We all know a bottle of Pepsi can substitute for a bottle of Coke, especially when you are thirsty. If you ask a fanatic, you may get a different answer: a Pepsi fanatic would prefer a plain glass of water to Coke if Pepsi is not available and vice versa. Based on the person you ask, you may get a different answer whether the two beverage behemoths compete.

But, we took a different approach and looked at how investors – people who buy stocks in Pepsi and Coke – think about the two companies. If two companies are in the same market and compete with each other, it is likely that some news that affects the industry is going to affect both the companies. So, if the news is good, both stocks are likely to go up and if the news is bad, both are going to sink.

To illustrate this effect, take a look at the Bank of America and Citibank. We plotted the weekly percent change in the stock price of these two companies for the last 20 years.

As we expect, we can see that regression line slopes upwards. When BAC drops, we see Citi also drops. When Citi climbs, we see that BAC also climbs.

How does the corresponding plot for Pepsi and Coke look like? As you may guess, we see something surprising.

Apart from the cluster on the regression line, you see two small clusters at the bottom and the left of the graph. What do these clusters say? The cluster at the left is when the Pepsi stock has tanked, but leaves the coke stock price unchanged. The other cluster is when Coke stock price has sunk but Pepsi is unscathed. If they really competed, we would have seen points in the graph on the far left corner of the regression line and all along.

We would love to hear from you on what you think about this anomaly or if you have any interesting explanations. Drop us a note!