Understanding the price-earnings (P/E) ratio, and a few of its limitations

The price-earnings ratio (P/E Ratio) is a ratio used to value a company that measures its current share price relative to its per-share earnings (profit per share). The ratio can be calculated as:

Market Value per Share (Share Price) / Earnings per Share (EPS)

Essentially, the P/E ratio is meant to show how much investors are willing to pay per dollar of earnings. In general, a high P/E suggests that investors are expecting higher earnings growth in the future compared to companies with a lower P/E.

The first question many have when seeing the above formula is “what happens to the P/E ratio when a company has negative earnings?”. This is an excellent question! Although companies with negative earnings can technically have negative P/E ratios, the standard practice is to display ‘N/A ‘, or show no ratio for that quarter’s P/E. However, a single quarter of negative earnings can skew the P/E for the next 3 quarters (when earnings are later positive and P/E is again displayed), since the metric adds the trailing 4 quarters of earnings; this means that a negative quarter can make the denominator (EPS) of the above equation very small, sending the P/E ratio through the roof. I’ll highlight a couple current examples below when I split out the companies with the highest P/E ratios, which may make this point easier to follow.

P/E ratio by sector (averages can be deceiving)

The below chart shows us the various average P/E ratios across sectors, for the largest ~500 publicly traded companies. At first glance, looks like Materials and Health Care companies have sky high P/E ratios:

average_pe_sector

However, if we bring in the standard deviation of the P/E values in each sector, we see that Materials and Health Care both have very high standard deviations, meaning some companies likely have unreasonably high P/E ratios (I’ll get into the cause of some of these outliers below).

 

p_e_with_std_dev

So, since our averages are skewed due to outliers, the median P/E would probably be a better metric to use:

median_pe_sector

As we can see, P/E varies pretty significantly between the different sectors, with Real Estate, Energy, and Health Care topping the list. It’s worth noting that P/E ratios can vary across sectors due to differing ways in which companies earn money, and differing timelines in which that money is earned; as a result, P/E is best used as a comparative tool when considering investing in companies within the same sector.

That said, I did pull the top 10 companies with the largest P/Es below, in an attempt to highlight some of the reasons a company’s P/E can be several standard deviations above its sector average.

Which companies trade at the highest P/E ratios? (understanding unreasonably high P/E ratios)

pe_company_top_10-1

At first glance, one may look at Glaxosmithkline or Rio Tinto and think investors must either be crazy or expect a tremendous amount of future earnings growth, since the companies are trading at 1.2-3k multiples of earnings.  However, a deeper look shows that Glaxosmithkline had negative net income in 2 of the last 4 quarters, while both Rio Tinto and Kinder Morgan also had a negative net income quarter at some point in the last year. (note this is the example I mentioned I’d cover at the beginning of this post)

Basically, companies with sky high P/E fall into these basic groups:

  1.  The company had some negative earnings in the trailing four quarters, but not in the most recent quarter (if the earnings were negative in the most recent quarter, the general practice is for no P/E metric to be shown)
  2. The company has been squeaking out very small profit margins (and consequently a small EPS), while their share price is reasonably high
  3. The company is actually expected to grow its earnings by a large amount so investors are happy to buy large multiples of current earnings (this is the assumed case with Netfllix, Salesforce, and Amazon)
  4. That’s not all! Debt, leverage, earnings accuracy, one-time purchases, and more can impact P/E ratios as well

The takeaway from this is that P/E ratios should definitely be paired with other metrics to get the complete picture. Many financial metrics can skew for a variety of reasons, and this is helps illustrate how important it is to avoid simply looking at a single financial metric and pulling the trigger on an investment. 

Also, for those interested I pulled the 10 companies trading at the lowest P/E multiples (note they mostly fall into sectors that bottomed out our list above)

Which companies trade at the lowest earnings multiples? (are expected to grow the least) 

pe_company_bottom_10

 

The data

I used Fidelity’s stock screener to grab the data, though in hindsight would be easier to pull directly from Yahoo Finance. I then used Python to clean with Plot.ly to make the charts. Those interested in the iPython notebook can view it here.