The Week the Fed Hiked, the Yuan Crossed 6.70

The Week the Fed Hiked, the Yuan Crossed 6.70

On September 16, the US Federal Reserve raised its benchmark rate by 0.25 percentage points to 3.75–4.00%, its first hike since 2023.¹ Two days later, the Bank of Japan raised its policy rate to 1.25%, the highest level since 1995.² The European Central Bank, having already hiked in June, added another 0.25 percentage points in September.³

By the textbook, this should have been a week for the dollar to strengthen. The usual logic goes: when US rates rise, returns on dollar assets rise with them, and currencies in lower-rate countries weaken by comparison. China’s rates sit below America’s, and that gap keeps widening. And yet on September 18, the onshore yuan broke through the 6.70-per-dollar level, climbing all the way to 6.6957. That same day, the People’s Bank of China’s reference rate came in at its highest since February 2023.⁴

There’s an interview that tackles this mismatch head-on: a conversation between the Chinese securities-industry outlet Quanshang Zhongguo (“Brokerage China”) and Zhao Wei, chief economist at Shenwan Hongyuan Securities.⁵ It ranges widely—monetary policy, the yuan, AI, domestic consumption—but today I want to follow three threads from it: why the rate gap can’t explain the exchange rate, how AI has been propping up China’s exports, and why that export boom barely seems to reach household spending.

Same direction, different reasons for the hike

Zhao Wei starts by unpacking the phrase “synchronized global tightening.” All three central banks raised rates, she says, but for entirely different reasons.

The Fed is raising rates to press down on inflation that won’t budge even as growth holds up and demand stays strong. Europe is raising rates to stop import-price inflation—originally sparked by energy costs—from spiraling into a second-round effect where wages and prices push each other higher. Japan, pointing to rising wages and revived inflation expectations, is slowly exiting the ultra-easy policy it maintained for years. So the US and Europe are in the early stage of a new hiking cycle, making “compensatory” hikes to catch up for lost time, while Japan may be in the later stage of a cycle that’s been running since 2022.

The Fed’s own projections line up with this distinction. The median forecast among September FOMC participants was for 2.3% growth and 3.7% PCE inflation in 2026.⁶ Growth looks fine, but inflation stays high — that’s the picture. This is also why Zhao Wei, while calling the global economy in Q4 “quasi-stagflation,” distinguishes between the US, which she sees as relatively solid, and Europe, where the stagflationary character is more pronounced.

One number is worth flagging. Zhao Wei says markets are pricing in roughly three more Fed hikes over the next year. But the median among Fed participants puts the rate at 4.1% at the end of 2026, and still 4.1% at the end of 2027.⁶ That’s closer to one more hike this year, then holding steady. In other words, there’s a gap between what the market has priced in and the path the Fed itself has laid out. Zhao Wei adds the caveat that the actual path will depend on how inflation behaves.

The tail risk she watches most closely is oil. If oil prices spike again and lift household inflation expectations, a loop could form where wages and prices keep pushing each other up. That would force central banks to tighten even further, and demand would get squeezed that much harder.

Three Things the Rate Gap Can’t Explain

Now, the yuan. Zhao Wei says it’s no contradiction that the yuan is strong even as the US-China rate gap widens, and she points to three shifts.

First, the value of simply holding dollars is being repriced. Dollar assets have long commanded a premium for being safe and liquid—sellable anytime. This is called the convenience yield. Zhao Wei’s diagnosis is that as doubts grow over whether US fiscal finances can hold up and where policy is headed, this premium has shrunk. That changes how we should read America’s high interest rates. They may be high not just because the dollar is attractive, but because investors are demanding bigger compensation for bigger risk.

The second point is the most useful distinction in this interview. Split the money that moves exchange rates into flows and stocks. New carry-trade money chasing the rate gap is a flow. But the dollar assets foreign investors have already accumulated are a stock. Because that stock is so enormous, even a small uptick in hedge ratios—as investors try to trim currency risk—generates a huge volume of dollar-selling through forwards. Zhao Wei believes this selling can outweigh the new money drawn in by the rate differential.

Let’s try a hypothetical to get a feel for the scale. If foreign investors hold $10 trillion in dollar assets, raising the hedge ratio by just 1 percentage point generates $100 billion in dollar sales. The key point: the bigger the stock, the bigger the dollar amount a small percentage change produces. That $10 trillion is a figure chosen purely for illustration, not an actual estimate of the stock.

Still, what this logic explains is why the dollar weakens—not why the yuan specifically strengthens. Dollars sold for hedging purposes don’t flow only into yuan; they scatter across many currencies. The yuan-specific story has to be found elsewhere—in the currency conversion behavior of Chinese companies, which I’ll get to next.

Third, America’s policy path is less predictable than it used to be. Carry trades run on the belief that today’s rate gap will persist into the future. When that belief weakens, today’s rate gap no longer guarantees tomorrow’s returns, and the appeal of carry trades fades.

That’s why Zhao Wei expects the yuan to appreciate against the dollar over the next year, even with ups and downs along the way. On the US side, she points to a “fallacy of composition.” Fiscal expansion and tight monetary policy each look like they should support the dollar on their own. But when they overlap, interest burdens on government debt grow, fiscal and monetary authorities clash, and the likelier outcome isn’t a steadily strong dollar but one that swings up and down.

On the China side, she looks at conversion flows. Trade-surplus dollars flowing in get converted to yuan by companies, and this could compound with roughly $400 billion that firms held back from converting between 2022 and 2025 and are now releasing. CICC (China International Capital Corporation)‘s second-half outlook makes a similar point: once exchange-rate expectations shifted, companies grew more willing to convert and hedge. In just the first half of this year, the yuan rose 2.9% against the dollar.⁷

But Zhao Wei’s caveat is that it won’t move in one direction only. When companies convert their accumulated dollars ultimately depends on their own read of exchange rates and cash-flow needs. The direction may be appreciation, but the path could still be bumpy.

Half of export growth came from AI

Where does the trade surplus propping up the yuan actually come from? Zhao Wei’s numbers are fairly clear on this.

Since 2023, value added in the new economy — driven mainly by IT services — has grown at an average annual rate of 11.3%. Over the same period, infrastructure and real estate growth settled around 2%, and labor-intensive industries around 5%. On the demand side, AI-related products are said to account for 50% of 2026’s export growth. On the investment side, only high-tech industry investment showed a notable rebound, and on the consumption side, telecom equipment sales performed well. AI-driven price increases are reportedly leaving their mark on both producer and consumer prices.

These figures come from Zhao Wei’s research team. The raw data behind the claim that AI accounted for half of export growth wasn’t disclosed in the interview. Still, the interpretation that this year’s AI investment boom pushed up China’s exports and in turn propped up the yuan also appears in other Chinese economists’ analyses.⁷

To seize the AI opportunity, he identifies six tasks: computing infrastructure, data, commercializing “AI + the real economy,” a layered talent system, long-term investment capital, and tiered AI governance. Two of these reveal his underlying concerns most clearly. On computing infrastructure, he warns against focusing only on hardware while neglecting the domestic software stack. On “AI + the real economy,” he warns against staying stuck at pilot projects instead of building structures where actual money flows. Read together, this looks like a diagnosis that the chips and the demos already exist, but the ecosystem running on top of them — and the structure for actually making money — is still weak.

How much China’s AI servers earn, and why the cloud providers holding the most of those servers earn so little, was covered in Issue 233.

AI servers make 44% margins, so why does the cloud only make 11%?inlevel9.com

Beyond Income: What Else Is Holding Back Consumption

The second half of the interview shifts direction. While AI lifts exports and investment, China’s domestic demand stays weak. Zhao Wei attributes this to the step-by-step adjustment pressures of an economy in transition, pointing to slowing household income growth, squeezed corporate spending capacity, and weakened consumer sentiment.

The proposed solutions target the weak links in this transition. As the old economy—real estate, for instance—cools, the gap should be filled with investment in future growth areas like new infrastructure, energy-transition infrastructure, and digital infrastructure, easing the job pressure in construction and steel. Workers who lose their jobs should get employment support to move into sectors aligned with the direction of transition, like services. For local governments, Zhao argues that boosting central government support, reforming the local bond system, and broadening the tax base are necessary to restore fiscal spending’s weakened power to stimulate the economy.

What held my attention longest in this interview came next. Zhao says consumption is shaped not just by income but by leisure time and supply. Tourist destinations grow emptier in the off-season and overcrowded during holidays, she notes—and the experience degrades to the point where some households simply give up traveling altogether. The services sector, meanwhile, is undersupplied. Quality hasn’t kept pace with demand: there aren’t enough places to exercise, not enough people to hire for housework. Even if you pump money into the system, if the time to spend it is crammed into a handful of days and there aren’t enough decent services to spend it on, consumption won’t grow as much as expected.

Over the medium-to-long term, she’s optimistic—citing continued growth in per-capita GDP, a rising labor income share from improved income distribution, and room for the hukou-based urbanization rate (urbanization measured by household registration status) to climb further. As the biggest risk during the 15th Five-Year Plan (15·5) period, she points to local government finances: real estate developers’ land purchases have fallen sharply, drying up local governments’ revenue streams, and combined with rigid expenditures, debt cleanup, and a shortage of viable projects, this is dragging down fixed-asset investment. As for the most undervalued opportunity, she names the consumption market: the middle-class share of the population, the labor income share, and household time spent on consumption are all low by international standards—which means there’s real room for growth, and expanded visa-free entry is already boosting spending by foreign tourists.

I covered how China’s latest five-year plan aims to change its growth model rather than just its growth rate in Issue No. 48.

China’s Next Five Years: A Pledge to Change the ‘Structure,’ Not Just the Numbersinlevel9.com

Oswarld’s Lens

[Editorial suggestion — author confirmation needed] Read the interview straight through, and two different speeds emerge. AI products drive exports, and the dollars earned flow back through currency conversion, pushing the yuan upward. Meanwhile households are spending less, caught between slowing income growth, holiday clustering, and a shortage of service supply.

[Editorial suggestion — author confirmation needed] Zhao Wei discusses the six AI challenges and the consumption fix as separate matters, but the interview never spells out the link between them. The path by which money earned through AI should flow into wages and service-sector jobs — and from there into household consumption — looks like the question this interview leaves open.

[Editorial suggestion — author confirmation needed] For Korean readers, the practical takeaway is about how to read exchange rates. The old formula — “when U.S. rates rise, the dollar strengthens” — is no longer enough on its own to judge currency moves. This interview’s real value lies in showing that you also need to watch the trade surplus, the dollar reserves companies have been accumulating, and the hedge ratios of foreign investors, alongside the interest-rate gap. Still, it’s worth remembering this is the assessment of a single economist at a Chinese securities firm, and that key figures — like the exact export contribution — haven’t been released in raw form.

Closing

The three central banks moved in the same direction in September, but the yuan took a path that diverged from the textbook. Whether Zhao Wei’s explanation holds up is something the next few months of exchange-rate and currency-exchange data will tell us.

---

💬 Reader, when you’re reading exchange-rate news, which do you check first — the interest rate gap or the trade balance? Let me know in the comments.

📨 If you have a colleague who follows both rates and currencies, pass this piece along.

References & Further Reading

The author is Oswarld (Kwangseob Ahn). Current roles: Adjunct Professor at Sejong University, Strategy Consultant at INLEVEL9. Career, research, books, and recent work are kept current on the About page. Latest · July 2026: HEMA-2: A Consolidation-Aware Tri-Memory Architecture with Multi-Channel Scheduling for Lifelong Conversational AI.

📝 Glossary

• Convenience premium: The markup an asset earns simply for being safe and instantly convertible to cash. U.S. Treasuries and the dollar are the textbook example.
• Carry trade: Borrowing in a low-interest-rate currency to invest in higher-yielding assets denominated in another currency, pocketing the rate differential.
• FX hedge ratio: The share of foreign-currency assets whose exchange-rate risk has been locked in through forwards or similar instruments. Raising this ratio effectively means selling that much foreign currency in advance.
• Jiehui (结汇): The process by which Chinese companies convert foreign currency earnings into yuan at a bank.
• Fallacy of composition: The phenomenon where choices that are individually rational add up to a different, often worse, outcome collectively.
• Hukou-based urbanization rate: The urbanization rate measured by how many people hold urban hukou (household registration), rather than how many actually live in cities. Holding urban hukou is what entitles residents to full access to public services like education and healthcare.

Footnotes

1. CNBC, “Fed rate decision September 2026: Rates rise to 3.75%-4%”, 2026-09-16. https://www.cnbc.com/2026/09/16/fed-rate-decision-september-2026.html ↩

2. CNBC, “Bank of Japan raises interest rates to 31-year high, flags concerns over inflation”, 2026-09-18. https://www.cnbc.com/2026/09/18/japan-raises-rates-30-year-high-yen-jgb.html ↩

3. The June hike (deposit rate 2.25%) is from OrbitRemit, “ECB Holds Rates at 2.25%: ECB Rate Decision July 2026,” https://blog.orbitremit.com/ecb-rate-decision-july-2026/. The September 0.25%p hike is from PwC Plus, “Monetary policy decisions September 2026,” https://pwcplus.de/en/article/256769/monetary-policy-decisions-september-2026/ ↩

4. Xinhua Finance, “Yuan breaks above 6.70 against the dollar, central parity rate climbs steadily,” 2026-09-18. https://www.cnfin.com/hs-lb/detail/20260918/4471715_1.html ↩

5. China Securities Journal, interview with Zhao Wei of Shenwan Hongyuan Securities (“AI has become an important driver of China’s economy,” “Consumption may be the biggest opportunity of the 15th Five-Year Plan”). https://mp.weixin.qq.com/s/2Fx3kAErrIlOT794tpZyHQ All of Zhao Wei’s remarks and figures in the piece are drawn from this interview. ↩

6. MC Markets, “Federal Reserve Rate Hike: What the September 2026 Decision Means for Markets.” https://www.mcmarkets.com/en/insights/federal-reserve-rate-hike-september-2026-analysis ↩ ↩²

7. The Paper, “The yuan’s spot exchange rate against the dollar rose 2.9% in the first half of the year — where does it go from here?” 2026-06-30. https://www.thepaper.cn/newsDetail_forward_33484975 ↩ ↩²