The 'Past Performance' Contradiction

Includes: INTC, KO, PEP
by: The Investing Engineer

Past performance is not indicative of future results

If you've been doing any sort of investing for anywhere from 30 minutes to 30 years, you've heard this saying. I find it used most commonly to express something like:

Here's my analysis of company ABC based on their growth and valuation for the past 10 years, and why I think they will continue to perform. While you're buying this stock based on my recommendation, keep in mind that past performance is not indicative of future results.

I hope the contradiction is clear. It's impossible to predict the future, unless you're Nouriel Roubini, or you have some sort of "insider" knowledge that you can "trade" on. Yet, the past is all we have. The best we can do is look at it, whether it's historical dividend growth data from the last 30 years or the just released 3Q earnings report, and decide if all that knowledge we gathered will indicate future results. And that's what we do; we analyze the past performance, give legitimate reasons why the company will continue to perform, and then contradict ourselves by saying that in fact, past performance can't tell you anything about future performance, and therefore everything you read means nothing.

I'm writing this article to tell you that the past is an excellent source of predicting future results, and that there's no reason for buying into this contradiction. The saying is so often used that it is mistakenly believed to be true when in fact it is not. When Apple (NASDAQ:AAPL) sells a trillion iPhones and iPads and becomes the world's most valuable company, does anyone actually believe that previous sales can't be used to predict sales on the newest iPhone and iPad? Of course not! The iPhone 5S or iPhone 6 or whatever it may be will sell like, well, like iPhones. There will be ridiculous lines that people are camping out in and profits will soar and no one will be surprised when Apple posts huge quarterly profits.

This saying can't always be wrong, right? People are always buying companies that are doing poorly despite great past performance. Then, when they're wondering what happened, they think:

I should have known better. I bought this company because of past performance, but that's not indicative of future results.

The flaw isn't in the person buying based on past performance. The flaw is that the saying is too general to be used as any sort of guide, because all decisions are made on past performance. I suggest that it should be changed to this:

Past performance is not indicative of future results, given that the conditions that led to the past performance have changed

What I mean is that you have to look at what catalysts caused the past performance, rather than the performance itself, and see if those catalysts are still valid. If they are, past performance should tell you what future performance will look like. Going back to my Apple example, after they made boatloads of cash on the first iPhone, could I have predicted that they would make even more on the iPhone 3G? My first task would be to see what caused the huge sales of the iPhone. These catalysts for sales included a strong demand for innovative smartphones, little to no serious competition (only the Blackberry at that point, but that was aimed at the business person, not the regular consumer) and a cult following for Apple products, built on the foundation of the iPod, iTunes and the iMac. When the iPhone 3G was released, were these catalysts still there? Yes;

  • The demand for innovative smartphones had grown, and the 3G promised more innovation than its predecessor (App Store, 3G data, assisted GPS, etc.)
  • There was still little competition as the iPhone 3G was released in July 2008, while the first Android phone came 3 months later, in October
  • The cult following for Apple products had only grown as a result of the huge success of the first iPhone

As a result, the 3G sales beat the original, and they've continued to rise with each new release. The original iPhone sold 6.1 million units total. The iPhone 5 sold over 5 million units in the first three days. What's changed with the environment? Only competition, resulting from the strong growth of Android products. This has been offset by incredible growth in the number of smartphone users. Going back to the original question, can I predict that the next iPhone and iPad and the new iPad mini will make Apple lots of money, causing their stock price to increase? All of the catalysts leading to their original success are still there in some capacity, so I can say yes, without having to back myself up with a statement about past performance and future results.

What about future results for 10 or 20 years from now? It depends on the catalysts. If they're still there and Apple is still keeping up with them, yes, past performance will be indicative of future results. If sometime down the road the environment changes and smartphones are replaced with some newer technology that Apple doesn't innovate or keep up with, then past performance won't be able to help them. But past performance shows that they've been ahead of the technology curve where it matters, so it gives them a good chance of repeating that for the next few years.

Technology is tricky because it is always changing so let me give an example of how past performance is a gray area when determining future results. The most relevant company for this right now is Intel (NASDAQ:INTC). For the past few decades, their processors were in high demand. Everyone wanted software to run faster on their PCs and the increasing demands of software led to increasing demands for hardware, CPUs in particular. While PC growth was up, you could easily use Intel's past performance to predict future results.

The market is changing now. PC sales are slowly dropping while smartphone and tablet sales are skyrocketing. If you've been following Intel's stock (or not), you've seen it get hammered over the last few months. I wrote an article about this last month, and what I found from reading through the comments is that whether or not you sold Intel, or used the drop to buy more, your reasoning behind your move was based on past performance.

Intel clearly missed the boat on getting in on the smartphone market, and sales of their processors have dropped. Past performance is no longer indicative of future results because the conditions (demand for PC processors) that led to that past performance have changed. On the other hand, a CPU is a CPU, and Intel is the best at making them. Just because they don't have one in a popular smartphone yet doesn't mean that they're stuck selling obsolete products. With this viewpoint, the conditions for Intel's performance have only changed slightly (demand for small CPUs instead of the larger CPUs that Intel traditionally makes), so given enough time, past performance will again indicate future results.

These examples aren't groundbreaking economics theories. They're simply supply and demand. As long as a company is supplying what consumers are demanding, without any significant changes to the market, past performance will indicate future results. If the market changes and all of the sudden a company is supplying a product that fewer or no people are demanding, past performance won't tell you too much.

As a final example, I'd like to take a look at a company that's been supplying a product for over 100 years, where the catalysts that caused it to be successful haven't changed. This company is The Coca-Cola Company (NYSE:KO), and the product is, of course, Coca-Cola, introduced to the world in 1886. Here are the catalysts that caused its success:

  • People want a tasty beverage
  • They don't want to pay a lot for it

What has changed since 1886 that would cause this product to stop selling? Not much. There are health concerns, but Coca-Cola responded to this market condition by introducing Diet Coke and Coke Zero, plus thousands of other beverages in their repertoire. For a time the company was threatened by Pepsico (NYSE:PEP), but these catalysts stayed the same and Coca-Cola has effectively won that battle. In terms of past performance, this tells me that not only will the company have good future results 5 years down the road, but if you look 20 or 30 years in the future, past performance will still be a good indication of how the company is doing. The fact that the flagship product has over 100 years of past performance is very indicative of future results!

After writing this article, it seems obvious to me that the whole "past performance" thing is either too generalized or just wrong. What seems strange is that the saying reads more like a disclaimer than a tenet of investing. What comes to mind is the warning on a package of regular M&Ms, "may contain peanuts". The product isn't supposed to have peanuts in it, but if they mess up and someone with a severe allergy eats a peanut M&M in a package of regular M&Ms and dies, Mars Inc. can't say they didn't warn you. In the same sense, if a mutual fund manager convinces you to buy in, based on statistics of beating the market for the past 10 years, you can't hold him responsible for a loss because you were warned that past performance is not indicative of future results. Well, I'm telling you to take that saying with a grain of salt. If you study the market and think the reasons for past performance are still there, by all means use that to predict future results.

Please leave comments below. I'd like to hear what you think, especially if you find yourself protecting your stock analysis with this saying. If you think I'm completely wrong, don't be afraid to tell me. I have thick skin. Just be sure to back it up with a good counterexample.

Disclosure: I am long KO, PEP. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.

About this article:

Author payment: $35 + $0.01/page view. Authors of PRO articles receive a minimum guaranteed payment of $150-500. Become a contributor »
Problem with this article? Please tell us. Disagree with this article? .