A good company is a wonderful thing to own, but investors often forget to think about how much they paid for their favorite AI stock. No, I am not talking about the actual stock price, but what that price is in relation to the company’s earnings.

One key ratio to consider when looking at a stock is its price-to-earnings ratio, or P/E ratio. The P/E ratio is basically earnings multiple on a stock, and many investors like us use it to evaluate how “expensive” a stock is.

Suppose I put in front of you a little widget that kind of looks like a smartphone. And that widget prints out $1 for you every year, and you can take it to the bank and deposit it. It would keep doing so indefinitely. How much would you pay for it? If you paid $10 for it, then the widget’s P/E ratio is 10.

If you paid $20 for it, you probably bought it at an expensive price. But if I tell you that it produces $1 this year, and $1.1 the next, and $1.2 in the next… and so on, $20 may suddenly look like a good price. So whether a P/E of 10 is expensive or a bargain depends on the business fundamental and outlook.

In the stock market, the average P/E ratio varies by industry. It is common for technology companies to trade at high P/E ratios because investors are expecting innovations and high growth rates. It is also common for banks to trade at lower P/E ratios because, let’s face it – banks are less fascinating than graphic chips.

For most people just looking to conceptually understand what P/E ratio is, you can stop reading here. If you want to dive into more technical details or you just really like the way I write (in that case we can be friends), you can keep reading.

More technical fun stuff

Suppose you love Nvidia. Suppose you also like Berkshire Hathaway. As of February 9, 2024, a blissful Sunday evening as I am writing this blog – Nvidia stock is sitting at all-time high at $721.33 per share. Berkshire Hathaway’s A class stock, is sitting at a whopping $599,090 per share. It is the most expensive stock right now. Okay, maybe using the A share is a little unfair because the bid-ask spread is super high, but it helps drive home a point.

You might intuitively think that Berkshire is 830 times more expensive than Nvidia. Yahoo Finance would agree with you, as it thinks Nvidia is sitting near fair value, where Berkshire is over-valued. But if you pay attention to the P/E ratio of each company, you will see that Nvidia is trading at a P/E ratio of 95.29 and Berkshire’s A share is trading at a P/E ratio of 11.38. This means if Nvidia is just cranking out chips as it normally does and keep its earnings constant forever, it will take a shareholder 95 years to recuperate the cost of the stock. Meanwhile, if things remained constant at Berkshire, it would take an investor 11.38 years to make the money back. Now I don’t know if advances in medicine will allow an everyday investor to wait 95 years just to get his money back. So apparently something else is at play here – people are expecting different growth rates for Nvidia and Berkshire. This just shows you that a higher price tag doesn’t necessarily mean a more expensive stock. You should look at the P/E ratio and think – am I paying this multiple, or can I invest my money else where?

How is it calculated?

To fully understand a financial metric, I like to start from the mathematical calculation. I hope it’s the same for you if you have read this far.

Let’s start from earnings. To get earnings, you take a company’s top-line revenue, subtract all the expenses it took to run the company that year, and you have the company’s bottom-line earnings.

You then take the bottom-line earnings and divide it by the number of the shares held by all shareholders. This is the amount of earnings that is attributed to each share of ownership in the company, also known as earnings-per-share, or EPS.

Lastly, you take the current price of each share of common stock, and divide it by EPS. You will have the price-to-earnings ratio, which is the multiple of the price of each share on the earnings attributable to that share.

Why not use EBITDA instead?

You may know that some of the expenses reported in a given year are not cash expenses. In other words, they are not expenses paid out of pocket in that year. An example of this is depreciation.

You may also have heard of an earnings alternative called EBITDA, which stands for “earnings before interest, taxes, depreciation, and amortization”. EBITDA is a metric often used in pricing private business transactions, and some public companies public this figure in their annual reports also. They often call it “non-GAAP earnings” and put adjustments in them to show what the results of a normal year of operation would look like. It tells you about a company’s earnings, assuming everything remains status-quo, and without consideration of taxes and cost of debt.

There are a few problems of using solely EBITDA in evaluating a public company. First, not all companies report EBITDA in their annual reports; investors may have to pay a platform or calculate its EBITDA, which leaves more room for error and/or take more time.

Second, each company adjusts its EBITDA differently. There is a lot of room for subjectivity when adjusting this figure. In a private business transaction (mergers and acquisition), the adjusted EBITDA is calculated by a team of accounting experts through a process called “Quality of Earnings”. A significant amount of non-public information is considered through meetings the company’s senior management team. Such calculations are very expensive to conduct, and 100 teams might come up with 100 different adjusted figures. I spent 7 years doing this in different industries and handled transactions both small and large.

Finally, much of the excluded expenses, such as depreciation, are real expenses to run the business – even though they are non-cash. Depreciation is essentially the cost to use a piece of equipment in a given year. That piece of equipment might be purchased and placed in service years ago. A company will estimate the cost to use that equipment, and do its best to match the cost to use it with the revenue it earned. In accounting, this is called the “Matching Principle” and it is one of the first things aspiring CPA’s would learn. At the end of the day, equipment need to be replaced.

EBITDA may be a good number to consider, but you should take it with a grain of salt. For the reasons above and to make companies comparable, you may as well just look at the regular P/E ratio.

Disclaimer – the content of this blog is intended for educational purposes only. We do not know where NVDA and BRK.A will be a day from now, much less a year from now. We own NVDA in our personal accounts as well as in some managed accounts. We also own BRK.B in our personal accounts as well as managed accounts for clients. None of the stocks mentioned are intended as financial advice. Also, my understanding and opinions may be flawed. Because they are – just a person’s understanding and opinions. There are people much smarter than me.

Screenshots from Yahoo Finance