Four Days After We Called the Yen Cheap, It Hit ¥153: A 99%-Priced BOJ Hike, Bessent's “I Am the House,” and How to Reprice a Japan Plan Built at ¥160

Last Friday's article on Yokohama's incentives for foreign companies leaned on one number: a yen near ¥157 to the dollar, which made yen-denominated setup costs historically cheap for anyone funded in dollars or euros. We stand by the incentives. The number aged faster than we expected. On the morning of September 8, the yen reached ¥153.3 — its strongest level since mid-February. In July it had touched roughly ¥164, a four-decade low, and triggered the first joint yen-buying intervention by Japan's Ministry of Finance and the US Treasury since 1998, an operation Bank of Japan data put near $36.6 billion. By early August the yen had drifted back past ¥158, and our argument at the time was that an intervention buys a level, not a trend.

What changed in September is that the inputs to the trend moved, not only the level. On September 2, Governor Kazuo Ueda said that with underlying inflation approaching 2%, the Bank of Japan needs to pay greater attention to upside risks, and that a rate hike is on the table at every meeting — including this month's. Rate futures now price a move from 1.0% to 1.25% at the September 17–18 meeting at roughly 99%. A Reuters poll sees the policy rate at 1.75% by the second quarter of 2027, and Reuters sources say the board is weighing a faster pace than the roughly twice-a-year rhythm it has kept since leaving negative rates.

The data behind that shift arrived all at once. Real wages rose 2.4% in July, the seventh straight monthly gain and the biggest since May 2021. Nominal earnings rose 4.7%, the fastest since January 1997, and base pay 4.1%, the fastest since April 1992. Second-quarter GDP was revised up to 1.4% annualized. Producer prices rose 7.6% in August, above the 7.4% forecast — and yen-based import prices were up 24.8% on the year. That last figure is the weak yen's own invoice, and it is the invoice the BOJ has decided to stop paying.

Then came the sentence that made the week. Speaking at Southern Methodist University in Texas on September 8, Treasury Secretary Scott Bessent said “I am the house now,” explaining that when the United States intervenes on the yen he has good insight into what the Bank of Japan and Japanese policymakers will do. US Treasury Secretaries do not usually talk that way about another country's currency. Read plainly, it says the cheap end of the yen's range now has a named, coordinated defender — Bessent working alongside Finance Minister Satsuki Katayama — and that traders betting on a return to ¥164 are betting against both governments.

It is worth being precise about what this does and does not mean. By September 11 the yen had slipped back to around ¥154 after firmer US producer-price data lifted the dollar. One month of yen strength is not a regime. The Federal Reserve's path, oil prices and Middle East risk will all keep moving the rate, and nobody — including us — knows where it will sit in March. Anyone who tells you the yen is going to ¥140 is selling something.

What has changed is the shape of the risk. In July, the surprises for a dollar- or euro-funded company planning Japan costs ran mostly in its favour: every leg down in the yen made the office, the hires and the lawyers cheaper. Today three forces lean the same way at once — a central bank tightening in steps, wages running ahead of inflation for the first time in years, and a US Treasury openly unwilling to let the yen fall back to ¥164. The cheap end of the range is defended. The expensive end is not. That asymmetry, not a point forecast, is what a sensible plan should price.

Markets have already started pricing it. The Nikkei 225 closed near 64,500 on September 10, down about 4.4% over the month, as a firmer yen squeezes the exporters that carry the index. For a foreign company, the equivalent adjustment is less visible but just as real: every budget, price list and business case that was quietly built on a ¥158–¥160 yen now rests on an assumption that two governments are working against.

Start with the cost side, because it is the one most entry plans get wrong. Take a first-year Japan budget of ¥60 million — a registered office, three local hires, a judicial scrivener, certified translations and a modest launch campaign. At ¥160 to the dollar that is $375,000. At ¥153 it is about $392,000, 4.6% more. At ¥145 it is about $414,000, roughly 10% more. None of that is a reason to cancel a Japan entry. All of it is enough to break a budget that was approved at the tight end.

And the yen figure itself is rising, not just its dollar value. With base pay growing 4.1% a year, the three hires in that budget will cost more than a salary survey from 2025 suggests. A Japanese subsidiary's yen credit line gets more expensive with each BOJ step: still cheap next to dollar or euro borrowing, but no longer free, and a plan that assumed near-zero local financing costs should be rerun.

Now the side that gets far less attention. If you sell into Japan and bill in dollars or euros, a stronger yen is a price cut you did not have to give. A $100,000 annual contract costs a Japanese buyer ¥16.0 million at ¥160 and ¥15.3 million at ¥153 — ¥700,000 less, without you touching your price. After a year in which yen-based import prices rose by double digits, Japanese procurement teams have spent months resisting every foreign price increase. The currency has just handed them some relief, and it has handed you room — to hold price at renewal, to win a deal that stalled on cost, or to finally move a customer from a pilot to a contract.

Four practical moves follow. First, re-run the Japan budget at three rates — ¥145, ¥153 and ¥160 — and get approval on the worst of them, so a strong-yen quarter does not become a crisis meeting. Second, for yen costs you are already committed to over the next twelve months — deposits, leases, first-year salaries — ask your bank about forward cover or pre-funding; this is ordinary treasury practice, not a currency bet, and the right answer depends on your own situation. Third, if you sell into Japan, check which currency each contract is in before renewal season rather than after. Fourth, re-base your salary assumptions on 2026 wage data, not on the last hiring round.

The same fortnight brought the other half of the picture. At the end of August, the Ministry of Economy, Trade and Industry submitted a record fiscal 2027 budget request of ¥7.7859 trillion. Around ¥4.5 trillion of it sits in a new growth-investment framework that, unusually, carries no spending ceiling. The largest item inside it is about ¥1.4 trillion for AI, semiconductors and robotics. Next come roughly ¥900 billion for investment support to small and medium-sized enterprises, ¥680 billion for critical minerals and ¥220 billion for naphtha supply. Rapidus, the 2-nanometre chip venture, is slated for another ¥150 billion of equity through the Information-technology Promotion Agency, taking the government's stake to ¥400 billion, with total state support including subsidies expected to reach about ¥3 trillion by fiscal 2027.

A request is not a budget. The final numbers will be settled in year-end negotiations with the Ministry of Finance and passed by the Diet before the fiscal year starts on April 1, 2027, and some lines will shrink. But the direction is not in doubt, and it is an interesting one: the government is pushing companies to invest in exactly the year that money is getting more expensive. Rising rates on one side, subsidized capital spending on the other.

For most foreign companies selling into Japan, the ¥900 billion matters more than the headline chip money. Mid-sized Japanese manufacturers, logistics firms and service businesses — the companies facing the sharpest labour shortage and the wage growth described above — will have partly subsidized budgets for equipment, digitalization and labour-saving systems. That is precisely what foreign AI, automation and software vendors sell. The ¥1.4 trillion, meanwhile, pulls in suppliers around fabs, AI data centres and robotics lines, where European and American component and tooling makers already have a foothold.

Then there is timing, which is where most foreign vendors lose. Japanese companies draft their April-start budgets between October and January. A vendor who is not in a buyer's plan by January usually waits for the next fiscal year — or for a supplementary budget. So the window that matters this autumn is not the currency window. It is the budget window, and it opens now.

For two years the weak yen let foreign companies think of Japan as a market on sale. We used that framing ourselves, as recently as last week. It is now a liability. A Japan plan that only works at ¥160 is a plan that does not work, because ¥160 is the one level two governments have said, in different words, they will not tolerate.

The good news is that the structural case for Japan never depended on the currency. Real wages rising for seven straight months means Japanese consumers and employees with more to spend. A labour shortage deep enough to push base pay up at the fastest rate in three decades means sustained demand for automation and AI. A government willing to lift the ceiling on growth investment means subsidized buyers. And a record ¥53.3 trillion stock of inward foreign direct investment, with an official target of ¥120 trillion by 2030, means Tokyo still wants you in. All of that survives a yen at ¥145.

What we are telling clients this month is simple. Price the plan at the strong end of the range. Sell into the subsidy cycle rather than around it. And localize now, because the Japanese buyer drafting next year's budget in November will read your proposal in Japanese or not at all. That is the work Medusa Japan does from Osaka — bilingual go-to-market, localization and AI-enabled products built for Japanese buyers — and this autumn, the calendar matters more than the exchange rate.