The Excuses, Tested

Four reasons Americans give for why the housing crisis is nobody's fault, and the four countries that disprove them.

The American housing conversation has a feature you don't find anywhere else in our politics: near-universal agreement that the situation is terrible, paired with near-universal agreement that nothing can be done about it.

That's genuinely unusual. Argue about taxes and everyone has a plan. Argue about health care and everyone has a plan, most of them bad, but they exist. Argue about immigration and you'll get more plans than you wanted. Housing is the one where the conversation reliably terminates in a shrug — where people who agree on nothing else will nod together at the observation that this is just how it is now.

Fatalism is the only genuinely bipartisan position in American housing. It is also, and I don't think this is a coincidence, extremely convenient for the roughly two-thirds of American households sitting on the appreciating asset.

So I'm not going to argue policy this week. I'm going to take the four excuses that do the heavy lifting — the four explanations that let everyone agree nothing can be done — and check each one against a country that ran the experiment. Not a country that's simply nicer to live in. A country that removed the specific variable Americans blame and got a result.

An excuse, after all, is just a policy position that has learned to dress as a weather report.

Excuse one: "It's our mortgage system."

This is the conclusion most people drew from the last installment, and it's the wrong one. If the thirty-year fixed is a federal artifact, if the guarantees and the deductions and the securitization pipeline are all policy choices, then surely the distortion lives there. Strip out the subsidies, price the risk honestly, and the market clears.

The test: Canada.

Canada has no mortgage interest deduction. None. Not reduced, not capped — it doesn't exist. There is no thirty-year fixed; the standard product is a five-year term amortized over twenty-five, which means renewal risk lands on every borrower, repeatedly, for the life of the loan. Most provinces have full recourse, so defaulting doesn't just cost you the house, it exposes everything else you own.

Every single demand-side feature Americans blame for their bubble, Canada lacks.

Toronto and Vancouver are worse than almost anywhere in the United States.

Verdict: mortgage design determines who bears risk and how quickly a rate shock reaches a kitchen table. Those are serious questions — they're distributional questions and financial-stability questions and they decide who gets destroyed in a downturn. They are not affordability questions. The instrument does not set the price of the land underneath the house.

But here's the part that keeps this from being merely a correction, because the natural follow-up is: fine, so subsidies don't matter?

They matter enormously. Just not in the direction people assume.

Look at Britain, which has now made its Mortgage Guarantee Scheme permanent — first-time buyers in at 95% loan-to-value on a 5% deposit, with the government standing behind the rest. That is a state deliberately injecting purchasing power into the hands of buyers in a market where supply cannot respond.

Which is the seller's escalator I described last week, running in a foreign accent. The subsidy functions precisely as designed. The buyer's improved payment capacity gets bid into the sale price, capitalized into the land, captured by the seller at closing, and the entry cost rises for whoever comes next.

Britain has our disease and a worse mortgage. Which at least establishes that the disease isn't the mortgage.

Excuse two: "It's global. This is happening everywhere."

This is the big one, and it deserves the most attention, because it does more political work than the other three combined. The logic runs: if it's happening everywhere, it's structural. If it's structural, it's nobody's fault. If it's nobody's fault, nobody has to vote on anything, and we can all go back to watching our home equity appreciate in peace. Amen.

Start by killing the premise, because it's false. The crisis is not global. It is startlingly, specifically narrow.

Where is housing catastrophically unaffordable relative to local incomes? The United States. The United Kingdom. Canada. Australia. New Zealand. Ireland.

Look at that list for a second. Six countries. Common law legal tradition. Municipal or local control over land use. A national civic ideology that treats homeownership as the definition of arrival.

That is not a global phenomenon, but a family resemblance, and family resemblances have causes. When six countries with the same legal architecture produce the same result, that isn't weather. It's a recipe.

The test: Japan.

Japan is the clean experiment because it changes exactly one variable, and it's the variable nobody in the American conversation wants to discuss.

Zoning in Japan is set nationally. Twelve use categories, defined by the national government, applied across the country. A municipality administers them. A municipality cannot invent a thirteenth category to protect a neighborhood's character, cannot impose a discretionary review process to run out the clock on a project, cannot let a hearing room full of existing residents decide whether a building gets built.

The pen is held somewhere the neighbors can't reach it.

The result: Tokyo routinely permits more housing units in a year than the entire state of California. Real rents in Tokyo went roughly sideways for three decades.

Now, the objection, which I'll handle here rather than in the comments: Japan's population is falling. True, nationally. But Tokyo's population grew across the period in question, and grew substantially — people moved there in enormous numbers, for exactly the reasons people move to San Francisco, Sydney, Toronto, and Dublin. Tokyo absorbed a generation of internal migration and its rents stayed flat.

That's the entire finding. Depopulation cannot explain it, because the depopulation happened in the places people were leaving. The place they were arriving is the place that stayed affordable, and it stayed affordable because it built.

Verdict: the variable isn't finance. It isn't culture. It isn't demographics or density or geography or the mysterious appetites of foreign capital. The variable is whether the people who already live somewhere hold a veto over whether anyone else may.

Six countries share that veto. Six countries have the same crisis. Japan doesn't have the veto and doesn't have the crisis.

Excuse three: "The tax code is what it is. You'd never get it changed."

This one has the best empirical support of the four, right up until June.

The distortion, first. Australia stacks two provisions. Negative gearing lets a property investor deduct rental losses against their wage income — not just against rental income, against salary. The capital gains tax discount, introduced by the Howard government in 1999, then taxes the eventual gain at half rate.

Put them together and you have built a machine for converting salary into concessionally taxed capital gain, funded by leverage, and pointed overwhelmingly at existing dwellings. It adds nothing whatsoever to the housing stock while bidding directly against first-home buyers at Saturday auction, with the tax office effectively subsidising the bid. Add that Australia is a floating-rate market with essentially no long fixes, so every Reserve Bank decision lands in household budgets within a quarter, and you have a market engineered for maximum pain transmission and maximum investor advantage simultaneously.

Now the political history, which is the actual point of this section.

Labor campaigned on almost exactly this reform in 2016. Lost. Campaigned on it again in 2019. Lost, and the reform was widely blamed for the loss — it became the standard cautionary tale, cited by every strategist in the English-speaking world as proof that you cannot touch housing tax preferences and survive.

That is about as close to a demonstration of political impossibility as democratic politics ever produces. Two elections. Two defeats. A settled consensus that the thing could not be done.

In June 2026, Australia did it.

Negative gearing on established residential property is abolished from 1 July 2027, applying to properties purchased after budget night on 12 May 2026, with losses offsettable only against rental income or future property gains. The 50% capital gains discount is replaced with cost base indexation plus a 30% minimum tax on net gains — and not just for property, across all capital gains assets. It passed the Senate on 25 June.

What changed was not the argument. What changed was the price.

Everything held as of budget night is grandfathered completely. Existing investors can negatively gear until they sell, indefinitely. Eligible new builds keep both concessions, so the incentive survives — it just now points at construction instead of at bidding up the existing stock. And the whole thing sits on a fourteen-month fuse.

Read that structure carefully, because it's the least comfortable sentence I'll write in this series:

They bought off every incumbent asset holder, completely, in advance. Not one person who currently owns anything loses a single dollar. The entire cost of the reform falls on people who haven't bought yet — who are, conveniently, not yet a constituency anywhere, and who cannot vote in the jurisdiction that is pricing them out.

Verdict: it is not politically impossible. It is politically expensive, and the price is a total indemnity for everyone who already holds the asset. That is the only kind of housing reform that passes, and the reason it's the only kind tells you exactly who holds the veto.

Which is next week's subject.

One caveat I'd rather supply myself than have supplied for me: Grattan's modelling of the 2019 version of this reform put the effect on prices in the low single digits. This is a fairness and composition reform — it changes who's bidding and how the tax burden falls — far more than it's an affordability cure. Which, if you've been following the argument, is exactly what you'd expect. A tax fix can't fix it, because the binding constraint isn't the tax code.

The results start arriving in 2028. I'll be watching, and so should you. Fingers crossed!

Excuse four: "There's no alternative. Ownership is the only security."

The test: Germany.

German homeownership runs around 47%, the lowest in Western Europe, and it got there on purpose. Down payments of 20 to 30% are standard. There is no mortgage interest deduction for owner-occupiers. Property transfer taxes run 3.5 to 6.5% of the purchase price, which makes the transaction itself expensive enough to think hard about.

Every incentive Americans consider structurally necessary to a functioning housing market: absent.

And German renters are not sitting in a waiting room. Leases are indefinite. Eviction protection is real. Rent increases are tethered to a published local reference rent. Renting in Germany is a tenure — a legitimate, permanent, secure way to house a family — rather than a failure state you're expected to escape.

Which produces a political consequence Americans rarely think about: Germany does not have a mass electoral constituency organized around housing scarcity, because the median German household is not holding two-thirds of its net worth in a single leveraged asset whose value depends on nobody building nearby.

Denmark, briefly, because it's the elegant one and nobody copies it. Danish mortgages run through a covered bond system under a strict balance principle: your loan corresponds to an identifiable bond trading in a public market, and you can prepay by buying that bond back at market price. So when rates rise and the bond trades below par, you can retire your debt at a discount. Your liability shrinks as your house value falls.

An automatic stabilizer, built into the household balance sheet, doing its work without anyone having to pass emergency legislation.

Compare the American product from last week: we get the option in one direction only, and it has now become a cage.

But here is the caveat, and it matters more than everything else in this section. German tenant protections work because they are distributing a supply that exists. Germany builds. The protections determine how a sufficient stock gets allocated and on what terms.

Layer tenant protection onto a construction freeze and you do not get Germany. You get security for incumbents and a sealed door for everyone else — the same failure mode as demand-side subsidy, arriving from the opposite direction. Any policy that improves the position of people already housed, without adding units, gets capitalized into the price of getting in.

German tenure law without German construction volume is just California with better manners.

What the excuses are protecting

Four excuses. Four tests. Four failures.

It isn't the mortgage — Canada has none of our subsidies and worse outcomes than we do.

It isn't global — it's six countries with a shared legal architecture, and Tokyo grew while its rents didn't.

It isn't politically impossible — Australia did it in June, and we now know precisely what it costs.

And it isn't the absence of an alternative — Germany chose a different bundle and has calmer politics to show for it.

Now notice what all four excuses have in common.

Every single one of them relocates the cause somewhere no voter can be held responsible. The financial system. Globalization. Political gravity. Human nature. Each explanation places the mechanism at an altitude far above anyone's actual conduct, in a realm where things simply happen and no one is choosing.

Not one of them mentions a planning commission.

That isn't a rhetorical accident. That's the entire function. The excuses exist to keep the conversation flying at a height where the real decision never comes into view — the decision made on a Tuesday night, in a municipal building, by a body most Americans could not name, attended by a couple dozen people who already live there.

Every country on that list of six has the same crisis and the same legal architecture, and the piece of that architecture they share is the one nobody argues about: the people who already live somewhere get to decide whether anyone else may.

Next week, the part where I stop picking on Texas.

Next in this series: San Jose is at twelve times median income. So is the problem really the Republicans?

Like what you've read, but not able to switch to paid? Why not buy me a coffee instead! All money goes to help defray the expense of running the site.