I. Zero-Sum Games: The Starting Point of Economic Distortion

When the vast majority of investment is engaged in zero-sum games over existing assets, the result is not economic development but economic distortion. The essence of the zero-sum game is competing for shares of a relatively fixed “pie”—capital does not flow toward truly creating new value, but rather changes hands repeatedly among existing assets. The gains of participants often come from the losses of other participants, not from the generation of new wealth.

After the 2008 global financial crisis, the world economic system entered an unprecedented era of excess liquidity. Four financial levers—U.S. Federal Reserve Quantitative Easing, China’s Four-Trillion-Yuan Stimulus, Japan’s zero-interest-rate global carry trade, and unanchored cryptocurrencies—together shaped a global structure in which the scale of financial assets far exceeds the real economy. These enormous sums did not flow toward the foundational research and investment that could promote incremental advancement of human society as a whole, but instead engaged in continuous competitive games among existing assets. And real estate—an asset class with a global total value of $393.3 trillion—is the largest gaming table of this zero-sum competition.

II. The Four Financial Levers: The Root of Global Excess Liquidity

2.1 U.S. Federal Reserve Quantitative Easing (QE)

The Fed’s balance sheet expanded from approximately $0.9 trillion at the end of 2007 to a peak of nearly $9 trillion in April 2022 (Federal Reserve data, published January 2025). The Bank for International Settlements (BIS) found in research published in 2017 that QE’s impact on the stock market was ten times its impact on real economic output, suggesting a severe disconnect between valuations and fundamentals. Former Fed dissenter Thomas Hoenig noted that QE produced an “allocation effect,” redirecting capital from real investment on “Main Street” to speculative activity on “Wall Street”—manifested in over $1 trillion in annual share buybacks by corporations, rather than hiring or R&D. By 2015, total buybacks exceeded total capital expenditures for the first time.

2.2 China’s Four-Trillion-Yuan Stimulus

The ¥4 trillion stimulus plan announced by Beijing in November 2008 saw the central government commit only ¥1.22 trillion—roughly 30%. The remainder was borrowed from banks through Local Government Financing Vehicles (LGFVs). In 2009 alone, approximately ¥4.7 trillion in additional bank loans flooded the system, of which LGFVs absorbed about ¥2.3 trillion. After these loans were disbursed to enterprises and local governments across various industries, they converged on the same destination: the real estate market. Hua Min, Director of the Institute of World Economy at Fudan University, commented: “The Chinese economy has been hijacked by the real estate industry.”

2.3 Japan’s Zero-Interest-Rate Global Carry Trade

The Bank of Japan pushed its policy rate to near-zero and even negative territory, making the yen the world’s cheapest funding currency. According to China Merchants Bank Research Institute (December 2024 report), as of the end of 2023, the broad yen carry trade had reached ¥1,420 trillion (approximately $9.3 trillion), equivalent to 236% of Japan’s total economic output. This is the largest hidden leverage pivot in the financial system—no single regulatory body monitors its full picture. When the Bank of Japan raised interest rates by just 0.15% in August 2024, it triggered a 30% crash in Bitcoin within 48 hours, exposing global assets’ deep dependence on this hidden leverage.

2.4 Unanchored Cryptocurrencies

Total cryptocurrency market capitalization reached a historic high of $4.4 trillion in Q4 2025 (CoinGecko, 2025 Annual Report), with CoinGecko tracking over 17,500 cryptocurrencies. This market cap corresponds to no factories, patents, land, cash flows, or physical commodities—its valuation depends entirely on trust in the ecosystem.

2.5 The Compounded Global Effect of Four Layers

From top to bottom, anchoring grows weaker and speculation grows purer. BIS data shows that as of mid-2024, the notional value of global OTC derivatives had exceeded $700 trillion—the “value tokens” circulating in the financial system are more than 7 times humanity’s total real economic output.

III. The Structural Distortion of GDP Measurement

The four levers described above inflated global asset prices. And GDP—the core indicator of global economic health—partially counts this inflation as “growth” on two levels, thereby creating an illusion that the economy is still healthy and providing systemic cover for banks to continue lending at peak valuations.

3.1 NDP vs GDP: The Depreciation Blind Spot

GDP measures total output, including investment spending used to replace worn-out and obsolete equipment. But this spending does not increase the economy’s productive capacity. Available data shows that the average country’s NDP (Net Domestic Product) is approximately 10% to 20% lower than GDP. As investment shifts toward shorter-lived information technology assets, depreciation’s share of GDP has been steadily increasing in most OECD countries.

3.2 The Nature of Paper Wealth

McKinsey Global Institute (published October 2025) reveals: from 2000 to 2024, household wealth increased by approximately $400 trillion, but only about $100 trillion of that was new wealth created through cumulative net investment—three-quarters of the growth came from paper appreciation of assets and inflation. Of the approximately $620 trillion in global real assets as of end-2024, productive asset growth amounted to only 18 points (indexed to 1995 = 100), while financial assets expanded by over 100 points—the financial system’s rate of expansion is 5–6 times the real economy’s capacity for organic growth. Global debt is near its all-time high, reaching 2.6 times GDP (McKinsey, October 2025).

IV. Bank Credit: The Temporal Blind Spot of Collateral Valuation

GDP’s structural overstatement provides banks with systematic misdirection: macro indicators show the economy still growing, so banks’ risk models assess the overall environment as “safe” and continue to lend against collateral valued at peak market prices. This valuation method contains a fundamental temporal blind spot.

4.1 Instantaneous Pricing vs. Temporal Decay Function

When banks lend, their valuation of collateral is a time-slice—assessing “what this asset is worth on the market at this instant.” A loan may have a 30-year term, but the collateral’s valuation is anchored to the market price at the moment the loan is originated. The temporal discount function of collateral value is almost entirely absent from lending decisions.

4.2 The Financial Accelerator: Positive Feedback Spiral

The self-reinforcing feedback between credit creation and financial asset market value—the “financial accelerator”—constitutes a coupled nonlinear process: asset prices rise → collateral valuations increase → banks lend more → more capital chases the same asset class → prices rise further. This positive feedback amplifies booms on the way up and collapses on the way down.

4.3 Building Lifespan ≈ Loan Duration: The Mathematical Contradiction

Former Chinese Vice Minister of Housing and Urban-Rural Development Qiu Baoxing (public statement circa 2010) demonstrated with data: Chinese housing has a useful lifespan of only 25–30 years, less than one-quarter of the UK’s 132-year building lifespan and less than half of the US’s 74 years. China’s maximum mortgage term is also 30 years—by the time the loan is fully repaid, the building has reached the end of its physical life.

4.4 Negative Terminal Value of Collateral

When a building reaches the end of its useful life, not only does building value reach zero, it generates demolition costs, construction waste removal fees, and site remediation expenses. The case of Guangzhou Jiqun Street Building No. 2 (reported March 2024) shows residents paying ¥4,600 per square meter for demolition and reconstruction of their aging building.

V. Real Estate: Humanity’s Largest Financial Bubble Anchor

5.1 A $393 Trillion Global Asset

Savills research data (end-2024 data, published September 2025) shows global real estate at a total value of $393.3 trillion, representing approximately 76% of all global assets and nearly four times global GDP. Residential property alone is valued at approximately $286.9 trillion. China leads at approximately $74 trillion (23.5% of the global total), with the United States at $52 trillion close behind (20.7%). China’s housing market value reached a peak ratio of 414% of GDP in 2020 and fell to 343% by 2023 due to the downturn (Ren Zeping team, October 2024 report), but remains higher than the US’s 155%, Japan’s 300%, and Germany’s 263% during the same period.

5.2 Shelter Value vs. Investment Value: The Essential Definition of a Bubble

Housing is humanity’s most basic and rigid need. This absolute rigidity makes housing a naturally “reassuring” form of collateral for banks—it cannot run, cannot hide, cannot be taken away. Banks are therefore willing to provide the longest terms (30 years) and highest ratios (70–100%) of lending. But it is precisely this rigidity that makes real estate the perfect “parasitic host” for financial leverage. The pricing anchor for shelter value is the present value of future rents—the discounted total of all future years’ rental income for a given property. The pricing anchor for investment value is the game-theoretic bet on “the next buyer’s offer”—essentially a Ponzi structure.

5.3 Four Countries, One Script

Four countries, four time points, one script—shelter value is amplified by financial leverage into investment value, the bubble inflates to unsustainable levels and bursts, leaving behind decades of debt deleveraging.

VI. The Unique Morphology of China’s Real Estate Bubble

6.1 The Distorted Structure of Valuation and Debt

According to estimates by the Ren Zeping team (report published October 2024), total Chinese housing market value in 2023 was ¥432.5 trillion ($63 trillion), down from a 2021 peak of approximately ¥450 trillion. As of mid-2025, approximately ¥120–150 trillion had evaporated (based on multiple estimates). Personal mortgage loan balance of ¥37.68 trillion (PBOC, end-Q4 2024 data, published February 2025) is merely the tip of the iceberg—adding developer loans (¥13.56 trillion, end-Q4 2024), housing provident fund loans (¥7–8 trillion), local hidden debt (¥14.3 trillion as of end-2023, disclosed by the Ministry of Finance in November 2024), and LGFV platform debt (¥40–65 trillion), the total debt directly linked to real estate conservatively exceeds ¥100 trillion.

6.2 Cumulative Sales 2018–2024 and Sunk Household Savings

The ¥37.68 trillion in bank loans is only part of the household wealth locked up by real estate. The larger portion is hidden in sales volumes.

Of the ¥101 trillion in cumulative sales, bank loan balance accounts for only ¥37.68 trillion. The gap of approximately ¥63 trillion represents household savings directly invested in the form of down payments, full cash purchases, parental support, and intergenerational wealth transfers. This ¥63 trillion does not appear on bank loan statements, but it has been locked into real estate that is actively depreciating. When ¥120–150 trillion in market value evaporates, this ¥63 trillion in household savings is consumed first—because in the liquidation waterfall, down payments and equity always rank behind bank loans.

6.3 The “Cold Treatment” Model

China’s political system ensures that there will be no American-style rapid deleveraging (bank failures, asset auctions, government takeovers), nor a Japanese-style passive market digestion. This produces a third form of bubble deflation—no crash, no restructuring, no clearing. Continuous M2 expansion creates a liquidity illusion, banks maintain the appearance of healthy balance sheets through measures such as “¥500/month” loan modifications, and local governments restrict judicial auctions to stabilize surface-level housing prices. Nationwide, the number of mortgage defaulters has reached 1.87 million (up 35.6% YoY), with the default rate surging 130% to 3.7% (according to the China Banking and Insurance Regulatory Commission and CRIC, 2025 data). Nearly 1 million properties are in judicial auction, with a transaction rate below 20%.

The latest 2025 data further confirms the persistence of this pattern: the six major state-owned banks’ combined personal mortgage loan balance was ¥24.48 trillion (end of 2025), down ¥711.5 billion from end-2024—the decline has widened for three consecutive years (¥556.9 billion in 2023, ¥621.1 billion in 2024, ¥711.5 billion in 2025), with the second half accelerating (only ¥107.8 billion reduction in H1, approximately ¥603.7 billion in H2). Concurrently, in the first two months of 2025, total new household loans were only ¥54.7 billion, down 86% from ¥389.4 billion in the same period of 2024. Meanwhile, new household deposits during the same period reached ¥6.13 trillion, up 6.8% YoY, with total household deposits reaching ¥157.38 trillion by the end of February 2025.

Lending is collapsing while deposits are surging—this is not a signal of “economic stabilization” but a classic characteristic of balance sheet recession: households have stopped borrowing to consume and are desperately saving to repay debt.

6.4 The Compounding Hemorrhage Mechanism

The ¥37.68 trillion in outstanding personal mortgages (plus approximately ¥7–8 trillion in housing provident fund loans), calculated on an equal payment basis at an average interest rate of approximately 3.4%, rigidly drains approximately ¥3–4 trillion in cash flow from household consumption capacity each year (estimated based on end-2024 outstanding balance and interest rates).

But ¥3–4 trillion per year is not a “fixed deduction”—it is a compounding, self-reinforcing contraction engine that stacks year after year. The consumption multiplier in economics (typically 1.5–2.5×) means that every ¥1 drained from consumption ultimately reduces GDP by ¥1.5–2.5. Moreover, each year’s hemorrhaging occurs on an economy already weakened by the previous year’s hemorrhaging, amplifying the effect year by year:

By 2025, the compounding effect may bring actual economic losses to approximately 10% of GDP. Yet GDP on the surface still “grows at 5%” because the government is offsetting the damage with fiscal deficits, M2 expansion, and infrastructure investment. Rhodium Group analysts estimate that China’s actual economic growth rate in 2024 may have been only 2.4%–2.8%.

6.5 K-Shaped Divergence: The Rift Between Financial Illusion and Physical Reality

M2 expands at 9.7% (average growth rate from January 2022 to March 2025, PBOC data), while full-year CPI is only 0.2% (2024, NBS published January 2025), and PPI has been negative for 31 consecutive months (as of March 2025). The upper arm of the K is the financial side—M2 is rising, bank balance sheets are being maintained, government bonds are expanding. The lower arm is the real economy—incomes are falling, consumption is shrinking, businesses are losing money, buildings are decaying, but debt remains unchanged to the last penny.

In an environment without inflation, the real burden of debt is not eroded by inflation. If inflation were 5% (normal growth period), the real burden of ¥37.68 trillion would automatically shrink by approximately ¥1.9 trillion per year, falling to approximately ¥23 trillion over ten years. But in a deflationary environment with CPI = 0.2%, this “invisible repayment” is nearly zero—every penny of debt must be repaid with real income. A household with ¥5,000 monthly payments and ¥15,000 monthly income in 2021 had a debt-to-income ratio of 33%; by 2025, with income cut to ¥10,000, that ratio becomes 50%—the monthly payment hasn’t changed by a single yuan, but the burden has risen from one-third to one-half.

6.6 Housing Replacement: The Overlooked Intergenerational Debt Spiral

Nearly all economic models ignore a structural debt inflation mechanism—housing replacement. The physical deterioration of buildings after 25–30 years is not an investment choice but a survival necessity: roofs leak, elevators fail, insulation peels, structural safety grades drop, forcing families to replace their housing after 25–30 years. This is not market behavior but survival behavior—yet every replacement cycle creates new, larger debt.

A typical cycle: buy a ¥1 million home in 2010 (¥700,000 mortgage) → by 2035 (25 years later), total payments of approximately ¥1.5 million (¥400,000 principal, ¥1.1 million interest), remaining loan approximately ¥300,000 → building must be replaced due to physical deterioration → sell old home (may only fetch ¥500,000 after discount) → after repaying ¥300,000 remaining mortgage, ¥200,000 in hand → new home costs ¥2 million (larger, better quality) → ¥600,000 down payment (¥400,000 gap must come from savings) → new mortgage of ¥1.4 million, another 30 years of payments until 2065.

The debt mathematics of each replacement: old mortgage ¥700,000 → new mortgage ¥1.4 million (doubled); interest timer resets to zero (starting again from the most interest-heavy phase); household savings wiped out once more to cover the down payment gap. This process requires no speculative behavior—it is driven entirely by the physical deterioration of buildings. A person needs to replace their housing at least 1–2 times in a lifetime, each time triggering a debt reset and amplification. When compounded across generations, this creates a debt treadmill that generation after generation can never outrun—especially in a deflationary environment where inflation does not erode debt.

VII. The American Model: Rapid Crash, Securitization Contagion, and QE Life Support

China’s “cold treatment” represents the third type of real estate bubble deleveraging model in human history—distinct from both America’s rapid crash and Japan’s slow digestion. To understand the costs and trajectory of this unique model, we must first examine the two precedents that have already run their full course.

7.1 The Mechanics of Subprime: Zero Down Payment + Securitization + Leverage Stacking

After the dot-com bubble and 9/11, the Fed cut rates 13 consecutive times to 1%. Low interest rates fueled a real estate frenzy, with financial institutions rolling out “zero down payment,” “no income verification,” and “interest-only” products. New subprime loan originations surged from $120 billion in 2001 to $600 billion in 2006, rising from 5% to 20% of total originations (Wuhan University Institute of International Studies review; Wikipedia Subprime Crisis entry). By 2008, government-sponsored enterprises held approximately $5.1 trillion in total mortgage debt (Baidu Baike “US Subprime Crisis” entry, citing primary sources).

But the most distinctive feature of the US subprime crisis was not the loans themselves, but the three-layer securitization stack on top of them: subprime loans were packaged into MBS (Mortgage-Backed Securities) → the harder-to-sell tranches of MBS were repackaged into CDOs (Collateralized Debt Obligations) → CDOs were then insured and speculated upon through CDS (Credit Default Swaps). By 2008, on a base of $1.5 trillion in subprime loans, Wall Street had created over $2 trillion in MBS and over $10 trillion in CDOs. The total debt covered by CDS was estimated at $33 to $47 trillion (Baidu Baike citing Marxism Studies, Issue 11, 2008). Leverage ratios reached 1:30 and higher.

7.2 The Scale of the Crash

Starting in 2004, the Fed raised rates 17 consecutive times, and home prices fell 20–35% from their 2006 peak. Between 2007 and November 2008, American households lost more than one-quarter of their net worth. Housing equity fell from a 2006 peak of $13 trillion to $8.8 trillion by mid-2008. Total retirement assets declined 22%. Savings and investment assets lost $1.2 trillion. Pension assets lost $1.3 trillion. Total losses amounted to approximately $8.3 trillion (Wikipedia Subprime Crisis entry, based on Federal Reserve and BEA data). The IMF estimated that direct global financial losses could reach $945 billion. Lehman Brothers went bankrupt, Bear Stearns and Merrill Lynch were acquired, Fannie Mae and Freddie Mac were placed under government conservatorship, and AIG was in distress.

7.3 Deleveraging Model: Government Takeover of Leverage + Unlimited QE Life Support

The US chose the “rapid crash → government bailout → QE backfill” model. A $700 billion emergency rescue plan (TARP), Fed balance sheet expansion from $0.9 trillion to $4.5 trillion (QE1–QE3), and zero interest rates maintained for 7 years (2008–2015). In essence, the US government used taxpayer money to take over financial institutions’ toxic assets, then injected massive liquidity into the financial system through QE. The household sector’s leverage was transferred to the government sector—US national debt expanded from approximately $10 trillion in 2008 to over $36 trillion by 2025 (US Treasury data).

This model’s “advantage” is speed—from crash to housing price bottom took approximately 5 years (2006–2011), with full recovery in approximately 7–8 years. But the costs include: dramatic government debt expansion, QE creating a new round of asset bubbles (the “golden decade” of US equities), further widening wealth inequality, and deep dependence on low interest rates and money printing—as the need for even larger QE during the 2020 pandemic proved.

VIII. The Japanese Model: 30 Years of Slow Hemorrhaging

8.1 Formation and Burst of the Bubble

In the 1980s, following the Plaza Accord’s sharp appreciation of the yen, the Bank of Japan dramatically cut interest rates to stimulate the economy, fueling a nationwide asset bubble. Banks lent at 100% or even above appraised values on real estate loans to both businesses and individuals. At its peak, the land beneath the Imperial Palace gardens in Tokyo was said to be worth as much as the entire state of California. From Tokyo to Osaka to Nagoya to small provincial cities, everyone was borrowing to buy property and speculate on land—total participation, total geographic coverage, total leverage. On December 29, 1989, the Nikkei index touched its historic peak of 38,957, and the bubble began to burst.

8.2 The Mechanism of Balance Sheet Recession

Japanese economist Richard Koo defined Japan’s prolonged post-bubble stagnation as a “balance sheet recession”: when asset prices collapse but debt remains unchanged, the primary objective of businesses and households shifts from “profit maximization” to “debt minimization.” The corporate sector shifted from being a borrower on the scale of 12% of GDP to a net saver of 11%—this 23-percentage-point swing plunged the economy into a deflationary spiral. Even with interest rates at zero, businesses would not borrow because they were busy repaying debt, not expanding.

Japan chose the “no crash, slow digestion” model. Banks did not fail on a large scale (with covert government support), zombie companies survived for extended periods, and non-performing loans took over a decade to gradually write off. The cost was staggering: from 1995 to 2025, Japan’s nominal GDP fell from $5.55 trillion to $4.27 trillion, and its global share dropped from 17.8% to 3.6%. An entire generation’s wealth and youth was consumed by debt repayment and deflation.

8.3 The Return After 30 Years

Outside of Tokyo and other hotspots, Japan’s housing prices have essentially returned to shelter-use value—millions of vacant homes, rental yields of 4.7–5.6% (a reasonable range), and a population broadly averse to leverage. However, Tokyo, due to yen carry trade capital repatriation, sees foreign investors accounting for approximately 40% of new apartment transactions (E-Housing, October 2025 report), forming a new localized bubble. Bubbles can transform from “whole-nation bubbles” to “top-tier-only bubbles,” but as long as the four global leverage structures remain operational, financialization’s erosion of shelter value will never fully cease.

IX. Comparison of Three Models and China’s Choice

The core difference among the three models lies in who bears the cost of the bubble. Under the American model, financial institutions and the government (taxpayers) bore most of the losses, and homebuyers suffered “limited losses” due to zero down payments. Under the Japanese model, corporations bore the primary deleveraging pressure, with households bearing the burden indirectly through wage stagnation amid deflation. Under the Chinese model, households directly bear the greatest losses—¥63 trillion in sunk down payments/savings + ¥37.68 trillion in rigid monthly payments + continually shrinking property values—and there is no personal bankruptcy system to provide an exit.

China entered this cycle at its starting point in 2021. Projecting from the last cohort of peak buyers (entering in 2021 with 30-year mortgages), debt deleveraging will continue until 2051. And China faces more difficult conditions than Japan: the primary debtors are households rather than corporations, there is no personal bankruptcy system, building quality is worse (25–30 year lifespan vs. 74 years in the US), and the population is already declining. The only cushions are an urbanization rate (67%) with remaining headroom and a strong manufacturing base.

X. Conclusion: From Zero-Sum Games to Incremental Creation

The complete chain of argumentation in this paper forms a six-layer causal derivation, validated through the comparative analysis of deleveraging models in the US, Japan, and China:

Layer One (Macro-Financial): Four levers inject enormous liquidity globally → the $1,700 trillion global balance sheet inflates to 17× GDP.

Layer Two (Measurement Distortion): GDP overstates by 10–20% by not deducting depreciation; three-quarters of wealth growth is paper appreciation → creates an illusion of economic health → provides systemic cover for banks to keep lending.

Layer Three (Credit Transmission): Banks lend against instantaneous prices without modeling temporal decay of collateral → pro-cyclical positive feedback → credit scale decouples from real value.

Layer Four (Physical Reality): Building lifespan of 25–30 years ≈ loan duration → collateral approaches zero value by repayment → demolition costs push terminal value negative → replacement demand creates a new, larger round of debt.

Layer Five (Bubble Anchor): Housing rigidity + financial leverage = the world’s largest single-asset bubble ($393.3 trillion) → the portion of price that deviates from shelter value is 100% financial bubble.

Layer Six (Economic Hemorrhage): 30-year rigid debt in a deflationary environment = a monthly auto-executing hemorrhage mechanism → ¥3–4 trillion annually drained from the consumption cycle → compounding effects deepen year by year → K-shaped divergence → more households enter a pure-liability phase → consumption contracts further → the spiral deepens.

These six layers form a closed loop returning to the starting point of Chapter One: real estate is the largest vehicle of zero-sum games. Over $300 trillion globally is locked into this asset class that creates no incremental value, with over ¥100 trillion of Chinese household wealth alone sunk within it. If this capital and talent were liberated and redirected toward controlled nuclear fusion, quantum computing, gene therapy, fundamental scientific research—directions that truly create civilizational increments—the face of the human economy would be entirely different.

But liberation requires debt deleveraging. The question is not whether to deleverage, but whether to do so proactively and in an orderly fashion, or to passively wait 30 years for chronic hemorrhaging to complete the process naturally. Japan spent 30 years proving the cost of the latter path—nominal GDP regressed, global share fell from 18% to 4%, and an entire generation’s wealth and youth was consumed in debt repayment. China stands at the beginning of the same road, and time does not favor those who avoid deleveraging.