The capital adequacy ratio (CAR) is the core regulatory metric for banks; simplified, it is the ratio of shareholders’ equity to asset size. When the numerator (shareholders’ equity, growth constrained by profits at approximately 6%) persistently lags the denominator (asset size, growth approximately equal to loan growth at approximately 13%), CAR will inevitably decline continuously, forcing banks to raise external capital every few years. Yet all three replenishment paths are blocked under real-world constraints, and when the additional drain from NPL write-offs is layered on top, the popular claim that “bank profit growth is too high” is difficult to sustain under accounting logic.
The Framework As It Stands
This section is compiled from research draft notes: the original framework’s structure, terminology, and key formulations are preserved, with editorial bridges and supplementary factual notes; diagrams are drawn by the compiler following the original text’s structure.
Core Issue and Three Threads
This framework answers a question that cuts to the heart of the matter: “Are a bank’s earnings enough to sustain its minimum capital adequacy ratio?” The entire lecture is not a value judgment; it is a chain of accounting-constraint reasoning: it reduces the capital adequacy ratio to an identity (shareholders’ equity / asset size), shows that when the numerator (shareholders’ equity, profit-constrained growth approximately 6%) persistently falls behind the denominator (assets / loans, growth approximately 13%), CAR inevitably declines, forcing banks to raise external capital every few years; and that all three paths to replenishing capital (retained earnings, new equity issuance, debt instruments) are blocked under real-world constraints. Layering on the additional drain from NPL write-offs, the popular claim that “bank profit growth is too high” does not hold up under accounting logic.
Main conclusions:
-
CAR ≈ shareholders’ equity / asset size; maintaining it constant requires numerator growth ≈ denominator growth. In practice, profit growth is approximately 6% (≈ numerator growth) and loan growth is approximately 13% (≈ denominator growth) — the denominator grows at twice the numerator’s pace, so CAR inevitably declines and banks must raise external capital every few years.
-
Three replenishment paths all blocked — new issuance is constrained by “nearly all bank stocks trading below 1x P/B + CSRC and SASAC prohibition on SOEs transferring assets below 1x P/B”; debt instruments are capped by Basel’s capital-quality limits; ultimately the only path leads back to shareholders’ equity (retained earnings), but retained earnings at 6% cannot keep pace with 13%, so retained earnings alone cannot maintain orderly operations without new issuance — making it difficult to argue that bank profit growth is too high.
-
The solution to the real-economy financing difficulty is not to compress bank profits but to develop non-bank financial channels. Capital adequacy regulation also cannot revert to the 1980s–90s “lend freely if you have money” approach; China, as the world’s second-largest economy, must comply with globally consistent regulation, and Basel also provides a risk management framework.
-
Capital replenishment demand in recent years is stronger than before because NPL write-off activity has accelerated — 1 yuan of NPL requires 1 yuan of provision to write off, almost a 1-for-1 drain on capital (approximately CNY 2 trillion resolved in 2018, CNY 1.4 trillion in 2017; the on-book balance still stands at over CNY 2 trillion).
Thread A — The Capital Adequacy Identity Drives the Entire Lecture
The accounting skeleton of the entire lecture is a simplified identity, continuous with the balance-sheet logic of The Two-Tier Banking System: T-Accounts and Loans Create Deposits: CAR is mostly just above 12%, rarely reaching 14%; simplified: CAR ≈ shareholders’ equity / asset size. Shareholders’ equity has only three sources (original share capital, new share issuances, retained earnings), and asset size is driven mainly by loans. Judgment rule: to keep CAR unchanged, the numerator (shareholders’ equity) must grow at roughly the same pace as the denominator (asset size). Reality: bank profit growth is approximately 6% (≈ numerator growth) and loan growth is approximately 13% (≈ denominator growth) — the denominator grows at twice the numerator’s pace, so CAR inevitably declines and banks must raise external capital every few years. Between 2010 and 2011, some bankers publicly pledged “this is the last fundraising — no more capital raises for the next ten years,” but under this growth gap that commitment cannot be kept, unless loan growth slows to match profit growth — and from the perspective of maintaining normal economic growth velocity, loan growth falling to single digits is not realistic in the foreseeable future.
Thread B — All Three Capital Replenishment Paths Blocked → Replenishment Dilemma
Given that retained earnings at 6% cannot keep up with loans at 13%, can new issuance or debt instruments supplement the numerator? All three paths are blocked. Judgment rules: ① New issuance blocked — CSRC and SASAC regulations stipulate that SOE asset transfers (including new issuances) cannot be below 1x P/B (i.e., cannot be below net asset value); virtually all listed banks are state-controlled enterprises owned by the central or local government, and the vast majority of listed banks trade below 1x P/B, making it impossible to replenish capital via new equity issuance. ② Debt instruments constrained — currently only perpetual bonds, convertible bonds, and similar debt-type instruments can be used to supplement capital, but Basel sets quality requirements: the share of supplementary capital and non-core Tier 1 capital cannot exceed a certain ceiling, and ultimately the path leads back to shareholders’ equity. ③ Therefore, with no ability to issue new equity, the only option is retained earnings — and under the 6%/13% configuration, retained earnings alone cannot maintain orderly bank operations without new issuance. All three paths are blocked; it is therefore difficult to argue that bank profit growth is too high.
Thread C — Capital Doubly Depleted (Balance Sheet Expansion + Write-Offs)
The drain on the denominator side is not limited to new loan origination. Judgment rule: capital replenishment demand in recent years is stronger than in the past, because NPL write-off activity has accelerated. The write-off mechanism: NPLs are written off against provisions, provisions come from earnings, and profits are reported after provisioning. NPL write-offs drain capital faster than new loan origination — 1 yuan of NPL requires 1 yuan of provision to write off (when provisions are exactly at the statutory requirement, write-offs must be replenished with new profits; if provisions exceed the statutory requirement, the excess is counted as supplementary capital and can be converted directly into core capital when no new provisioning is made), almost a 1-for-1 drain on bank capital. Commercial banks resolved approximately CNY 2 trillion in NPLs in 2018 and CNY 1.4 trillion in 2017; cumulative write-offs since 2006 amount to at least tens of trillions, while the current on-book NPL balance of commercial banks still stands at over CNY 2 trillion.
The value of this framework is that it elevates the moralized question “Is bank profit growth too high?” to an accounting-constraint argument of “CAR identity (numerator 6% < denominator 13%) + three replenishment paths blocked + NPL write-off drain.” The conclusion is not “bank profits are inherently high by right” but rather: “Under the existing identity and regulatory constraints, these earnings barely maintain the minimum CAR and support normal loan growth — the real solution is to develop non-bank finance and uphold Basel regulation, not to compress bank profits.”
Key Data Anchors (timestamp approximately 2018/2019)
| Indicator | Value |
|---|---|
| Current CAR | Mostly just above 12%; rarely reaching 14% |
| Simplified identity | CAR ≈ Shareholders’ equity / Asset size |
| Numerator growth (profits) | approximately 6% |
| Denominator growth (loans) | approximately 13% (denominator is twice the numerator) |
| 1x P/B red line | CSRC/SASAC: SOE asset transfers (including new issuance) not below net asset value |
| 1 yuan NPL | Requires 1 yuan provision write-off, nearly 1-for-1 drain on capital |
| 2018 NPL resolution | approximately CNY 2 trillion; 2017: CNY 1.4 trillion |
| On-book NPL balance | Currently over CNY 2 trillion |
Reasoning Structure
flowchart TD A[Core Question: Can a Bank's Earnings Sustain Its Minimum CAR?] A --> B[Thread A: Capital Adequacy Identity] B --> B1[CAR ≈ Equity / Asset Base — Currently 12%+] B1 --> B2[Maintaining CAR Requires Numerator Growth ≈ Denominator Growth] B2 --> B3[Reality: Profit ~6% Numerator / Loans ~13% Denominator] B3 --> B4[CAR Inevitably Declines — External Financing Required Every Few Years] A --> C[Thread B: Three Replenishment Paths All Blocked] C --> C1[Retained Earnings: 6% Cannot Keep Up With 13%] C --> C2[New Issuance: P/B Below 1x + CSRC/SASAC Prohibit SOE Transfer Below 1x P/B] C --> C3[Debt Instruments: Perp Bonds/Convertibles — Basel Cap on Capital Quality — Ultimately Falls Back on Equity] C1 --> C4[Retained Earnings Alone Cannot Sustain Orderly Operations Without New Issuance — Hard to Say Bank Profit Growth Is Too High] C2 --> C4 C3 --> C4 A --> D[Thread C: Capital Doubly Depleted] D --> D1[New Loan Expansion — Denominator 13%] D --> D2[NPL Write-offs: 1 Yuan NPL Requires 1 Yuan Provision — Nearly 1-for-1 Drain on Capital] D2 --> D3[Replenishment Demand Stronger Than Before — 2018: ~CNY 2T Resolved / ~CNY 2T+ Remaining on Books] C4 --> E[Meta-Framework Conclusion] B4 --> E D3 --> E E --> E1[Solution = Develop Non-Bank Financial Channels — Not Squeeze Bank Profits] E --> E2[CAR Regulation Cannot Be Abandoned — Basel = Global Integration + Risk Framework — No Return to 1980s–90s Free-for-All Lending]
Compiler’s Perspective
Coordinates: Category = Banking and Real Estate / axis_h = Fa / axis_v = Why It Is So
Connecting the Level: The entry point of this framework is three numbers — 6% (profit growth), 13% (loan growth), 1:1 (NPL write-off’s drain ratio on capital). Those who judge banks with the old framework often stop at the intuitive level of “is the profit high?”: seeing absolute net profit in the hundreds of billions feels like a windfall. But plug in this identity — numerator 6% cannot keep up with denominator 13%, CAR inevitably declines, external capital raises are a passively required structural need — and the specific wrong move made by those using the intuitive framework becomes clear: they are comparing absolute profit levels, when the correct analytical interface is the growth-rate gap between numerator and denominator. Going one layer deeper: the triple lock (slow profits + new issuance blocked by P/B below 1x + debt instruments capped by Basel limits) is the true structure of the constraint — not just “profit growth is not high enough.” The other two legs are equally locked, something invisible to anyone who has only read profit headlines.
Relationship with Bank Entity Liquidity Risk: The Two-Dimensional LCR and Asset-Quality Judgment: the LCR framework focuses on the stock-safety-net of liquidity buffers; this framework focuses on the incremental-dynamic gap in capital adequacy — the two are complementary: the former judges “can the bank survive a liquidity stress event,” the latter judges “can the accounting cost of loan growth be continuously absorbed.”
See Also
-
The Two-Tier Banking System: T-Accounts and Loans Create Deposits
-
Bank Entity Liquidity Risk: The Two-Dimensional LCR and Asset-Quality Judgment
-
Three Evidence Chains on Net Interest Margin: Debunking Bank Windfall Profits
Sources
- Compiler’s draft z-0116 · collected 2026-07 (single-session denoised draft, 33 source-transcript anchors, miss=0)
- External course denoised draft: Banking sector lecture 4, data timestamp approximately 2018/2019, NPL resolution data through 2018