- Rising interest rates from the late 1960's through the early 1980s.
- Vast expansion of the land supply through suburbanization
- Rising urban crime pushing down prices in city centers
Where you think the "natural" price of housing is depends on which of these trends you expect to persist, versus which you think will peter out and reverse. Right now my assessment is that things like the amortizing mortgage, increasing land supply, and huge shifts in inflation/interest rates are largely played out.
Zoning will continue to be a strong factor in a handful of urban areas, including the one I happen to live in--in fact, a growing influence in places where outward expansion has hit the limits of car range and the local infrastructure, while zoning depresses the supply of available housing closer in.
I expect household formation to be a weak-to-negative influence on housing demand for the next decade or so; we can't delay marriage much longer than we already are, and the baby boomers are eventually going to have to start pricing assisted living. The effect of crime on urban house prices is steadily reversing itself.
Securitization is a wild card; right now, there doesn't seem to be a private market for mortgage securities, and it's anyone's guess how long the government will continue to subsidize cheap mortgage bonds.
All in all, I'd expect that the housing market will be dominated, as real estate brokers like to say, by "location, location, location". In other words, local factors will dominate any broad national trend. But I'm not talking about which areas are going to deliver terrific investment returns for their owners, so much as which areas can be expected to hold their value or grow a little bit.
But I have no faith in my ability to pick which neighborhoods those will be. That's why we're choosing our house by where we want to live, and how it compares to our rental cost, rather than trying to make money.
Update: D'oh! Commenters remind me of one I missed: women working, which led to higher purchasing power in good school districts.
There's another factor some have suggested: the mortgage interest tax deduction. That's actually complicated. Before 1986, all interest was tax deductible; when they did the tax simplification, they removed the deduction for everything except mortgage interest. So it's hard to say that this played a big role in housing prices.
You can argue that it interacted with other trends--that changes in the income tax made the mortgage interest deduction more valuable, or that as long-term self-amortizing mortgages became more common, and downpayments got smaller, interest played more of a role in the monthly carrying costs of a house, and that of course made the deduction more valuable. But while it's absolutely true that removing the deduction would devastate real estate prices (at least at the high end), given that income tax rates were falling over much of the postwar period, the tax deduction probably played at best an ancillary role in boosting prices.