5 ms·

“Overvalued” is speculation about the future disguised as a factual statement about the present. The difference is whether the price will eventually come crashi

“Overvalued” is speculation about the future disguised as a factual statement about the present. The difference is whether the price will eventually come crashing back down. If it will, it’s overvalued, otherwise it’s just the new normal. Of course, that means you can’t make that distinction with any kind of certainty until long afterwards.

mistermann 8y ago
Exactly. Overvalued implicitly implies "on a historical basis", but norms from the past are no guarantee of a reversion to the historical mean. It might be fair to say a reversion to the mean is likely, but it's not guaranteed, especially with the number and magnitude of changes to the input variables we've seen with real estate. One of the main changes is, in the past you were competing with local (employed) buyers, in many cities you are now competing with the entire world.
AnimalMuppet 8y ago
It could be speculation about the future. Alternately, it could be a claim about an (at least somewhat) objective standard of value. "Stocks are overvalued." How do you know? "The price to earnings ratio is too high." Now the question is, how good a measure is the price to earnings ratio? Is the "right" price to earnings ratio determined by the historical average, or does "right" change with changing circumstances? I don't have any solid answers about whether stocks are overvalued. (My personal guess is that, in a time of very low interest rates, a high P/E ratio makes the rate of return on stocks comparable to the rate of return on bonds, and so stocks may in fact not be overvalued. But I'm not an economist or a stock trader, and I could well be wrong.) My point is that there are some at-least-somewhat-objective measures of whether some things are overvalued.
- mikeash 8y ago
  There are objective measures, but how do you connect those objective measures to the price? In other words, how do you decide that a certain P/E ratio is a good objective measure, and counting the number of napkins in the company cafeteria is a bad objective measure? In the end, it’s just another layer on top of the speculation that the price will come down in the future.
  - AnimalMuppet 8y ago
    Well, for P/E ratio, we have a century or more of solid data. We also have a plausible reason why it might be a measure of how overvalued stocks are. For napkins in the company cafeteria, not so much (by either measure). The case against the P/E ratio boils down to "this time it's different". That can in fact be true... once in a while. Not nearly as often as it's claimed, though. That doesn't mean that P/E is absolutely correct - just that it's somewhat better than random noise.

mistermann 8y ago

Exactly. Overvalued implicitly implies "on a historical basis", but norms from the past are no guarantee of a reversion to the historical mean. It might be fair to say a reversion to the mean is likely, but it's not guaranteed, especially with the number and magnitude of changes to the input variables we've seen with real estate. One of the main changes is, in the past you were competing with local (employed) buyers, in many cities you are now competing with the entire world.

AnimalMuppet 8y ago

It could be speculation about the future. Alternately, it could be a claim about an (at least somewhat) objective standard of value. "Stocks are overvalued." How do you know? "The price to earnings ratio is too high." Now the question is, how good a measure is the price to earnings ratio? Is the "right" price to earnings ratio determined by the historical average, or does "right" change with changing circumstances? I don't have any solid answers about whether stocks are overvalued. (My personal guess is that, in a time of very low interest rates, a high P/E ratio makes the rate of return on stocks comparable to the rate of return on bonds, and so stocks may in fact not be overvalued. But I'm not an economist or a stock trader, and I could well be wrong.) My point is that there are some at-least-somewhat-objective measures of whether some things are overvalued.

mikeash 8y ago
There are objective measures, but how do you connect those objective measures to the price? In other words, how do you decide that a certain P/E ratio is a good objective measure, and counting the number of napkins in the company cafeteria is a bad objective measure? In the end, it’s just another layer on top of the speculation that the price will come down in the future.
- AnimalMuppet 8y ago
  Well, for P/E ratio, we have a century or more of solid data. We also have a plausible reason why it might be a measure of how overvalued stocks are. For napkins in the company cafeteria, not so much (by either measure). The case against the P/E ratio boils down to "this time it's different". That can in fact be true... once in a while. Not nearly as often as it's claimed, though. That doesn't mean that P/E is absolutely correct - just that it's somewhat better than random noise.