June–September 2024: The G10 Yen Carry Unwind and Seven-Cross Synchronous Decline

Mid-session on June 5, 2024, USD/JPY traded above 155. Nine weeks later it had dropped through 142 — an 8.6% decline in the world's most liquid currency pair. But USD/JPY was not alone. Seven G10 yen crosses fell simultaneously over the same 68 trading days, a scale and breadth that made this the decade's most textbook carry-unwind episode in G10 forex.

The carry trade — borrowing cheaply in Japanese yen to fund higher-yielding positions in currencies like the Australian and New Zealand dollars — had built up for years in an environment of near-zero Japanese rates and elevated global yields. When that environment began to shift, crowded positioning turned from a source of steady carry income into a structural liability. The USD/JPY Historical Replay: 2024-06 to 2024-09 scenario captures this reversal across 68 daily bars and 12 decision points, four of them critical.

The scale of simultaneous movement — seven crosses declining together over 68 trading days — reflects how crowded positioning had become during the preceding low-volatility period. The CAD/JPY Historical Replay: 2024-06 to 2024-09 scenario illustrates this crowding: when the initial crack appeared, cascading stop-loss activity and position liquidations accelerated the move. No single currency pair moved in isolation; the entire carry basket felt the pressure.

The trigger was not a single event but a gradual shift in expectations around Bank of Japan policy. As the BoJ signaled a move away from its ultra-loose stance, the interest rate differential that had funded years of yen-funded carry positions began to narrow in expectation — not yet in actual rates, but enough to undermine the carry trade's fundamental premise. When crowded positioning meets a macro catalyst, what follows is rarely gradual.

The spread of the unwind across different crosses varied by their carry characteristics. AUD/JPY, NZD/JPY, and USD/JPY — the highest-yielding pairs in the basket — saw the steepest losses, consistent with their greater carry intensity. CHF/JPY, which functions more as a safe-haven instrument than a carry vehicle, recorded the shallowest decline. This gradient reveals how carry intensity determines vulnerability: positions built on high interest rate differentials face the sharpest re-pricing when those differentials compress in expectation.

Institutional carry desks faced mark-to-market losses on long positions across the carry basket, triggering deleveraging that spread losses across correlated pairs. Liquidity providers widened spreads or stepped back precisely when execution mattered most, amplifying volatility through positive feedback between forced selling and deteriorating market depth. The mechanism is well-documented in carry crises: low-volatility crowding masks tail risk, compresses cross-pair correlations during the buildup, and then reverses sharply when conditions change.

In the aftermath, institutions revised stress-test scenarios to capture simultaneous unwind of crowded carry positions. Regulatory frameworks tightened scrutiny of yen-funding strategies and their cross-border transmission channels. The structural vulnerability — positions that feel stable in low-volatility regimes but crystallize rapidly on a macro catalyst — became embedded in the post-event institutional memory of the G10 FX market.

For decision trainers, the core tension here is not about predicting the direction of carry. The AUD/JPY Historical Replay: 2024-06 to 2024-09 scenario demonstrates that even correct directional calls become unmanageable when position sizing ignores the drawdown that precedes the move. The discipline of systematic position sizing and pre-defined exit rules matters more than the directional conviction. The USD/JPY Historical Replay: 2024-06 to 2024-09 shows how liquidity can evaporate rapidly under stress, turning a manageable position into a forced exit regardless of conviction.

Data coverage for the attached replay scenarios extends through September 8, 2024. Each member covers 68 daily bars representing the full data span of that particular cross, while this narrative focuses on the central carry-unwind episode and its market structure consequences.