I find this sort of data fascinating. In Detroit, median house price is $38K and median household income is $26K, for a ratio of roughly 1.5 In San Francisco, with a median house price of $1.1 million and median household income of $81K, the ratio roughly 14.7
What should the ratio be? price/income = (price/rent) times (rent/income). Figure that in a city where there is plenty of housing supply, you might spend get by with spending only 1/5 of your income on rent. In an city where housing is scarce, you might need to spend 1/2 your income on rent.
A price/rent ratio is sort of comparable to price/earnings ratio on stocks. It should tend to be lower than stock P/E’s, because of high transactions costs, property taxes, HOA fees, and depreciation.
Historically, according to Robert Shiller, the housing price-rent ratio is around 10 or 12. So, if I take numbers on the low side, namely 1/5 for rent/income and 10 for price/rent, I get a price/income ratio of 2.0. Only Detroit is below that.
If I use high values, of 0.5 for rent/income and 12 for price-rent, I get a price/income ratio of 6.0 Of the 27 cities, 15 fall within a range of 2.0 and 6.0, which means that they are in line with historical values.
Another 6 cities have values ranging from 7 to 9. There are another 5 cities with ratios over 9, topped off by San Francisco.
Of course, the marginal buyer of the median home is not necessarily the person with median income. I think that is safe to say that in the cities with ratios over 9, median-income residents are not the marginal homebuyers. In Miami, for example, my guess is that a fair number of home buyers reside in other cities, and that could raise the price/income ratio within the city. But otherwise, those high ratios could be a sign that speculation is getting out of hand.
P/E can be high for two reasons: 1) inflated expectations/bubbles, 2) market anticipates future earnings growth will be high. Hard to pick those two apart.
My point is price/income in cities like San Francisco is partly due to #2. That is to say, because NIMBY is expected to continue. So the future is bright for the price of existing houses if little new housing gets built.
Has Schiller or someone like that already done some work to tease those two apart? If so, that would be useful to publicize. Since I lived in San Francisco for 10 years, and now live about 25 miles south (house with big mortgage), I have intuitions. But not sure. I suspect NIMBY continuing is being priced in. And a large part of what’s going on with this ratio. But honestly have little confidence.
I agree; high numbers indicate that either:
1. the housing is serving a separate purpose than ‘housing’
[could be investment, education, or prestige]
2. or that it is serving a separate population with different demographics
Some of this data is skewed by the definition of ‘city’. If you’re looking only at property within city limits, this can mean wildly different things for the affordability of housing in the broader area. The city of Boston, for example, is a tiny postage stamp of (mostly) very dense expensive housing in the center of a somewhat more affordable area. The city limits of LA are vast, encompassing both the dense urban areas of Century City and downtown, but also very suburban areas like Canoga Park.
What if we adopted targets such that the income/home price (per square foot) ratio should trend down every year?
Crap, I mean trend up.
The larger the metropolitan area, the larger the ratio because congestion limits commuting distances and these are dominated by central city prices, and there is a strong trend in the data due to this. The standout is Chicago which is low for its size. Not all places are equal.
It’s also worthwhile to consider commuter geography. Coastal cities all have natural commuting bottlenecks, while “non-NIMBY” cities curiously are all surrounded by flat land as far as the eye can see.
Chicago is deceiving for a few reasons.
1. There are large geographic swaths of the city that are really ghetto. This skews the average/median home price downward.
2. Most middle and upper middle class folks want absolutely nothing to do with the public schools in Chicago. Most of the wealth is in the suburbs, and Chicago has very good commuter rail service (Metra) for an American city.
3. Chicago and its suburbs pay the highest property taxes in the nation. Paying $15,000 to $25,000 a year on a $700k house is common. And there are no caps or limits —they will keep going up. This is a hidden cost that significantly affects the affordability of housing in Chicagoland, and prices reflect this reality.
I took a real estate class in college 30 yrs ago where the instructor said to not buy a house more than 2.5 times income. That is possible in many cities outside of the coasts because the median home buyers income is more than the income of all households.
Rent prices (perhaps standardized per square feet and somehow quality adjusted) are of course determined by supply and demand.
There’s a lot of emphasis on the supply side, too regulated so too little and too inelastic.
But the demand-side question is why are people willing to bear the burden of such high ratios and pay such sky-high prices to live in those places, when there are plenty of cheaper and nicer places.
A large part of the answer is always, “because it’s still better than any of the alternatives,” which in general means, “because that’s where the lucrative jobs are, both now, and likely in the future.” It’s really hard to earn a lot of money in the middle of nowhere without being a prestigious tenured university professor in some small college town. Having known a few of those, I can tell you that with high incomes and low cost of living they live really, really well. But there are not many opportunities for that.
Most good jobs are in the top big cities, and people have little choice but to bid up real estate prices to the very limit of indifference with some lower-paying, best-alternative job in some cheaper location.
So, while supply restraints are important, that’s not the only way to interpret the high variance of these ratios. What they are telling us is that it’s really hard to earn a lot of income outside the few ‘winner’ cities, that many high earners are a kind of captive audience, and that as a result, urban politicians are able to deploy Blagojevich logic when it comes to questions of the quality of local governance.
Immigrants also drive up rents, because they tend to settle in big “winner” cities. If I had my druthers, immigrants would have to live for at least their first 10 years in the USA outside of our top 25 (50?) metropolitan areas.
If you [I]really[/I] want to live here and become an American, you’re gonna have to move to a real American town and make a real effort to assimilate. And insofar as our “winner” cities are generating economic rents for workers, why in the world would we let strangers from foreign countries in on the game? Let’s stop being so naïve and giving away the store.
Whoa Arnold, you are missing an important distinction. The price / rent multiple is not analogous to a P/E ratio. Rather, it is similar to an Enterprise / revenue multiple. The distinction is important because, as an investor in apartment properties, I can tell you that the Net operating income margin (roughly analogous to an EBITDA margin for C-Corps is much, much higher in San Francisco than in other places. For example, the NOI margin in San Francisco is usually about 75%. In Merced California, it is about 30%. Once you subtract Capital Expenditures, the 30% is cut almost in half. Basically what I’m saying is that the net cash flow, after Capital Expenditures, in many “C” and below locations is basically zero. Thus, the price has to go very low to entice investors to buy there. This is not irrational behavior.
That said, I personally think that investment in the core SF Bay Area is overvalued.