The Structure of China's Total Social Financing

Total Social Financing (TSF) is a financial system asset-side statistical indicator first created by the People’s Bank of China in 2010, measuring the total volume of funds provided by the financial system to the real economy. Together with M2 (the liability side), it forms a two-perspective framework for observing monetary transmission mechanisms. Of its four statistical principles, the residency principle and the financial-system principle contain theoretical flaws; the framework concludes that digital currency will be the ultimate successor to TSF.

The Framework As It Stands

This section is compiled from research notes: it preserves the structure, terminology, and key formulations of the original framework, including editorial bridging and supplementary external facts; diagrams are drawn by the compiler following the structure of the original text.

Core Thinking: Cutting the Economy in Two

The framework’s fundamental perspective divides the socioeconomic system into two halves:

• The financial system: the system that produces and manages the special commodity called money, including the central bank, commercial banks, brokerages, insurance companies, trusts, and funds.
• The real economy: the consumers of money, including corporations, individuals, and the government.

The framework emphasizes: the government must be counted as part of the real economy (as a consumer of money), and cannot be listed as part of the financial system — this differs from the mainstream Western assumption that “governments can print money.”

The “wealth warehouse” metaphor: the real economy works every day → entrusts wealth to the financial system (the warehouse) → the warehouse keeper issues receipts (currency) → the receipts are M2. The essence of monetary devaluation is institutional breach of trust: people no longer believe in the receipt’s redemption capacity, and the entire society’s chain of trust collapses. Repo and Shadow Money reveals the micro-mechanism by which the warehouse keeper issues “duplicate receipts” off-balance-sheet.

M2 vs. TSF: Liability Side and Asset Side as Complements

• M2 = liability side of the financial system: total money stock is known, but “where the money went” is opaque (all types of digital balances look the same). One of the core lessons of the 2008 financial crisis was that regulators could not trace money flows.
• TSF = asset side of the financial system: audits the “shelves” — looks at who holds asset-liability relationships with financial institutions, thereby inferring structural information about the direction of capital flows (by region / by industry / by financing structure).
• TSF was first created in 2010 by Sheng Songcheng and colleagues at the People’s Bank of China; to date, China is the only country using this statistical measure.

The framework corresponds to two schools of monetary transmission mechanism:

• The monetary view (focuses on the liability side / mainstream): money outflow → interest rates → investment/consumption → aggregate demand/aggregate supply → prices
• The credit view (focuses on the asset side / minority): interest rate policy → asset prices → investment/consumption → aggregate demand/aggregate supply → prices

TSF corresponds to the credit view, supplementing the structural information that M2 cannot provide.

TSF’s Four Statistical Principles and Their Flaws

Principle	Content	Theoretical Assessment
Residency principle	Covers only domestic financial system funds to domestic entities (excludes FDI, external debt, foreign exchange reserve flows)	❌ Flaw: foreign capital inflows actually affect domestic monetary transmission
Financial-system principle	Excludes central government bonds + local government bonds (except LGFV bonds)	❌ Flaw: assumes government = money issuer, valid only in China; inapplicable where the central bank is independent (e.g., the Fed)
Consolidation principle	Financial institutions consolidated into one for statistical purposes	✅ Logically sound
Incremental statistics principle	Quarterly/monthly increments as the primary measure, stocks as supplementary	✅ Logically sound

The framework concludes: only the consolidation principle and the incremental statistics principle are genuinely sound; the residency principle and the financial-system principle are theoretically inconsistent. The implicit assumption of the financial-system principle (that the government holds money-issuance authority) points to the government–central bank boundary issue in China’s monetary policy — an assumption that does not hold in the U.S. system where the Fed is independent.

TSF’s Four Components (2018 Definition)

1. On-balance-sheet loans: RMB loans + foreign currency loans (including discounted bills)
2. Off-balance-sheet financing: entrusted loans + trust loans + undiscounted bankers’ acceptances
3. Direct financing: non-financial corporate bonds + equities
4. Other financing: insurance indemnities + loan write-offs + asset securitization + local government special-purpose bonds + …

Newly incorporated from July 2018: loan write-offs + asset securitization by deposit-taking financial institutions; loan write-offs in 2018 were approximately RMB 600–700 billion, growth of 30%+ over 2017.

Not included: P2P / private equity / equity crowdfunding / internet finance.

flowchart TD
    A["Systems thinking: cut the economy in two<br/>Financial system = produces money<br/>Real economy = consumes money"] --> B["Wealth warehouse metaphor<br/>Real economy submits wealth → warehouse issues receipts<br/>Receipts = M2"]
    B --> C["M2 limitations<br/>Liability side: can't see where money went<br/>Lesson from 2008 crisis"]
    C --> D["TSF: PBOC first created 2010<br/>Audits shelves from asset side of financial system"]
    D --> E["Four statistical principles"]
    E --> F["✅ Consolidation principle<br/>Financial institutions consolidated"]
    E --> G["✅ Incremental statistics principle<br/>Quarterly/monthly as primary"]
    E --> H["❌ Residency principle<br/>Excludes FDI/external debt/forex flows<br/>Theoretically inconsistent"]
    E --> I["❌ Financial-system principle<br/>Excludes central/local gov't bonds<br/>Assumes government = money issuer"]
    D --> K["TSF four components"]
    K --> L["On-balance-sheet loans<br/>RMB + foreign currency"]
    K --> M["Off-balance-sheet financing<br/>Entrusted loans + trusts + undiscounted acceptances"]
    K --> N["Direct financing<br/>Corporate bonds + equities"]
    K --> O["Other financing<br/>Insurance indemnities + loan write-offs + asset securitization + local gov't special bonds"]
    D --> P["Theoretical basis: monetary transmission<br/>Monetary view vs. credit view<br/>TSF corresponds to credit view"]
    D --> Q["Future replacement: digital currency<br/>Real-time tracking of every transaction<br/>Resolves both monetary and credit sides"]

Key Data Cross-Section (2018–2019)

• 2018 TSF outstanding stock: approximately RMB 205 trillion, year-on-year growth rate approximately 10.4%
• January 2019 TSF increment: approximately RMB 4.6 trillion; February 2019 increment: approximately RMB 0.7 trillion (significant Chinese New Year effect)
• Q3 2018 regional divergence: Guangdong 17,000+ billion (highest) / Beijing 12,000+ billion / Shanghai 4,116 billion / Shanxi 900+ billion / Jilin 1,100+ billion

Future Replacement: Digital Currency

The framework concludes: central bank digital currency (CBDC) naturally resolves both the monetary side and the credit side — every transaction is recorded in real time, every unit of currency is traceable. In the digital currency era, TSF will no longer be needed — lower cost, higher efficiency, zero theoretical flaws. The digital RMB (e-CNY) has been piloted since 2020 in Shenzhen, Suzhou, Chengdu, Xiong’an, and other locations; the 2022 Beijing Winter Olympics completed an international pilot. China and US Payment Systems contains a dedicated analysis of the strategic competition in digital currency and payment infrastructure.

Compiler’s Perspective

Coordinates: Category = Observation Indicators & Signals / axis_h = Qi / axis_v = What It Is

Connecting layer

The divergence between M2 and TSF has practical implications: when M2 year-on-year rises while TSF year-on-year does not move in sync, money is stuck circulating within the financial system (interbank lending, not reaching the real economy) — a diagnostic marker for “monetary easing without stimulus,” not a green light for “easing is working.” Conversely, if TSF surges but M2 does not follow, it means direct financing (corporate bonds/equities) is expanding — a sign of market vitality, not monetary illusion. Analysts who look only at M2 will err in both directions. Loan write-offs alone in 2018 amounted to approximately RMB 600–700 billion; if not incorporated into the TSF measure, actual deleveraging pressure would be significantly underestimated. This data was hidden before the July 2018 definition revision, and the definition change itself was a policy signal.

Proprietary increment

TSF’s residency principle and financial-system principle contain a structural asymmetry: the residency principle omits foreign capital inflows (FDI/external debt/foreign exchange reserve flows), while the financial-system principle’s assumption (government = money issuer) holds when China’s central bank is subordinate to the State Council system, but does not hold in a system where the central bank is independent. This means: China’s TSF cannot be directly compared on a cross-sectional basis with U.S. monetary statistics — TSF’s “credit perspective” is closer to the credit view, and the U.S. Fed system has no equivalent indicator. When cross-national researchers apply U.S. monetary statistics logic to interpret China’s TSF, they will systematically misjudge the true influence of the sovereign and local government bonds excluded by the “financial-system principle.”

See Also

• Repo and Shadow Money
• The Dollar Circulation System
• China and US Payment Systems
• The Stagflation Risk Framework

Sources

• Compiled notes z-0028 · collected July 2026
• PBOC Monthly Total Social Financing Statistics, pbc.gov.cn
• Sheng Songcheng, Theory and Practice of Total Social Financing, China Financial Publishing House
• Digital RMB (e-CNY) pilot program, pbc.gov.cn