Currency-Hedged ETFs: What the Hedge Actually Costs You Per Year

You compare two funds tracking the same international index. One is hedged, one isn’t. The fee tables say 0.35% and 0.15%, so the hedge looks like it costs 0.20 percentage points a year — twenty dollars per ten thousand. A year later the hedged fund has trailed the unhedged one by more than two points, and nothing in the prospectus explains where the money went.

The fee gap is real, but it’s the small half of the price. The rest is the short-term interest rate gap between the dollar and the currency being hedged, and it flows through the fund’s net asset value without ever appearing on a fee line. Visible cost: 0.05 to 0.30 points a year, plus a couple of basis points for rolling the contracts. Invisible cost: anywhere from roughly minus 3% to plus 3% a year. That second number is why two funds with an identical stated fee difference can finish a year four points apart.

Why the expense ratio can’t tell you

The expense ratio covers what the fund charges to run itself. The SEC’s breakdown of mutual fund and ETF fees and expenses sorts those into management fees, 12b-1 fees, and other operating expenses. Every bucket is a payment to somebody for a service.

The hedge is not a service. It’s a stack of currency forward contracts the fund enters, holds for a month, and replaces. Gains and losses on those contracts land in the fund’s NAV directly. They are not a fee, so nobody is required to quote them to you, and no fund does.

So when someone says “the hedge costs fifteen basis points,” they have priced the floor mats and handed you the keys.

Three components, then:

The middle row is the whole game.

Where the hidden number comes from

Covered interest parity sets the price of a currency forward. If the foreign currency’s short-term rate sits below the dollar’s, the forward price of that currency is above spot — and a fund that sells the currency forward to hedge collects that difference every time it rolls. If the foreign rate sits above the dollar’s, the fund pays it.

First-order version, from a US dollar investor’s seat:

Annual carry ≈ (US short rate) − (foreign short rate)

The exact form is (1 + r_US) / (1 + r_foreign) − 1. The gap between the two versions is small under about 5% but not zero, and it’s the reason a couple of numbers below don’t land on round figures.

Nothing here is a forecast. Carry is the one part of the equation you can read off a screen before you buy.

Put $50,000 in and follow it

Say you buy $50,000 of a hedged international index fund and hold it for twelve months without trading. Every assumption stated:

• Hedged fund expense ratio 0.35%; the unhedged version of the same index charges 0.15%. Fee gap 0.20 points.
• US one-month rate: 4.00%.
• Foreign one-month rate: 6.00%.
• Forward roll cost: 0.05% a year.
• Hedge ratio 100% of NAV, reset monthly.

Fee drag is easy. $50,000 × 0.20% = $100.

Carry is where the money is. Approximated, 4.00% − 6.00% = −2.00%. Exactly, (1.04 / 1.06) − 1 = −1.887%, which on $50,000 is −$943. Roll cost takes another $25.

Total cost of the hedge: $1,068, or 2.14% of the position. The fee you compared in the prospectus is $100 of that — nine percent of what you actually paid. The carry is $943, or eighty-eight percent.

Now change exactly one input and leave everything else alone. Foreign rate 2.00% instead of 6.00%. Carry becomes (1.04 / 1.02) − 1 = +1.961%, or +$980. Fee and roll still cost $125. Net result: the hedge pays you $855, a gain of 1.71%.

Same fund. Same fee table. Same $50,000. A 3.85-point swing in outcome, driven entirely by a variable that appears nowhere in the document you used to choose.

The axis you’re optimising is the wrong one

Vary both inputs and the asymmetry gets blunt. Rows are the expense ratio gap between hedged and unhedged share classes; columns are the rate differential (US minus foreign). Figures are annual net cost as a percentage of the position, including 0.05% roll. Positive means it costs you.

Read across any row and the answer moves six percentage points. Read down any column and it moves 0.60. The horizontal axis is ten times more powerful than the vertical one, and central bank policy rates move by more than a point on a routine basis while expense ratios sit still for years.

Which makes the usual advice — pick the cheaper hedged fund — an exercise in optimising the weak variable. Choosing a 0.15-point fee gap over a 0.45-point one saves 0.30% a year, forever, reliably. A single one-point move in relative policy rates is worth 1.00%, in whichever direction it happens to go.

That doesn’t make the fee pointless. It’s the one component you control, the one you can verify before you click buy, and the one that never surprises you. It just isn’t the driver.

Notice the middle column too. At a perfectly neutral rate differential the hedge still costs 0.25% a year with a 0.20-point fee gap. There is no free version.

What differential makes it break even

Set total cost to zero and solve. Break-even differential = fee gap + roll cost. With the $50,000 example’s 0.20-point gap and 0.05% roll, you need the US short rate to sit 0.25 points above the foreign rate. A near-trivial hurdle.

It stops being trivial at the wide end. A 0.45-point fee gap with a 0.20% roll spread needs +0.65 points. A 0.60-point gap with 0.40% roll needs a full point. That last combination is emerging-market territory, and it compounds badly: hedging a high-yielding emerging currency means paying away a rate gap that is often four to eight points, through forwards that trade at wider spreads than developed-market ones. Emerging-market currency hedges are structurally expensive for a dollar investor in a way that developed-market hedges are not.

Five years, one position, nothing changed but rates

Scale the example up to $100,000, hold the fee gap at 0.20 points and roll at 0.05%, and run it through five hypothetical rate environments. The rate paths are illustrative and chosen to span a realistic range, not a prediction.

Five-year total: +$3,997, or +4.0% cumulative. Year 3 alone produced 93% of it. Years 1 and 4 were noise. Year 5 handed back nearly a third. The carry figures use the exact form, which is why year 3 shows $3,980 rather than a flat $4,000.

Hold this position through years 1 and 2 and you’d conclude hedging is basically free. Hold it through year 5 alone and you’d conclude it’s expensive. Both conclusions come from the same fund, the same prospectus, the same 0.20-point fee gap.

What the hedge does and does not remove

A hedge takes out one source of variance and leaves everything else exactly where it was. If a foreign index falls 20% in local terms and the currency doesn’t move, hedged and unhedged both fall about 20%. The hedge did nothing, because there was nothing for it to do.

Back to the $50,000 position, carry and fees excluded here to isolate the mechanism:

The unhedged column uses (1 + index) × (1 + fx) − 1, so the cross term is in there. That’s why the last row shows $13,250 and not $12,500.

In two of five scenarios the hedge cost you money in the outcome, entirely separate from its running cost. Removing currency exposure narrows the range of results in both directions. That is the point of it, and it is not the same thing as protection.

The related mistake is assuming hedged and unhedged share classes of one index should land in roughly the same place. They shouldn’t. At an average +1.5% annual carry, cumulative carry compounds to +4.6% over three years, +7.7% over five, +16.1% over ten — and that’s before the currency itself moves. Add a 12% spot move against you and a ten-year divergence of 28% between two funds holding the same stocks is ordinary arithmetic, not an anomaly.

Where these numbers will mislead you

Monthly resets go stale. A fund that sets its forward notional on the last business day of the month is fully hedged that day and drifts from there. If the underlying index rallies 6% mid-month, the hedge still covers only the starting notional, so roughly 6% of the position rides unhedged until the reset. Over a volatile year that alone can produce 0.3% to 0.8% of tracking difference, in either direction.

Covered interest parity does not hold exactly. Forward-implied rates and actual cash rate gaps have diverged persistently since 2008, most sharply around quarter-end when bank balance sheet constraints bite. The Bank for International Settlements has documented this cross-currency basis in its work on foreign exchange and derivatives markets. Where the basis runs against a dollar hedger, add roughly 0.10% to 0.40% a year. Every calculation above assumes parity holds perfectly. It doesn’t.

Broad funds hold many currencies. A developed-markets fund might carry twenty currency exposures. The relevant differential is a weighted average across all of them, and a single column won’t capture it. Funds publish the hedge in aggregate, not per-currency with weights.

Rates move mid-year. Everything here assumes a static differential for twelve months. Central banks change policy several times a year. Realised carry is the path average, not the starting value.

Partial years accrue partial carry. Carry accrues daily. A three-month hold gets roughly a quarter of the annual figure.

Capital controls widen everything. Non-deliverable forwards trade at wider spreads, and the implied rate can sit several points from the onshore policy rate. Treat the roll-cost range above as a floor for those currencies, not an estimate.

What this doesn’t tell you

Whether to hedge. The cost is arithmetic. The decision depends on your liabilities, your horizon, and how much return variation you can sit through without selling, none of which is in a spreadsheet.

The cost of trading the ETF itself. Hedged share classes are usually smaller and thinner than their unhedged siblings, and secondary-market spreads can run two to five times wider, on top of any premium or discount to NAV. For a buy-and-hold position that’s a one-off. For anyone trading with any frequency it can exceed the annual fee gap outright.

Tax. Forward contract gains inside a fund can generate distributions with a different character than equity gains, and that depends on the fund’s structure and on the account holding it.

And whether the currency moves your way. The carry is knowable in advance. The spot move is not, and nothing above pretends otherwise.

FAQ

How much does a currency hedged ETF cost per year?

Expense ratio gap, plus roll cost, plus or minus the interest rate differential. With a 0.20-point fee gap and a neutral differential, that’s 0.25% a year — $250 on $100,000. If the foreign short rate sits 3 points above the US rate, the same position costs about 3.25%, or $3,250. If it sits 3 points below, the hedge pays you roughly 2.75%.

Is the hedged or unhedged version of an ETF cheaper?

The unhedged version always carries the lower stated expense ratio, usually by 0.05 to 0.30 points. Total cost is a different question. At a +2-point differential in the dollar’s favour, a hedged fund with a 0.30-point higher expense ratio still nets out about 1.65% ahead before any currency move.

What is a typical expense ratio difference between hedged and unhedged ETFs?

Across mainstream developed-market index funds, 0.05 to 0.20 percentage points. Emerging-market hedged products run wider, often 0.25 to 0.50 points, because the hedging instruments cost the fund more to trade. On $50,000, a 0.20-point gap is $100 a year.

Does currency hedging reduce my returns?

Only when the rate differential runs against you, or when the currency would have moved in your favour. When US and foreign short rates are both near zero, carry is close to nil and the hedge costs roughly its fee. When the US rate sits several points above the foreign rate, the carry is positive and the hedge adds to return before any spot move.

How often do currency hedged ETFs roll their forward contracts?

Most reset monthly, on or near the last business day. Some roll weekly, a few daily. Monthly is standard because it trades roll cost against hedge accuracy. The consequence is a hedge notional that can be up to 30 days stale, which typically produces 0.3% to 0.8% of annual tracking difference against a perfect hedge, either direction.

At what interest rate gap does hedging break even?

Break-even equals fee gap plus roll cost. With a 0.20-point gap and 0.05% roll, the US short rate needs to be 0.25 points above the foreign rate. With a 0.45-point gap and 0.20% roll, 0.65 points. Below that line, the hedge is a net drag before the currency does anything.

Where to look next

Pull the prospectus fee tables for a hedged fund and its unhedged twin and write down the gap in basis points. Then find the current short-term policy rates for the US and for the fund’s dominant currency exposure, subtract, and find that column in the cross-condition table above. That’s your row and your number.

After that, two documents settle how far reality drifts from the arithmetic: the strategy section, for the fund’s stated hedging frequency and hedge ratio, and the quote screen, for the bid-ask spread on the hedged share class next to the unhedged one.

This article is general information, not financial advice. See our disclaimer.

More in Costs and Fees · all topics · calculators · how this was checked