Why are house prices so low?


Today’s daily chart in the Economist magazine details house prices and some related series for a number of countries, in many cases back to 1975. Eyeballing the data there are a few things that I find surprising. There is still a lot of talk of bubbles in Australian house prices, yet if you look at real house price levels they are only about 2.5 times 1975 levels – roughly the same increase as Britain, and an annual increase of only around 2.8%. In the case of Australia that is close to real per capita GDP growth, which is about what you’d expect. Interestingly, in 2006 real prices in both Australia and the US were about double 1975 levels, since which time US prices have of course fallen significantly so that they are now only about 20% higher than in 1975. Maybe Australian house prices are still too high, but it seems more likely to me that US house prices are currently too low. Of course there are some explanations for the very low rates of real price appreciation in the US – slow growth in median wages being one factor. But the population is still growing, and the US of course has a much larger population than Australia, so it is surprising to see so little in the way of house prices increases. And if you want to see what declining populations do to house prices look at the series for Japan – the issue there is not so much that the property bubble bust in 1990, but that prices have continued to fall for the past 22 years.

5 Responses to "Why are house prices so low?"
  1. I’m not sure this statement is quite correct.
    “In the case of Australia that is close to real per capita GDP growth, which is about what you’d expect.”

    Prices appear to be about 75% higher that what you might expect from GDP per capita growth

    Also, I’m not sure what your point is with US prices being too low.  Because you would have have every reason to argue the same about Japanese home prices in the mid 1990s, yet they continued to fall for another 15 years or so.  

    In fact, the rate of population growth in Japan was on a massive decline during its boom years (1980s) , so I’m not sure what population has to do with prices, unless you assume that there is some limiting factor on new housing construction (also, Japan’s population peaked around 2006). 

    Finally, you seem to be comparing many countries with known housing bubbles that are now deflating, and also from an arbitrary starting date – the picture will look very different depending on your starting point.   

  2. Agree with most points, but your point about prices being 75% higher is also blighted by an arbitrary starting point, being 1953. According to your graph real prices do not move between 1953 and 1975 (since the index is still at 100), a period during which real per capita GDP growth in Aust was around 3%. So that would be enough to make prices too low again, if you use 1953 as your starting point. Your graph is interesting, and you could equally argue that the puzzle in Aust is why prices go for such long periods without moving – up until 2000 for example. Lots of candidate explanations, such as liquidity constraints etc. Of course all of your and my arguments require a proper model to explore, which I’m not going to do on a blog post. I still think the data is interesting and in the case of the US surprising.

  3. This doesn’t seem right, I bought in 1991, and my house cost approximately 2.5 times my annual salary, twenty years later, my house costs approximately 6.75 times my annual salary and I’ve definitely moved up the percentiles over that 20 years. 

  4. I agree with the other commenters – most Australian house prices are still above what longrun “fundamentals” should lead you to expect.  However I agree that many house prices in many parts of the US must be near their bottom – in fact I’ve recently put my money where my mouth is on that and bought some exposure to New York residential property for my private super fund.

%d bloggers like this: