The United States may be buying yen. It should not pretend the fix is permanent

THE UNITED STATES Treasury may be buying yen. Or at least it has told banks to expect nothing less. On July 31st, the Treasury notified several banks, via the Federal Reserve Bank of New York, that it may intervene in the yen market and that the banks should "stand ready for future action", according to Reuters. The warning came a day after Japanese authorities had already stepped in, pulling the yen off four-decade lows against the dollar. Japan's top currency diplomat, Mr Atsushi Mimura, confirmed that American support "goes beyond psychological support".

A casual reader might assume Washington is doing a favour for a beleaguered ally. The real story is less charitable and more interesting. Currency intervention is not an act of friendship. It is an exercise in risk management. The United States does not prop up foreign currencies out of generosity. It does so when letting them fall becomes more costly than holding them steady.

To understand what is happening, it helps to know how the plumbing works. The Treasury cannot simply print yen. Intervention is usually executed through the Exchange Stabilization Fund, a $100bn account within the Treasury that can be used to buy or sell foreign currencies. But dollar liquidity -- the actual dollars that have to be moved -- must pass through the Federal Reserve. The Fed has maintained a dollar liquidity swap line with the Bank of Japan since 2013, a standing arrangement that allows the two central banks to exchange currencies on demand. The "rate checks" that dealers reported -- requests for indicative dollar/yen quotes -- are the usual precursor: a feeler to gauge how much resistance intervention would meet before committing to a trade.

The yen has been sinking for structural reasons, not because of a temporary market panic. In June the United States Federal Reserve held interest rates at 3.5% to 3.75%. The Bank of Japan, for all its recent tightening, was at just 1% after its June 16 hike. Traders have priced in further Fed hawkishness, driven by inflation from the oil shock of the United States's war with Iran, yet the rate gap has narrowed even as the yen fell. As of early July the dollar index was up 3% this year, after tumbling 9% in 2025. But the interest-rate gap alone no longer explains everything. As Lazard, an investment bank, points out, through 2023 the US-Japan rate differential explained roughly 90% of the variance in the yen's value. Since then the gap has narrowed, yet the yen has depreciated by about 15%. Rising Japanese inflation expectations have filled the gap: the market now prices in the view that the Bank of Japan is lagging the cycle.

Japan tried to intervene on its own. In April and May, the Ministry of Finance deployed a record ¥11.73trn ($73bn) to buy yen -- nearly double its largest previous effort. The yen rallied briefly, then returned above the intervention level within six weeks. The market had simply grown too large, and the underlying forces too strong, for one small country to resist on its own.

So why has the United States now joined in?

The most obvious answer is geopolitical. Japan is America's principal ally in the Indo-Pacific and a bulwark against Chinese expansion. A collapsing yen would raise the cost of imported energy and food in Japan, deepen domestic discontent, and risk destabilising the Takaichi government. A political crisis in Tokyo would weaken the alliance precisely when the United States needs it. Currency stability is, in this reading, a proxy for regime stability.

The second answer is supply chains. Japan sits at the heart of semiconductor, automotive and defence manufacturing networks that the United States has spent years trying to decouple from China. A dysfunctional Japanese economy complicates that project. American firms with Japanese suppliers and customers have a direct interest in the yen not collapsing.

The third answer is more prosaic. Treasury Secretary Scott Bessent has called the yen "very undervalued" and praised the Takaichi government for enacting "strong policies". Mr Bessent has a particular fondness for arguing that markets misprice assets. Whether he genuinely believes the yen is undervalued or is finding a rationalisation for a geopolitical move is immaterial. The result is the same.

The trouble is that none of these arguments addresses the fundamental problem. Japan is not a country with a temporary currency dislocation. It is a country with a deep structural crisis. The population is shrinking. The workforce is ageing. Public debt exceeds 250% of GDP, the highest ratio among advanced economies. Productivity growth is stagnant. As Professor Tan Kong Yam of Nanyang Technological University writes, the yen's weakness is not merely a monetary phenomenon but a reflection of these deeper weaknesses. Exchange rates ultimately reflect economic strength; they do not determine it.

This is where the analogy with the 1985 Plaza Accord is tempting but misleading. Then, the United States and four allies met at the Plaza Hotel in New York to agree on an orderly depreciation of the dollar, which had become dangerously overvalued. The dollar fell sharply, and the yen appreciated, contributing to the asset bubble that burst and sent Japan into decades of stagnation. Today the tables are nominally turned: the dollar is strong and the yen is weak. But the United States is not now asking Japan to accept a permanently stronger yen. It is asking the market to pause. These are very different things.

The danger of Treasury intervention is not that it will fail today. It probably won't. The yen jumped to 159 from a Thursday low of 163.65 on the news alone. The danger is slower and more institutional. Every time the United States uses the credibility of the dollar to shore up another currency, it reinforces the expectation that it will do so again. That creates moral hazard. It tells investors that when a currency reaches the point of political sensitivity, the Fed and the Treasury will backstop it. That is not a problem in a single episode. It becomes one when it becomes a habit.

To be sure, the United States has an interest in global financial stability that sometimes requires it to act as an implicit guarantor of the international monetary system. The dollar swap lines deployed during the 2008 financial crisis and the 2020 pandemic lockdowns were justified on similar grounds. The difference then was that the problems were cyclical -- a seizure of credit markets, a demand shock -- and could be addressed with liquidity. The yen's weakness is not a liquidity problem. It is a solvency story disguised as an exchange rate.

The better answer would be for the United States to support Japan in ways that address the structural deficit rather than the exchange rate. That means encouraging, not resisting, the Bank of Japan to normalise its policy further. It means supporting fiscal consolidation -- however politically painful for the Takaichi government. It means investing in the productivity gains, through technology transfer, trade agreements and regulatory alignment, that would make Japanese assets attractive to foreign investors for reasons beyond carry-trade mechanics. These are slower, harder, more politically difficult measures. They are also the only ones that last.

Currency intervention is not wrong. It is merely temporary. The United States should not pretend otherwise. The swap line with the Bank of Japan is a useful tool for managing market disorder. But it cannot fix a 250% debt-to-GDP ratio, a shrinking workforce, or a central bank that has spent three decades fighting a war it cannot win. The yen will test the limits of this arrangement again. When it does, the question will not be whether the Treasury intervenes. It will be what Washington has done in the intervening months to make intervention unnecessary.

Until then, the arithmetic is simple. Japan needs the United States more than the United States needs Japan, but neither country can afford for the arrangement to be tested beyond its breaking point. The market always wins eventually. The only question is when.