EUR vs USD Box Spreads: Is the Gap Real Money?
If you earn in euros and invest in dollars, the two box curves present an obvious temptation. Euro boxes price well below dollar boxes. Borrow the cheap currency, convert, buy the expensive asset, pocket the difference. It is one of the most-asked questions we get, and the answer is a clean no — but the reason is worth understanding, because it tells you what the decision actually turns on.
The gap is real. The arbitrage is not.
Both curves are live on the front page, so you can check this yourself on any given day. At the time of writing the twelve-month points sat here:
1.9% on a €500,000 borrow is €9,500 a year. That is not a rounding error, and the instinct that something is being left on the table is a reasonable one.
The catch is that you cannot spend euros in a dollar-denominated portfolio. Somewhere in the chain you have to convert — and if you intend to repay the box in euros at expiry, you have to convert back. The rate at which you can lock that return trip today is the forward rate, and the forward rate is not free.
Covered interest parity, checked against real borrowing
Covered interest parity says the forward premium between two currencies has to equal their interest-rate differential. If it did not, you could borrow in one currency, lend in the other, hedge the currency risk completely, and bank a riskless profit — which is exactly the trade arbitrageurs remove.
The formula is simple: F / S = (1 + r_USD) / (1 + r_EUR). Run our two box rates through it:
Five basis points. The euro discount is, to within measurement noise, precisely the price of the forward. Convert at spot, lock the return leg, and you end up where you started.
What makes this a stronger demonstration than the textbook version is the input. Box spreads are about the cleanest synthetic borrowing rate a private investor can observe — collateralised, exchange-cleared, no credit spread, no relationship pricing. Two of them, in two currencies, reproduce covered interest parity to five basis points. That is the theory doing its job.
So what are you doing if you skip the hedge?
Most people asking this question are not hedging the return leg. They borrow euros, convert once, and hold dollar assets. That is a coherent position — it is simply not an arbitrage.
It is a carry trade: you have sold euro-dollar exposure forward and are being paid roughly 1.85% a year to carry it. The payment is not a discount for cleverness. It is compensation for a risk you have taken on, and the market is pricing that risk at 1.85%.
Carry trades are a real and well-documented strategy. They also have a characteristic shape: long stretches of small steady gains, punctuated by fast, large losses when the funding currency rallies. The returns are negatively skewed, which is the statistical signature of being paid a premium to absorb something unpleasant. Anyone who carried short-JPY or long-MXN positions through a sharp unwind knows what the tail looks like.
Whether that is a trade you want is a legitimate question. It is a different question from whether the box rate is cheap.
Why EURUSD in particular leaves so little on the table
Parity holds more tightly in some currency pairs than others. EURUSD is close to the tightest case available: both legs are deep, both currencies are freely deliverable, and the swap market that enforces the relationship is enormous.
The residual that does persist is the cross-currency basis — typically a few to a few tens of basis points for EUR, and it widens predictably around quarter-ends and year-end when bank balance sheets tighten. It is not an inefficiency waiting to be harvested; it is the price of balance sheet, and capturing it requires being a bank.
Emerging-market pairs are where the wider deviations live, and where carry is genuinely compensated rather than arbitraged away. That is a different asset class with a different risk profile, and it is not what is happening between SPX and ESTX50.
The variable that actually decides it
If the rate comparison nets to roughly zero, the choice is not about the rate. It is about what your liability is denominated in relative to everything else on your balance sheet.
- A dollar box against a dollar portfolio is a matched position. Asset and liability move together; currency moves largely cancel.
- A euro box against a dollar portfolio is a mismatched one. It costs less in headline terms and adds a currency exposure sized to the whole loan.
- If you would ultimately repay from euro income, a euro liability matches your income even where it mismatches your assets.
There is no universally correct answer — it depends on which mismatch you would rather carry, and that is a question about your circumstances, not about the curve.
Two costs that do differ, and are not FX
Parity washes out the rate. It does not wash out everything.
Commission per unit borrowed. An SPX contract carries a 100 multiplier against ESTX50's 10, so the same borrow needs roughly a tenth as many SPX contracts. Even at a higher per-contract fee, dollar boxes usually work out cheaper per euro raised — often by several times. Compare fees per unit of notional, never per contract.
Tax. This is the one most likely to change the answer. Currency gains and losses on the conversion can be a separate taxable item from the box carry itself, with their own treatment and their own holding-period rules. An after-tax comparison can point somewhere quite different from a pre-tax one, and it is highly jurisdiction-specific. Worth a specific question to your tax adviser rather than a rule of thumb.
The short version
The 1.9% gap is real, and it is already spoken for. Hedge the currency and it disappears into the forward. Do not hedge it and you are running a carry trade that happens to be funded by a box spread. Choose the currency that matches your balance sheet, then compare commissions and tax treatment — that is where the differences that survive actually are.