[swexbe]

You have the cause and effect reversed. Areas densify because they become more expensive. This is the classic “more medicine causes higher mortality” Simpson's paradox.

Edit: Also, looking at one country for less than a generation is not conclusive enough to make a general rule.

[cyberax]

> You have the cause and effect reversed. Areas densify because they become more expensive.

The thing is, with Simpson's paradox you'd expect at least _some_ inversions. You'll have at least _some_ people who are eating a lot of medicine and not dying.

There are _no_ examples of higher density leading to lower prices in the US. This strongly suggests a causal relationship.

And some examples are rather extreme. The number of units in Seattle has been growing by about 2.2% YoY for the last ~12 years. This is already at the upper range of the possible construction rate. Yet Seattle's price growth curve has been basically mirroring SF.

> Edit: Also, looking at one country for less than a generation is not conclusive enough to make a general rule.

I don't have data for more than 25 years and for other countries. And probably you can find counterexamples somewhere in Brazil.