A systematic debunking of two popular claims — “large bank size = sucking the lifeblood out of the real economy” and “banks account for 41% of economic profits” — using a two-step methodology: Claim 1 stems from misreading a ranking metric (global bank rankings are by Tier 1 capital; bank assets consist primarily of loans, meaning more support for the real economy); Claim 2 stems from A-share sample bias (banks have an outsized weight in A-shares; restoring the 2015 input-output table full-economy definition, the financial sector’s operating surplus is only 20% of the total economy, not 41%).
The Framework As It Stands
This section is organized from the compiled research draft: the original framework’s structure, terminology, and key expressions are preserved, including editorial bridging and externally sourced factual annotations; charts were drawn by the compiler according to the structure of the original text.
Epistemological foundation
This framework treats agnosticism as the philosophical premise of economics: every person grasps only partial, incomplete information; the essence of markets is to allow each person to bring their incomplete information to the table and assemble it with others — the more participants contribute to the assembly, the more complete the picture becomes. No one commands truth across time and space; we are only expressing things closer to reality at a given moment, subject to later revision. This premise explains why the entire framework insists on “returning to the specifics of indicators and data” rather than taking emotional positions. Against the reflexive total dismissal of “your seat determines your head”: your own experience is what you know best, and no one is barred from expressing it; the outcome reached by all participants through collision may not be optimal for each individual but is not the worst (as with market sellers asking prices and buyers bargaining down).
Claim 1: Large bank size does not mean sucking the real economy dry
The popular view infers “banks have sucked the real economy dry” from the facts that “the four largest banks in the world are all Chinese, and many domestic banks that would otherwise be obscure have made the global top-1000 ranking.” This inference is incorrect; the key is understanding the ranking metric. Global bank rankings are by Tier 1 capital, and financial regulators impose minimum Tier 1 capital requirements; larger Tier 1 capital corresponds to larger asset size (since every bank fully deploys its capital to maximize capital efficiency). corporate bonds), and cash reserves. Therefore large bank size precisely indicates that banks are providing more support to the economy, while other financial channels are severely underdeveloped; if other channels were fully developed, banks would not be so large. The essence of the problem is the underdevelopment of other financial channels, not that banks have sucked the real economy dry.
Claim 2: “Banks account for 41% of economic profits” is a statistical illusion
This figure originates from listed bank net profits of RMB 1.5 trillion in 2018, equaling 41% of total listed company profits. This data suffers from severe sample bias: the industry distribution of A-share listed companies is not a perfect mapping of the entire Chinese economy; A-shares more closely resemble a financial-sector index (“if financial stocks don’t rise, the index can’t rise”). Key data for comparison: in 2018, listed banks accounted for 76% of total banking-sector assets (close to 80% of banking assets are inside listed companies); no other industry has such a high listing rate. Listed real-economy companies cover less than 30% of all enterprises; in 2018, listed real-economy company profits equaled only 38% of concurrent national above-scale industrial enterprise profits — large numbers of profitable real-economy enterprises are not among listed companies. Restoring the 2015 input-output table: the financial sector’s enterprise operating surplus is only 20% of total enterprise operating surplus across the whole economy, far from 41% (operating surplus and profit have slight definitional differences but are broadly equivalent). Listed bank profits account for 82% of all bank profits (consistent with their asset share of 78%), while listed industrial enterprise profits account for only 24% of national above-scale industrial profits — precisely because non-bank enterprises’ listed asset coverage is far below that of banks, the statistical illusion of a disproportionately high bank profit share under the listed-company definition arises.
Methodological takeaway
What matters most in analysis is detail and a thorough understanding of data. There is a case of an internet sector leader at the Pujiang Forum who squared off against a bank chairman, asserting directly that “banks take more than 50% of the entire economy’s profits”; the chairman, lacking command of the data details, was left speechless. The same “bank share of profits” proposition yields 41% under the A-share definition and 20% under the full-economy definition. The difference lies not in position but in definitional detail.
Debunking checklist (4 categories):
| # | Category | Checklist question |
|---|---|---|
| 1 | Epistemological | Does the claimant believe they hold “truth across time and space”? Do they refuse to bring incomplete information to the table for assembly? |
| 2 | Ranking metric definition | What ranking metric does “large size” use (Tier 1 capital?)? Does large capital necessarily imply large assets? Are assets primarily loans (supporting the real economy)? |
| 3 | Profit sample definition | Is the denominator for “X% of profits” the whole economy or the A-share listed sample? Is that sample distorted by the high weight of the financial sector? What does the input-output table full-economy definition yield? |
| 4 | Data detail | Does the claim provide only crude figures such as “earning RMB 30 billion per day” or “over 50%“? Are there asset/profit share ratios and listing-coverage rates for cross-verification? |
Key data anchors (A-share profit definition 2018 / input-output table definition 2015):
| Data | Value | Definition / time point |
|---|---|---|
| A-share listed bank net profits / all listed company profits | 41% (RMB 1.5 trillion) | 2018 A-share definition (popular claim) |
| Listed banks / total banking-sector assets | 76% (close to 80%) | 2018 |
| Real-economy enterprise listing coverage | Less than 30% | 2018 |
| Listed real-economy company profits / national above-scale industrial profits | 38% | 2018 |
| Financial sector operating surplus / total enterprise operating surplus | 20% (≠ 41%) | 2015 input-output table (full-economy true value) |
| Listed bank profits / all bank profits | 82% (asset share 78%) | — |
| Listed industrial enterprise profits / national above-scale industrial profits | 24% | — |
| Erroneous claim (case) | “Banks take more than 50% of the entire economy’s profits” | Pujiang Forum debate (debunked) |
Compiler’s Perspective
Coordinates: Category = Banking and Real Estate / axis_h = Fa / axis_v = What It Is
Connecting layer
The belief that “the financial sector’s profit share is too high” is extremely widespread among Chinese investors and is supported by a surface-level compelling number: 41% of A-share profits in 2018. The problem is not that the number itself is false, but that the wrong denominator was chosen — this 41% has as its denominator the total profits of A-share listed companies (a sample with an oversized financial-sector weight), not total enterprise profits across the whole Chinese economy (the input-output table definition).
An analyst who draws a whole-economy conclusion from the A-share definition will make systematic errors in the following specific steps: interpreting “listed banks account for 82% of profits” as “banks are encroaching on real-economy profits,” while simultaneously ignoring the asymmetric fact that real-economy enterprise listing coverage is less than 30%. This is not a question of position but of sample design — A-shares are not a random sample of China’s real economy; they are a biased sample with an oversized financial-sector weight.
The exclusive claim of this framework: between the A-share definition (41%) and the full-economy input-output table definition (20%) there is a systematic 2.05× overstatement, rooted in the structural asymmetry between bank listed-asset coverage (76%) and real-economy enterprise listed-asset coverage (<30%) — the problem is not with profits themselves, but that the denominator has been shrunk. Total Social Financing and Broad Money M2: Accounting Definitions and Divergence also exhibits the mechanism of “choice of definition causing a shift in the meaning of data,” and the two share a structurally analogous pattern at the methodological level.
See Also
-
The Two-Tier Banking System: T-Accounts and Loans Create Deposits
-
Total Social Financing and Broad Money M2: Accounting Definitions and Divergence
-
From Microprudential to Macroprudential: The Post-Crisis Regulatory Paradigm Shift
-
The Theory of Cognitive Algorithms: Integrating Deduction, Induction, and Dialectics
-
Finance Value-Added as a Share of GDP: A Four-Factor Accounting Debunking
Sources
- Compiled draft z-0113 · collected 2026-07
- “External course (collected with identity removed), lecture 1.1: Large bank size — have banks sucked the real economy dry?”