7 ms·

I've never seen market valuation expressed as market cap as % of GDP. I'm not an economist, so I'll leave the detailed arguments to them. But it would be at lea

by runako 9y ago

I've never seen market valuation expressed as market cap as % of GDP. I'm not an economist, so I'll leave the detailed arguments to them. But it would be at least useful to explain why you think this is a meaningful metric as compared to those typically used to measure market valuation (e.g. P/E ratios etc.). Your graph also ties your valuation metric to the 2000 peak and the 2008 peak. However, there were crashes in 1990 and 1987 as well. Should readers conclude that the 1987 peak level was also too high, and that therefore the last ~30 years have also been too high as well? (Abstaining from investing in the stock market at levels above the 1987 crash would have resulted in the loss of tremendous opportunity for wealth creation.) There are a lot of opinions implicitly expressed in this site; it would be good to try to make those explicit.

amelius 9y ago
You should always use separate validation data anyway.
truffle_pig 9y ago
Good points, I might go into a bit more detail for how I calculate the risk for each factor. Thanks for the feedback
wheaties 9y ago
One issue I see here is how much of the S&P500 (or general market cap) are we attributing to a US-centric view of global companies? That is, if a company is "in the US" but economically are not.
JumpCrisscross 9y ago
> market valuation expressed as market cap as % of GDP This metric makes little sense for this use case. Consider two countries. They are identical in every way except in Country A 90% of the companies are publicly-traded while in Country B 10% are. Country A will have a market cap to GDP 9x Country B's. Does that mean Country A is 9 times overvalued relative to Country B? The objection works in-country, too. Saudi Aramco is going public in New York or London [1]. This will lift one of those market's aggregate capitalisation by up to $1 trillion. Does this mean that market will necessarily become overpriced? The answer to both question is of course not. Market cap to GDP tells you the degree to which a country has developed public markets. Not anything interesting about the levels in those markets. [1] https://www.bloomberg.com/view/articles/2017-04-05/aramco-ipo-is-just-the-first-step-for-saudi-arabia https://www.bloomberg.com/view/articles/2017-04-05/aramco-ip...
- javajosh 9y ago
  Nit pick: only about 5% of Saudi Aramco is going public, the total company is worth $1-2T (although this is a matter of some debate, as you might imagine), so aggregate market cap will only increase by $50B. [0] https://www.cnbc.com/2017/04/24/saudi-aramcos-valuation-reportedly-off-by-500-billion-or-roughly-amazon-and-netflix-combined.html https://www.cnbc.com/2017/04/24/saudi-aramcos-valuation-repo...
  - JumpCrisscross 9y ago
    > aggregate market cap will only increase by $50B If you list 5% of a $1 trillion company on a stock exchange, the aggregate market capitalisation goes up by $1 trillion. (Float goes up by $50bn.) Market capitalisation is price per share times shares outstanding [1]. Float is price per share times publicly-trading shares [2]. [1] https://www.fool.com/investing/small-cap/2005/04/29/quotoutstanding-sharesquot-vs-quotfloatquot.aspx https://www.fool.com/investing/small-cap/2005/04/29/quotouts... [2] http://www.investopedia.com/terms/f/floating-stock.asp http://www.investopedia.com/terms/f/floating-stock.asp
    - javajosh 9y ago
      You sound knowledgable about these matters, and I am not. In my ignorance, "agg. market cap" sounds like an easily gamed number. For example, I could start an exchange and ask every company in the world to list 10^-10 of their shares, and so become the largest exchange in the world by agg. market cap. (Perhaps this doesn't come up in practice because most firms list in only one market, and so agg. market cap becomes a useful measure of how much of the economy a given exchange touches, in some sense.)
      - milcron 9y ago
        It does come up in practice. It's actually a common scam. When only a small fraction of shares are on the market, it's quite easy to manipulate the price higher... buying pressure goes a relatively long way. If you can get 1% of your shares to be worth $100k, it now appears as though you are a $10 million company. This makes you appear reputable. Drive hype about your "valuable" company, and once the stock is sufficiently pumped, dump your shares for profit.
Ended 9y ago
In fact unless I'm missing something the units don't match up: Total stock market value has units $ GDP has units $/year So expressing their ratio as a % is misleading.
- AnimalMuppet 9y ago
  The ratio has units of years - how many years of GDP does the value of the stock market represent? How long will it take the economy to produce the value assumed by the stock market valuations? Sure, that's not a pure number, but it's also not a totally nonsense number. At a minimum, it can be compared to historical values of the same number.
- rb808 9y ago
  A good similar example would be speed as a % of distance. Eg going on a trip 240 miles at 60 miles/hour is 0.25. Same thing.
jbrozena22 9y ago
Market cap as % of GDP is a statistic famously used by Warren Buffett https://www.advisorperspectives.com/dshort/updates/2017/08/03/market-cap-to-gdp-an-updated-look-at-the-buffett-valuation-indicator https://www.advisorperspectives.com/dshort/updates/2017/08/0...
georgeecollins 9y ago
Here is the case for looking at Market Cap / GDP: Warren Buffet looks at the total market cap vs GDP because it is free from the influence of corporate accounting, unlike the more traditional P/E ratio. The case against: economic activity in the US has been concentrating in larger companies (less small companies being started for example) and there is more "financialization" in general. So you would expect the ratio of market cap / GDP to go up for those reasons as well.
Iv 9y ago
It makes some sense if you consider market caps to be expectation of returns. All other things being equal (like the percentage of the GDP made by publicly traded companies) it makes sense to expect the market to expand at about the same speed as the GDP. That this ratio doubled in 8 years is worrisome. The thing is that an alternative analysis is to look at the linear regression since 1980 and it seems clear that a background upward tendency does seem to exist, so we can see 2009 as an adjustment to the 2008 crisis and 2017 as only slightly above the expected trend.
fernly 9y ago
In particular, people's evaluation of a company are not (wholly) based on its current annual output, whereas the GDP is exactly the amalgam of all companies' output. People evaluate a company at least in part on the assets it holds, intellectual property, and future earning power, none of which are captured by the GDP.

by runako 9y ago

amelius 9y ago

You should always use separate validation data anyway.

truffle_pig 9y ago

Good points, I might go into a bit more detail for how I calculate the risk for each factor. Thanks for the feedback

wheaties 9y ago

One issue I see here is how much of the S&P500 (or general market cap) are we attributing to a US-centric view of global companies? That is, if a company is "in the US" but economically are not.

JumpCrisscross 9y ago

> market valuation expressed as market cap as % of GDP This metric makes little sense for this use case. Consider two countries. They are identical in every way except in Country A 90% of the companies are publicly-traded while in Country B 10% are. Country A will have a market cap to GDP 9x Country B's. Does that mean Country A is 9 times overvalued relative to Country B? The objection works in-country, too. Saudi Aramco is going public in New York or London [1]. This will lift one of those market's aggregate capitalisation by up to $1 trillion. Does this mean that market will necessarily become overpriced? The answer to both question is of course not. Market cap to GDP tells you the degree to which a country has developed public markets. Not anything interesting about the levels in those markets. [1] https://www.bloomberg.com/view/articles/2017-04-05/aramco-ipo-is-just-the-first-step-for-saudi-arabia https://www.bloomberg.com/view/articles/2017-04-05/aramco-ip...

javajosh 9y ago
Nit pick: only about 5% of Saudi Aramco is going public, the total company is worth $1-2T (although this is a matter of some debate, as you might imagine), so aggregate market cap will only increase by $50B. [0] https://www.cnbc.com/2017/04/24/saudi-aramcos-valuation-reportedly-off-by-500-billion-or-roughly-amazon-and-netflix-combined.html https://www.cnbc.com/2017/04/24/saudi-aramcos-valuation-repo...
- JumpCrisscross 9y ago
  > aggregate market cap will only increase by $50B If you list 5% of a $1 trillion company on a stock exchange, the aggregate market capitalisation goes up by $1 trillion. (Float goes up by $50bn.) Market capitalisation is price per share times shares outstanding [1]. Float is price per share times publicly-trading shares [2]. [1] https://www.fool.com/investing/small-cap/2005/04/29/quotoutstanding-sharesquot-vs-quotfloatquot.aspx https://www.fool.com/investing/small-cap/2005/04/29/quotouts... [2] http://www.investopedia.com/terms/f/floating-stock.asp http://www.investopedia.com/terms/f/floating-stock.asp
  - javajosh 9y ago
    You sound knowledgable about these matters, and I am not. In my ignorance, "agg. market cap" sounds like an easily gamed number. For example, I could start an exchange and ask every company in the world to list 10^-10 of their shares, and so become the largest exchange in the world by agg. market cap. (Perhaps this doesn't come up in practice because most firms list in only one market, and so agg. market cap becomes a useful measure of how much of the economy a given exchange touches, in some sense.)
    - milcron 9y ago
      It does come up in practice. It's actually a common scam. When only a small fraction of shares are on the market, it's quite easy to manipulate the price higher... buying pressure goes a relatively long way. If you can get 1% of your shares to be worth $100k, it now appears as though you are a $10 million company. This makes you appear reputable. Drive hype about your "valuable" company, and once the stock is sufficiently pumped, dump your shares for profit.

Ended 9y ago

In fact unless I'm missing something the units don't match up: Total stock market value has units $ GDP has units $/year So expressing their ratio as a % is misleading.

AnimalMuppet 9y ago
The ratio has units of years - how many years of GDP does the value of the stock market represent? How long will it take the economy to produce the value assumed by the stock market valuations? Sure, that's not a pure number, but it's also not a totally nonsense number. At a minimum, it can be compared to historical values of the same number.
rb808 9y ago
A good similar example would be speed as a % of distance. Eg going on a trip 240 miles at 60 miles/hour is 0.25. Same thing.

jbrozena22 9y ago

Market cap as % of GDP is a statistic famously used by Warren Buffett https://www.advisorperspectives.com/dshort/updates/2017/08/03/market-cap-to-gdp-an-updated-look-at-the-buffett-valuation-indicator https://www.advisorperspectives.com/dshort/updates/2017/08/0...

georgeecollins 9y ago

Here is the case for looking at Market Cap / GDP: Warren Buffet looks at the total market cap vs GDP because it is free from the influence of corporate accounting, unlike the more traditional P/E ratio. The case against: economic activity in the US has been concentrating in larger companies (less small companies being started for example) and there is more "financialization" in general. So you would expect the ratio of market cap / GDP to go up for those reasons as well.

Iv 9y ago

It makes some sense if you consider market caps to be expectation of returns. All other things being equal (like the percentage of the GDP made by publicly traded companies) it makes sense to expect the market to expand at about the same speed as the GDP. That this ratio doubled in 8 years is worrisome. The thing is that an alternative analysis is to look at the linear regression since 1980 and it seems clear that a background upward tendency does seem to exist, so we can see 2009 as an adjustment to the 2008 crisis and 2017 as only slightly above the expected trend.

fernly 9y ago

In particular, people's evaluation of a company are not (wholly) based on its current annual output, whereas the GDP is exactly the amalgam of all companies' output. People evaluate a company at least in part on the assets it holds, intellectual property, and future earning power, none of which are captured by the GDP.