The Data Point That Changed Everything
July new yuan loans: -340 billion yuan.
This is not a seasonal dip. This is the largest monthly contraction on record — and it confirms what the property data has been screaming for 18 months: China is in a balance sheet recession.
For context:
- June: +210B yuan (seasonal)
- July 2025: +345B yuan
- July 2026: -340B yuan (record contraction)
- 12-month trailing: +13.2T vs. +18.5T a year ago (-29%)
The credit impulse has collapsed. And in a system where credit is the transmission mechanism, this is the signal that matters.
The Transmission Chain: Property → LGFV → Sovereign → Global
Step 1: Property (The Trigger)
Commercial property values: -40%+ from peaks in Tier 1 cities (Beijing, Shanghai, Shenzhen, Guangzhou).
40-year commercial leases granted in the 1990s-2000s are expiring 2030s-2040s. Renewal requires:
- ≥70% of depressed benchmark land price
- Payable upfront or installments (>1 year)
- Not perpetual — deniable for "planning changes/violations"
Cash flow math: Typical office NOI CNY 150M → renewal cost CNY 1.4B upfront → annualized CNY 280M > NOI → negative carry → rational owner walks away.
This is not cyclical. It is structural. Residential has a "social stability put" — commercial does not. No bailout coming.
Step 2: LGFV (Local Government Financing Vehicles)
LGFV swap quota: 90% used.
LGFVs were the primary buyers of land (revenue source for local govts). With land sales collapsing (-50%+ YoY), LGFVs:
- Cannot service existing debt
- Cannot buy new land
- Are being "swapped" into explicit local government debt (quota nearly exhausted)
Result: The contingent liability becomes explicit sovereign liability. The "implicit guarantee" is being tested.
Step 3: Sovereign Absorption → UST Sales → Higher Global Yields
This is the global transmission mechanism (Pillar 2: China Credit Impulse):
Property collapse
→ LGFV distress
→ Local govt explicit debt issuance
→ PBOC forced to absorb (or allow rates to spike)
→ CNY pressure (capital outflow risk)
→ China sells UST reserves to defend CNY
→ UST yields rise globally
→ Dollar strengthens
→ EM funding stress
→ Global liquidity tightens
Evidence: China's UST holdings have declined from $1.1T (2021) to ~$770B (2026). The pace accelerated in 2024-2025.
The $148B Commercial Lease Expiry Cliff
This is the forward-looking catalyst that makes the -340B loan print structural, not cyclical.
| Metric | Detail |
|---|---|
| Total exposure | $148B (commercial lease renewals 2030-2045) |
| Annual renewal volume | ~$8-12B/year (ramping) |
| Renewal cost | ≥70% of benchmark (depressed) |
| Payment terms | Upfront or ≤1 year installments |
| Haircut vs. peak value | 60-70% |
No precedent for resolution. The Guangzhou/Shanghai template: pay 70% or lose asset. No restructuring framework. No bankruptcy court with teeth for state-owned enterprises.
CNY: The Pressure Valve
USD/CNH: ~7.25-7.30 (watch 7.35 crisis threshold)
The PBOC has managed CNY via:
1. Daily fix (stronger than spot, signaling resistance)
2. State bank dollar selling in onshore market
3. Cross-border flow controls (quiet tightening)
4. Offshore CNH bill issuance (drain liquidity, raise CNH HIBOR)
But: With loan contraction accelerating, property in freefall, and LGFV quota exhausted, the PBOC faces a trilemma:
- Defend CNY (sell UST, tighten domestic liquidity) → deflationary
- Ease domestic (cut RRR/MLF) → CNY weakens, capital outflow accelerates
- Let CNY slide → imported inflation, UST selling, global spillover
Current choice: "Appropriately loose" (MLF/RRR cuts) but no broad stimulus. This is the "slow bleed" path — CNY gradually weakens, UST selling continues, global yields grind higher.
Positioning Implications
From the Unified Framework (Aug 16)
China Credit Impulse = Pillar 2 — one of four pillars driving global macro.
Current positioning:
- Short China property (4%): Excellent asymmetry (5:1) — structural bear
- Short China banks (4%): NPL cycle turning, LGFV exposure
- Long USD/CNH (5%): 7.35 crisis threshold, carry + trend
- Short Copper/Iron Ore (3%): China demand destruction
Updated Tactical View
| Asset | Position | Trigger to Add | Trigger to Reduce |
|---|---|---|---|
| China Property (short) | 4% | New lows in sales floor space | Policy bazooka (broad stimulus) |
| China Banks (short) | 4% | NPL ratio >2.5% | Recapitalization announced |
| USD/CNH (long) | 5% | Break 7.30 → target 7.35 | PBOC fixes >7.20 consistently |
| Copper (short) | 3% | LME inventory >300kt | Green capex demand surprise |
What Could Break the Thesis (Bear Case for the Bear Case)
- Broad fiscal stimulus (>3% GDP, not targeted): Property developer bailout, LGFV full recap, consumption vouchers
- PBOC QE: Explicit UST buying / yield curve control → CNY stabilizes, global yields drop
- US recession: Fed cuts aggressively → dollar weakens → CNY breathing room → China eases without outflow risk
- Property floor: Sales stabilize at 60% of peak → confidence returns → credit impulse recovers
Probability: <15% combined. The structural forces (demographics, overcapacity, debt saturation) are too entrenched.
My Honest Take
The -340B loan print is the most important macro data point of 2026 so far.
Not NVDA earnings. Not Jackson Hole. Not even the Treasury buyback.
Because it confirms that the world's second-largest economy is contracting its balance sheet — and in a credit-based system, that is a deflationary impulse that transmits globally via the dollar funding market.
The West thinks in terms of "China stimulus coming." China is delivering balance sheet recession. The gap between expectation and reality is where the money is made.
Position: Maintain China shorts. Add to USD/CNH on 7.28-7.30 pullbacks. Watch UST foreign custody data weekly — if China selling accelerates (>$20B/mo), the global duration bear market has a new, powerful driver.
The "slow sucking sound" (FT) of AI capex funding is matched by the slow bleeding sound of China credit contraction. Both push long yields higher. Both are structural. Neither is priced.