The Historical Relationship Between the Economy and the S&P 500, Part 1
by Mike Kimel
The Historical Relationship Between the Economy and the S&P 500, Part 1
This post is the first in a series on the historical relationship between the economy and the stock market. Data used in this post is the adjusted close of the S&P 500 going back to 1950 and quarterly nominal GDP going back to the same date. Because quarterly GDP figures measure the economy at the midpoint of the quarter, the S&P 500 for February, May, August and November are considered the analogous “quarterly” S&P 500 figures.
The following graph shows how the two series have evolved since 1950.
A simple eyeball test does indicate that historically, the two series have mostly moved together. And in fact, the correlation between the two series has been 93.5%, which is extremely high.
But which series has led and which has followed? For that, we look at the next graph:
The black line shows the correlation between GDP and the S&P 500 x quarters hence, where x = 0, 1, … 24. In other words, it shows the correlation between GDP and the S&P 500 in the same period, the correlation between GDP and the S&P 500 lagged by a quarter, the correlation between GDP and the S&P 500 lagged two quarters, etc. As the black line shows, historically, GDP seems to have led the S&P 500 by about 16 quarters, or about four years; the correlation between GDP and the S&P 500 16 quarters out is about 94.1%. One possible reason… historically, when the economy grew, it encouraged more investment in corporations, but in many instances, investments take a long time (four years) to show up in ways that boost the market.
The red line looks at the relationship the other way – it looks at the correlation between the two series with the S&P 500 leading GDP. It shows that the correlation between the S&P 500 and the lagged GDP reached a peak of 95.2% when GDP was lagged by 10 quarters, or about two years. One possible explanation… historically, a growing stock market has made people feel wealthier, which in turn caused them to buy more stuff.
The conclusion from this… going back to 1950, GDP has led the S&P 500 and vice versa.
In the next post in this series, I’ll take a look at how this relationship has changed over time.
—————————-
A few notes…
1. I’m not a financial advisor. I strongly suggest against making investments based on anything written above.
2. If you want my spreadsheet, drop me a line via e-mail with the name of this post. My e-mail address is my first name (mike), my last name (kimel – with one m only), and I’m at gmail.com.
Could it be, since the correlation is similar whether GDP is leading or trailing, that there is a third variable, unaccounted for, that is influencing both GDP and the S&P? Or is plus or minus 2 to 4 years just withing the variability of the data?
Obviously, other variables matter. But… its hard to imagine that hits, say GDP now, the S&P 16 quarters later, and then GDP again 10 quarters after that.
I think it makes more sense to assume they affect each other, in each case with a lag.
Mike
slight variation on your explanation:
GDP leads stocks: people have more excess money , buy stocks.
stocks lead GDP: people “expect” rise in GDP, buy stocks.
So I guess what you are saying is that the interact in a way that they reenforce each other.
jerry
no. i think what i am saying is that the stock market follows the growth in the economy except when it anticipates it.
it does not lead to the growth. there may be some investing (i’m sure there is) that leads to growth. but that isn’t what the stock market is all about.
I do think there was some back and forth during the period graphed. Follow up posts are going to show something a bit different later.
What is the difference between anticipating and leading? Doesn’t anticipation lead to leading? Isn’t the result the same? Or am I just being a nitpicker?
Since WW II the long term trend growth rate of the S&P 500 earnings per share has been 7%.
Currently we are approaching that trend line for both operating earnings and reported earnings after they fell sharply below that trend line during the great recession.
The other component of the S&P is its valuation, or PE. From 1957, when the S&P dividend yield fell below the 3 month t bill, till about 1995 the S&P 500 PE moved tightly in an inverse relationship to interest rates.
Since 1995 the PE has displaced much more volatility and much wider deviations from where interest rates suggest the PE should be.
Interestingly, as earnings continued to grow at about a 7% rate in the 1960s when inflation caused nominal GDP growth to accelerate over 7% profits share of GDP fell. Since the early 1980s nominal GDP growth has only averaged some 5% but earnings growth continued to bounce around the 7% trend so
profits share of GDP started rising. Moreover, as inflation and interest rates fell on a secular basis since the early 1980s the market PE has also risen. Consequently both profits and the S&P have risen relative to GDP since the 1960s.
Currently the S&P on trailing earnings has a PE of about 15, well below what the traditional relationship to interest rates imply it should be. This suggest the market is cheap. But maybe it isn’t. Maybe as nominal GDP growth remains under 5% it will be impossible for the S&P 500 to sustain the old 7% earnings trend. In this case maybe the market is starting to discount a lower long term earnings trend.
Maybe that is why the PE appears lower so the market actually may not be cheap.
The reverse of this happened in the 1990s when the market started discounting a higher long term growth rate and the PE rose into the upper 20s, much higher than the traditional relationship to rates suggested.
jerry
not being a nit picker. the difference is in the concept of causation. people buy stocks when they expect the economy to improve “and stocks to go up.” they also buy stocks when they have a lot of money they want to “save but make money on.” one is anticipating, the other is following. neither of them are the stock market “leading” (causing) growth.
i think you are having trouble letting go of the idea that buying stocks is an “investment” that “creates wealth (growth). i can’t say that that is “not true.” i can say that i am saying that it is not true.
(being a little funny with you here. the difference is that my saying it is not true is not a claim that i know it is not true.)
spencer
you undoubtedly know more that i do about this. but the language you use sounds almost “magical”.
in what sense does “the market” discount a… trend. if this only means that “people” are expecting a trend… i might learn to accept the language, but i suspect that language tends to mislead even the people who use it into believing “the market” knows something.
Make believe you are buying a small business. If the biz is a growth biz you will have to pay a lot for it because it will earn you a lot in future earnings. Not so if it is a low growth biz.
So if the stock market starts believing companies will have trouble growing, and perhaps having difficulty paying increasing dividends, because the economy is growing at a slower pace than the historical norm, then they will be less willing to pay a high price for stock.
That’s assuming we have rational investors, of course, and the Fed hasn’t driven their animal spirits into a frenzy with ZIRP and QE.
Cedric
if your comment is addressed to me, i think you are saying the same thing i am. my objection to spencer was the language “the market discounts..” because that seems to me to imply a kind of magical knowledge residing in “markets” as opposed to simply saying “more people will expect..”
i suspect that spencer means exactly what i mean. but i have heard many people talk as if “the market” were sentient if not “all wise.”
i see no evidence that “investors” in stocks are rational. even if they have a system. this does not mean that there are not people who look at a company and understand what it sells, and who buys what it sells, and what conditions will lead to the kind of profits that result in dividends or higher share prices…. but i don’t believe such people drive the kind of stock market “averages” that we are talking about here.
Perhaps more can be learned about the relationship between GDP and stock prices by using the quarterly percentage change for each series rather than nominal levels.
hAL2000 — The rates of change generate interesting results.
For example, the corelation between the S&P and EPS is very strong, implying there is a strong relationship between earnings and the market.
But on the other hand the correlation between the change in ernings and the change in the market is essentially zero, implying there is no relationship between the two.
Spencer– This often happens with time-series data. For example, two price series might both roughly double over 10 years, and one might assume they have some sort of relationship with each other. However, the doubling might be due to general inflation that impacts both series, but their relationship to each other, especially from a cause and effect standpoint, is non-existent.
With time-series data, if there is no relationship between the incremental changes in two variables, either direct or lagged, it might be that they are both being influenced by something else, but they both may be independent of each other.
There is something wrong with your math. I have worked on a project doing correlations for the last three years and in doing so I have looked at literally hundreds of thousands of graphs of correlations and I can tell you just from eyeballing it that the graph above does not represent anywhere near 0.94 correlation and further. Also, where the two lines diverge around 1970 if why the correlation number increases slightly as you “adjust to see which is leading which.”
There is a problem with your math, something rounding too soon or something.
The Angry Bear writes:
“One possible reason… when the economy grew, it encouraged more investment in corporations, but in many instances, investments take a long time (four years) to show up in ways that boost the market. “
Putting aside that the economy (“it”) doesn’t do anything, rather that individuals decide about investment; the Angry Bear’s analysis fails.
Stock prices adhere to the one true invariant law of economics — The Law of Prices — the same as any produce. The Law of Prices holds that the winning bids of demand in the face of supply set the price.
Stock prices rise only when more money is bet relative to the float (money increases the float remains constant or falls; money increases faster than an increasing float).
Had the Angry Bear looked at something relevant like real disposible personal income rather than something meaningless as GDP, he or she could be onto something.
When real DPI rises, more money is available directly and indirectly for margin leverage.
Also, the Angry Bear fails to account for what percentage of the S&P 500 gets held by foreigners in any year and thus fails to account for foreign real DPI growth.
In short, the Angry Bear fails and his or her analysis amounts to foolery.
I really appreciate your post and you explain each and every point very well.Thanks for sharing this information.And I’ll love to read your next post too.
Regards
Gold Silver Tips
I really appreciate your post and you explain each and every point very well.Thanks for sharing this information.And Ill love to read your next post too.
Regards
Gold Silver Tips
I like to spend my free time by scaning various internet recourses. Today I came across your site and I located it is as one of the best free resources available! Well done! Keep on this quality!
iso 9000