Bank of Japan September rate decision, a quarter point rise to 1.25% is 84% priced after the yen surge

Upcoming catalystiThe event that could move the exchange rate.

The Bank of Japan interest rate decision on 18 September 2026, at the end of a two day Monetary Policy Meeting. The Bank decides whether to raise the uncollateralised overnight call rate from 1.00% to 1.25%. It is decided by the Bank's Policy Board and announced by the Bank of Japan in Tokyo, with a press conference by Governor Ueda afterwards.

Event date and timeiWhen the event is expected. Shown in your own time zone, or in UTC if you switch the clock at the top of the page. If the exact hour is not confirmed, you only see the date.

18 September 2026. The Bank of Japan does not publish a release time in advance, so the day is confirmed and the hour is not. At recent meetings the decision has arrived between 03:00 and 06:00 UTC, and the Governor's press conference has begun at about 06:30 UTC. The timing itself is read as a signal in this market, because a decision that arrives late usually means a divided board.

Main currency pair affected

USD/JPY. This is the deepest market for the yen and the pair where a change in Japanese policy is expressed with the least interference. The Federal Reserve decides on 16 September, two days earlier, so the dollar side of this pair has already digested its own event by the time the Bank of Japan speaks, which leaves the reaction on the 18th cleaner than it looks. The pair rejected was EUR/JPY. It carries the same yen leg with no Federal Reserve decision anywhere near it, which is a real argument for it, and it was rejected on liquidity. It trades in a thinner book, it moves further on identical news, and the widening in the spread around a Tokyo morning announcement is material. GBP/JPY was rejected for a stronger reason: the Bank of England meets in the same week, which puts a second central bank on the other side of the pair.

Current priceiWhere the exchange rate stood when this write-up was made, and when that was.

154.37

Expected directioniWhich way we expect the token price to go after the event. Positive means it rises, Negative means it falls, Mixed means it could go either way.

Positive. The pair rises, which means the dollar strengthens against the yen. This is not a view that Japanese policy is turning back. It is a view that six sessions and 666 pips have already paid for the decision, that a quarter point still leaves the dollar paying 2.375 percentage points more than the yen, and that a rise delivered without a change of pace gives a heavily positioned market a reason to take profit.

Action before the eventiWhat we suggest you do before the event. Buy, Watch or Avoid.

Watch. Do not buy this pair before the announcement. The decision is 84% priced, the outcome is two sided, the release has no published time so you cannot size for the minute, and the chart is in a clean downtrend that this call is arguing against. The reward for being early here is small and the cost of being early is a position held through an unscheduled announcement. What to wait for is the press conference rather than the decision. If the rise arrives and the Governor gives nothing on timing, the case for the bounce is made and the pair can be bought with the 152.88 low as the level that proves it wrong. If he confirms a faster pace, this call is void and the trade is not worth taking in either direction until the pair settles.

Why it mattersiHow the event reaches the price. Through demand, supply, how much the network is used, fees, or the rules the token trades under.

The chain runs like this. The Bank of Japan raises to 1.25%, which lifts Japanese short term yields, which narrows the two year yield gap against the United States from 253 basis points, which makes holding yen less costly, which supports the yen and pushes this pair down. That is the chain the market has already traded, and it is why the pair is 666 pips lower than it was on 1 September. The chain that has not been traded runs the other way. A gap of 2.375 percentage points after the rise is still an enormous incentive to own dollars and owe yen. Japanese inflation excluding fresh food is 1.8% and below target, which gives the Bank every reason to move once and wait. American growth has slowed to 1.5% annualised and Japanese growth was revised up to 1.4%, so the growth story now favours the yen, but that is a slow driver and it is already in the six session move. Both sides of this pair are hawkish at the same time, which is unusual and is the thing most likely to be misread. If the Federal Reserve raises on 16 September and the Bank of Japan raises on the 18th, the rate gap ends the week exactly where it started.

Catalyst confidenceiHow sure we are that the event happens as described. Scored out of 100. Higher means more sure.

99%. The meeting is on the Bank of Japan's published calendar for 17 and 18 September and a decision is announced at every meeting, so the event itself is as close to certain as this market offers. What is not certain is the outcome. My own view is that a quarter point rise to 1.25% is about 80% likely, a little below the 84% the swaps market prices, because Japanese inflation excluding fresh food is still under target and the board has shown before that it prefers to wait for one more confirmation than it needs.

Price-reaction confidenceiHow sure we are that the pair moves the way we expect. Scored out of 100. Higher means more sure.

58%. The direction call is genuinely two sided and the number reflects that rather than hedging it. The case for a bounce rests on how much is already priced, and that is a strong argument. Against it sit a downtrend the chart says is intact, a Federal Reserve decision two days earlier that this card cannot see the result of, a Japanese inflation print hours before the announcement, and clear reporting that a faster pace is under discussion inside the Bank. Any one of those can turn this the other way.

Expected moveiHow far we think the price moves, and how long the move may last.

Our Expected Move +60 to +140 pips, which is +0.39% to +0.91%. That range is set against an average daily range of about 112 pips, so the low end is a normal session and the high end is a decision day that goes this card's way. It is deliberately below what the last six sessions have produced, because a relief bounce in a downtrend is usually smaller than the move it interrupts. The comparable history is thin and worth saying so plainly. This pair has traded through only a handful of Bank of Japan increases in the modern cycle, the sample is too small to average into a figure, and the pattern within it is consistent rather than statistical: the yen has rallied into each of them and given most of it back within a fortnight, because the rate gap survived every one. Impact High. A rate decision with a live press conference is the largest scheduled event this pair gets, and on a pair whose average day is 112 pips a 150 pip session is a large one. The rating is about the event, not the pip count this card forecasts. Estimated Duration 3 days. The first reaction is over within hours of the press conference. What runs longer is the repricing of what the Bank does next, which is settled over the following week as officials speak and the swaps curve resets. If the decision changes the expected pace, the move is not a reaction at all and this duration is wrong in both directions.

What exactly is happeningiThe event itself in full. Who is involved, what changes, and the amounts and conditions that matter.

The Bank of Japan holds a two day Monetary Policy Meeting on 17 and 18 September 2026. The decision is announced on the second day. What it decides is the uncollateralised overnight call rate, which is the interest rate Japanese banks charge each other for lending overnight and the rate the whole yen interest rate curve is built on. The rate is 1.00% today. The Bank raised it to that level earlier in 2026 and then held it at the 31 July meeting, where one board member voted for 1.25% instead. The question on 18 September is whether the rest of the board joins that member. Forex traders care because Japan has spent most of the last three decades with interest rates at or near zero. Every step away from that is a step toward the yen paying a real return for the first time in a generation, and toward the end of the trade where investors borrow cheap yen to buy something that pays more. A quarter point does not sound like much. In this currency, the direction and the pace matter more than the size. Two other things are decided at the same meeting and both can move the pair as much as the rate itself. The vote split tells you how close the board is to the next move. The Governor's press conference, which normally begins at 06:30 UTC, is where the pace is set out in words rather than in a number.

Market expectationsiWhat most traders currently expect to happen. Bullish means they expect a rise, Bearish a fall, Neutral neither.

Bearish for the pair, which means the market's base case is that USD/JPY falls. Overnight index swaps price roughly an 84% probability of a quarter point increase to 1.25% on 18 September. An overnight index swap is a contract whose price moves with expected central bank rates, so it is the cleanest read on what traders are actually paying for. That figure has risen from roughly 75% to 80% at the start of September. Three pieces of evidence built it. Japan's July labour cash earnings rose 4.7% against a year earlier, against a 3.9% consensus and 4.0% the month before, the fastest pace since January 1997. Real wages, which take inflation out, rose 2.4%, the largest gain since May 2021 and the seventh consecutive monthly increase. Japan's second quarter GDP was then revised up to a 1.4% annualised expansion from the 1.1% first estimate, helped by a smaller fall in business investment than first reported. Governor Ueda has said publicly that a rate rise is on the table at every meeting including this one. Reporting from people familiar with the Bank's thinking says it is looking at September and at a faster pace of tightening than its recent rhythm of roughly two increases a year. Note what the consensus is and is not. It is a survey median and a swaps price, not a decision. The Bank has surprised in both directions before, and the individual forecasts behind a median like this one disagree more than the median suggests.

How much is already priced iniHow much of the expected outcome the exchange rate already reflects before the event happens.

High. USD/JPY closed at 160.193 on 1 September and at 153.536 on 9 September. That is 666 pips, or 4.16%, in six sessions. The single largest day was 3 September, which travelled 366 pips. Look at what moved in that window. The move was not a dollar story. It began with the wage data, extended on the GDP revision, and extended again as the swaps price for a September increase climbed. That is the market buying the hike in advance. The two year yield gap tells the same story from the other side. Japan's two year government bond yields about 1.84% against about 4.37% in the United States, a gap of 253 basis points in the dollar's favour. A basis point is one hundredth of a percentage point. That gap has been closing, and it has been closing because Japanese yields have risen rather than because American yields have fallen. Positioning is the last piece and it points the same way. The futures report covering positions to 1 September showed leveraged funds net short 102,188 yen contracts, about 24.8% of open interest. Yen futures rise when the yen strengthens, so a large yen short is the same thing as a crowded long position in USD/JPY. That is the fuel a 666 pip fall burns. The report covering the fall itself has not been published, so how much of that position survives is not something you can check today. So the hike is close to fully in the price. What is not in the price is the pace after it.

What else lands that dayiOther news and events on the same day that could push the price around, and whether they help this trade, bury it, or work against it.

18 September is not a clean day, and one item on it matters more than the rest. Japan's August national consumer price index is published at 08:30 Japan time, which is 23:30 UTC on 17 September, a few hours before the Bank announces. The same measure that half the argument for a rise rests on lands first, on the same currency, with the decision already written. A strong print gives the press conference a harder edge. A soft print gives the doves a line to use and can move the pair before the decision is out. The Federal Reserve decided two days earlier, on 16 September. That is close enough to matter. If the Federal Reserve raised its own rate on the 16th, the dollar arrives at the Bank of Japan meeting stronger and the rate gap barely narrows even after a Japanese increase. If it held, the dollar arrives weaker and a Japanese increase lands on a softer counterparty. 18 September is the third Friday of the month, which is the quarterly expiry for equity index futures and options in the United States, and the day large currency option expiries cluster on. Expiry flow is mechanical. It has nothing to say about Japanese monetary policy and it can still pin or drag a rate for hours. United States weekly jobless claims are at 12:30 UTC, which is after the decision and after the press conference. It is a small release on a day like this. What to expect on the day. The pair will probably have moved twice before the Bank of Japan says anything, once on the Japanese inflation print overnight and once as Tokyo positions into the decision. Treat the first hours as noise about the release rather than a verdict on the decision. The decision itself has no published time, so you cannot plan an entry to the minute, and the press conference at 06:30 UTC is where the pace is actually set. If you are holding into the European afternoon, remember that the flow you are trading against is expiry related and will be gone by Monday.

Bullish scenarioiWhat would have to happen for the price to rise, and why the market would read it that way.

The Bank raises to 1.25% and frames it as a step rather than the start of a run. The statement keeps the language about acting gradually, the vote is close to unanimous, and Governor Ueda declines to point at the next meeting. The market reads a rise it had already bought and no promise of another, and the pair recovers toward the 155.30 to 156.00 area as the short dollar position built over the last six sessions is closed. A weak Japanese inflation print that morning makes this outcome more likely, because it gives the Governor a reason to sound careful. A hold at 1.00% belongs here too, and it is the larger version of the same move. Roughly 84% of a rise is in the price, so removing it would force a fast unwind and could carry the pair through 156.00 toward the 20 day average near 157.97.

Neutral scenarioiThe outcome that simply meets expectations and moves the price very little.

The Bank raises to 1.25%, the statement reads much as expected, and the press conference gives nothing new about timing in either direction. The pair moves 40 to 60 pips in each direction through the morning and closes the week close to where it started. Expiry flow does more to the rate than the decision does.

Bearish scenarioiWhat would have to happen for the price to fall. A delay, a disappointment, or supply nobody expected.

The Bank raises to 1.25% and signals that the pace is quickening. That could be an explicit reference to a further move this year, a hawkish dissent asking for more, an upgrade to the inflation forecast, or the Governor confirming in the press conference that the roughly two rises a year rhythm no longer applies. This is the outcome the reporting has already flagged as under discussion, so it is a live risk rather than a tail. In that case the pair breaks the 152.88 low and opens the February 2026 floor at 152.26.

Result needed for a big moveiWhat the outcome has to be for the pair to move a lot instead of a little.

A move worth more than 150 pips needs one of two things. Upward, it needs the rise not to happen, or to happen with language that closes the door on the next one. With 84% priced, a hold alone is worth well over 150 pips. Downward, it needs a stated change of pace. A quarter point rise on its own is not enough, because that is the number already in the swaps price. What would be new is a second rise being put on the table for 2026, or an inflation forecast revision that implies one. The Japanese inflation print at 23:30 UTC on 17 September is the number to watch for it: core inflation excluding fresh food at or above 2.0% against the 1.8% July reading would make the hawkish version much easier for the board to write.

Other currency pairs affected

EUR/JPY, expected direction positive, meaning the euro rises against the yen. It carries the same yen leg without the Federal Reserve decision sitting two days in front of it, so it is a cleaner read on the Bank of Japan alone. It is also thinner than USD/JPY and moves further on the same news, so it is noisier to hold through a release. GBP/JPY, expected direction positive. Same yen leg again, and the most volatile of the three. The Bank of England also meets in the same week, which puts a second central bank into the other side of the pair and makes it a worse expression of this event rather than a better one. AUD/JPY, expected direction positive. This pair is the yen carry trade in its purest form, since the Australian dollar has been the classic thing to buy with borrowed yen. It reacts to global risk appetite as much as to Japanese policy, so it is a good confirmation of whether the carry trade is being rebuilt and a poor way to trade the decision itself.

Main risksiWhat could go wrong, and what would prove this call wrong.

Overall risk is High. The pace risk is the main one and it runs against this call. Reporting already says the Bank is considering a faster rhythm. If that is confirmed on the day, the pair does not bounce, it breaks. The Japanese inflation print lands hours earlier on the same currency and can set the tone before the decision is out. It can also make this analysis stale between now and then. The Federal Reserve decision on 16 September changes the dollar leg before the yen leg ever moves, and this card cannot know that outcome. A surprise there is a bigger move in this pair than the Bank of Japan is likely to produce. The decision has no published time. A position sized for a known minute is sized wrongly here. Quarterly expiry can hold the rate still through the London morning and then release it, which produces a move that looks like a reaction and is not one. The trend is strongly against this call. Betting on a bounce inside a 666 pip fall is the trade with the worst record in this market, and the fact that a chart is oversold has never stopped it becoming more so. Liquidity around the announcement is thin in the way that matters. The decision often lands while London is asleep and New York is closed, so the first move happens on the fewest participants of the day and is frequently reversed once Europe arrives.

What the market may be missingiSomething we think investors have overlooked or read wrong.

The rate gap is what this pair actually trades on, and a rise to 1.25% barely touches it. Do the arithmetic. The Federal Reserve target range is 3.50% to 3.75%, a midpoint of 3.625%. A Japanese rise to 1.25% takes the gap from 2.625 percentage points to 2.375. After the largest yen policy shift in a generation, holding dollars still pays you well over two percentage points a year more than holding yen. The market has spent six sessions trading the direction of Japanese policy and has priced very little about its destination. Even at the faster pace being discussed, the gap does not close in 2026 and probably does not close in 2027. That is why previous Bank of Japan increases have produced sharp yen rallies that faded within weeks, and why the pair is higher today than it was twelve months ago despite a rising policy rate the whole way. The second thing being underweighted is what the wage number actually says. Nominal earnings at 4.7% is the strongest since 1997 and it is genuinely new. But Japanese inflation excluding fresh food is 1.8%, which is below target, and it is the inflation number the Bank writes its forecast around. A board looking at strong wages and soft prices has a reason to move once and then wait, which is precisely the outcome this card is positioned for.

Every figure below uses price data through the 9 September close at 23:59 UTC. The rate quoted at the top of this card is the live rate at the time of writing on 10 September, which is why it sits above the last close.

TrendiWhether the price is above or below its own 50-day and 200-day averages. Above both is an uptrend, below both is a downtrend.

Short term: Bearish. Medium term: Bearish. Longer term: Neutral. Over the last month the pair fell from 159.294 on 14 August to 153.536 on 9 September, which is 576 pips or 3.61%. The high in that month was 160.392 and the low was 152.880, a range of 751 pips. Over three months it fell from 161.804 to 153.536, which is 827 pips or 5.11%. The high across that window was 163.985 and the low 152.880, a range of 1,110 pips. Over twelve months it is still higher, up from 147.085 to 153.536, which is 645 pips or 4.39%. The twelve month high is 163.985 and the low is 145.476. So the long trend is up and the last three months have taken back most of this year's gain. The structure of the last six sessions is a straight decline, not a correction inside an uptrend.

Market structureiThe shape of the chart. Rising highs and rising lows mean an uptrend, falling ones mean a downtrend, and neither means the price is stuck in a range.

Lower highs and lower lows since late June, which is a downtrend. Sellers control this pair. Picture it simply. The pair kept failing at lower levels on the way down, 163.99 in late June, 160.39 at the start of September, then 156.75 on 4 September, then 156.25 on 7 September. Each attempt to bounce stopped sooner than the last. That is what a downtrend looks like from the inside. The one thing that has changed is the speed. Six sessions produced 666 pips. A market that moves that fast is not distributing, it is repricing, and repricing moves end abruptly rather than fading.

Support and resistanceiPrices where buyers have stepped in before, and prices where sellers have. A price often stalls or turns at these.

First support is 152.88 to 153.00. The 8 September low of 152.880 and the 9 September low of 152.937 sit inside it, and those are the only two sessions in the last thirty to trade below 153.20. A daily close below this zone confirms the break and points at the next one. Second support is 152.26 to 152.60. The 152.259 low from February 2026 anchors it, and that is the last floor the pair built before this year's advance. A close below 152.26 would put the pair at its lowest in more than a year. First resistance is 155.30 to 156.00. The pair traded through this area on 3, 4 and 7 September on the way down, so it is recent supply with sellers still in it. It was also a level in early May 2026. A close above 156.00 would be the first sign this fall is over. Second resistance is 156.75 to 157.97. The 4 September high of 156.75 sits at the bottom of it and the 20 day average at 157.965 caps it. Reclaiming this zone would end the short term downtrend. Major resistance is 159.50 to 160.40. The 1 September close of 160.193 and the 2 September high of 160.392 are here, and so is the 50 day average at 159.578.

VolumeiWhether trading is picking up or dying down. Rising volume behind a price move means more people agree with it.

Increasing. Spot currency trading has no central exchange, so there is no true global volume figure for this pair and anyone quoting one is quoting something narrower. The figures here are tick counts from an aggregated price feed, which count how many times the quote changed rather than how much money traded. They are a fair measure of participation and they are not turnover. The twenty day average is about 267,600 ticks a session. 3 September, the 366 pip day, registered 321,629, which is 20% above that average. 8 September, the session that made the low, registered 320,601, also 20% above. 9 September registered 291,155, 9% above. Participation confirms the decline. Both of the two largest down days came with the two heaviest tick counts of the period, which says the move was taken seriously rather than drifting through a thin book.

RSI and momentumiWhether the price has been rising or falling hard lately, and whether that push is getting stronger or running out.

Weak. The 14 day relative strength index closed at 21.6 on 9 September. The index measures how much of the recent movement has been upward, on a scale of nought to 100, and readings under 30 are conventionally called oversold. Below 25 is rare on this pair. It says the fall has been one directional almost without pause, which is exactly what six consecutive lower closes looks like. Do not read that as a buy signal on its own. Oversold means the move has been fast, not that it has finished, and a currency in a genuine policy repricing can hold a reading in the twenties for a fortnight. What it does say is that momentum is stretched enough that a decision matching expectations is more likely to produce a bounce than a continuation, because the sellers who wanted this level already have it.

Moving averagesiThe average price over the last 20, 50, 100 and 200 days. Trading above them is generally a healthy sign, below them a weak one.

A moving average is the average closing rate over that many days, and nothing more. The 20 day average is 157.965 and falling. The live rate is 359 pips below it, which is 2.28%. The 50 day average is 159.578 and falling. The rate is 521 pips below it, which is 3.26%. The 100 day average is 160.112 and falling. The rate is 574 pips below it, which is 3.59%. The 200 day average is 158.787 and now flattening after rising for most of the year. The rate is 442 pips below it, which is 2.78%. Price is below all four and the shorter averages have crossed below the longer ones during this fall. That arrangement is bearish and it is also the arrangement that produces the sharpest counter moves, because a rate this far under its own averages has a long way to travel before it meets resistance from them.

VolatilityiHow much this pair swings day to day right now, compared with how much it usually swings.

High. The average daily range over the last twenty sessions is about 112 pips, which is 0.73% of the 154.37 rate. The 14 day average true range is 118 pips. Both are well above what this pair does in a quiet month, and the 3 September session alone travelled 366 pips. Implied volatility from the currency options market around this decision is not something this card can quote. That data was not available from a source worth citing, and a number invented for the sake of the heading would be worse than the gap. For position size, treat 112 pips as an ordinary day and assume the decision day is larger. A stop closer than about 120 pips to your entry is inside the noise and will be taken out by movement that means nothing. Size the position so that a 150 pip move against you is survivable, rather than widening the stop and keeping the size.

Breakout and breakdown levels

Upside confirmation is a daily close above 156.00. That clears the 3 to 7 September supply and opens 156.75 and then the 20 day average at 157.965. A close above 157.97 after that would end the short term downtrend and put 159.50 to 160.40 in reach. Downside confirmation is a daily close below 152.88. That takes out the low of this move and opens the February 2026 floor at 152.26. Below 152.26 there is very little structure until the low 150s. Both levels come from where this pair has actually traded, not from a round distance away from today's rate.

Technical biasiWhat the chart alone suggests, ignoring the news. Rated Bullish, Neutral or Bearish.

Bearish. Three facts drive it. Price is below the 20, 50, 100 and 200 day averages and all four of the shorter ones are falling. The structure is lower highs and lower lows since late June with no higher low yet. The two heaviest participation days of the period were both down days. This disagrees with the Expected direction given above, which is positive, and you should not have to notice that on your own. The chart describes what has happened. The direction call is about what an 84% priced decision does to a market that has already made the move. I weight the event more heavily than the chart for the next one to three days, because a scheduled decision of this size overrides trend for that window, and I weight the chart more heavily after it. If the decision passes without a change of pace and the pair still cannot close above 156.00, the chart is right and this call is wrong.

What would invalidate this biasiThe price level or chart change that would prove the technical read wrong.

A daily close above 157.97, the 20 day average, would end the bearish chart reading and turn it Neutral. A close above 160.40 would turn it Bullish. The other invalidation is not on the chart at all. If the two year yield gap stops narrowing and starts widening again, meaning Japanese two year yields fall back while American two year yields hold, the fundamental engine of this decline has reversed and the chart will follow within days. If instead the pair closes below 152.26, the bearish bias is confirmed rather than invalidated, and the positive direction call above is the thing that was wrong.

Interest rate differential

The Federal Reserve target range is 3.50% to 3.75%, a midpoint of 3.625%. The Bank of Japan policy rate is 1.00%. The gap is 2.625 percentage points in favour of the United States dollar, which is the base currency in this pair. A currency paying more interest is, all else equal, the one investors prefer to hold, because holding it earns the difference. The gap is narrowing. A Japanese rise to 1.25% on 18 September takes it to 2.375 percentage points. If the Federal Reserve also raises on 16 September the gap returns to 2.625 and nothing has changed at all, which is a real possibility this week and one the market has spent very little time on.

Central bank biasiWhether the first currency's central bank is leaning toward higher rates, which is hawkish, or lower ones, which is dovish, compared with the second's.

The Bank of Japan is Hawkish. It held at 1.00% on 31 July with one member voting for 1.25%, the Governor has said a rise is on the table at every meeting including this one, and reporting from early September says a faster pace is under discussion. The Federal Reserve is Hawkish in absolute terms and divided about acting. It has held the range for five consecutive meetings, three members dissented in July in favour of raising, guidance on 28 August left an increase open, and comments on 3 September said further improvement in inflation could justify holding. Both are hawkish, so the question is which is more so, and right now that is the Bank of Japan. Expressed for the base currency's central bank against the quote currency's, the Federal Reserve stance is Dovish relative to the Bank of Japan. That favours the yen and therefore points the pair lower.

Expected rate cuts and hikes

Overnight index swaps price roughly 84% for a quarter point Japanese rise to 1.25% on 18 September. Some measures put it as high as the mid nineties. Market pricing for the Federal Reserve on 16 September is close to a coin toss, roughly 58% to 62% for a quarter point increase. The repricing room is on the American side. A Japanese decision that is 84% priced can only surprise in one direction and only by a limited amount. A Federal Reserve decision at 60% can surprise either way, and the United States August inflation report on 11 September is what decides it. That is worth holding on to. The larger surprise available in USD/JPY this week is not the one this card is about.

Bond yield differential

The United States two year Treasury yields about 4.37% against about 1.84% for the Japanese two year government bond. That spread is about 253 basis points in the dollar's favour. A basis point is one hundredth of a percentage point. At ten years the United States yields about 4.78% against about 2.91% in Japan, a spread of about 187 basis points, also favouring the dollar. Two year yields track what a central bank is expected to do, so the two year spread is the one that moves this pair. It has been narrowing, and it has narrowed because Japanese yields have risen rather than because American yields have fallen. That is the healthier version for the yen and the reason this move has held better than previous yen rallies. The spread still favours the dollar by a wide margin, which is the point made under what the market may be missing.

Inflation trendiWhether price rises are speeding up, holding steady, or slowing down.

Stable. Japan's July headline inflation was 1.9%. The measure excluding fresh food, which is the one the Bank writes its forecasts around, was 1.8%. Excluding fresh food and energy it was 1.9%. All three sit below or at the 2% target. United States headline inflation slowed from 3.5% in June to 3.4% in July, and core inflation slowed from 2.6% to 2.5%. Neither side is accelerating. American inflation is drifting down from a higher level and Japanese inflation is holding just under target. For the pair that is close to neutral, and it is why the argument for a Japanese rise rests on wages rather than on prices.

Economic growth trend

Weakening, meaning the base economy is weakening relative to the quote economy. United States real GDP grew at a 1.5% annualised rate in the second quarter of 2026, down from 2.1% in the first. Japan's second quarter was revised up to a 1.4% annualised expansion from 1.1%, helped by business investment falling less than first estimated. Those two numbers are now almost the same, and they were not three months ago. The American growth advantage that supported the dollar through the first half of the year has closed. That is a genuine shift and it has had less attention than the rate story.

Risk sentimentiWhether investors are feeling brave or frightened. When they are frightened, money tends to move toward the dollar, the yen and the Swiss franc.

Neutral. Equity markets are near their highs, the volatility index has been in the mid teens rather than spiking, and credit conditions are calm. This matters for these two currencies more than for most. Both the dollar and the yen gain when investors are frightened, so a risk shock does not have an obvious effect on this pair. What a calm market does do is keep the carry trade alive, and the carry trade is a bet against the yen. If that calm breaks in the next week the yen gains for reasons that have nothing to do with the Bank of Japan.

Market positioning

Long-heavy in USD/JPY. The futures report covering positions to 1 September showed leveraged funds net short 102,188 yen contracts, about 24.8% of open interest. Yen futures rise when the yen strengthens, so a large short in yen futures is the same position as a crowded long in this pair. That data is published with a lag and 1 September is the last date it covers. Everything in the 666 pip fall since then is invisible in it. A move that size against a position that crowded normally forces a large part of it out, so the honest reading is that the crowding was extreme before the fall and cannot be measured today. That cuts both ways for this card. If the position has already been unwound, the fuel for a further yen rally on a hawkish surprise is gone, which supports the positive call. If it has not, the same surprise has a lot left to squeeze.

Carry trade attractiveness

Unattractive. A carry trade earns the interest rate difference between the two currencies and loses money quickly if the exchange rate moves the wrong way. Holding dollars against yen pays about 2.625 percentage points a year, which is roughly 0.22 percentage points a month. The pair moved 3.61% against that position in the last month. A year of carry was erased in six sessions. With the average daily range at 112 pips, a narrowing rate gap and a policy decision in the window, the income does not come close to paying for the risk right now. That is a statement about the trade, not about the direction. The carry being unattractive is one of the reasons the position was unwound so violently, and the unwinding is what this card expects to pause.

Wage pressure

High. Japan's July labour cash earnings rose 4.7% against a year earlier, the fastest since January 1997, against a 3.9% consensus and 4.0% the month before. Real wages, which adjust for inflation, rose 2.4%, the largest gain since May 2021 and the seventh consecutive monthly increase. This is the single number the case for a rise is built on. The Bank has said for two years that it needs wage growth to become self sustaining before inflation can hold at target without stimulus, and 4.7% with seven straight months of real gains is the closest Japan has come to that in three decades. United States wage growth is running near 4%, which is normal for that economy rather than exceptional. The comparison favours the yen, and it favours it for a durable reason rather than a cyclical one.

Core inflation pressure

Normal. Japan's July inflation excluding fresh food and energy was 1.9%, and the measure excluding fresh food alone was 1.8%. Both are at or under target and neither is accelerating. United States core inflation was 2.5% in July, down from 2.6% in June and from higher readings earlier in the year. Underlying price pressure is modestly higher in the United States and falling, and close to target in Japan and flat. Neither central bank is being forced by prices. The Bank of Japan is moving because wages give it permission, not because inflation gives it no choice, and that distinction is the reason a single rise followed by a pause is a realistic outcome.

Central bank intervention risk

Low, and the direction of that risk has flipped. Japan's Ministry of Finance intervenes to defend the yen when it is falling too fast, by selling dollars and buying yen. Those episodes have come when the pair was pushing into the high 150s and above. At 154.37, with the pair already 666 pips lower in six sessions, there is nothing for that policy to defend against. What matters is what the absence removes. For most of the last two years, buying USD/JPY carried the risk that an official would sell into you without warning. That risk is not present at this level, which makes a bounce easier to hold than it would be 500 pips higher. The reverse risk, intervention to weaken a yen that is rising too fast, is not a live concern. Japan has not done that in this cycle and a stronger yen is closer to what its policymakers say they want.