I agree, although it was not always thus, nor does this necessarily hold true in all locations. My father bought a house in 1960 for $50,000 (in 1960 dollars) and sold it 30 years later for $850,000 (in ~1990 dollars). He then moved to a cheaper part of the country (he was retired by then), bought a house for about $100,00 and had a nice chunk of change to augment his pension during retirement.
$50.000 in 1960 was about $221.000 in 1990, so in real terms the house increased an amazing 4,6% a year. I guess the surrounding area changed dramatically in the 30 years he owned the house (more urban, closer to good services). With a good link to labour market the area/house would increase in tandem with US economy and income - wich was great
Since the middle ages house prices have correlated with income, and since then longer periods with real income gains over 1-2% (as 1960-1990) has been extremely rare. My guess is it will be rare the upcoming 500 years also. And - even after a house price bubble fueled financial crisis, house prices are high (compared to income) thanks to extremely low interes rates.