Unlock the Editor’s Digest for free
Roula Khalaf, Editor of the FT, selects her favourite stories in this weekly newsletter.
The writer is a senior fellow at the Council on Foreign Relations
Not long ago, there was a near-consensus that Japan’s intervention in the foreign exchange market wouldn’t work because the country’s interest rate differential with the US — which propelled the yen to record lows — was simply too powerful.
That consensus reflects somewhat dated research showing that, without capital controls, intervention only works when it is co-ordinated and backed by broader monetary and fiscal policy changes. This conventional wisdom is embedded in the IMF’s core model for assessing the impact of intervention. That model assumes that intervention is completely ineffective for large, open, advanced economies.
Yet Japan’s intervention set the floor on the yen this summer. The country’s intervention in October 2022 did the same. Theories about the general ineffectiveness of intervention need to be updated.
There are three reasons why intervention may be more effective than the financial conventional wisdom holds. First, in Japan’s case, the government is a major player in the market. Most analysts look at intervention relative to market turnover, but this daily churn is often largely a function of market players trading among themselves in response to more modest real flows.
The government of Japan has a ton of foreign assets — the Ministry of Finance has $1.2tn in reserves, and the Government Pension Investment Fund has another $800bn or so. These assets are unhedged, and account for more than half of Japan’s total net foreign asset position. The government of Japan is thus the biggest financial beneficiary from yen weakness.
Think of it this way: if the winner of a big financial bet never takes profits, that impacts the market — as the government’s gains are large relative to $50bn-$100bn in foreign purchases that often comes from “private” Japanese investors. There is another corollary: when Japan’s Ministry of Finance sells dollars bought at ¥80 or ¥100 to the dollar for ¥150 or ¥160, it books a hefty profit. That profit runs somewhat against the standard market view that intervention “wastes” scarce foreign exchange.
Second, many studies of the ineffectiveness of intervention focus on the wrong variable. Intervention tends to work not by sustainably strengthening a currency, but by credibly setting a floor under the market. For example, if the government is expected to intervene at ¥162 against the dollar and if the Japanese currency is currently trading at ¥160, the distribution of likely returns is skewed: bad news for the yen won’t lead to significant yen depreciation, but good news for the yen could lead to a big appreciation.
Intervention to set a ceiling on an appreciating currency works much the same way — the market knows the currency won’t be allowed to get much stronger, but nothing stands in the way of depreciation. That’s why so many Asian countries were able to engage, successfully, in “competitive non-appreciation” during the decade of foreign currency manipulation from 2003 to 2014.
Third, in Japan’s case, the government’s actions can send a signal to a much broader group of regulated institutions that have to decide whether to partially or completely hedge their foreign bond holdings. Quasi-public institutions — Japan Post Bank, the Norinchukin Bank, and the investment fund for the small savings banks (Shinkin Central Bank) — collectively hold close to a trillion dollars of foreign bonds. The extent to which these bonds are hedged impacts the markets. The same goes for the nine big life insurers, who have gone from holding about $360bn of hedged bonds and $240bn of unhedged bonds in the fiscal year to March 31 2020 to around $185bn hedged and $215bn unhedged at the end of the 2023 fiscal year. While the government stayed out of the market, these institutions let their hedge ratios slip to cut costs and boost profits.
None of these arguments deny the reality that interest rate differentials do matter for countries with open financial accounts. High US short-term rates make hedging expensive. High US long-term rates make unhedged holdings of dollar bonds attractive. But the government of Japan is potentially a big, not small, player — and intervention can still shape the near-term distribution of risk and return.
Uncoordinated foreign exchange intervention is harder for large, open, advanced economies — especially those that lack the enormous firepower of Japan’s government. But after recent experience, it is a mistake to assume that it can never work.
