[mikekchar]

Have a look at this graph: https://research.stlouisfed.org/fred2/series/ATNHPIUS22420Q

It's housing prices in Flint, Michigan. It hit a peak of $172K in 2005. By 2011 it was $106K, or about the same price as in late 1995. Currently it is at $140K, which is essentially the price in 2000. Assuming it goes at its current rate, I guess it will be about the year 2020 before it recovers to its 2005 high.

Now, it's not really fair to call this a "bubble" since externalities are at fault. However, if you look at the increase (1.6x in 10 years) against an inflation rate of 3% (1.3x), you can see that housing was over valued during that period. I'm pulling the inflation rate from http://www.usinflationcalculator.com/inflation/historical-in..., and if anything overestimating it.

If you look at today's price it has a multiplier of 1.3 from the 1995 price, so it is still slightly undervalued, but should catch up to inflation in the next few years.

My point is that despite the credit crunch being the underlying cause of the crash, the market had been growing at nearly twice inflation for 10 years. 11 years after the crash the market is still recovering and it will be a couple of years before you get to reasonable prices. So you can go a very long time before unsustainable growth will collapse. This leads to a very, very long time for recovery.

I live in Japan. It is 2016. The market still has not recovered from 1992. If the world ever gets into an energy crunch, I think we might well look at the last 50 years or so as being a "bubble".

Edit: fixed inflation rate link

[acchow]

What does Flint, Michigan have to do with a tech bubble? Or any other bubble for that matter.

[mikekchar]

How does a house gain value? You can renovate it (real growth), or you can hold on to it and hope that someone will pay more for it. People don't think about it, but houses are relatively liquid assets. Let's pretend they are tulips, just for fun.

Imagine that you bought some tulips and just by hanging on to them for a while, you can realise a profit. This would be cool because everyone could put all their spare money into tulips and then turn around and sell them for a profit. Because the price of tulips keep going up (and people want to spend some of their gains on other things), eventually people will be able to buy less and less of them. However we might be able to raise salaries so that people can afford these tulips endlessly. In this way the inflation rate will exactly match the increase in price in tulips.

If we can't raise salaries to match the increase in the price of tulips, eventually people will be priced out of the tulip market and demand will dip. This will cause the price to fall. If people start to think, "Hey wait a minute. I'm not guaranteed to make a profit with these tulips after all", the price can fall a lot. If people start realising that they need to take a loss on their tulips so that they can afford to eat today, the price can tumble. How far can it fall? Mostly it depends on how clever people were for keeping the tulip bubble going. The more clever they were, the worse the potential fall. Essentially, it is likely to fall to the point at which the price of tulips escaped from inflation -- because it is a liquid asset and the need for tulips hasn't increased substantially over that time period.

Demand can influence the price of houses (and obviously did in Flint), but the degree to which the market dropped was a result of how overcooked the housing market was. Beware. Flint was never as overcooked as some markets are and people were never as clever about keeping the values high as some markets are.

Can a bubble last 10 years or more and then wipe out all of those gains? Absolutely. I only picked Flint because young people are likely to have heard of the problems there. You could also look at the housing market in London in 1991/1992.

Can the tech bubble burst and wipe out 10 years of gains? Sure. No problem. That's the only way I could interpret the person's question, "Can a bubble that has lasted 10 years still be called a bubble?" Definitely.

[eru]

> How does a house gain value?

To be more mathematical, your property is the house plus the land under it.

People will not pay more for the house than replacement value. If you keep it in good order, you can keep that value up.

The land can't be `replaced'. So for pricing we look at the whole future income stream discounted to today's dollars at some appropriate interest rates.

That income stream is, yes, basically what other people are willing to pay to use that piece of land.

What people can afford to pay for rent is basically what's left over after they paid other things. You can see it as an auction. That's why land values in silicon valley are so high. (Exacerbated by the fact that local regulation there makes it almost impossible to substitute capital for land, ie you can't build up.)

[mikekchar]

But it's also very important to understand that there is a ceiling. There will be a point at which Silicon Valley startups will stop increasing salaries. People will no longer be able to buy a house. Demand will drop and so will prices. If this occurs at the same time interest rates rise, people will not be able to afford to renew their mortgage and the house of cards will tumble. If the prices are really high and nobody has paid off any of the capital on their properties (as in London in 1991/92 and Tokyo where people had multi-generational mortgages), people will not be able to afford to sell because they owe more than they can sell the house for. But they will be forced to sell because they can't afford the new mortgage. There will be a rash of personal bankruptcies and unless someone steps in, very, very nasty things will happen. In London, the Japanese bought up the land (ironically just before their bubble burst). In Tokyo the government stepped in.

I don't know when, but everything is lined up "nicely". Ridiculously low interest rates, sky high prices, insane salaries. Even if SV companies move to a cheaper location to save on salaries it could trigger the collapse. Seriously not looking forward to a time when the fed raises interest rates to protect a falling dollar... Etc.

[eru]

Oh, definitely.

That's why I am in favour of taxing land values (as a proxy for unearned land rent), and the central bank targeting nominal GDP levels.

The former policy dampens land price bubbles and raises taxes in the most economically efficient way possible; the latter avoids real shocks in one part of the economy taking the whole house of cards down.

Ideally, no single company would then be too big to fail.

http://www.economist.com/blogs/freeexchange/2015/04/land-val...

http://www.economist.com/blogs/freeexchange/2011/10/monetary...