Canada's Housing Bubble

Jan 20, 2010 Nick Rowe worthwhile.typepad.com

This is an inchoate post. That's not really an apology. Economists' ideas about bubbles aren't very clear. We can define a bubble theoretically, but we can't explain why they sometimes exist, sometimes don't exist, or why they sometimes start and stop existing. And we are not very good at identifying bubbles, even perhaps in hindsight. And even when we do say something is or was a bubble, all we are really saying is that we can't think of any plausible fundamental explanation of the time-path of prices, as Scott Sumner argues. That doesn't mean there wasn't one; just that we don't know what it is. (Scott's been doing more great posts recently than I can hope to keep track of.)

For what it's worth, I do think that bubbles are sometimes real, even though that belief is about as unfalsifiable as a belief that UFO's are real. In fact, we should probably drop the term "bubble" and replace it with "UFP", for "Unexplained Flying Prices".

And, for what it's worth, I believe in bubbles because I believe humans tend to follow faddish beliefs in lots of things, and I don't see why their beliefs in asset values should be any different.

"Bubble" is a metaphor, of course. But there's nothing wrong with using metaphors in science. How useful they are depends on how far you can push them.

Real bubbles are unstable; they burst when you prick them. They don't spontaneously revert to their original size. Soap bubbles aren't like tennis balls. If the bubble metaphor means anything, it has to mean that. If asset price bubbles aren't unstable, and don't burst when you prick them, or re-inflate immediately, then the bubble metaphor is useless.

How do you know if something was a bubble? If you prick it and it bursts, it probably was a bubble. If you prick it and it goes back to the original size, it probably wasn't.

We can (at least in principle) decompose the price of any asset into a fundamental component and a bubble component. The fundamental component evolves over time according to the expected present value of the dividends, rents, or whatever other benefit holding the asset indefinitely would yield the owner. And the bubble component must be expected to grow (or decline, if a negative bubble) exponentially at the rate of interest.

The expectations about the fundamental component are expectations about something that has an objective existence apart from beliefs about it. The expectations about the bubble component are expectations about expectations. Like Tinkerbell the fairy, they only exist if we believe in them; otherwise they die. That's what should make bubbles unstable. If others stop believing in them, there is no reason for us to believe in them.

If anything should cause a bubble to burst, it should be a fall in the price of the asset, especially a fall that is larger than any fall in the fundamental value of the asset. Because that is ipso facto evidence that others have stopped believing in the bubble.

US house prices fell last year, and show no signs of recovery yet. Canadian house prices fell last year, and have now recovered. Bubbles don't re-inflate themselves; they pop. Tennis balls return to their original size after they get hit, or even punctured; but then tennis balls aren't bubbles.

Now some might argue that the Bank of Canada's low interest rates have re-inflated the bubble. But if the metaphor "bubble" means anything, it means you can't just re-inflate it after it's burst. It's supposed to be unstable.

In any case, low interest rates are something that affects the fundamental value of an asset, not the bubble component. Sure, short term interest rates almost certainly won't stay this low, but long term interest rates already reflect the fact that short term interest rates will likely rise considerably. And long term interest rates are lower than they were before the fall in house prices. Maybe you could argue that there was a bubble in Canadian house prices; that bubble has now popped; but the fall in long term interest rates has raised the fundamental value by enough to compensate for the popped bubble; so the net effect on prices is about zero. But higher unemployment, and lower expected growth in real incomes, should have reduced the fundamental value of houses, and may well have offset any effect from lower interest rates.

I don't know if there was a bubble in Canadian house prices, but if there was one, it shouldn't be here now.

Could a bubble appear in future? Yes, certainly. And temporarily rising fundamental values might be the trigger for a future bubble. (Or might not, given that we can't predict bubbles). But I find it hard to see how the rise in prices from Spring 2009 was a bubble. Bubbles are supposed to burst when pricked, not re-inflate spontaneously.

In the US, with non-recourse mortgages, there is no problem in explaining why buyers would pay very high prices for houses. Get a 100% mortgage: if prices go up you win; if prices go down the bet's off, because you walk away from the house and mortgage. Panglossian expectations are rational with that payoff function. The puzzle there is why anyone would lend them the money, not why they borrowed it to pay very high prices. If there was a bubble, it was a lenders' bubble, not a house buyers' bubble.

Looking around the world, I see some asset prices in some countries have gone down and stayed down. But other asset prices in other countries have gone down, then bounced back up again. US vs Canadian houses are just one example. They can't all have been bubbles.