and an unhedged or single-tranche-hedged business absorbs the full move into margin. That's the one that ends careers.

Common mistakes

Several patterns turn up repeatedly in mid-market programmes.

Hedging budgeted volume rather than confirmed plus probable exposure. The budget is an aspiration. The hedge book needs to track actual exposure, not the version of the world the planners wanted.

Letting ratios drift. The framework only works if the bucket ratios are actually maintained. Treasury teams under pressure sometimes skip the monthly top-up, and the book quietly under-hedges. After 6 months of skipped layers, the front bucket might be at 40% instead of 75%, and nobody noticed until the auditor raised it.

Concentrating roll dates. A programme that consistently uses end-of-month maturities ends up with all forward maturities clustered on a single day each month. If a payment slips by even a day, the resulting roll is mechanical but creates unnecessary cost-of-carry adjustments. Spread maturity dates across the month.

Over-engineering the back end. Bucket 4 covers 9-12 months out, where forecast confidence is lowest and ratios are smallest. The temptation to add structured products, exotic options, and bespoke instruments is highest. Resist it. The expected dollar value of "optimising" Bucket 4 is small, and the operational and audit cost is real.

Failing to update the policy. Hedging policies written for 2018 conditions can quietly become inappropriate by 2024. Refresh annually. The board approval for the refresh forces the conversation about whether the framework still suits the business.

Closing

Layered hedging isn't intellectually fancy. It works because it removes most of the discretion from a process that humans handle badly when they have discretion. A treasury function executing a 75/50/30/15 framework month after month, with disciplined forecasting and modest governance, will outperform a more sophisticated programme run on intuition over most market cycles.

For $5-50M programmes, this is the sweet spot. The notional is large enough to justify the operational overhead. It's small enough that simple forwards do most of the work, with light governance and standard documentation. The framework is a finished, working answer to "how should we hedge," and once it's installed, hedging stops being a strategic question and becomes infrastructure.

The teams that adopt frameworks like this and stick with them tend to share a trait: they've stopped trying to be right about FX. They've decided that being consistent matters more than being clever, and they've built the operational discipline to execute that decision. The market will do what the market does. Their margin doesn't have to.