After winning the Nobel Prize, Fama said something to the effect that he doesn't know what bubble is and it's all meaningless.
To which Shiller said "What is a bubble? You [Gene Fama] said nobody defines it. So I will define it. A speculative bubble is a fad. People get excited sometimes. Too excited... Prices start going up, they start talking, the newspapers start writing about it, more and more people pile in to a market and they push prices up more and it goes on for a while. eventually it breaks and the bubble bursts."
To which John Cochrane, Fama's colleague at Chicago, said "That's not a "definition." That's an explanation, a theory. A definition tells you in an operational way what pattern in the data describes "bubble." An explanation is a theory that predicts the defined phenomenon."
About a year ago, a family friend had e-mailed me asking what exactly bubbles are. Our e-mail exchange contains my view of what a bubble is:
Maybe you could answer this question for me. I'm reading a book the repeatedly refers to a "bubble" in housing prices.
What is the technical definition of a bubble? It sounds as though the price of something is greater than its value. But I thought that the value of something is what people are willing to pay for it and, as such, I don't understand how the price of something can be higher than its value.
Wouldn't another way of looking at the big increase/decrease of the price of a particular is it used to be worth much more and now it is worth much less?
Sorry for taking so long to respond. I am in los Angeles this week for my brother's wedding and things have been pretty hectic. This is a very good question. I used to discuss bubbles in my finance class at Colgate. I promise a fuller answer after I return to Seattle next week. But the quick version is that these are investment assets and not pure consumption goods. Assets provide a stream of consumption over time. So its value at any time depends on people's beliefs about the future. I'll write in more detail next week.
Have a good weekend.
Here's the fuller explanation that I promised. When I said a house is an asset that provides a stream of consumption over time, [...] what I meant was that a house provides shelter not just this year, but in
future years as well. In that sense, a house is like an apple tree that yields an apple each year for many years. Houses, apple trees, refrigerators and cars are all examples of assets, because they are long lived and provide a stream of benefit (consumption) over time.
The correct price of an apple tree is, loosely speaking, the sum of the prices of apples that it will provide over time. But the future price of apples is not known for certain today. So we have to make a guess about the future. If apples will be expensive in the future, then apple trees should cost more. If apples will be cheap in the
future, then apple trees should cost less. This need to guess about the future opens the door to incorrect prices. The price of an apple is never wrong because it simply reflects values and tastes (as you had noted). But the price of an apple tree could be wrong because the apple tree contains future apples and we don't know for sure what the prices of apples will be tomorrow. We might pay a lot for an apple tree thinking that apples will be very expensive tomorrow. But when tomorrow comes around, we may find that that apples aren't quite that
expensive, and so we overpaid for that apple tree. In other words, the apple tree was a bad investment.
Just being wrong about the future, however, does not imply a bubble. Nobody can perfectly predict what apples will cost tomorrow. The term bubble implies a more egregious sort of error: assets are being overpriced relative to our expectations about the value of the future services provided by the asset. In other words, apple trees are
selling for more than what people think apples will cost in the future. If you ask people what apples will cost in the future, they say a dollar. But they buy an apple tree for millions of dollars. If you ask them why they are doing this, they say the apple tree will become even more expensive in the future. In other words, people are
buying at an inflated price because they think the inflated price will become even more inflated in the future. The price of the asset has lost touch with the value of the consumption that the asset provides.
For a house, the underlying service provided is shelter in this and future years. The price of that shelter is the cost of renting that house in this year and future years. So the price of the house today should (loosely speaking) equal the cost of renting that house forever. When the price of the house becomes greater than the rent people expect that house to command, we have a bubble. By that calculus, I don't think we have a bubble in housing prices right now. Prices of houses are reasonably aligned with rents on the Eastside at
least, which is where we are looking right now. [Note: In my post of 2/28/2013, I referred to house prices as a bubble. That was loose talk. House prices are artificially inflated by the Fed's policies, but that still doesn't make it a bubble.]
Let me know if you have any more questions. This is a fascinating topic.
It seems that you require there to be rents to determine if housing prices have a bubble or not. However, I get the impression from some economists that there might be a bubble even if *both* housing prices and rents are high, no? (And what is "high"? Today's high might be tomorrow's low.)
There cannot be a bubble in rents for the reason that you described:
it's simply the value of the service being provided. One cannot
speculate by renting a house, any more than one can speculate by
eating an apple. If people are saying there's a bubble in rents, they
may be using the term bubble in a loose sense to mean that rents are
temporarily high and will go down in the future (what you had
originally proposed). I would not refer to that as a bubble.
Rents tomorrow may be higher or lower than today's rents. No way to be
sure. One can only guess.
There can be a bubble if both housing prices and CURRENT rents are
high, if FUTURE rents are expected to be low. By high and low, I just
mean relative to each other. A numerical example might help. An asset
that provides a service that is worth X dollars a year forever, should
have a price X/r where r is the interest rate. So if you have a
constant real interest rate of 3%, and a house that will rent every
year for $2,500 real dollars per month net of maintenance (so $30,000
real dollars per year), the price should be $30,000/.03=$1 million. If
the house is priced more than that, it is too expensive. If it's less
than that, it is too cheap. This of course is a highly simplified
model that makes a lot of assumptions (constant real interest rate of
3%, constant real rents of $2500, no taxes, etc.). But I hope it gives
you a general sense of the calculation.