USD/JPY holds a 157.30–160.30 band into the September Federal Open Market Committee (FOMC) meeting, with a base case of 160.30 by September 30, 2026, a bull case of 163.91 and a bear case of 156.68. The mechanism is carry-to-volatility, not the rate differential — which has narrowed, not widened, since the intervention.

USD/JPY traded at 159.19 at 07:34 UTC on August 19, 2026 (FXRatesAPI), against a European Central Bank (ECB) reference fix of 159.70 on August 18. Since the August 3 low of 156.68, the two-year United States–Japan yield differential has narrowed by 18.9 basis points, yet the pair has risen 1.93%. That gap between rates and price is the whole call, and the Jackson Hole symposium on August 27–29 is the event most likely to close it.

Key Levels:

• USD/JPY: 159.19 spot, taken 07:34 UTC on August 19, 2026 — FXRatesAPI; ECB reference fix 159.70, August 18, 2026
• Base case target: 160.30 by September 30, 2026 — the 50% retracement of the 163.91–156.68 intervention drop, measured on ECB daily reference fixes
• Bull case target: 163.91 — the July 28 fix, the 2026 high; requires a hawkish Warsh keynote and a Bank of Japan (BOJ) hold on September 18
• Bear case target: 156.68 — the August 3 post-intervention low; requires a BOJ hike to 1.25% on September 18
• Major support: 158.39 — 23.6% retracement of the same drop; then 157.41, the August 4 ECB fix
• Major resistance: 159.44 — 38.2% retracement; then 160.00, described by State Street Global Advisors as “a political line in the sand”
• Invalidation level: a daily close above 161.15 (61.8% retracement) or below 157.28, either held for two consecutive sessions

Methodology: what this call is built on and what it is not

Price history is ECB euro reference rates converted to USD base, January 2 to August 18, 2026, cross-checked against an intraday quote at 07:34 UTC on August 19. Yields are official: the US Treasury daily par yield curve and the Japanese Ministry of Finance JGB rate file, both to August 18. Positioning is the Commodity Futures Trading Commission (CFTC) Commitments of Traders (COT) legacy report to August 11. Volatility is realised, not implied: annualised standard deviation of daily log returns on the fix series. That is the caveat — a daily-fix series understates intraday range, so these volatility numbers are a floor, not a ceiling.

The rate differential narrowed and the pair went up anyway

The standard yen story is a rate-gap story: US rates high, Japanese rates low, money flows one way. That story has been correct for four years. It is not what happened in the past fortnight.

Variable	August 18, 2026	August 3, 2026	Change	Direction vs USD/JPY
US 2-year yield	4.19%	4.25%	-6 bp	Yen-supportive
JGB 2-year yield	1.691%	1.562%	+12.9 bp	Yen-supportive
2-year differential	249.9 bp	268.8 bp	-18.9 bp	Yen-supportive
US 10-year yield	4.71%	4.70%	+1 bp	Neutral
JGB 10-year yield	2.934%	2.824%	+11.0 bp	Yen-supportive
10-year differential	177.6 bp	187.6 bp	-10.0 bp	Yen-supportive
USD/JPY (ECB fix)	159.70	156.68	+1.93%	Yen-negative

Sources: US Department of the Treasury daily par yield curve; Japan Ministry of Finance JGB interest rates; ECB euro reference rates. Time window: August 3–18, 2026. Differentials computed by TheIndustrySpread.

Every single rate input moved in the yen’s favour and the yen still lost 1.93%. The resolution is volatility. Measured on ECB fixes, annualised realised volatility in USD/JPY ran at 4.66% across the 20 sessions to July 29, spiked to 11.87% across the intervention window of July 30 to August 3, then collapsed to 2.87% across the 11 sessions from August 4 to August 18. Divide the 249.9 basis point two-year differential by that 2.87% and the carry-to-volatility ratio is 0.87, against 0.54 immediately before the intervention and 0.36 on a 60-session volatility window. In other words, the carry trade is not being paid more; it is being charged less. A position that yields the same but moves half as much is, on any risk-adjusted sizing rule, a bigger position. That is why 159 was reclaimed without the rate gap widening by a single basis point.

“The intervention successfully reset market psychology and demonstrated an unusually strong degree of U.S.-Japan policy coordination. What it has not yet done is eliminate the yield advantage supporting the dollar.”

— Masahiko Loo, senior fixed income and currency strategist, State Street Global Advisors (CNBC)

The mechanism: a cleaner book is a more dangerous one

The joint operation did something the rate differential did not. It emptied the speculative book. CFTC data show non-commercial accounts holding a net short yen position of 163,412 contracts on July 28, the week USD/JPY fixed at its 2026 high of 163.91. By August 11 that net short was 42,085 contracts — a 74.2% reduction in three weeks, with open interest down from 432,366 to 391,874.

COT report date	Non-commercial long	Non-commercial short	Net position	Open interest
July 21, 2026	107,590	259,715	-152,125	423,796
July 28, 2026	101,271	264,683	-163,412	432,366
August 4, 2026	147,228	192,701	-45,473	419,393
August 11, 2026	134,188	176,273	-42,085	391,874

Source: CFTC Commitments of Traders, Japanese yen futures, Chicago Mercantile Exchange (CME), legacy futures-only report. Time window: July 21 – August 11, 2026.

This cuts against the intervention’s own purpose. A crowded short is fragile — it unwinds violently on a shock, which is what the Ministry of Finance and the US Treasury engineered between July 30 and August 3. A three-quarters-empty short book is not fragile. It has no overhang to squeeze and room to rebuild, so the 159 handle rests on a lighter base than the 163 handle did. The next leg needs no new speculators, only the absence of a shock.

The steelman against all of this: it treats the yen as a rates asset when it may be a capital-flows asset. Crédit Agricole CIB argues the deeper problem is an “asymmetry of investment power”, with US artificial-intelligence capital expenditure absorbing global savings while Prime Minister Sanae Takaichi’s investment programme has yet to land. On that reading no BOJ path fixes the yen and the bear case is too generous to Tokyo.

What the model misses

Three limits. First, realised volatility is backward-looking; a ratio of 0.87 describes what the trade has cost, not what it will cost, and it mean-reverts violently. The 2024 analogue is instructive — that August’s yen carry unwind followed a period of similarly compressed volatility, and the compression was the warning, not the reassurance.

Second, the framework says nothing about a second intervention. Official flow is not a price-taker, and a repeat operation would put the pair below the bear case within hours whatever the two-year differential is doing. Treasury Secretary Scott Bessent and Finance Minister Satsuki Katayama have both said they are ready to act again.

Third, September’s risk is asymmetric in a direction most models are not calibrated for. On August 19, 2026 Polymarket’s Fed decision market — $37.2 million of volume — priced no change at 71.5%, a 25 basis point hike at 28.5% and a cut of any size at 1.3%. Chair Kevin Warsh has removed forward guidance from Fed communication, which is why the Jackson Hole keynote is a volatility event: it is the only scheduled chance to observe the reaction function.

“The yen’s weakness is a misalignment reflecting Japan’s inconsistent economic policies, not market disorder as the ministry of finance would have it known.”

— Mark Sobel, Vice Chair and Chief Economist, OMFIF (OMFIF)

What would invalidate this call

The 157.30–160.30 band and the 160.30 base case break if any one of these four signals fires:

• Two consecutive daily closes above 161.15 — the 61.8% retracement. Above it the drop is retraced on the standard reading and 163.91 becomes the base case, not the tail.
• The BOJ raises the policy rate to 1.25% on September 18. The board held at 1% on July 31 by 8-1, Hajime Takata dissenting for a hike. A move would compress the two-year differential through 225 basis points and drag carry-to-volatility below 0.55 even at current volatility.
• Realised volatility above 6% annualised on a 10-session window. The ratio falls to roughly 0.42, back inside the pre-intervention regime, and the risk-adjusted case disappears without the rate differential moving at all.
• A second joint intervention. Any operation the size of July 30 – August 3 puts the pair below 156.68 mechanically. Loo calls 160.00 “a political line in the sand”, so the trigger and the resistance are the same number.

What to watch next

The July 28–29 FOMC minutes are due August 19, 2026 on the Fed’s three-week convention — the first read on how divided the committee was at the hold. Japan’s July national consumer price index follows on August 21; the BOJ said on July 31 that core inflation was likely to accelerate “clearly above” 2% from the second half of its 2026 fiscal year. The Bureau of Economic Analysis publishes the second estimate of second-quarter gross domestic product (GDP) and July Personal Consumption Expenditures (PCE) together at 08:30 on August 26, the day before Jackson Hole opens. The symposium runs August 27–29 at Jackson Lake Lodge on the theme “Financial Innovation: Implications for Payments and Policy”, with Warsh’s keynote on Friday, August 28. Then the decisions that settle it: the FOMC on September 15–16 with a Summary of Economic Projections, and the BOJ on September 17–18.

Frequently asked questions

Why does a narrowing rate gap not strengthen the yen?

Because carry traders size positions on risk-adjusted return, not raw yield. A 249.9 basis point two-year differential against 2.87% annualised realised volatility gives a better ratio than a 268.8 basis point differential against 4.66% volatility did in July. The gap got smaller and the trade still got more attractive.

Did the July 30 – August 3 intervention fail?

It succeeded at what it was designed to do and failed at what markets wanted. It cut the speculative net short by 74.2% and drove USD/JPY from 163.91 to 156.68, but changed none of the policy settings that produce the differential — so half the move was given back within two weeks.

What is the market pricing for the September FOMC?

As of August 19, 2026, Polymarket’s $37.2 million Fed decision market priced no change at 71.5%, a 25 basis point hike at 28.5%, and a cut of any size at 1.3%. The live debate on this committee is hike-versus-hold, not cut-versus-hold — a distinction that matters for anyone extrapolating from previous years.

Why does Jackson Hole matter more than usual this year?

Chair Kevin Warsh took office on May 22, 2026 and has removed forward guidance from Fed communication. His August 28 keynote is one of the few scheduled chances to observe the reaction function directly, making it a volatility event rather than a rates event — and volatility is the variable this call turns on.

What level brings Tokyo back into the market?

State Street’s Masahiko Loo has described 160.00 as “a political line in the sand” and said he “would not rule out another intervention, particularly if the move becomes rapid or disorderly”. The observable trigger is therefore less the level itself than the speed at which it is reached.

How does this sit against the previous house call?

TheIndustrySpread’s August 13 piece argued for 163.50 by September 30 on the intervention’s half-life. This call is materially less dollar-bullish for one reason: the volatility regime has since compressed further, and a compressed regime is a mean-reverting one. Both calls share the same invalidation logic at 161.15.

Related coverage on the same policy complex: Japan’s 10-year JGB to 3.10% by Q4 2026, the DXY Warsh hawkish-repricing case, the Nikkei 225 BOJ-flinch case and the September FOMC hike-unwind case.

This article is informational analysis only and is not financial, investment, or trading advice. Foreign-exchange, commodity, and equity markets are highly volatile and can lose substantial value rapidly. Leveraged products carry total-loss risk and may exceed the initial margin posted. Past performance and historical correlations do not guarantee future results. Do your own research and consult a regulated financial adviser before making any investment decision.