Carry strategies attracted renewed attention in 2026 as rate dispersion across currencies remained meaningful and market microstructure evolved. Pure carry, however, leaves traders exposed to directional shocks. This guide walks through a practical, implementable process to combine interest‑rate carry with price momentum while using 3‑month forwards or swaps to hedge undesired spot exposure and crystallise funding costs. The article gives step‑by‑step instructions for data, signal construction, portfolio construction, execution, cost modelling, backtesting pitfalls and live risk controls.

Why hedge carry with forwards?

Traditional carry captures the interest differential earned by borrowing in low‑yield currencies and investing in higher‑yield currencies. But a long spot position in a high‑yield currency exposes you to spot depreciation that can swamp carry returns. Using a 3‑month forward (or FX swap) lets you lock in the forward premium/discount that reflects the interest rate differential, isolating funding/carry from spot moves. That enables a purer signal—earn funding while hedging directional risk—or allows a partial hedge to retain some spot beta.

Strategy overview — what you will build

  • Universe: 10–20 liquid G10 and selected liquid EM spot pairs quoted vs USD (or your base currency).
  • Instruments: spot for momentum signals; 3‑month forward contracts (or FX swaps) for hedging and capturing carry.
  • Signals: 1) Carry score from forward points/interest differential; 2) Momentum score from 1–3 month spot returns or trend filter.
  • Portfolio rule: combine normalized carry and momentum ranks into a composite score, select top/bottom N pairs, and take positions hedged with 3M forwards.
  • Risk controls: per‑pair caps, portfolio VaR, maximum gross exposure, and regular rebalancing/roll management every month or quarter.

Step 1 — Required data and where to get it

Collect these time series at daily or business‑day frequency:

  • Spot mid rates (EOD) for all pairs in your universe.
  • 3‑month forward points or outright 3M forward rates.
  • Overnight/3M money market rates for each currency (OIS or Libor/term rates as available) to validate forward‑interest relationship.
  • Historical transaction costs: bid‑ask spreads, typical dealer forward spreads, and settlement fees for forwards/swaps.
  • Event calendar (central bank rates, data surprises) for out‑of‑sample sanity checks.

Sources: bank dealer feeds, market data providers (Bloomberg/Refinitiv), central bank published term rates, and broker tick histories for realistic spreads. Always document data licensing and timestamps for reproducibility.

Step 2 — Compute the carry score

Carry can be measured in a few equivalent ways. Using 3‑month forward points is practical because forwards incorporate the negotiated funding differential:

  1. Calculate 3M forward premium/discount in annualised percent: approximate annualised carry ≈ (F/S − 1) × (12/3), where F is the 3M forward rate and S is spot.
  2. Alternatively, convert forward points to implied interest differential using the covered interest parity discrete form: F = S × (1 + r_dom × dt) / (1 + r_for × dt), solving for r_dom − r_for.
  3. Rank currencies by this annualised carry from highest positive (earn) to most negative (pay).

Example (hypothetical): Spot EUR/USD = 1.0800, 3‑month forward = 1.0740 → F/S − 1 = −0.00556, annualised ≈ −2.22% (implies EUR pays carry vs USD). Be explicit whether your base is USD or local and keep sign conventions consistent.

Step 3 — Compute the momentum score

Momentum removes static exposure to carry blowups. Use a short momentum horizon to align with a 3M hedging cadence:

  • Compute 1M and 3M spot returns and normalize (z‑score) within each point‑in‑time cross‑section.
  • Optionally apply a volatility‑adjusted momentum: momentum_i = return_i / realised_volatility_i (90‑day).
  • Use a trend filter: require momentum > 0 to take a long exposure in the currency (i.e., earn carry and have positive trend) or use momentum to tilt positions.

Step 4 — Combine signals and form positions

Construct composite score: Composite = w_carry × rank_carry + w_mom × rank_mom. Typical weightings start at w_carry=0.6, w_mom=0.4; test alternatives.

Selection rules:

  • Long positions: top N currencies by composite score where carry > threshold and optional momentum filter holds.
  • Short positions (if permitted): mirror for bottom N for market‑neutral exposure.
  • Hedging: enter a 3‑month forward contract that neutralises the spot exposure (i.e., sell forward the currency you're long to lock subsequent conversion at F).
  • Position sizing: equal risk by volatility or equal notional; cap per‑pair notional to limit concentration.

Step 5 — Execution and roll mechanics

Practical execution choices:

  • Use outright 3M forwards or a sequence of rolling 1‑month swaps depending on liquidity and available tenors. 3M forwards are common for corporate/treasury counterparties; retail/futures traders may use calendar swaps or futures where available.
  • Place spot trades for momentum capture if you want residual spot exposure; if not, enter spot and simultaneous forward to be flat at trade date.
  • Roll schedule: typically roll forwards 5–10 business days before maturity to avoid end‑quarter squeezes and wide spreads. Automate roll execution with limit orders to control slippage.
  • Account for settlement margin and bilateral collateral when using OTC forwards—funding and credit limits affect sizing.

Step 6 — Transaction cost and slippage modelling

Realistic backtests must deduct:

  • Bid‑ask spread on spot trades and the dealer spread on forward outright quotes.
  • Commission or clearing fees and any FX swap swap‑points charged over mid for smaller trades.
  • Slippage: model slippage as a function of notional and ADV (% of daily volume).
  • Roll cost: difference between successive forwards and expected cost if roll dates coincide with market stress.

Incorporate a conservative execution cost buffer (e.g., 5–15 bps per trade) when evaluating historical returns.

Step 7 — Backtesting best practices

  1. Use realistic historical forwards; if only forward points are available, reconstruct using spot and published forward points to reflect actual traded quotes.
  2. Avoid look‑ahead bias: base decisions only on information available at rebalancing time.
  3. Handle corporate actions: currency redenominations, market closures, and regime changes need treatment or exclusion.
  4. Survivorship: include delisted or illiquid pairs historically present in your universe to avoid upward bias.
  5. Risk‑adjust: present returns net of financing and transaction costs, and report Sharpe, Sortino, drawdown, and tail metrics (e.g., 95% conditional drawdown).
  6. Walk‑forward test: validate parameter stability by rolling in‑sample/out‑of‑sample tests (e.g., three years in, one year out).

Step 8 — Risk management and limits

Key controls to implement in live trading:

  • Gross exposure cap: limit total notional across all hedged positions (e.g., 3× equity).
  • Per‑pair cap and sector/country concentration limits for EM exposures.
  • Stop frameworks: rather than per‑trade stops, use portfolio stress limits—if 5% intraday drawdown across portfolio, pause new positions and reduce size.
  • Counterparty and settlement risk: diversify forwards across 2–4 dealers and monitor rolling exposure and collateral postings.
  • Liquidity contingency: predefine exit plans if forward markets widen (>x historical spread) or central bank intervention risk rises.

Monitoring and performance attribution

Track these daily and weekly:

  • Carry P&L vs momentum P&L decomposition.
  • Forward roll cost and realised carry income.
  • Correlation of strategy returns with spot benchmarks (e.g., DXY) and global rate moves.
  • Exposure metrics: net and gross notional, weighted realised vol, and VaR.
  • Dealer spreads and slippage trends—if execution costs rise, pause scaling in.

A worked hypothetical example

Hypothetical, simplified single‑pair example to clarify mechanics:

  1. Universe: EUR/USD. Spot S = 1.0800. 3M forward F = 1.0740.
  2. Compute annualised carry: (F/S − 1) × 4 = (1.0740/1.0800 − 1) × 4 = (−0.005556) × 4 ≈ −2.22% → EUR pays carry vs USD.
  3. If composite score ranks EUR as a top short candidate and you short EUR spot 10m notional and enter a 3M forward to buy EUR at 1.0740, you lock the forward receive USD at that rate — you crystallise the implied carry and neutralise subsequent spot moves for the 3M window.
  4. P&L drivers over 3M will be: (i) realised forward roll (difference between forward at inception and forward at close if you unwind early), (ii) counterparty costs and spreads, and (iii) any basis if OIS/term rates diverge under stress.

Scaling to a portfolio repeats this for selected pairs with weightings and risk limits.

Common pitfalls and how to avoid them

  • Ignoring forward liquidity: narrow historical forward quotes can widen dramatically under stress—test with stressed spread scenarios.
  • Over‑leveraging: carry looks stable until a sudden risk‑off wipes gains; cap leverage and enforce drawdown limits.
  • Neglecting counterparty exposure: forwards are OTC; diversify and net exposures across dealers.
  • Overfitting parameters: prioritize economic sense (carry captures rate premium; momentum captures trend) over curve‑fitting to past returns.

When this strategy performs best

Carry‑tilted, hedged implementations typically perform well in stable rate‑differential regimes with moderate volatility, where funding premia persist and market participants do not rush into safe‑haven flows. Momentum overlay helps by avoiding trade entries into carry winners that are already in rapid secular decline.

Final checklist before going live

  1. End‑to‑end backtest with transaction costs and roll rules.
  2. Operational readiness: dealer connectivity, collateral lines and trading permissions for forwards.
  3. Predefined risk limits, monitoring dashboards and escalation procedures.
  4. Paper‑trade for a full roll cycle (3 months) to validate execution and P&L attribution.

Combining carry and momentum with a disciplined forward‑hedging program yields a strategy that captures funding premiums while materially reducing unwanted spot beta. The mechanics require attention to forward liquidity, execution costs and counterparty management. Done correctly, the approach offers a practical, repeatable way for FX traders in 2026 to harvest interest differentials without leaving themselves exposed to one big directional shock.