Ratios are commonly analyzed by investment professional and retail investors alike. We utilize them as a simplified way of judging performance or valuation for a company. One of the most frequently analyzed ratios is the P/E (price to earnings).
For the purposes of this study the price, the reported share price on the date the study, was commissioned. We have several options for use of "earnings" in the denominator of the ratio. Earnings as reported for the trailing 4 quarters of SEC issued filings, excluding any extraordinary accounting items, is used in this study. This is commonly referred to as trailing P/E.
Graham and Dodd in 1936 were one of the first analysts to publish the use of Price to Earnings as a screening tool in analyzing stocks. For decades before their publishing though, investors have been concerned about the "price" they are paying for some income stream; "earnings." This multiple, as it is called, will typically dictate how expensive a stock is and provide for a relative comparison against other securities.
We use a simple and objective method for testing the predictability of the P/E ratio. Understanding our process and the rules it abides by is important in interpreting the following results.
Testing the performance of a ranking system by P/E ratio
For the time period January 1st, 1999 to July 25, 2013
We begin by pulling every company in the S&P 500 (^GSPC). Those companies are then put in order from lowest P/E ratio to highest. They are divided into 10 separate buckets, each bucket having 50 companies in them. We create a fictional portfolio that is equally invested in those 50 companies, or 10 buckets, and track the performance of those 10 buckets for a period of 30 days. At the end of the 30 days the companies and their new P/E ratios are reshuffled into their new corresponding buckets. The 10 bucket portfolios are created again. This process is done every 4 weeks for a period of 13 years and the performance of bucket 1 through bucket 10 is tracked. The only difference between the buckets is the average P/E ratio within each.
To judge performance, we calculate the outperformance of one bucket versus the next. If the performance is statistically significant between each bucket, then we can determine that this ratio is important in making an investment decision. Throughout this research article you will see a colorful bar chart that display this process. (The first red bar will always indicate the S&P 500 performance.)
To remain objective and free from common data analysis issues, there are several restrictions on our quantitative analysis:
- $5 minimum share price
- 4 week rebalance frequency
- Time Period Studied: January 1, 1999-July 25, 2013
**Data analysis utilized in this article takes the inverse of the P/E ratio, called the earnings yield. For data calculation purposes it creates a smoother output, as it deals with companies with negative earnings more effectively.
- The aggregate performance for investing based solely on Price to Earnings Ratios is depicted above.
- The study determines that P/E ratios are predictive in excess returns and should certainly be included in consideration for stock investments.
- The payoff for this ratio is inverse linear; the lower the ratio is, the better.
What you Need to Know:
A portfolio consisting of the lowest Price-to-Earnings stocks over the last 13 years has returned 12.1% on average per year, outperforming the S&P 500 by 9.9%. This is considerable outperformance from just making one simple choice in your investments: Buy low P/E companies.
Regardless of your outlook for the market as a whole, we also found significance for the Low P/E model across bull and bear markets. Since the beginning of our study for this article, there have been two of each.
Let's take a look at the Low P/E Model now and see what the recommendations are. Below is a table of 20 companies in the S&P 500 (SPY) with the lowest P/E ratios.
"But wait," you say, "those P/Es don't make sense with your ranking." True, you are used to seeing P/Es listed as we have them here, quoted from Yahoo Finance. But recall earlier where we mentioned our methodology for calculating true operating earnings. We don't count "extraordinary items" as reports by the company, because we don't believe it to be indicative of true performance. Right or wrong, that's our method in this case.
A Quick Profile of a Few Low P/E Stocks:
Ford Motor Company (F): The Detroit automaker has fared significantly better than the municipality in which it is headquartered with a P/E below 12, significantly lower than historical ratios for auto manufacturers, normally around 14x. The Street is expecting little to no growth. Interestingly enough, though, every time auto sales numbers come out, they are caught off guard with a happy surprise, auto sales are doing quite well. Our estimate for EPS growth next year is 17% (2013 to 2014 FY). We wouldn't call that a no-growth environment.
Yahoo! Inc. (YHOO): There is no secret that this is distressed investing at its finest. Operationally, Yahoo is trying to find its place in the land of internet services and search providers. It may never be able to compete with Google (GOOG) on search, it may never compete with eBay (EBAY) in the online shopping experience, but it is one of the most visited landing pages across the entire web. There is certainly an intangible value to that breadth; even though it does not have a cornerstone product in the market, this quest inherently adds value. Let's not forget that through all this speculation, the company still produces nearly $5 billion in revenue each year -- a huge accomplishment.
Marathon Petroleum (MPC): Marathon is a predictable story. It transports and refines one of the most necessary products in Americans' daily lives: oil. Take a look at its top line sales growth over the last few years (20%+), and couple that with the aggressive share buyback started in 2012 (3% of outstanding shares) and you can see how it will deliver strong bottom line earnings growth. Typically, an investor would be concerned about a declining business when looking at low multiple stocks, but I believe Wall Street has missed something with MPC. The ratio is even 30% less than competitors Chevron (CVX) and Exxon (XOM).
There have been some in-depth stories written on Seeking Alpha regarding these profiled companies. Our goal here was not to tell the top-to-bottom story on each business, but simply to profile the necessity for looking at value in the low P/E category. It is certainly a factor that has predictive power in future returns, and one you should add to your analysis toolkit.
Additional disclosure: Business relationship disclosure: AlphaStreetResearch is a team of Independent Research Analysts and Providers. This article was written by Hunter Orr, our Director of Research. We did not receive compensation for this article (other than from Seeking Alpha), and we have no business relationship with any company whose stock is mentioned in this article.