Graphs of Stock Market Historical Ratios
March 24, 2012 9:13 PM   Subscribe

Market Capitalization-to-GDP is an indicator described by Warren Buffet as "probably the best single measure of where [stock market] valuations stand at any given moment." Here is a historical graph of this indicator along with twenty other historical indicator charts covering the US S&P, Japanese Tokyo Stock Exchange and Indian Sensex indices. The indicators include P/E ratios and dividend yields. Also of note: which currencies are under or overvalued according to purchasing power parity.. posted by storybored (16 comments total) 22 users marked this as a favorite
 
Interesting, because Felix Salmon recently posted a similar chart and interpreted it differently: not as a measure of valuation but as a measure of the relative size of the public equity markets compared to other sources of capital.

It seems like these two factors interfere with each other; you need to find an independent measurement of one and correct for it in order to make the other interpretation, right?
posted by pete_22 at 9:52 PM on March 24, 2012 [1 favorite]


However, because the data is quarterly, we filled in the other months with the prior ratio adjusted for the change in stock prices. Because nothing says "quantitative analysis" like half-assing it.
posted by phooky at 9:55 PM on March 24, 2012


The upper graph is the actual ratio, what is the lower (black line) graph?
posted by Chuckles at 10:01 PM on March 24, 2012


However, because the data is quarterly, we filled in the other months with the prior ratio adjusted for the change in stock prices. Because nothing says "quantitative analysis" like half-assing it.

Yeah, unless a to-the-day update of the entire domestic product of all 300+ million Americans is available, there's absolutely no point in doing analysis at all involving the GDP. I remember that one day last month; someone released a delightful cat video that had me and my coworkers chortling for a good two minutes - on work time. If you haven't taken that into account, it's just not possible that you are doing solid quantitative macroeconomics.
posted by Homeboy Trouble at 10:17 PM on March 24, 2012 [5 favorites]


Behold, gringos! I have Australian dollars! All the colours of the rainbow serpent! Dance, monkeys! DANCE! BWAHAHAAAAHAAHhHAA!
posted by obiwanwasabi at 10:32 PM on March 24, 2012


As Kiwis we met two Austrayans in New York last fall; wandering around Manhattan asking, in that way our cousins can by saying "how much is this in real money?"
posted by Samuel Farrow at 10:37 PM on March 24, 2012 [1 favorite]


HT, that's the opposite of what I meant. If you're showing the ratio of two values and one dataset is only available with quarterly granularity, you should only show quarterly values. It's like significant figures in arithmetic; their results should only reflect the precision of the least precise input. Otherwise they're misrepresenting their data as more exact than it is. It's sloppy, but I probably wouldn't have snarked about it I hadn't read their "About" page.
posted by phooky at 10:42 PM on March 24, 2012 [2 favorites]


So I guess 1975-1985 were the years to get into the market? Damn, missed it by that much.
posted by adamdschneider at 11:18 PM on March 24, 2012 [1 favorite]


Yeah, but you've got to factor in this somehow.
posted by hank at 3:37 AM on March 25, 2012 [2 favorites]


Couple of thoughts:

1) Warren Buffet is brilliant when it comes to public markets. He has made an outstanding career of operating via public markets, and it is not a surprise that he considers formal, sanctioned metrics to be the key indicators of the economy. Thus, whatever he says is best taken within that lens.

2) The concept of public market cap-to-GDP has a few significant limitations:

• It seems to represent the attractiveness of public markets to capital more than anything else. When public markets are attractive, capital flows into them. When they are not attractive, capital flows out of them.

• It does not represent other aspects of the economy -- private markets, government spending, informal economies. These are substantial aspects of the economy that are discounted by this thinking. Thus, back to Point 1), this graph is most interesting if you limit your view to the public aspect of the economy.

• In light of the previous point, this is a great retrospective measure that is intellectually interesting, however it does not seem to have great practical application, as the unmeasured parts of the economy are equally as relevant. It's a bit like trying to lose weight and solely looking at volume of fat consumed. Whilst that definitely a relevant measure, so are carbs, proteins, etc.

3) The informal economy is a significant force. Looking at the final graph on this page illustrates that for the past 50 years, the informal economy represents 10 - 25% of the economy itself. Without taking that into account in terms of GDP, the GDP metric is relatively useless. It's trying to determine the absolute speed of your car by comparing it to another car. You don't have an objective baseline metric, thus any derivative metrics will be inaccurate at best.

4) With regards to the currency value calculations, many of the 'overvalued' countries have very strong public sectors (high tax rates) and many of the 'undervalued' counties have massive informal economies.

The summary seems to be that this work takes the view of the American 1% and uses that to try and explain the American economy, and a few others. The result falls a bit short of being relevant to anything.
posted by nickrussell at 6:04 AM on March 25, 2012 [2 favorites]


The upper graph is the actual ratio, what is the lower (black line) graph?

Wondering the exact same thing. Dominant culture, I'll make you a deal: I'll listen to your complaints about how low kids wear their pants if you'll start labeling your axes.
posted by compartment at 6:38 AM on March 25, 2012 [3 favorites]


The lower graph looks like a log-scale of the S&P index. I must spend too much time looking at the market 'cause I recognized it immediately.
posted by Justinian at 9:15 AM on March 25, 2012


Here you go, compartment.
posted by Riki tiki at 9:27 AM on March 25, 2012 [2 favorites]


Wait, but that's not ... that isn't ...

Actually, come to think of it, it is an improvement over the unlabeled axes I was talking about.
posted by compartment at 1:48 PM on March 25, 2012


I find it really frustrating to talk about the stock market, or any other economic system, with such restricted timescales - 1980s onwards? 1950s onwards? Why not 1880s onwards? Surely one draws entirely different conclusions about economics if one isn't getting the "whole picture" - or at least as much of the picture as is possible?

Sure, you might argue that "the 1910s were entirely different from the 1950s because a different economic champion was in place, so it doesn't compare." OK, then, but if you think the Chinese are going to be in charge from 2025 onwards, then you can't extrapolate from the 1950s-2000s period either - it cuts both ways!
posted by alasdair at 3:15 PM on March 25, 2012 [1 favorite]


Great point, alasdair. The main reason we don't see earlier date ranges is because of the lack of good, accessible data sets. The Wall Street Waltz has a handful of charts from the 1800s (check out chart 76 - Gov't Expenditure from 1790-1953 ). There's also a chart on the South Sea Bubble of the 18th century. The prize though goes to Chart 50 - Prices in South England for the years 1290-1950.
posted by storybored at 9:28 PM on March 25, 2012


« Older "I think I just really wanted that last hold...   |   SAT anyone? Newer »


This thread has been archived and is closed to new comments