Two Powerful Forces Are Shifting Global Liquidity in 2026

The world no longer has a single global liquidity cycle, the USD cycle; it now has two. A Chinese bloc where liquidity is ample, yields are pinned near record lows and fiscal firepower is exported through the Belt and Road initiative. Its impact is associated to domestic and regional Asian demand for previous metal, export of majority of manufactured goods, and domination of various day to day industries. Then there is the dollar bloc (including the offshore Eurodollar market), where finance, banking and global services industries benchmark against. The US Treasury has increased the size of its long-end liquidity-support buybacks, which may affect liquidity and pricing in those sectors. The assets responses, including USD, CNY, gold, bitcoin, the Aussie, EM currencies are reacting separately to the variation of impacts.

Argument: Short-rate and liquidity conditions have decoupled into two global systems with different transmission mechanisms.

	Bloc A — CNY	Bloc B — USD / Eurodollar
Policy rate	1Y LPR 3.00%, 5Y LPR 3.50% — unchanged 12 straight months	Fed funds 3.50–3.75%, held 29 July
Direction of surprise	Easing bias, liquidity-led	Hiking bias, three dissents for an immediate hike
10Y govt yield	~1.71% (vs 20-yr average 3.39%)	4.65–4.71%
30Y govt yield	~2.50%	5.20% — a 19-year
Reach	Domestic + regional	Domestic + offshore Eurodollar
Fiscal channel	BRI: $126.3bn engagement in H1 2026	Debt buybacks; $1.2trn interest expense FY-to-date

Section 1: The CNY liquidity towards Gold

Ample domestic liquidity plus a pinned curve leaves yuan holders with a low real return, and short-term funds have likely rotated into precious metals, gold in CNY terms turning up before the dollar leg of the move. the public data that corroborates it is the official-sector bid, not retail flow: central banks bought 288.9 tonnes in Q2 2026, up 62% year-on-year, and did so while the price was falling. Full year forecast 895.4 tonnes.

Chart 1: GOLD vs CNY has been a strong reflection of Chinese liquidity, and both retail and official demand for the precious metal because of debasement of the CNY.

Chart 2: 10yr Chinese Sovereign Curve movement ETD 2026. Liquidity injection and bond demand.

Section 2: The dollar leg: $40 trillion and a bid from the issuer

The sequence, tightly

1. 18 Aug 2026  gross federal debt hits $40,047,425,768,420.22. Split: $32.27trn public, $7.78trn intragovernmental. It arrived roughly two fiscal years ahead of the CBO's May 2023 projection, five months after $39trn and 4.5 years after $30trn.

2. Same week: the 30Y yield touches its highest since 2007 on fiscal anxiety and Iran escalation risk.

3. 19 Aug 2026: Treasury announces liquidity-support buybacks in the 10–30Y sectors rise from a $2bn cap to at least $4bn per operation, effective 9 September through 4 November.

4. Yields drop: 30Y from 5.26% to 5.18% intraday, closing 5.196% (−9bp); 10Y 4.68% → 4.63%, closing 4.647% (−5bp).

5. 20 Aug 2026: Bessent signals operations could exceed $4bn and that Treasury intends to make a market in the long end. Yields give back much of the move.

6. 21 Aug 2026: DXY 98.82, a three-month low near 98.50 intraday, −2.28% on the month.

   
   

Funding debt with more debt doesn’t solve any liquidity or credit issue

A $2bn increase per operation against a $32trn market. The market reaction looks less like a response to the mechanical effect of the operation than to what it implied might follow: pricing for broader intervention to come. The manner of the announcement reinforced that reading, breaking Treasury's usual pattern of steady, well-flagged communication about its borrowing plans. The structural objection is the one that matters, though. Buybacks do not change deficits. Retiring the long end still requires issuance to fund it, most likely more bills, which shortens the average maturity of the debt and pushes the refinancing problem forward rather than solving it. this is a maturity swap, not money creation. It does not add bank reserves. Financed with bills, it exchanges long debt for short debt. It improves liquidity and supports long bond prices; it does nothing for the credibility question underneath.

Chart 3: The depreciation and credibility argument

The question everyone is asking is that, are we in a quantitative easing, or FED expansionary environment? If it’s only the Treasury buying the long end of the sovereign curve and reissues short end to fund it, and that a currency (The USD) should depreciate under that weigh of escalation credit vulnerability from a raising liquidity while the central bank worries about inflation. Chart 3 shows the movement of USD vs CNY during a period when US government bond yields are raising (Chart 4) while Chinese government bond (Chart 2) yields are dropping.

Chart 4: 10-year US Sovereign Curve movement ETD 2026. Rasing inflationary fear and credit profile exit.

Section 3: Where the money went since US treasury buyback

The August scoreboard

Asset	Move	Level
Bitcoin	+~29% on the week — biggest in over two years	~$62,800 → $80,000
Total crypto cap	+$150bn in a single day	above $2.45trn
Gold	+~5% on the week, highest since mid-May	above $4,600
Silver	+3.9% on the day of the announcement	~$62–66
DXY	−2.28% on the month	98.82
AUD	Multi-month high	—
MSCI EM currency index	All-time high on 6 Aug	—
MSCI EM equities	+2.6% in the week to 14 Aug, best since June	—
Nasdaq	Essentially flat on the announcement day	26,180
Crude	Four-week high	—

The crypto move

The crypto move was not pure allocation. Roughly $2.7bn of shorts were liquidated inside 24 hours, and CoinGlass put total bearish liquidations above $4bn across the rally. Spot ETFs did draw real money — around $1bn over the stretch, including a $517m single day for BTC and $189m for ETH, the biggest in months. So: a genuine flow, amplified by a squeeze. Potentially an reallocation of leverage and short-term resources from the AI complex back to other risky assets? The AI complex has been de-rating since July on its own logic (crowded positioning, capex scrutiny, memory oversupply),

Section 4: Our analysis

4.1: The short-term impact of US treasury buyback will not last on yield, given the FED positioning. 

Support: the Fed held at 3.50–3.75% on 29 July with a 9–3 vote, Hammack, Kashkari and Logan all dissenting for a hike, the first triple dissent since September 2016. CPI is 3.4% against a 2% target. Oil is at a four-week high with US pressure on Iran limiting hopes of a Hormuz reopening. Much less likely of a Quantitative stimulus or easing cycle from the US central bank.

4.2: Dollar weakness comes from financing credibility, not from yields. 

Support: the buyback pulled yields down and the dollar down together. DXY −2.28% on the month while the 30Y remains above 5.19% and real yields sit near 3.03%. The creditability of the DXY and USD denominated assets are losing its investor-based confidence relative to alternative assets.

Dollar weakness to continue despite elevated US yield.

4.3: Reserve-manager demand for gold may point to diversification away from USD exposure. 

Support: the official-sector gold bid, 288.9 tonnes in Q2, +62% y/y, bought into falling prices, which is the behaviour of a mandate rather than a trade. Nearly 900 tonnes forecast for the year.

4.4: Gold, crypto, AUD, SGD and EM assets etc are the beneficiaries.

Support: all of them outperformed in the same week, on the same catalyst.

4.5: The political overlay. 

Trump's crypto support is a policy tailwind now; November's midterms could remove the Congressional leg of it. A lame-duck presidency is a weak forecasting variable, and the honest version of the argument is that it matters mainly in a crash scenario. The CLARITY Act cloture vote scheduled for 15 September could elevate the digital asset market to new highs.

Section 5: What breaks the thesis

Three genuine risks, stated without hedging:

1. Oil. A sustained spike forces the Fed hawkish, real yields rise, and the entire hard-asset trade unwinds. Currently live given Iran.

2. The squeeze was the story. If crypto gives back the week's gains once forced buying is exhausted, the "flows into hard assets" narrative was mostly derivatives plumbing.

3. China's decoupling is weakness, not strength. Yields at 1.7% with the LPR unchanged for a year reflect weak credit demand, property strain and deflation risk. A curve that rallies because nobody wants to borrow is not the same as a curve that rallies because capital is arriving.

   
   

About Signafi Capital Management

Traditional finance and digital assets no longer operate in separate worlds. Our clients allocate across both, often simultaneously, and they need partners who understand institutional standards as deeply as they understand digital innovation. Signafi derives from signal and finance: clarity and precision in markets where complexity is increasing and the boundaries between financial systems continue to evolve. It reflects our mission of providing institutional investors and sophisticated wealth holders across Asia-Pacific with access to both traditional and digital markets, combining innovation with institutional discipline. It reflects our ambition: to become the region's most trusted institutional partner as finance continues to evolve.

General Disclaimer:

The content provided in this blog is published by Vector Capital Management Pty Ltd trading as Signafi Capital Management (“Signafi”) for general informational and discussion purposes only. It does not constitute legal, financial, tax, or investment advice and should not be relied upon as such. Any opinions, views, commentary, predictions or market views expressed are those of the author and do not necessarily reflect the views of Signafi.

The information provided does not take into account any person’s objectives, financial situation or needs and is not intended to constitute financial product advice or a recommendation to acquire, hold or dispose of any financial product or digital asset. No representation or warranty is made as to the accuracy, completeness or timeliness of the information, and Signafi accepts no liability for any loss or damage arising directly or indirectly from reliance on the content.

The content may refer to economic trends, markets or regulatory matters that are subject to change. Readers should undertake their own research and seek independent professional advice tailored to their individual circumstances before making any decisions.