Land NOT Housing Bubble

Housing prices depend upon the cost of building which has been remarkably constant over the decades.  Land prices depend on population and income and speculation and is extremely variable.
Yglesias:

There was, obviously, a huge boom in the price of land in the United States of America during this period. But was there really an extraordinary boom in housebuilding?

At the height of the “boom” we were adding units about as quickly as we were adding them in the late 1970s, when the total population was smaller and China’s “opening up” was just a glimmer of an idea of a possibility. If the Federal Reserve was trying to engineer a homebuilding boom it didn’t really work.

Most graphs I have seen of housing starts are truncated around 1990.  That makes the recent 'housing' bubble look like an extreme expansion of housing supply.  But it was actually a slower increase than previous increases and it did not top out as high as in the early 1970s when the US population was much smaller.  The most notable feature of this graph is the recent bust, not the boom.   The housing bubble was in the price of housing which is mainly driven by the price of land: