[a008t]

Houses have intrinsic value, which comes from future rental cashflows. Let's ignore the premium that some people would pay over that just because they want to be owners of a house.

Then, it seems to me that house prices can be approximated as a function of rents, interest rates (determining required return on owning a house), and anticipated growth in house prices (which again can be decomposed into growth in rents, reduction in interest rates and 'irrational' growth). Using London as an example. Rental yields appear to be around 3.5% in London at the moment, which suggests to me that a good deal of anticipated growth is priced in. But where can that growth come from? Rents are a function of incomes and are unlikely to outpace inflation. Interest rates seem to be more likely to go up than down. So it seems that house prices are unlikely to rise, and once that is realized by the market, prices may even come down as participants stop anticipating growth just because "London house prices always go up".

So a bubble means that house prices are disconnected from fundamentals, that some irrational assumptions are built into the price. I am not entirely sure that there is a bubble - it seems to assume that you need to assume that there is also a sovereign debt bubble, i.e. the "risk-free rate" or the treasury yields are irrationally low. I am not sure if that is the case or not, but that is the subject for a different discussion.