Monthly Bond Market Liquidity Tracking: The Six-Item Excess Reserve Framework

The monthly bond market liquidity tracking framework refers to an analytical method that estimates the looseness or tightness of bond market funding in a given month by tallying the monthly injection and withdrawal of excess reserves across six line items on the central bank’s balance sheet, and then uses market interest rates as the ultimate cross-verification indicator. The framework’s core inference is: in the two-tier banking system, banks’ willingness to invest in bonds is always a given (the opportunity cost of holding excess reserves is too high), so the true binding variable is the abundance of excess reserves.

The Framework As It Stands

This section is compiled from research drafts: it preserves the original framework’s structure, terminology, and key formulations, including editorial bridging and supplementary factual annotations; the diagrams are drawn by the compiler according to the original textual structure.

Core Topic and Dark Threads

This framework moves bond market liquidity analysis from “abstract principles” into “practical application”: the core liquidity of the bond market depends on funding conditions in the interbank market, and what banks use to invest in bonds is excess reserves (the surplus after statutory reserves are set aside).

The main judgment: for the bond market, investment willingness is constant (the opportunity cost of holding excess reserves is too high, so banks inevitably lean toward investing in bonds); therefore the true binding variable is not “willingness” but “having the money” — i.e., the abundance of excess reserves. Bond market liquidity tracking thus reduces to tracking the monthly net change in excess reserves.

Three dark threads:

• Dark Thread A — Liquidity = Willingness + Having the Money: On the bond market side, willingness is constant, so the question falls to “having the money” (excess reserves); this is the key conceptual step that operationalizes the abstract notion of “liquidity.”
• Dark Thread B — Interest Rates Are the Ultimate Liquidity Indicator: All liquidity changes ultimately show up in interest rates; even when other data are unavailable, directly watching market interest rates is the simplest and most effective tracking method (ex-post, but high-frequency and real-time).
• Dark Thread C — Precision Honesty: This estimation method “will not be particularly precise” — some data lack authoritative sources, some values are estimates; the social sciences always have unforeseen variables, and mastering the principles does not guarantee profits.

Argument Structure

1. The primary trading participants in China’s bond market are in the interbank market (where banks trade base money and bonds); retail investors enter indirectly through bond funds and bank wealth management products.
2. Banks invest in bonds using excess reserves (the surplus after statutory reserves are set aside); the reserve interest rate is low → hoarding excess reserves is not worthwhile → banks tend to invest in bonds.
3. Key inference: Bond market investment willingness is constant; the true constraint is the abundance of excess reserves (“having the money”).
4. Monthly tracking = tallying each channel’s injection/withdrawal of base money and computing the net change in excess reserves.
5. The 6 line items that affect excess reserves (using central bank balance sheet terminology):
   • ① Foreign exchange holdings: capital inflows → inject base money; outflows → withdraw
   • ② Claims on other depository corporations (relending, MLF/SLF, etc.): net injection > 0 → loose
   • ③ Government deposits (fiscal): deposits into the Treasury → withdraw from circulation (tight); government spending → inject into circulation (loose)
   • ④ Statutory reserve deposits: deposit expansion → more statutory reserves required → consumes excess reserves (tight)
   • ⑤ Currency issuance (cash/M0): cash withdrawal → excess reserves↓; redeposit → excess reserves↑
   • ⑥ Central bank non-bank deposits (Alipay/WeChat Pay): small in scale, not a primary focus under normal conditions
6. Higher excess reserve ratio → looser market; lower → tighter; monthly estimation of month-end excess reserves (ratio).
7. Interest rates are the more direct liquidity indicator: high rate = tight, low rate = loose; can only be tracked ex-post, but high-frequency and real-time; they are the ultimate result of all liquidity changes.
8. Anomalous situation: excess reserve ratio calculated high, but interest rates are also high — banks hold funds as a precautionary measure (fearing the central bank will tighten liquidity) rather than deploying them at low rates; this means “having the money” does not necessarily mean “putting it to use.”

Reasoning Chain

flowchart TD
  A["Bond market core in the interbank market<br/>Banks invest in bonds using excess reserves"] --> B["Willingness is constant (reserve rate is low, hoarding is suboptimal)<br/>→ Constraint = abundance of excess reserves"]
  B --> C["Monthly tally of 6 items: base money injection/withdrawal<br/>FX holdings · Relending · Gov deposits · Statutory reserves · Currency issuance · Non-bank deposits"]
  C --> D["Aggregate → Month-end excess reserves (excess reserve ratio)<br/>Ratio↑ = loose, ↓ = tight"]
  D --> E["Interest rate = ultimate/simplest indicator<br/>High = tight, Low = loose (ex-post, high-frequency)"]
  E --> F["Anomaly: High excess reserve ratio + High interest rate<br/>= Precautionary motive: hoarding funds without deploying"]
  D -.Precision limits.-> G["Results imprecise: partial data are estimates<br/>Social science has unforeseen variables; mastering the principles ≠ profits"]

Key Course Examples (Numbers Are from an Example Month, Not Current)

• March example: reverse repos drained ¥30 billion; relending (broad) injected over ¥50 billion; final calculation showed excess reserves increased by ¥1.24 trillion.
• M2 up ¥3.8 trillion in March: derivative sources centered on loan-created deposits (~¥2.9 trillion) plus fiscal spending; end-uses accumulated as corporate demand deposits (+¥2.8 trillion) / individual deposits (+¥1.9 trillion) / non-bank deposits (−¥200+ billion); both sides must balance in aggregate, but the intermediate fund-flow paths cannot be tracked with publicly available data.
• The M2–Excess Reserve Ratio relationship: calculating the excess reserve ratio requires M2 (the scale of deposits determines required reserve needs); M2 is an aggregate indicator of multiple liquidity sources (e.g., equity market funds come from M2).

⚠️ The figures above are course calculations for an example month and are not current data; making a current-period judgment requires taking the current month’s central bank balance sheet data and independently verifying it.

Monthly Tracking Checklist

#	Indicator / Line Item	Data Source	Direction Interpretation	Misread Risk / Fallback
1	Change in foreign exchange holdings	PBoC balance sheet	Inflows → inject base money (loose)	Data lag
2	Claims on other depository corps (relending MLF/SLF/reverse repos)	PBoC balance sheet / public operations	Net injection > 0 → loose	Reverse repo maturities must be netted
3	Government deposits (fiscal)	PBoC balance sheet	Deposits into Treasury → tight; spending → loose	Large swings at quarter-end / tax-collection months
4	Statutory reserve occupation	Derived from M2 / deposit scale × required reserve ratio	Deposit expansion → more reserves required → consumes excess (tight)	Requires M2 data
5	Currency issuance (M0 / cash)	PBoC data	Cash withdrawal↑ → excess reserves↓; redeposit → excess reserves↑	Spring Festival cash surge is seasonal
6	PBoC non-bank deposits (Alipay / WeChat)	PBoC data	Small in scale, not a primary focus	Negligible
7	Excess reserve ratio (aggregate items 1–6 to estimate month-end excess / deposits)	Self-calculated	Higher = looser	Imprecise; partial values are estimates
8	Market interest rates (ultimate indicator)	Interbank rates (repo rates, etc.)	High = tight, Low = loose	Ex-post only, but high-frequency and real-time

Cross-verification rule: The excess reserve ratio and interest rates should move in the same direction (more excess reserves → lower rates). If the excess reserve ratio is high but interest rates are also high, the interpretation is the anomaly of “banks precautionarily hoarding funds without deploying them” — the excess reserve ratio alone cannot be trusted. Each monthly estimate must carry a precision caveat; do not issue “precise forecasts.”

Compiler’s Perspective

Coordinates: Category = Observational Indicators and Signals · axis_h = Qi (Instruments) · axis_v = What It Is

Connection Layer:

The framework’s greatest methodological contribution is not the six line items themselves, but rather decomposing the vague concept of “bond market liquidity” into a calculable (though imprecise) monthly estimation procedure — converting “tight or loose” from a qualitative sensation into a two-dimensional coordinate of “excess reserve ratio range + interest rate cross-verification.”

The specific form of the error: looking only at whether the central bank “net-injected” in the current month (positive net of reverse repos + MLF), and concluding “funding conditions are loose this month” — this ignores the offsetting effect that large-scale government deposit inflows into the Treasury at quarter-end or tax-collection months have on excess reserves. For example: if government deposits increased by ¥800 billion at end-March (concentrated tax-filing deposits), while the central bank net-injected ¥500 billion, the six-item net calculation would show actual excess reserves declining, and funding conditions would be on the tight side. Analysts who focus on a single line item would be misled by qualitative signals.

Exclusive incremental insight: In the March example, “excess reserves increased by ¥1.24 trillion” and “M2 increased by ¥3.8 trillion” are numbers at two different levels and should not be compared directly — the former belongs to the base money layer (excess reserves are a line item on the central bank’s balance sheet and directly determine the looseness or tightness of interbank market liquidity), while the latter belongs to the broad money layer (including loan-derived deposits, primarily reflecting the scale of credit expansion). Conflating the layers of these two numbers is the root cause of the erroneous inference that “a large M2 increase = abundant market liquidity”; M2 expansion does not automatically equal abundant excess reserves — the two can move in opposite directions within the same month.

See Also

• Modern Money Creation: Money as Debt — Definition of excess reserves and the loan-creates-deposit mechanism: the conceptual foundation for the six-item estimation in this framework

• The Structure of China’s Total Social Financing — Viewing credit expansion sources through broad money M2, and its relationship to the statutory reserve occupation in excess reserve ratio estimation

• The Fed’s Balance-Sheet Reduction (QT) Mechanism — For comparison: the Fed actively manages base money through its balance sheet, complementing the perspective here of passively tracking via six line items

• Bank Entity Liquidity Risk: The Two-Dimensional LCR and Asset-Quality Judgment

• Central Bank Policy Tools and Monetary Policy Transmission

Sources

• Compiled draft z-0184 · collected 2026-07
• People’s Bank of China Monetary Authority Balance Sheet (monthly public release: line items including foreign exchange holdings / claims on other depository corporations / government deposits / currency issuance)
• China Interbank Market Dealers Association DR007 (interbank pledged repo rate) public data — real-time source for the interest rate ultimate indicator