← Back to blog

What Hong Kong's housing crisis teaches us about infrastructure failure

Hong Kong has the world's best transit system and some of the world's worst housing affordability. What happens when one kind of quiet infrastructure enables the failure of another.

I have written before about Hong Kong as the clearest proof of concept for quiet infrastructure — the MTR as the world's most operationally reliable transit system, built through fifty years of accumulated decisions rather than singular vision. That argument holds. But there is a version of Hong Kong I did not address in that piece, and it has been sitting with me since.

Hong Kong has the best transit infrastructure in the world and some of the most dysfunctional housing infrastructure in the world. Both things are true at once. The city that taught me the most about what quiet infrastructure can achieve is also the city that taught me the most about what quiet infrastructure, in the wrong hands, can enable.

The numbers

Hong Kong consistently ranks as one of the least affordable housing markets on earth. The median home price relative to median household income has hovered between twenty and twenty-five times earnings for over a decade — a ratio that makes London look accessible and makes any meaningful concept of housing as infrastructure functionally impossible for most residents.

The cage homes. The subdivided flats. The residents of public housing estates waiting fifteen, sometimes twenty years for a unit. The young professionals who have left — not because Hong Kong stopped being a remarkable city, but because the city's housing system stopped functioning as a system and became something else: a mechanism for transferring wealth upward, quietly, without announcement, without ceremony, without a press release.

This is what happens when quiet infrastructure serves the wrong end. The MTR is quiet infrastructure for residents. Hong Kong's land system is quiet infrastructure for landlords and the government entities that control land supply. Both are quiet. Only one functions in the public interest.

The land system

Hong Kong's government owns virtually all the land in the territory. It releases land for development through a leasehold system — developers bid for government land, build on it, sell the units, and the government captures a large portion of the value at the point of land sale. This system funds a significant part of the government's revenue. It has done so for decades.

The incentive this creates is not complicated. A government that funds itself through land sales has a structural interest in keeping land values high. High land values require constrained supply. Constrained supply is achieved by releasing land slowly, by maintaining development restrictions on the substantial proportion of Hong Kong's territory designated as country park, and by allowing the planning system to operate at a pace that ensures scarcity is never seriously addressed.

None of this is announced. There is no press conference at which a government official states that housing will be kept unaffordable to preserve land revenue. The system simply functions — quietly, reliably, day after day — in the interests of those who designed it.

Quiet infrastructure is not inherently good. It is infrastructure that works without requiring your attention. The question is always: works for whom? — from a forthcoming chapter in The Quiet Infrastructure

The MTR connection

There is a direct link between Hong Kong's transit excellence and its housing dysfunction, and it runs through the MTR Corporation's business model.

The MTR is funded substantially through property development. The corporation acquires development rights above and around its stations, builds residential and commercial projects, and uses the profits to cross-subsidise operations. This is the rail-plus-property model, and it is frequently held up as a financing innovation — a way of building world-class transit without requiring ongoing government subsidy.

It is a financing innovation. It is also, structurally, a mechanism that ties the financial health of the transit system to property values remaining high. An MTR that depends on property development profits has an institutional interest in the conditions that make property development profitable. Those conditions include constrained supply. High prices. A housing market in which land is scarce and development rights are valuable.

The transit system that functions as quiet infrastructure for residents is sustained by a financial model that requires the housing system to function as quiet infrastructure for landowners. The two are not in simple conflict. They are interlocked. This is what makes Hong Kong so instructive and so uncomfortable as a case study.

What this means for the quiet infrastructure thesis

When I write about quiet infrastructure, I am describing a disposition toward urban systems: the preference for delivery over announcement, for function over communication, for the accumulated reliability of things that simply work. I use Hong Kong's MTR as my clearest example of what this looks like at scale.

But Hong Kong also demonstrates the limit of the thesis as a normative claim. Quiet infrastructure is not inherently good. It is infrastructure that works without requiring your attention. The question — the question I have not always been clear enough about in my writing — is works for whom.

The land system works. The leasehold mechanism functions with the reliability and invisibility of excellent infrastructure. The government captures land value, funds itself, maintains fiscal stability. The developers who win the land auctions build the units, sell them at prices the system makes possible, and generate the returns the model requires. This is quiet infrastructure. It is also, for the majority of Hong Kong residents who cannot afford to participate in it, a system that works against them — quietly, without announcement, year after year.

This is the version of infrastructure failure that I find most interesting and most difficult to write about. It is not the failure of the Tel Aviv light rail — the thirty-year delay, the missed connection, the promise that took a generation to keep. That failure announces itself. The buses do not come. The trains do not run. The city teaches you, loudly, what it has not built.

Hong Kong's housing failure is the opposite. The system functions. The land auctions happen. The units get built. The prices rise. The cage homes continue to exist alongside towers that sell for twenty million dollars. The failure is silent, structural, and entirely legible to anyone who looks at it — but it requires looking, because the system does not announce its failures. It is quiet all the way down.

The lesson for other cities

I am not drawing a simple lesson from this. Hong Kong is not an argument against the rail-plus-property model, or against government land ownership, or against any particular mechanism in isolation. The mechanisms are not the problem. The problem is the political economy that determines whose interests the mechanisms serve.

What I am arguing is that quiet infrastructure — the disposition I have been advocating in this newsletter and in The Quiet Infrastructure — needs to be understood as a description, not a prescription. Describing how a system works tells you nothing about whether it should work that way. A system can be remarkably effective at transferring wealth upward without anyone noticing. That effectiveness is not a virtue.

The cities I spend most of my time writing about — the mid-size cities between 200,000 and 800,000 people that define my research focus — are almost all dealing with housing questions of their own. Not Hong Kong's question (scarcity manufactured through land supply restriction) but related ones: informal settlements that function outside the formal infrastructure system, housing stock that accumulated during a period of higher population and now serves a declining one, affordability pressures created by limited construction capacity rather than deliberate scarcity.

In all of these cases, the question I keep returning to is the same one Hong Kong forces. Not: does the system work? But: who does the system work for? And: is that the same group the system is supposed to work for?

The answers are rarely the same. The gap between them is where the interesting infrastructure analysis lives.