Yen Rips to a 7-Month High as the Carry Trade Unwinds — 155 Is the Line Before the Mid-140s

The Carry Unwind and the Mid-140s Scenario

The carry trade is the dominant speculative force in this pair, and it is unwinding faster than expected.

The mechanics are straightforward. Investors borrow yen at low cost and invest in other currencies and assets offering higher yields. That trade has been extraordinarily profitable for four years because the funding cost was near zero and the yen kept depreciating, delivering return on both legs.

Rising Japanese rates attack both legs simultaneously. Higher funding costs reduce the carry, and a strengthening yen turns the currency leg from a tailwind into a loss.

Higher Japanese rates weaken the justification for maintaining large short-yen positions. If the Bank raises rates in September and maintains hawkish communication, the rationale for selling the yen weakens further.

The scale of outstanding positioning is what makes the downside path asymmetric. Years of one-directional yen weakness built an enormous short base, and unwinding it requires buying yen regardless of what the fundamentals say on any given day.

That is the mid-140s scenario. It does not require the rate differential to compress to a level that justifies 145. It requires enough leveraged shorts to be forced out.

The move is also no longer purely a positioning story. Early hints of capital repatriation and expectations of a faster tightening pace, combined with US pressure on Japan, are all contributing.

Traders are currently choosing to retrace their steps ahead of the pivotal central bank meetings, which is why volume and volatility have both risen while the direction has stayed one-way.

The self-reinforcing dynamic is the risk in both directions. Stop-losses accelerate the decline on the way down. If the Bank of Japan delivers a hike with dovish guidance — signalling that 1.25% is terminal — the same leverage reverses, and a squeeze back toward 157 would be equally fast.

Repatriation: Pension Funds and the JGB Rotation

The flow story underneath the positioning story is potentially more durable, and it centres on where Japanese institutional money is invested.

Speculation over a potential shift in the Government Pension Investment Fund's asset allocation has provided support for the yen. Separately, potential rotation from foreign bonds into Japanese government bonds among domestic investors, including pension funds, could support the currency.

The logic is arithmetic rather than sentiment. Japanese institutions hold enormous foreign bond portfolios accumulated over decades when domestic yields offered nothing. As JGB yields rise with the policy rate, the relative case for holding unhedged foreign debt weakens, and the case for bringing capital home strengthens.

The scale is what makes this significant. Japanese institutional foreign bond holdings run into the trillions of dollars. Even a modest reallocation percentage translates into currency flows that dwarf daily speculative volume.

Unlike the carry unwind, repatriation is not a positioning phenomenon that exhausts. It is a structural reallocation that would proceed over quarters and years as domestic yields become progressively more attractive.

That is the strongest argument for a durable yen recovery rather than a positioning squeeze, and it is why the current move has drawn more attention than prior yen rallies.

The qualification is that this has been anticipated for two years without materialising at scale. Japanese institutions have been slow to rotate, partly because the hedging costs on foreign bonds already reflected the differential and partly because domestic yields remained low in absolute terms even after the increases.

At a 1.25% policy rate, JGB yields would still sit far below Treasury yields at 4.80% on the ten-year and 5.27% on the thirty-year. The unhedged carry on US Treasuries remains substantial.

The rotation becomes compelling at Japanese policy rates closer to 2%, not 1.25%. That is a 2027 story rather than a 2026 one, unless the back-to-back path Takata described materialises.

The 2024 Precedent and Why This One Is Better Telegraphed

The reference point everyone is working from is August 2024, and the comparison cuts both ways.

The stakes of a disorderly unwinding were laid bare in 2024, when a Bank of Japan rate hike sent the yen higher, forced traders to abandon their carry trades and sent shockwaves across global markets for days.

That episode is why the current move is being watched so closely well beyond the FX market. A yen squeeze does not stay contained in USD/JPY — it forces liquidation of whatever the borrowed yen was funding, which in 2024 meant global equities and high-beta risk assets across multiple continents.

The critical difference this time is telegraphing. The 2024 hike was a surprise against dovish positioning. The September 2026 hike is priced at 97% with a month of lead time, which means the leveraged shorts most vulnerable to it have had time to reduce.

That should mute the disorderly component. Markets that see a move coming for four weeks do not produce the same cascade as markets that are ambushed.

The evidence supports partial de-risking already. USD/JPY has fallen more than 4% in September without any corresponding stress in global equities. The S&P 500 (SPX) slipped only 0.35% to 7,691.55 on Tuesday and the VIX held at 15.46 — nothing resembling the cross-asset contagion of 2024.

The counterargument is that the 2024 unwind also began quietly before accelerating. Positioning data lags, and the size of the outstanding short base is estimated rather than observed.

The scenario that would replicate 2024 requires a hawkish surprise on top of the priced hike — a 50-basis-point move, or guidance explicitly committing to October. Either would force the remaining shorts out simultaneously into a market that had already absorbed the expected outcome.

The probability of that is low. The consequence if it happens is high. That combination is precisely what options markets should be pricing, and it is why implied volatility in yen crosses has risen while spot volatility in EUR/USD has collapsed.

Oil at $99 Is the Yen's Hidden Problem

There is a fundamental force working directly against the yen that the current momentum is ignoring.

Japan imports nearly all of its energy. Brent crude climbed 2.3% to $99.22 on Tuesday, its highest since 24 July, after Houthi drone and missile strikes on Saudi Aramco facilities wounded 73 people and halted operations at several sites. WTI reached $94.60. European natural gas hit a three-and-a-half-year high above €73 per megawatt-hour, with Qatar under extended force majeure.

Every dollar of increase in energy prices is a direct terms-of-trade transfer out of Japan. A deteriorating trade surplus or rising import costs — particularly energy — weakens the yen mechanically through the current account.

The paradox is explicit in the current setup. The yen carries a long-standing reputation as a safe-haven currency, and during periods of global uncertainty investors typically buy it. But in 2026 the Middle East conflict has pushed oil higher, which weighs on Japan's import-heavy economy and has at times worked against the yen's usual safe-haven bid.

So the same geopolitical escalation that should generate yen buying on risk-off grounds is generating yen selling on terms-of-trade grounds. Those two forces have partially cancelled all year.

Japan is also competing directly for LNG cargoes. Japanese and Korean buyers are scrapping for shipments against European demand as both regions restock before winter, which means Japan is paying at the margin for scarce energy in a market where the marginal price has doubled.

That import bill lands in the trade balance over the coming quarters and is the strongest fundamental counterweight to the rate-differential story.

It also complicates the Bank of Japan's position. Energy-driven inflation is the least desirable kind to tighten into, because it is a tax on households rather than a sign of demand strength. A committee raising rates into an oil shock risks compounding a real income squeeze.

If crude sustains above $100 into winter, the terms-of-trade drag becomes large enough to cap the yen's recovery regardless of what the Bank does with rates.

Forecast Dispersion Runs From 149 to 165

The professional forecasting community disagrees about this pair more than about any other major, and the spread is worth mapping because it defines the plausible range.

The bearish USD/JPY camp targets 149 for year-end, roughly 2.5% below the current 152.89 low. That case rests on faster Bank of Japan tightening, the carry unwind, and scope for Japanese institutions to repatriate capital as domestic yields become attractive.

The bullish USD/JPY camp targets 164, arguing that structural dollar demand from Japanese corporates and persistent carry flows offset the rate-differential narrowing. One twelve-month forecast was revised to 165 from 155 on 6 July 2026 — a call made when the yen was at four-decade lows and one that has been badly wrong-footed since.

Aggregate bank forecasts range 150 to 164.

That is a 15-yen spread, or roughly 10%, on the same instrument for the same date. Currency forecasting is difficult everywhere, but a dispersion that wide on a G3 pair indicates genuine structural uncertainty rather than ordinary disagreement.

The dispersion also explains the price action. When professional views span 10%, the market has no anchor to mean-revert toward, and momentum runs further than fundamentals justify in both directions.

The forecasting record on this pair specifically should temper conviction. The 165 revision on 6 July preceded a 7% move the other way. Positions entered at 159.70 with a 164 stop have gained 4.3% in under a week.

Neither outcome was well-anticipated by consensus a month ago.

The practical implication is that anyone using published targets as a reference point should treat the 149-to-165 band as the plausible range rather than any point inside it as a forecast. That range spans the entire distance from a full carry unwind to a full resumption of dollar strength.

The determining variable is not currently a forecastable one. It is whether Governor Ueda's press conference on 18 September validates or deflates the back-to-back framing.

Levels: 155 and 157.48 Above, 150.83 and the Mid-140s Below

Consolidate the map.

Resistance begins at 153.50, where the pair traded in the early European session before extending lower. Above that, 155 is the structural level whose loss defined this move and whose reclaim would repair the short-term damage. The 100-day simple moving average and the Bollinger midline both sit above current price, and price has to trade back through both to change the daily bias.

Beyond 155, the 157.48 area corresponds to a prior high cluster and represents the most important upside barrier. Above it, the 159.70 level is where the current dominant short position was established, and 164 sits at the stop level and at the top of the aggregate forecast range.

Support: 152.89 is Tuesday's low. Below it, 150.83 corresponds to a long-term upward trendline, and a return to that zone would put the trendline at risk. The 150 round number sits alongside it and marks the bottom of the aggregate forecast range.

Beneath 150, the 149 year-end target is the first published objective, roughly 2.5% below the current low. The mid-140s is the positioning-driven extension cited if the unwind continues, and 145 represents the practical floor of that scenario.

Percentage distances from 152.89: resistance at 155 is 1.4% up, 157.48 is 3.0% up, 159.70 is 4.5% up. Support at 150.83 is 1.3% down, 149 is 2.5% down, 145 is 5.2% down.

Those are tight increments for a pair moving 1% a day, which means the entire mapped structure can be traversed inside a week if the meetings deliver a surprise.

Momentum is unambiguous on the daily timeframe. Price holds well below the Bollinger midline and the 100-day average with a clearly bearish near-term bias, and there is no oversold divergence signalling exhaustion.

The single most important level to watch is 155. Reclaiming it converts this from a structural break into a failed breakdown and would suggest the carry unwind has run its course. Until that happens, rallies are corrective within a downtrend.

Verdict: A Positioning Unwind Wearing a Fundamentals Costume

USD/JPY at 152.89 with the yen at a seven-month high has moved more than 4% in six sessions on a repricing that is real but smaller than the price action implies. The trigger is genuine: money markets now put 97% odds on a Bank of Japan increase to 1.25% on 18 September, up from 52% a month ago, with board member Hajime Takata explicitly stating that a 25-basis-point step is not necessarily set in stone and that back-to-back hikes are possible — against a July vote that rejected 1.25% by 8–1 and a June hike that passed 7–1. But the arithmetic does not support the magnitude. The Fed's 3.625% midpoint against the Bank of Japan's 1.00% is a 262.5 basis point differential, and if both central banks hike next week — the Fed on 16 September at roughly 58% to 60% odds, the Bank of Japan on 18 September at 97% — that differential sits exactly where it started. The 100-basis-point-to-five-to-eight-yen historical relationship implies three to five yen of decline from the compression delivered so far. The pair has fallen considerably more. The excess is the carry unwind, which is why the stated downside path is framed around the size of the outstanding short-yen position rather than around any rate target, and why the mid-140s is described as achievable without a corresponding fundamental case. Working against it: Brent at $99.22 and European gas above €73/MWh are a direct terms-of-trade hit to an economy that imports nearly all its energy, and Japanese buyers are competing head-on with Europe for the same scarce LNG cargoes. The path: hold below 155 and the sequence runs 150.83 and the long-term trendline, then 149, then the mid-140s on a hawkish 18 September press conference — 1.3% to 5.2% of further downside. Reclaim 155 and the 100-day average and this becomes a failed breakdown with 157.48 and then 159.70 back in play. Bias is bearish USD/JPY below 155 and neutral above it, and the honest framing is that a 97%-priced hike carries almost no surprise capacity on the decision itself — everything now rests on whether Ueda validates the back-to-back path or deflates it, with a 149-to-165 forecast dispersion telling you how little anyone actually knows.

That's TradingNEWS