Checking
Beijing

Historical reading · v230 · 2026-10-18. Read the current understanding and full revision history →

Historical reading v230 · 公开专题 · 货币政策与人民币

Monetary Policy & RMB

Monetary policy and the renminbi — PBOC rate moves, RRR cuts, the exchange rate, liquidity, LPR, and cross-border RMB.

Updated
2026-10-18
Revision
v230
Evidence
10 cited documents
Access
Public reading and history
Graphite editorial illustration of the curved headquarters of the People's Bank of China in Beijing.
Policy transmissionPeople's Bank of China · Beijing

Historical reading v230 · v230 · 18 Oct

Reserves drop is valuation, not defence; RMB plumbing reaches the Gulf

The PBOC's stance is unchanged and the newest data keeps confirming it: the September reserve decline was a price effect, not a defence bill, and the renminbi's international build-out has extended from CIPS bookkeeping into Gulf central-bank swap and settlement arrangements. That combination — no instrument moved, no dollars sold, plumbing widened — is the whole story of this cycle, and it is why external readers should not read the $3.4trn headline as stress.

Foreign exchange reserves were $3,400.3bn at end-September, down $38.1bn or 1.11% on August, and SAFE attributes the entire move to a firmer dollar and lower global asset prices 1. Translation and mark-to-market drain no domestic liquidity and require no policy response. A genuine defence looks different: reserves falling alongside heavy spot dollar selling, state banks absorbing the other side, and the corporate settlement ratio lurching as firms front-run the fixing. August showed the opposite — the corporate FX settlement ratio was 61.5%, 2.7 points below the January-July average — and the flow data now reinforce it. August goods-and-services trade ran a $94.4bn surplus on a $424.3bn export base 2, and the domestic FX market turned over $3.87trn (26.28trn yuan) in the month, of which derivatives were $2.74trn, roughly 71% 3. Firms are hedging their exposure, not liquidating it.

The internationalization layer is where accumulation is actually happening, and it has just acquired a Gulf arm. At the fourth China-GCC Central Bank Governors' Meeting in Bangkok on October 14, governors agreed to keep pushing local-currency cooperation, deepen cross-border payment links, expand two-way investment and connect financial market infrastructure; Pan Gongsheng separately renewed the bilateral local-currency swap with Qatar Central Bank and met the UAE, Qatari and Omani governors 4. This matters beyond the communiqué language: the Gulf prices its oil in dollars, so local-currency settlement and swap lines there are a real substitution channel, not a symbolic one. It is the same logic as CIPS, which ended September with 214 direct and 1,651 indirect participants across 133 countries and regions, having added 4 direct and 53 indirect participants in the quarter against 17 indirect withdrawals and 14 newly qualified direct holders not yet live 5. None of that requires a rate or RRR move.

The limits remain structural. First-half current-account surplus of $378.0bn was offset almost entirely by a $345.2bn capital-and-financial-account deficit 6 — earned inflows recycled outward. The honest depth gauge is still the currency composition of external debt: $2,498.2bn of total liabilities at end-June, of which local-currency debt was 95,609bn yuan ($1,403.8bn), or 56%, with short-term debt at 59% and central-bank debt at just 4% 7. That share is holding, not surging. And Pan's October pledge to make renminbi bonds "globally widely recognized and used eligible collateral" still carries no eligible-issuer list, haircut schedule or cross-border custody detail 8.

The easing-versus-stability tradeoff is being managed by refusing to let the currency do adjustment work. Pan restated the "moderately loose monetary policy" (适度宽松的货币政策) at the G20 finance ministers and central bank governors meeting in Bangkok on October 15, and told the same meeting that global imbalances cannot be corrected by exchange-rate adjustment, that the exchange rate bears no simple linear relation to the current account, and that surplus and deficit countries must both adjust 9. He repeated the argument to European Commissioner Dombrovskis, ECB President Lagarde and Eurogroup President Pierrakakis in Washington the next day, with the EU side conceding that adjustment requires all stakeholders 10. That is a deliberate decoupling: as long as the yuan is not the adjustment instrument, monetary policy can stay loose without generating a defence bill — and the September reserve print is what that decoupling looks like in the data.

The live risk is communicative, not operational. If the dollar stays firm, the headline reserve number keeps falling and will be read abroad as covert defence even when SAFE's own attribution says otherwise, and $3.4trn is a psychologically loaded line. For external readers and their agents, nothing here justifies repricing yuan exposure off a policy surprise. The base case remains incremental liquidity support, a firm grip on the fixing, refusal to let external-adjustment arguments drive the currency, and quiet internationalization via CIPS, Panda issuance, local-currency debt and now Gulf swap lines. Watch four things: whether the 14 CIPS direct-qualification holders actually go live; the third-quarter external-debt release and whether the renminbi share holds above 56%; whether the Qatar swap is drawn or enlarged and whether Gulf settlement shows up as new CIPS direct participants; and any month in which reserves fall by a magnitude SAFE cannot attribute to valuation and asset prices.

What changed in this reading

The internationalization layer acquired a Gulf dimension: the fourth China-GCC central bank governors' meeting in Bangkok on October 14 agreed to push local-currency cooperation, cross-border payments and infrastructure linkage, and Pan Gongsheng renewed the bilateral local-currency swap with Qatar Central Bank 4. New flow data reinforce the no-defence reading — August goods-and-services surplus of $94.4bn 2 and FX market turnover of $3.87trn with derivatives at ~71% 3. Nothing on rates, RRR, the fixing or reserves attribution changed.

How the evidence changed

Cited documents

Comparing v229 with v230. Band width counts distinct cited documents.

Evidence between two readings7 citations retained, 3 newly cited and 1 no longer cited. The document lists below provide the same information.7 documents cited in both readings1 documents no longer cited3 newly cited documentsv229 · 8v230 · 10Added · 3Left · 1

Newly cited means new to this reading, not newly published. No longer cited does not mean retracted. Citation changes do not establish a policy change.

Evidence behind this reading

10 cited
Published by

China Snapshot, a publication of CHINA SNAPSHOT LIMITED (Hong Kong). Operated by Tristan McInnis, Managing Partner, Inner Chapter, Shanghai. Chinese sources machine-translated by DeepSeek; interpretation drafted by a language model under our editorial standards; corrections to support@china-snapshot.com.