The yen rally makes a BoJ hike less likely, not more.

That runs against the read on the tape, so let me put the mechanism on the table. The only inflation channel the Bank of Japan has ever reacted to with any consistency is imported cost, transmitted through the currency. Domestic wage-price dynamics get discussed; the import bill gets acted on. A 3.37% move in USD/JPY over five sessions, with EUR/JPY down 3.33% and GBP/JPY down 3.43%, does more to compress that channel than any plausible policy step at the next meeting. Brent at 99.097 and WTI at 93.102, up 6.58% and 7.09% on the week, is precisely the shock that would have forced the Bank's hand a year ago. Priced in a stronger yen, much of it is neutralised. Appreciation is a substitute for tightening, and a central bank that has spent a decade avoiding the latter will take the former gratefully.

The uniformity of the move is the second reason to doubt the policy story. When a terminal rate genuinely gets remarked, the crosses separate, because each counterpart carries a different path. Three crosses moving within a tenth of a percentage point of each other over a week is one leg being closed, not three differentials being repriced. Carry unwinds look like this. They also rebuild, because the funding advantage that created the position has not been withdrawn.

What deserves more attention than the currency is the long end. Bund futures fell 0.32% on the day and gilt futures 0.52%, against 0.12% for the ten-year Treasury contract. Japanese capital is the marginal bid for duration in Europe, and a yen that stops bleeding changes the arithmetic of hedged foreign paper before it changes anything in Tokyo. The pressure shows up in Berlin and London before it shows up in the JGB curve, which is exactly the sequence that makes the trap durable: the Bank can suppress its own long end and export the adjustment.

On this side of the Pacific, the Fed's balance sheet stood at 6,737,204 million dollars and rose in the latest week while the effective funds rate sat unchanged at 3.63%. The statement will describe a policy on hold. The asset side is doing something else, and it usually does it first. Ten-year breakevens at 2.35% with the thirty-year at 5.25% tell you the market wants term premium, not inflation protection, which is the correct instinct when the fiscal path is the binding constraint.

Position accordingly: the yen strength is a squeeze that fades, and the trade with legs is short duration in the periphery of the Japanese bid, meaning gilts and Bunds rather than Treasuries. The BoJ will not spend this gift on a hike. It will bank it.

The yen rally is real repricing, and the Bank of Japan should not follow it.

Start with what did not happen. There is no stress in this move. High yield spreads sit at 2.65 and tightened on the day. Gold fell 0.54%, copper rose 0.59%, and the dollar did nothing against anything else: EUR/USD up 0.05% over five sessions, GBP/USD down 0.06%, AUD/USD up 0.73%. This is not a risk-off cascade or a broad dollar turn. The yen moved alone, 3.37% against the dollar and 3.43% against sterling in five sessions, and it moved while Brent gained 6.58% to 99.097 and WTI gained 7.09% to 93.102.

That last combination is the part worth sitting with. Japan imports its energy. A five-day crude move of that size is a straightforward hit to the terms of trade, and the currency appreciated through it. Markets do not do that unless they have decided the policy rate is going somewhere. So the repricing read is right, and the argument that yen weakness is a sideshow to be waited out does not survive a week where the currency strengthened into a fuel bill going the wrong way. The exchange rate has become the transmission channel that matters in Tokyo.

Where I part company with the desk consensus is on what the BoJ should do about it. Crude at these levels is a supply-side price, and it is the same kind of price that Japan has spent three decades importing without ever generating the domestic wage-price dynamic that would justify a tightening cycle. If the Bank ratifies this move, it tightens into an energy shock it cannot offset, and it does so at the exact moment the currency is already doing the disinflationary work for free. A stronger yen is a transfer to Japanese households who buy imported food and fuel, paid by the export complex that has run the adjustment in the other direction since the 1990s. Take the transfer. Do not also raise the price of credit.

The consequential channel is not Tokyo anyway. It is the global long end. Japanese capital funds duration everywhere, and the US 30y is already at 5.25 against a 10y at 4.77 and a 10y-2y spread of 0.41. That shape is a term premium story, not an inflation story: breakevens are 2.35 and have not moved. If the yen keeps appreciating and domestic yields keep repricing, the marginal Japanese buyer of foreign long bonds gets scarcer, and the steepening comes from the back end rather than from Fed cuts.

Position for that. Long-end supply pressure in dollars, not a carry-unwind panic. And if the BoJ hikes on the back of a crude spike, it will spend 2027 explaining why it tightened into a slowdown it manufactured.

This is not the carry unwind. Look at what did not move.

The yen leg is uniform: USD/JPY 154.352, off 3.37% over five sessions, EUR/JPY off 3.33%, GBP/JPY off 3.43%. Near-identical magnitudes across three crosses means the move is entirely in the yen, and the dollar contributed nothing. EUR/USD is up 0.05% on the week. Cable is down 0.06%. So this is one currency being bought back, not a dollar event and not a risk event.

A genuine deleveraging shows up in the funded assets first. It has not. AUD/USD is up 0.73% on the week. Rand is stronger, 15.9962, up 0.72%. Peso, zloty and forint all firmer against the dollar over five days. High yield OAS sits at 2.65, tighter by a basis point. Gold is down 0.97% at 4406.065, silver down 0.58%. Nothing in that set is consistent with forced position reduction. In an actual unwind, the high-carry EM complex goes first and hardest, because it is the same book. It went the other way.

What did sell is duration, globally and quietly. Ten-year Treasury futures down 0.36% on the week, Bunds down 0.49%, gilts down 0.23%. US 10y at 4.77 against effective funds at 3.63, thirty-year at 5.25. That combination, yen bid and foreign long end offered, with credit and EM untouched, is the domestic Japanese real-money leg, not the levered speculative one. Money coming home sells long bonds and buys yen. It does not sell Mexican carry, because it never owned it.

Consensus over the next few days will treat this as the opening move of a broader unwind and position for contagion into EM and credit. That trade is early and it is expensive, because the entry is into a bid tape. The pain is in the long end, where the marginal Japanese buyer of 5.25% thirty-year paper reprices his hedge cost and steps back. Ten-year breakevens at 2.35 are not what is driving 30s.

The part that does not fit the frame: WTI up 7.09% to 93.102 and Brent up 6.58% to 99.097, against a yen that rallied 3.37%. Japan imports the barrel. Terms of trade moved sharply against the currency and the currency rose anyway. Crude stocks drew 4,450 MBBL and cracks are elevated, so the oil move is physical and self-contained. But the two prices are supposed to talk to each other and this week they did not.

Trade the long end. Leave EM alone.