Good morning. The US manufacturing PMI survey shifted into expansion last month, ending 26 consecutive months of contraction. New orders were up, including export orders, and inventories were down. This is sorely needed positive news for the country’s struggling manufacturing sector. Donald Trump thinks tariffs will provide support to industry in the long term. But what about the short term? Email us: robert.armstrong@ft.com and aiden.reiter@ft.com.
China’s tariff volley
Unlike Canada and Mexico, which scored delays on US tariffs by opening negotiations with Donald Trump, China chose to retaliate. Yesterday, Beijing put 10-15 per cent tariffs on US liquefied natural gas, coal, crude oil and farm equipment. China also opened an investigation into Google, put export controls on rare metals, and added two US companies to its national security blacklist. Most analysts have played down the impact of these tariffs. But that is the point: they were intended to cause only a little pain, demonstrating China’s resistance while not escalating tensions.
The tariffs affect less than 10 per cent of US imports to China — a big contrast with Trump’s across-the-board tariffs. And by focusing on energy, China’s leaders are counting on large, flexible energy markets to ease the pain for both US and Chinese consumers. China’s response, however, is not toothless; the tariffs will enact short-term pain on the US economy. China is the world’s biggest coal importer, producer and consumer, and has recently increased its consumption of US coal:
In 2024, 11 per cent of US coal exports went to China — but that was less than 1 per cent of its total coal consumption. China is already expected to reduce its coal use, making the change trivial to China. But it will take time for US coal miners to find new buyers, potentially causing Trump some political pressure in coal producing states.
A similar point can be made for crude oil and LNG. China has become a larger US gas and oil importer in recent years:
In both cases, too, US exports are only a small fraction of Chinese consumption. And for crude oil, China is already expected to hit peak demand in 2027. The pain will be felt more in the US, as supply chains will take time to adjust.
The other measures are more restrained. Google does very little business in China, and one of the two companies added to the security list was already under investigation. Beijing also put restrictions on various metals last year; US importers have already begun to adjust.
In the context of China’s broader challenges, this all makes sense. Its economy is struggling. Its leaders want to avoid US tariffs. These responses make Beijing look tough to its domestic audience and give it some leverage in potential negotiations, while not overreaching. From Tianlei Huang at the Peterson Institute:
I just don’t think it is in China’s interest to do that much. External demand is very important for China’s growth right now, given how weak domestic demand is. In 2024 . . . net exports contributed 30 per cent to GDP growth; but they were a drag in the prior year. Big US tariffs will definitely hurt China’s economy, particularly unemployment, wages and domestic demand. Given all those challenges at home, it’s really not in China’s interest to retaliate, at least not reciprocally.
Comments from Trump’s economic team suggest that the US is not interested in negotiating with China. Trump’s trade adviser Peter Navarro claimed that the president would speak with Chinese leader Xi Jinping on Tuesday, but that didn’t happen. When they do talk, common ground may be hard to find.
(Reiter)
QT and liquidity redux
We recently discussed how to best measure US financial system liquidity, in an attempt to guess when the Federal Reserve might end its quantitative tightening programme. As a liquidity proxy, we used the sum of bank reserves held at the central bank and the balance in the Fed’s reverse repo programme. Together, they represent how much money is available to US banks and money market funds.
It’s a crude proxy, though. Happily, the Fed recently came out with a guide to four other ways of looking at liquidity. To summarise:
The spread between the interest rate on reserve balances (IORB) and the effective federal funds rate (EFFR): IORB is one of the two rates set at FOMC meetings (the other being the offered rate on reverse repurchase operations). Together, the two help to control the EFFR, which is the market-determined rate for interbank borrowing, and by extension all borrowing rates in the economy. When financial conditions are tight and banks are clamouring for liquidity, the floating EFFR can go above the IORB, and did so before and during the 2019 reverse repo crisis:
Dispersion of rates on overnight interest rates: The EFFR is the volume-weighted median of rates charged in the overnight funding market. But when demand for reserves is high during or before a liquidity crunch, not all rates are above the IORB — there is often just greater dispersion, with more outliers above the IORB. Employees of the Fed have a way to look at the daily volume-weighted average, rather than the median, to gauge the dispersion. For lay market folk like us without access to the series, the authors of the paper propose looking at the 1st, 25th, 50th, 75th and 99th percentile of rates in daily Fed funds trading, and use an equation to tease out a daily coefficient. Here is there graph:
This measure also wobbled in 2019 and 2020, but looks stable now.
Repurchase agreement spreads to IORB: The Secured Overnight Financing Rate (Sofr), or the rate at which banks lend to each other against their Treasury holdings, and the Triparty general collateral rate (TGCR), the rate for lending against Treasury holdings in a more specific set of multi-party transactions, are also meant to be near the EFFR. In theory, if liquidity is ample, Sofr and TGCR should be just a little above the IORB, and should jump only when there is not enough liquidity, as in 2019:
The Sofr rate peaked above the IORB in September and December of last year, too, suggesting liquidity may be on the tight side:
But other factors — including changes to the weighted average maturity of Treasuries — could have had an influence.
Money market volatility. Money markets should be more volatile when there is not enough liquidity. The 15-day standard deviation of the EFFR provides a decent measure of money market volatility — and, better yet, it is leading. Volatility increased in the run-up to the 2019 repo crisis:
By all these measures, we still appear to have ample reserves, and QT is safe to continue. But all are imperfect, and the Sofr and demand curve readings only jump in the moment of market distress, not in the run-up. When it comes to QT, we are all still stumbling in the dark.
(Reiter)
One good read
FT Unhedged podcast
Can’t get enough of Unhedged? Listen to our new podcast, for a 15-minute dive into the latest markets news and financial headlines, twice a week. Catch up on past editions of the newsletter here.