On 3 November the People’s Bank of China said it would raise the FX reserve requirement ratio for financial institutions by two percentage points, from 4% to 6%, effective 15 November 2026. It is the first increase since December 2021 and it reverses precisely the 6%-to-4% cut of September 2023.
Most coverage treated this as a currency story. It is better read as a plumbing story: the PBOC changed how many dollars are allowed to sit inside the onshore banking system, and the market has ten days to reprice it.
What the ratio actually does
A reserve requirement ratio on foreign-currency deposits is the share of those deposits a bank must park at the central bank instead of lending out or converting into yuan. Raise it, and a slice of dollar liquidity is pulled out of the market and immobilised. Since onshore dollar supply is one side of the USD/CNY price, a smaller conversion flow means the yuan firms less easily.
Note what it is not. It is not an interest rate: it does not change what dollars cost to borrow, it changes how many exist. Quantity tools work at the margin and through expectation, which is why price moved on the announcement, long before any dollar was locked up.
On scale, CITIC Securities chief economist Ming Ming estimates the move freezes roughly $20-25 billion of foreign-currency liquidity at bank level. Against daily onshore turnover in the hundreds of billions, that is not a wall. It is a signal — and in a market with a daily administered fixing, the signal is the tradeable part.
Why now: capping speed, not reversing direction
The yuan gained 4.2% in the first nine months of 2026 and about 5% year to date — a strong currency against weak domestic demand, soft prices and exporters watching margins compress in local-currency terms every quarter.
The PBOC’s problem is the pace, not the level. Rapid appreciation does the same damage as rapid depreciation to an economy that needs stable trade terms, and it builds speculative positioning that must be unwound later. Two decisions in the same week make the mix explicit: this ratio hike, and a cut to the central bank’s share-buyback funding rate to 1%. Cheap money for equities, a slower currency, no change to the policy rate.
So read it as a brake, not a handbrake turn. Anyone claiming the PBOC is targeting a specific level, or has decided to weaken the yuan, is extrapolating from one instrument.
The fix is the second signal, and most readers missed it
Spot closed at 6.6765 on 3 November, down 307 pips. On 4 November the onshore rate slid to about 6.68, its weakest since 19 October.
Then the midpoint. The PBOC set the fix near 6.72 against consensus around 6.68 — its weakest fix since 23 October. That 40-pip gap is the informative part, because the fix is not a forecast. It is an administered price, and spot may trade only about ±2% around it each day. Two instruments were used on consecutive days: one removing dollar supply, one lowering the day’s centre of gravity for the yuan. Reading only the first gives you a one-day trade; reading both gives you the regime.
Where the pair sits now
USD/CNH traded at 6.67191 at 00:03 UTC on 12 November with the dollar broadly firmer — DXY 102.843, up 0.60%. AUD/USD was 0.68479, down 0.34%: a dollar-strength move, not a China-specific one. Hong Kong’s Hang Seng closed at 24,315.92 on 11 November, down 0.54%.
Be strict about attribution here. When the dollar rises against everything, a softer yuan is not proof the ratio is working. The clean test comes after 15 November, when the same broad dollar move produces a different yuan response.
The trade map
USD/CNH is the direct expression and it is available — the pair sits on the live instrument list alongside EUR/CNH, GBP/CNH and JPY/CNH. The detail most people get wrong is the pip. At 0.01 lots the position is 1,000 USD notional, so one pip is worth about $0.015, roughly a sixth of the pip value of 0.01 lots of EUR/USD. A 300-pip policy week is about $4.50 on a micro lot, not $30. Margin at 30:1 is roughly $33; at 500:1 about $2 — a difference that changes survival odds more than the direction call does.
AUD/USD carries China beta with tighter spreads and deeper liquidity than any CNH cross. Use it when the view is Chinese growth rather than the fixing mechanism.
HK50 / HSI, last at 24,315.92, is the equity channel: a slower yuan plus cheaper domestic funding is a mixed signal for Hong Kong-listed Chinese names, and index CFDs price that ambiguity faster than single stocks.
Invalidation, stated in advance: if the fix drifts back below 6.65 and spot follows, the rule failed and the rally has resumed. If USD/CNH holds above 6.72 into December, the brake is working.
The honest counter-case
Quantity tools have a poor record of ending currency moves. In September 2023 the same instrument was used in the opposite direction — cut 6% to 4% to lean against depreciation — and the yuan kept sliding for months. Reserve ratios slow a trend; rate differentials and positioning decide it.
Watch the offshore leg too: central bank bill issuance in Hong Kong drains CNH liquidity and raises CNH funding costs. If bill sales restart, that hits USD/CNH more directly than the onshore ratio does.
Calendar
The rule is effective Sunday 15 November. The first genuine test is the Monday 16 November fix, then the following week’s settlement data on exporter conversion. Anything before that is positioning, not evidence.