BUNK 37
OIL AND STOCKS SEESAW
Somme bunk oes in and out of style. This one was red-hot in recent years. But as I write in 2010, you don’t hear it as much. It will be back—probably the next time oil prices rise a lot, fast—or rise for a long time.
The fairly universal belief is high oil prices are bad for stocks. When oil is up, stocks go down—they’re negatively correlated. And to get higher stock prices, you want lower oil prices. But it’s bunk. (Interestingly, this is one you hear more when oil is high. When oil drops a lot in price, folks don’t say, “Phew! Everything is great now!”) Long term, they mean nothing to each other and one isn’t predictive for the other—that’s provable.

A Dollar Spent Is a Dollar Spent

The thinking is fairly intuitive (which by now you know to spot right away as a likely bad sign) and goes: We’re too oil dependent! Higher oil makes life more expensive—we need oil just to go to the grocery store or work, so we spend more on gas, heating, or air conditioning and less on other stuff. This slows the economy and means lower earnings and profits for firms as money gets sucked away from them and into our gas tanks—and it’s all ultimately bad for stocks.
First, borrow for a moment from Part 5 and think globally, and this belief starts seeming silly—even before the data crunching and checking history (which devastates this myth, as you’ll soon see). See it this way: A dollar spent in the global economy is a dollar spent in the global economy. If you spend $100 a month, the global economy doesn’t care if it’s on gas, tennis shoes, tax advice, or pet rocks. If you spend $30 on gas and $70 on everything else—that’s $100. But if gas rises, costing you $40, you have just $60 for everything else—except it’s still $100 spent. Maybe oil firms profit marginally more (which may be great for their stocks), and other firms less. Or maybe not. Maybe oil prices are being driven higher by transport costs resulting from huge demand for other gadgets. Or maybe those other firms instead gain via innovation, cost cuts, or something else giving them more profits—happens all the time. Capitalism is awesomely adaptive! But ignore that economic reality for a moment.

Long Term—No Correlation

Folks fear higher gas stalls the economy and dings stocks. So when oil goes up, stocks go down—that’s the fear. And lower oil means folks have more money to spend (so they think), so that’s supposedly better for stocks. But recent history contradicts that. Oil fell in 2001 when stocks dropped and we had a recession. And oil rose in 2003, 2004, 2005, all the way through 2007—right alongside stocks! When stocks started falling in late 2007, oil kept rising. Oil rose while stocks fell all the way through June 2008—just like the myth! Then after June, they both fell through year end, then both rose for most of 2009.
So are oil and stocks—gasp—positively correlated? Sure, sometimes. Sometimes they’re negatively correlated. Sometimes they’re nothing. But over long periods? There is nothing there. And it’s easy to check the data yourself—straight debunkery—there’s vast, free, easy-to-find prices for both stocks and oil. Figure 37.1 shows monthly returns for both US stocks and oil.
The correlation coefficient (which you can easily get on two data sets in Excel—if that intimidates, ask a teenager to show you how) is a number between +1 and -1 that shows how much two variables move together. The higher the number, the more positively correlated they are—zigging together at the same time and to the same degree. A number close to -1 means they’re negatively correlated—they zigzag the way the myth presumes oil and stocks do. A number close to zero means the two have no relation—one zigs and the other broccolis.
Figure 37.1 Oil Versus Stocks—No Material Correlation
Source: Global Financial Data, Inc., West Texas Intermediate Oil Price (US$/Barrel), S&P 500 total return from 12/31/1979 to 12/31/2009.
041
Since 1980, the correlation between oil and US stocks is -0.02. Meaningless—the two aren’t related in the long term at all. (By the way, you can do the same thing with oil and UK, German, Japanese, and world stocks, and get the same meaningful lack of correlation.) Now look at the R-squared—which is the correlation coefficient, squared. In statistics, that tells us how much of one variable’s movement can be explained by the other. Here, it’s 0 percent. Said another way, over long periods, anything and everything but oil contributes to the stock market’s movement. Anything and everything but! Repeat after me: “Anything and everything but.”
Can oil prices impact certain industries and firms more directly? Sure! Energy firms for a start. Particularly if they’re drilling for, refining, and/or selling oil or oil products. But overall, oil and stocks aren’t correlated—positively or negatively.

Short-Lived Correlation

That’s longer term. Over shorter periods, oil and stocks can have short spurts of intense positive or negative correlation. But any two bizarre variables can inexplicably move together (or oppositely) for short spurts. If you take baseball statistics (or any other totally non-related activity that has regular outputs) and run them against stock prices, you can find periods when they’re positively correlated and others negatively. But they’re meaningless and fleeting. I assure you, correlating changes in Nigerian rainfall with the Nasdaq 100 will work at some point in time. Means nothing.
Figure 37.2 Correlation Between Oil Prices and the S&P 500—There Isn’t One
Source: Global Financial Data, Inc., West Texas Intermediate Oil Price (US$/Barrel), S&P 500 total return from 12/31/1979 to 12/31/2009.
042
Figure 37.2 shows 12-month rolling correlations between oil and stocks—sharp, jagged, short-lived spikes up and down. Above the line means they’re positively correlated, below means negatively. There’s a long-ish period of varying degrees of negative correlation from 1987 through 1992—but even then it was wildly variable. Other than that, there’s no predictability except for the pure randomness.
What’s interesting is folks commonly commit a slew of cognitive errors when thinking about oil and stocks. Times when stocks and oil zigzag (below the line) confirm their belief. They say, “Yep! This proves it! They’re negatively correlated.” Or they seek out periods when oil and stocks were negatively correlated, saying, “See! That period from 1987 to 1992 proves it,” and conveniently block out the other periods—a cognitive error we’ve discussed elsewhere called confirmation bias. Periods when oil and stocks don’t zigzag (above the line), they reframe. They say, “Sure, oil and stocks don’t always move together. You have to look at longer time periods!” Fine—we just did. Longer term, there’s no correlation.

Supply and Demand

The fact is, oil prices are determined by supply and demand—same as stocks. Sometimes, higher prices are reflective of higher demand driven by a growing economy. Then, it would be perfectly normal to see oil and stocks rising together—the same general forces driving oil higher can in part drive stocks higher, and vice versa.
Or oil might be spiking because of some unforeseen supply disruption. That might not necessarily be great for the economy or stocks—and they could zigzag. Or maybe the disruption is short-term in nature, and stocks know it, so they do whatever they were going to do regardless of what oil is doing. Often, oil and stocks move together when the economy is suddenly stronger than expected, which is good for stocks and causes unexpected economic demand that ripples to energy usage. And sometimes the reverse occurs. But knowing where oil is going won’t tell you where stocks are headed—provably.
Oil and stocks have completely, wholly, and inherently separate supply drivers. And they have many, many demand drivers—some overlap, many don’t. Our economy is intensely complex—there are just no iron-clad “when X is up, sell stocks, and when X is down, buy stocks” rules that work long term.
..................Content has been hidden....................

You can't read the all page of ebook, please click here login for view all page.
Reset
3.235.182.206