Skip to content

2026-07-19 · Validated Research Team

Are FX swaps profitable? The spread is real, but the easy carry is not

FX swaps look like a simple way to collect the difference between two countries’ interest rates. That interpretation is incomplete. The forward exchange rate normally absorbs almost all of the rate differential, leaving a much smaller cross-currency basis after the currency risk is hedged.

Our verdict is inconclusive as a tradable strategy and unattractive for a typical retail account. The institutional profit mechanism is well documented, but we do not have executable institutional quotes, matched funding rates, collateral terms or trade-level results with which to validate a standalone implementation.

What an FX swap actually earns

An FX swap exchanges two currencies today and reverses that exchange later at a pre-agreed forward rate. Under covered interest parity, the forward adjustment offsets the difference between the two currencies’ interest rates.

Suppose dollars earn 5% and yen costs 1%. Borrowing yen and holding dollars appears to offer 4% annual carry. Once the exchange rate is fully hedged, however, the forward price should cost approximately that same 4% before transaction and funding costs. The residual opportunity is not the headline interest differential. It is the deviation from parity, called the cross-currency basis.

Where profits can exist

Profit source Economic reality Evidence state
Headline interest-rate differential Mostly offset by the forward price when FX risk is hedged No standalone edge
Cross-currency basis Can pay a small premium to institutions with unusually cheap funding and balance-sheet capacity Documented, not strategy-tested here
Dealer spread Fractions of a basis point can matter across very large client flow Institutional business model
Unhedged carry Keeps the rate differential but also keeps currency risk Directional FX trade, not arbitrage
Retail overnight swap Broker financing and markups can consume the apparent carry Requires broker-specific executable data

The Bank for International Settlements reported about $4 trillion of average daily FX-swap turnover in April 2025, making swaps the largest segment of the global FX market. Size does not imply a large return per dollar. It instead makes very small spreads economically useful to dealers operating at enormous scale.

Scale of the basis trade

The following is an illustration, not a live quote or backtest:

Notional Gross basis captured Gross annual dollars
$1 million 5 bps $500
$10 million 15 bps $15,000
$100 million 25 bps $250,000

One basis point is 0.01% per year. Funding spreads, spot and forward execution, collateral, credit charges, capital usage and operational costs must still be deducted. A nominal 15-basis-point opportunity can therefore be uneconomic for an outside investor while remaining useful to a bank with cheaper funding or offsetting customer flow.

Why the basis does not disappear

BIS research finds that covered-interest parity holds much more closely after each participant’s actual marginal funding rate is included. The most credible arbitrage access is concentrated among a narrow set of top-tier global banks. Other participants incur balance-sheet, collateral, counterparty, liquidity and mark-to-market costs that create a no-trade band around the apparent opportunity.

Dealer economics are also different from investor economics. A dealer can match clients who need opposite currency funding, collect a spread on both flows and reduce the amount it must hedge externally. BIS analysis of the 2025 market found increased internalisation of client trades and fewer interbank arbitrage opportunities. That supports a high-volume intermediation business, not an easy passive-return strategy.

What would be needed to validate it

A proper strategy test needs more than historical policy rates. It requires:

  1. Simultaneous executable spot and forward bid/offer quotes.
  2. The trader’s actual secured or unsecured borrowing rates in both currencies.
  3. Collateral haircuts, margin requirements and capital tied up.
  4. Roll timing, holiday calendars and settlement costs.
  5. Stress tests for quarter-end funding pressure and margin calls.
  6. Separate accounting for currency exposure if any leg is left unhedged.

Without those inputs, a chart comparing policy rates would overstate the attainable return.

Bottom line

FX swaps can be profitable as a large-scale dealer or funding-arbitrage business. They are unlikely to offer an attractive standalone edge to a typical investor after the forward adjustment and real funding costs. Unhedged carry may earn several percentage points in favorable periods, but it is a leveraged currency bet and can lose far more than its accumulated carry.

The honest label is therefore institutionally plausible, not validated for retail execution. We would only promote this to a tested strategy after obtaining synchronized executable quotes and account-specific funding data.

Sources


By Validated Research Team (Methodology v1.0 — 11-gate validation). Part of 11-gate validation. Backtests are not investment advice.