One thing that's often left out of these cost/benefit exercises is that owning your home is a good hedge against being priced out of the area in which you want

One thing that's often left out of these cost/benefit exercises is that owning your home is a good hedge against being priced out of the area in which you want to live. If housing prices surge and you're renting, then you might not be afford to continue to live in your rented home, and so you might need to move; insurance against this sort of disruption is worth more than $0 to pretty much everyone and so it should be included in the cost/benefit consideration. Put more generally, your model should consider not just the possibility that growth in your real estate market is 1% YoY for the next 30 years, but also 10% YoY and -10% YoY. The advantage of renting of course is that if housing prices collapse, then you either pay less or can pay the same and upgrade. > Between 1948 and 2004 the real increase in value in the U.S real estate market was less than 1% a year. True but not the whole picture; buying a house doesn't give you exposure to "the US real estate market" in aggregate, you get exposure to a hyper-localized slice of the market, i.e. the neighborhood in which you buy your house. For example in the Bay Area you're looking at something like 5-10% YoY increases over the last few decades, and in the same period I imagine that prices have gone down in Flint, MI. This would be like saying "the stock market increases by 8% YoY on average" in an article about "should I buy AAPL"; you can see that it's relevant information that should be considered but it's not sufficiently specific to stand as an argument on its own.