Author: Nebula Walker Date: 26AUG2026 MYTHOGEN ENGINE (mythogenengine.com)

📌 The cheapest goods in history are produced right next door, yet in Hong Kong, Shenzhen, inland China, and Taiwan, no one captured that cheapness. Where did it go?

Four Places Where No One Captured the Cheapness

1. A Contradiction Every Hong Kong Resident Knows

The world's factory sits just across the Shenzhen River, yet nothing in Hong Kong is cheap — not clothing, not food, not housing, not transport.

That alone is strange enough. In human history, there are probably very few instances where a place sat flush against the world's largest source of cheap manufactured goods while its residents' cost of living ranked among the world's highest.

Even stranger: the moment you log onto Taobao, things suddenly become cheap.

The same product costing several hundred HKD in a local retail channel costs a fraction of that when ordered directly and consolidated for shipping. This gap isn't caused by consumers lacking savvy — it exists every single day, everyone knows it, and it is absurdly large.

This gap is the entry point of this essay.

Because it proves one thing: the cheapness exists. It did not vanish at the factory gate, nor at the border — it traveled all the way to your doorstep, only to be intercepted in the final mile.

Intercepted by whom? Shop rents, import markups, and distribution layer upon layer. The savings never reached consumers; they became landlords' rental income. Hong Kong residents shopping on Taobao are essentially bypassing local land rent to pipe directly into the production source.

So the question can be asked on a larger scale: For twenty years, the world generated the cheapest flood of goods in human history. Into whose pockets did that cheapness ultimately flow?

I examined four places — Hong Kong, Shenzhen, inland China, and Taiwan.

The conclusion is ugly: No one in any of them actually captured it.

2. A Three-Minute Recap: Where Did Those Cheap Goods Come From

This section is entirely other people's work; I'm merely compressing it. The expanded version is in a separate essay (〈Your Fixed Deposits Subsidized German Car Buyers〉); those already familiar can skip to Section 3.

The source of cheap goods is not as simple as "government cash subsidies."

China's fiscal capacity actually shrank over time — economist Huang Yiping pointed out that fiscal revenue as a share of GDP fell from 36% in 1978 to 11% in 1996. Many local governments could barely cover civil-servant payrolls, let alone directly subsidize enterprises.

So the state rerouted through the financial system: suppressing formal-sector interest rates to extremely low levels and channeling credit allocation toward designated enterprises. In Huang Yiping's words, this was effectively a disguised subsidy.

A necessary qualifier, or anyone who has done business in China will demolish the argument in one sentence: Cheap capital was never universally available. Private SMEs face persistently high borrowing costs. Those who actually obtained near-zero-cost capital were SOEs, local government platforms, and policy-favored industries — the entities least required to deliver returns.

Who pays? Two channels. Deposit rates suppressed at or below inflation, and labor compensation chronically lagging output value. The same people, simultaneously as savers and workers, pay for that capacity. Michael Pettis has spent over a decade building the argument that "income is systematically transferred from households to the production sector."

And the capacity expansion can't be stopped, because both ends push simultaneously.

On the government end: after the 1994 tax-sharing reform, local fiscal revenues were largely siphoned centrally, leaving industrial attraction and land sales as the sole revenue lifeline. Layered atop that was the GDP-and-investment-based cadre promotion tournament (classic analysis from Zhou Li-An). Every province had incentive to replicate the same industry on its own territory.

The household end pushed even harder: People need jobs. Tens of millions of graduates each year need somewhere to go. Society can't supply that many positions, so people find their own way — crowding into whatever looks profitable, trying every industry. What they produce must be sold; if quality can't compete, the only lever is cheapness; to recoup costs quickly, the only path is dumping. Rock-bottom prices are grown this way, and local governments — seeing someone creating employment on their behalf — naturally support it.

So this is not a chess game. Nobody is playing chess. Thirty provinces each acting rationally, summing to absurdity.

On the other side, the United States processed the same surplus of money in the opposite way.

America's pathology is called financialization: capital returns inflated so high that companies profit without investing, money flowing toward buybacks and asset prices. China's side has capital costs suppressed so low that companies keep investing despite losses. Same lever, opposite directions; shared result: return on capital decoupled from investment decisions.

And the two sides interlocked into a closed loop: China's surplus flowed back into US dollar assets, suppressing American long-term rates (Greenspan's long-rate conundrum and Bernanke's "global savings glut" both describe this step); US debt-fueled consumption then absorbed China's excess capacity. Some call it Bretton Woods II; others, Chimerica.

The most critical part is the offset. Financialization pressed down the income share of American workers — this should have immediately detonated politically. But simultaneously, capacity-ization pressed down goods prices, so wages didn't rise but things got cheaper, and the surface standard of living held.

Two regimes each parked their most dangerous bomb in the other's house.

Up to here, all is existing literature. The real question begins in the next section: The dividend of suppressed goods prices — whose hands did it actually land in?

3. Hong Kong: First Dismantle the Ceiling, Then Complain the Roof Is Too High

Hong Kong's currency is pegged to the US dollar, meaning it surrenders its own monetary policy and directly imports American interest rates. The near-zero-rate era from 2009 to 2021 was absorbed wholesale — but Hong Kong received only the cheap capital, not American wage momentum, and not American-style land supply.

Cheap capital + a fiscal system sustained by land sales = one-way asset-price escalation. The goods side should have clawed back some benefit, but as noted, the dividend was intercepted at the last mile by land rent.

But if the story stopped here, it would fall into the most common explanation: Hong Kong is small, supply is constrained, therefore expensive.

This explanation has a chronological flaw: Hong Kong's land has always been scarce. Why was the post-handover period specifically the most extreme?

The answer is not on the supply side. It is that the ceiling was dismantled.

Public housing, subsidized retail premises, rent controls — the real function of these things was never merely to care for the poor. They propped up a non-commodified segment outside the market, and private rents had to compete against it. That was a counterparty; a cap.

The rent-control statutes themselves were an explicitly written ceiling: before abolition, permitted rent was capped at 90% of market rent, with biennial increases limited to 30%. Rent controls were abolished in 1998; security of tenure was abolished in 2004.

The dismantling proceeded item by item, a point worth spelling out since it was not a single act of privatization:

• 1998: Rent controls abolished
• November 2002: The "Nine Measures" immediately ended the Home Ownership Scheme, mixed development, private-participation subsidized housing, and Housing Society assisted self-purchase schemes; regular land sales replaced by application-triggered land sales
• 2004: Security of tenure abolished
• 2005: Tenant Purchase Scheme terminated; the Housing Authority packaged its shopping malls and car parks into Link REIT for public listing
• Public housing construction halted until 2011 before resumption was announced; new units were not available for sale until late 2014

Note that rental public housing continued throughout; the population residing in various forms of public housing still exceeds 40%. What was dismantled was not public housing itself, but four distinct price counterparties: the subsidized home-ownership ladder, publicly owned retail premises, rent controls, and a stable rhythm of land supply.

And the asset-disposal step was itself the financialization described in Section 2.

In accounting terms, it simultaneously did two things: converted a recurring income stream into a one-time capital gain, and removed the counterparty that had been holding prices down. As for public retail properties being packaged into a publicly listed REIT — that is the literal meaning of "converting physical assets into tradable financial claims," requiring no metaphor.

So Hong Kong was not merely affected by someone else's financialization. Hong Kong financialized its own public assets with its own hands.

Half-credit is due: the Nine Measures were not unprovoked. At the time, property prices had fallen roughly 70% from their 1997 peak, 100,000 small homeowners were in negative equity, and the government faced real pressure.

But the crisis ended after 2003, and none of those tools came back. The real mistake was not the emergency prescription but writing the emergency prescription into the long-term formulary.

As an aside, when the UK sold council housing to tenants, it didn't save money either: the government previously collected rent; afterwards it paid housing benefits to private landlords. The same money shifted from a revenue line to an expense line, with a rent-collector added in between. So "others did the same" was never justification — it merely shows the mistake was fashionable.

One last item this essay does not address must be noted.

During the same period, Hong Kong's demographic composition changed dramatically — large outflows and large inflows. This has substantive impact on the scale and composition of housing demand, and on the political base for "who would demand the ceiling be reinstalled." Properly addressing it requires another essay; this one will not expand on it.

But one point is unaffected by demographics: In a market without a counterparty, regardless of whether population rises or falls, rents have nothing to benchmark against. Demographics can explain the level of demand; they cannot explain why the cap disappeared.

And the offset itself has a natural ceiling.

Cheap goods can only offset the "buying things" portion of household costs. Hong Kong household expenditure skews toward housing, dining, services, education, and healthcare — precisely the things China cannot manufacture cheaply and that asset inflation makes ever more expensive.

The money you pay for a bowl of wonton noodles is mostly not the cost of noodles and shrimp — it's the rent on that shop unit. Nothing China can make cheap fits into that bowl.

Hong Kong is the only quadrant that loses on both fronts: the goods dividend intercepted by land rent, and the wall that had held back land rent — dismantled by its own hand.

4. Shenzhen: Producing the World's Cheapest Goods While Being One of the Country's Most Expensive Cities

This quadrant is where the problem is laid completely bare.

Shenzhen's property prices, shop rents, dining, and service costs all rank among China's highest. It produces the world's cheapest manufactured goods, yet is absurdly expensive itself.

The cause shares a root with Hong Kong: capacity-ization suppresses goods prices while inflating land prices at the point of production. Factories arrived, people arrived, supply chains rose — all of it capitalized into land; and local government is precisely the entity that monetized this process through land sales.

So a worker in Shenzhen is extracted three times over:

1. Wages below output value
2. Deposit rates suppressed
3. A large portion of remaining income surrendered to land rent

The first two are the Section 2 mechanism; the third is only visible from this quadrant.

The cheap goods he personally manufactured — not a cent of the dividend stayed in his own city.

A historical irony as an aside: mainland China's land-finance model was not homegrown — it was learned from Hong Kong by Shenzhen in the 1980s. So today's picture becomes: Hong Kong exported land finance to the mainland; the mainland exported cheap goods back to Hong Kong; and Hong Kong residents bear costs from both sides — full payment on the land-rent side, self-service on the goods side.

5. The Inland: Cheapness Follows Poverty

So where is it actually cheap? Go further inland.

But those places are cheap because there is no land-value premium, no service-sector demand, and no wages.

This dismantles a common defense: "At least overcapacity means people can afford things." Affordability requires being in a place where land rent hasn't yet caught up — which is precisely the place excluded from the entire growth process.

Cheapness is not a characteristic of the dividend zone; it is a characteristic of the exclusion zone.

Scale this up to the international level, and you get the "ladder" argument.

The popular narrative: developing nations have only one path to industrialization — start with shoes, garments, and toys; earn foreign exchange; cultivate a first generation of industrial workers; then climb upward. Japan, Taiwan, and South Korea walked this path; China itself did too. Now China has reached the upper floors but refuses to lower the ladder — competing with Germany in autos while simultaneously competing with Bangladesh in garments.

It sounds morally weighty. But it contains a hidden premise, and that premise is false: No one ever "lowered" the ladder.

Taiwan ceded garments, South Korea ceded footwear, Japan ceded textiles — none of these was generosity. Every case was wages rising until the business became unviable, forced out by costs. In all of history, no country has ever voluntarily exited an industry tier while it was still profitable.

The so-called ladder was actually an unwilling, price-driven yielding of position.

What China disrupted, therefore, is not anyone's goodwill but the mechanism itself. When the cost of capital is suppressed to near zero, rising wages no longer constitute a reason to exit — losses can be sustained on even cheaper capital, held out until the competitor dies first.

The ladder wasn't taken away by anyone — the law of physics holding the ladder up was switched off.

Framed this way, the argument need not appeal to morality, only point to price-mechanism failure. And the ladder is being dismantled from the other end too: economist Dani Rodrik's "premature deindustrialization" shows that developing nations are seeing manufacturing's share of employment decline at far lower income levels than Taiwan or South Korea achieved — because the same output requires ever fewer workers.

The ultimate final competitor is not China but the machine.

The same logic applies within China: using industry to absorb total employment, in an era where per-capita output has long surpassed per-capita need, is mathematically untenable. Developed nations once used the tertiary sector to absorb these people, and China didn't take that path — for reasons tracing back to Section 2: for services to flourish, people must first have money to consume, and consumer spending power is precisely what was suppressed in exchange for capacity.

Refusing to raise wages means domestic demand can't be sustained; unsustainable domestic demand means continuing to rely on exports to absorb capacity.

6. Taiwan: Far Looser Supply Constraints, Same Result

Taiwan hasn't had its ladder pulled away, because it stands on the upper rungs — wafer fabrication remains unconquered, so it isn't the one being squeezed out but a mandatory node on the supply-chain bloodline.

On the goods side, Taiwan also benefits: Chinese cheap goods flow in freely, without Hong Kong's problem of local land rent intercepting them.

In theory, Taiwan should be the only quadrant winning on both fronts.

But housing prices are equally disconnected from income.

And this is precisely its value in this comparison. Hong Kong's high property prices can be attributed to artificially constrained land supply. Taiwan's supply constraints are visibly weaker — not absent (the Taipei basin and science park peripheries have geographic and regulatory limits), but nothing resembling Hong Kong's monopolistic structure — yet the outcome is similar.

When the supply variable is substantially relaxed and the outcome doesn't change, the remaining explanation can only fall on the demand and holding side: Surplus capital with no outlet, plus holding costs near zero.

This is not a rigorous controlled experiment, but it suffices to strip the common explanation "land rent stems from land scarcity" of its exclusivity.

Taiwan has already made a move on this quadrant. The Hoarding Tax 2.0 passed its third reading in late 2023, took effect in July 2024, and was first levied in May 2025 — switching to nationwide aggregate accounting with non-owner-occupied rates raised to 2%–4.8%. The effectiveness is hotly debated with substantial arguments on both sides: one camp argues that under expected capital gains and tax-burden pass-through, forcing vacant units onto the market to lower prices and rents through higher holding costs may be wishful thinking by policymakers, lacking sufficient empirical support; the other camp notes that price growth in some regions has genuinely slowed and transaction volumes show a wait-and-see pattern, but emphasizes this reflects buyer-seller attitudes, not structural price correction.

This is the only quadrant among the four that has already acted, with the verdict still pending.

One more point must be stated clearly: the upstream itself is also being squeezed.

Technical rent is not captured by the entire supply chain but by a tiny number of links at a tiny number of companies. Hundreds of equipment, materials, packaging-and-testing, and design firms below all face a single anchor customer — so Taiwan has grown its own version of involution: not dozens of provinces competing for one consumer market, but hundreds of suppliers competing for one purchase order.

The mainland version is soft budget constraints plus decentralized entry, producing overcapacity; Taiwan's version is low bargaining power plus dispersed competition, producing margins stripped by the anchor customer. Different engines; structurally isomorphic.

Wage dividends are not evenly distributed either. Bonuses concentrate at a handful of companies in a handful of job grades; the rest mainly receive work hours. And that cash flow earned through time is immediately consumed by property prices around the science parks — the more successful the industry, the faster local property prices rise.

Proportionality must be maintained: in absolute terms, these remain the island's highest salary bands. The accurate statement is: absolute numbers are high, but relative to the hours invested and the share extracted by land rent, they are compressed flat. High pay and being squeezed can coexist.

This also explains why "if you're unhappy, just switch jobs" doesn't work here: changing jobs can change the employer but not the buyer structure, and not the housing prices.

Taiwan's dilemma is harder to untangle than Hong Kong's. Hong Kong needs to persuade a minority of asset holders — and that minority refuses. Taiwan's home ownership rate is high; the majority of voters are themselves asset holders, meaning curbing property prices requires the majority to voluntarily devalue their own assets. The former is more hopeless; the latter more embarrassing.

7. All Four Quadrants Laid Out: No One Captured It

	Goods	Land Rent	Net Outcome
Hong Kong	Expensive (unless self-importing via Taobao)	Extreme	Loses on both fronts
Shenzhen	Cheap	Extreme	Land rent consumes the dividend
Inland	Cheap	Cheap	But income is also low
Taiwan	Cheap (free inflow)	Extreme (near-zero holding cost)	Upstream squeezed by anchor customer on margins and by work hours on quality of life; everyone else bears the same land rent without that paycheck

All four places sit within the same system, and not a single quadrant captures the benefit of cheapness.

So someone must be capturing it, right? On the surface, it's consumers in the US and Europe who have been buying things for the past twenty years.

But they were simultaneously stripped of even more by asset inflation on their own side. And since a significant portion of wages' purchasing power was propped up by suppressed goods prices, the real purchasing power of Western households over these twenty years was actually lower than it appeared — the gap was filled by the suppressed wages and deposit interest of people on the other side.

What was called "middle-class living standards were maintained" was partly not earned by the middle class itself but subsidized by others. And those who subsidized it didn't capture anything either.

The result is a peculiar situation: Everyone is losing, yet no winner can be found.

8. All Roads Lead to Land Rent

Let me be clear about what I am and am not arguing.

I am not saying land rent caused all of this. The causes across the four places are entirely different: Hong Kong is dismantled non-market sectors plus the currency peg; Taiwan is near-zero holding costs; Shenzhen is land finance; the US is zoning regulation plus prolonged low rates. Collapsing them into a single cause commits the same error this essay's Section 2 criticized — attributing dispersed, multi-causal phenomena to a single villain, just with a different villain.

What I argue is the direction of flow, not causation: regardless of starting point, surplus ultimately flows into the same pocket.

The US turned surplus capital into housing prices; China turned surplus capacity into land values; Hong Kong got both at once; Taiwan's story is that the more successful the industry, the faster local property prices rise. The dividend of productivity, regardless of which side generates it and however far it detours, ultimately gets capitalized into land and collected by those who hold land.

This observation is itself not new — Henry George wrote in 1879 about why progress marches alongside poverty. What is new is merely that this time it has two upstream tributaries: one is asset inflation, one is overcapacity, and they were previously regarded as two opposing phenomena.

But this also indicates the direction of a solution, and the direction runs counter to intuition.

If the problem is that land rent has no counterparty, then the prescription is not "build more houses" — newly built houses are sold at market prices anyway, merely handing new supply to the same pricing mechanism. What is genuinely effective is rebuilding a segment outside the market: one that prices independently of market value, does not resell at market value, and is large enough that private rents must compete against it.

Hong Kong once had that segment and then sold it. Taiwan never built one — which is why near-zero holding costs are so lethal.

• Hong Kong: Once had a ceiling; dismantled it
• Taiwan: Never built a ceiling

Two different paths, one identical outcome: Nothing stands in land rent's way.

9. How Does This End

The closed loop is being severed. The US tariff playbook is already set; the EU's is not — how much Chinese capacity the world can still absorb is largely written in Brussels's forthcoming tariff and subsidy investigations. Germany has long been the most committed beneficiary and defender of trade with China; when the largest stakeholder itself begins losing money, the wall inside the EU that had been holding back protectionism will start to crack.

What happens after the severing is more accurately called not hemorrhage but withdrawal.

For twenty years, the West relied on Chinese goods prices to suppress inflation — that was not an external supplement but something that had grown into the metabolism itself. Once withdrawn, prices return to where they belong — and people will experience that process as "inflation out of control," will search for culprits, yet will struggle to accept the real answer: their wages' purchasing power was never actually that high.

China's withdrawal will be equally painful: without external buyers, it must for the first time face alone the capacity it created.

As for "when will China blow up" — that question has produced no answer for twenty years, because the system's capability lies not in avoiding losses but in slicing them thin and distributing them to those with no bargaining power: depositors receive below-market interest; suppliers eat extended payment terms; local finances eat rollovers.

A more useful question: Who is left to absorb the losses? When is that list exhausted? This question has observable proxy indicators — actual payment terms from automakers to suppliers, default rates on commercial acceptance bills, LGFV bond rollover ratios and refinancing costs, the direction of household deposit migration.

Watching these is more reliable than watching any official statement.

And for Taiwan, the truly crucial metric to watch is not housing prices but how much of the upstream technical rent's future growth has already been pre-consumed by property prices. The day those questions all come due simultaneously — that trigger sits on the semiconductor cycle chart, not in housing market data.

10. Conclusion

Someone once said a very clean line: Surplus is not abundance. Abundance means you can use it all; surplus means you cannot. Abundance is wealth; surplus is waste.

We spent twenty years pushing two economies' greatest surpluses — one of capital, one of capacity — to their limits, and produced the cheapest flood of goods in human history.

Then the person sewing garments in Shenzhen, the person paying exorbitant rent in Hong Kong, the person working overtime deep into the night in Hsinchu, and the person who lost their factory in Detroit — none of them actually captured that cheapness.

Their commonality is not nationality but that none of them sits on the rent-collecting side.

And the most embarrassing point is this: Hong Kong once had a wall standing in between, and dismantled it itself; Taiwan's wall was never built in the first place. So when someone says "this is a global structural problem and we are powerless" — part of that is true, and part of it is simply an unwillingness to admit it was a choice.

Notes: What This Essay Owes

The easiest mistake in commentary is presenting what you've read as your own discovery. So let's settle accounts.

Not mine:

• China using suppressed interest rates and biased allocation to transfer household income to the production sector — Michael Pettis has argued this for over a decade; Huang Yiping provided the fiscal-capacity-decline causal thread
• The tax-sharing system and promotion tournaments as generators of redundant investment — Zhou Li-An
• The US-China savings–surplus–low-interest-rate closed loop — Bernanke's "global savings glut," Greenspan's long-rate conundrum, and subsequent Bretton Woods II and Chimerica formulations
• Premature deindustrialization — Dani Rodrik
• Progress marching alongside poverty, surplus ultimately flowing to land rent — Henry George, 1879
• Overcapacity as a global systemic risk, and the "Absolute Advantage Economy" formulation — Michael Froman in Foreign Affairs, September/October 2026 (published online August 13), concept borrowed from Yasheng Huang
• Employment as a hard constraint, "using industry to absorb total employment is a dead end," the "withdrawal" metaphor, and the first challenge to "kicking away the ladder" — from anonymous participants in a public discussion thread who reached key points earlier than most credentialed commentators

Mine:

• The four-quadrant comparison of Hong Kong, Shenzhen, inland China, and Taiwan, and the conclusion derived from it: no one in any quadrant captured the cheapness
• Using the Taobao price gap from everyday experience to prove the dividend exists but is intercepted at the last mile
• Taiwan as a quadrant where supply constraints are substantially relaxed yet the outcome is identical, used to weaken the "land rent stems from land scarcity" explanation
• Taiwan's upstream monopsony structure as a structurally isomorphic variant of involution
• Rewriting the explanation of Hong Kong property prices from "insufficient supply" to "non-market sectors dismantled item by item," and identifying the disposal of public assets as itself an act of financialization
• The ladder was never lowered — it was toppled by costs: replacing a moral narrative with a price-mechanism narrative
• And assembling all the above components into a single machine

Whether assembly counts as original is debatable. But if this essay has value, the value lies in the assembly, not the parts.

---

This is Part 2 of a trilogy. Part 1, 〈Your Fixed Deposits Subsidized German Car Buyers〉, addresses how cheap goods were engineered; Part 2 addresses where that cheapness ultimately flowed; Part 3, 〈Capital Won't Leave, People Won't Leave〉, addresses why exit mechanisms failed. Series landing page: 《Exit Mechanism Trilogy: The Cost of Cheapness》.

Capacity and trade data cited from Michael B. G. Froman, "The Next Global Economic Crisis Could Be Made in China: How Overcapacity Ends," Foreign Affairs, September/October 2026; fiscal ratios and financial repression mechanisms from Huang Yiping's public lectures at the National School of Development, Peking University; soft budget constraint from Kornai János.