What are PE Ratios and Do They Matter?

I recently read Benjamin Graham’s book, “The Intelligent Investor.” In the book, he points out that a lot of metrics used to value stocks have little inherent meaning. It got me thinking about some “common wisdom” that goes flying around the internet, and made me want to dig more deeply into some details.

Specifically, I’ll look at the PE ratio and what it means by looking at the stock of a well-known company. Also, whenever I write “currently,” I mean as of Friday, April 25, 2025. If I update this anything at a later point in time, I will note the new dates.

But first, a disclaimer!

The information on this web page is for general informational purposes only and should not be considered personalized financial advice. I am not a financial analyst or professional; if you are seeking investment advice, then consult with a qualified financial professional before making investment decisions. I don’t directly hold any shares of Oracle of Palantir, the two companies discussed here, but I may own them through an ETF or other managed fund. I’m making a best-effort to get all numbers correct, but I am manually copying them from various sources and may make mistakes.

Okay, now let’s talk about the PE ratio: what is it, how is it used, and does it have any meaning?

The PE ratio is the ratio of a company’s price per stock to its earnings per stock (which is just the net earnings divided by the number of shares). Most companies only release earnings information four times a year, and the PE ratio takes the last four earnings statements to calculate PE over a year’s worth of earnings. The filings of companies listed on the U.S. exchanges are available on the SEC’s website.

Common knowledge says that buying something cheap is better than buying the same thing for more money, so we want to buy stocks that have a low price relative to their earnings. Why? Because we would like the company to pay us back for owning their shares.

If you purchase a stock with a crazy PE ratio of 1/1000 (meaning that they earn $1000 of profit for every dollar of share cost), the company could theoretically compensate you for the entire cost of the share by distributing 1/1000 of their yearly earnings as a dividend to all shareholders. In fact, they could easily pay that much every year and you would become filthy rich in no time. Everyone else would notice too, and soon the buying frenzy would push the share price higher. That would increase the PE ratio as well, and soon the world would once again be in balance.

A stock with a crazy PE ratio of 1000/1 would have to give all of its earnings to its shareholders every year for 1000 years to compensate you for the cost of a share. Hypothetically, the inverse of the previous scenario should occur: people will sell the shares because their price is outrageous, and eventually the PE will go down until it reaches something reasonable.

So why do people buy high PE stocks? The PE ratio is necessarily backwards looking (since we don’t know future earnings yet), which is a problem as a metric since we live our lives going forwards. For a stable company with constant growth, the current PE is a good predictor of the future PE; if the current PE is too high, then there is a good chance that the future PE will be too high as well. In that case, using the PE as an indicator of the reasonable value of a stock makes sense.

But what about a new company? For example, what about Palantir (PLTR), which currently closed with a share price of $112.78 and a PE ratio of 596.40?

The company’s revenues in the four quarters of 2024 were, in millions, $634, $678, $725, and $828. As people on internet would say, “number go up.” If that growth trend holds, then the shareholders will be swimming in cash in a few years.

But wait – do we expect the growth to continue at the same rate, or will it go up by just $200 million a year? Will the rate accelerate? Or will it slow?

PE ratios can get crazy when the future earnings of a company are expected to be much greater than past earnings because the growth is so uncertain. Maybe the company is finally landing its first contracts, or perhaps it has finished paying for an expensive facility that will allow it to print money. Whatever the case, the shareholders of Palantir are expecting the earnings to go up like a rocket.

I can’t see the future, so I can’t say how the story with Palantir will play out, but we can look backwards at another stock: Oracle Corporation (ORCL), founded in 1977. Oracle started by selling database software, and has expanded to cover different slices of the enterprise software pie since its simpler origins.

Today, Oracle’s stock closed at $138.49, and it really took off in the late 1990s when the Internet was becoming the next hot thing. At the end of 1995, if you had correctly predicted that Oracle would survive and emerge as a leading software company then you could have gotten a 13.43% yearly return on investment. However, buying its stock wasn’t always a good deal: there have been plenty of years where its price plummeted, and a couple of years where your returns would have been downright lousy.

Here’s a table with a bunch of numbers going back to 1995:

Some notes:

• There were two 2:1 stock splits in 2000, and 3:2 stock splits in 1999, 1997, and 1996. The price (from marketwatch.com) should be adjusted for those.
• The EPS (earning per share) is the value taken after the quarterly report in November¹.
• Yearly Growth is the growth of the closing stock price from the last day of the prior year.
• CAGR is Cumulative Average Growth Rate, or the average yearly rate of return. I’ve assumed that dividends are reinvested in the stock, but I’ve calculated them as reinvested on the last day of the year to simplify my calculations.

Taking a look at the annual growth rate, it’s pretty good (at least up until today – the stock price could fall off a cliff tomorrow for all we know). The stock has gone through a few rough years. The worst drop, in 2001, was more than 50%. That didn’t quite wipe out the gains of the prior two years when it jumped from $7.19 to $29.06, but if you bought in 1999 or 2000 you would be kicking yourself – and your spouse might be kicking you out of the house. The 10-year treasury yields at the end of 1999 were 6.45%, so an investor who bought before 1999 would have been better of selling their stock, paying taxes on their gains, and investing the rest in federally tax-free treasuries. In the time since the 1990s, yields on U.S. treasuries have been terrible, but if they roar back to life one day it’s a good idea to remember that they are a safe alternative to risky stocks.

So this has been a great buy-and-hold investment for most years. For the off years, is the PE ratio a warning? Yes, sort of, maybe.

Notice that the PE ratio was high in 1999 when the EPS was twenty cents. The next year, the EPS jumped to $1.10 and the PE fell to 26.42 (because the earnings had risen more than five times – the share price was about the same). So, in fact, those who predicted the jump in earnings and bought or held through the high PE were not wrong. The people who predicted that earnings would remain at that level made a mistake though, as the EPS dropped to forty cents in the next year. That caused a plummet in the stock price as people were not willing to hold as they waited for the EPS to rise again. It wouldn’t be until 2008 that Oracle’s earnings per share again rose above a dollar.

Should you have sold after the crash, then? Probably not. You would have lost another 21.8% if you held through 2002 after the 52% beheading in 2001, but the annual growth rate since then has ended up being more than 10% if you held on through today. Selling may have made sense, either for tax purposes or if you couldn’t afford any more losses.

Should you have sold at the top? Yes. To reiterate, if you sold anywhere in 1999 or 2000 you could have invested in tax free treasuries and been happy.

Should you have bought in 1999 or 2000? No. The high PE was a warning sign in 1999, but the PE in 2000 looked fine after the huge jump in earnings. To realize the underlying risk, you would have had to question the sustainability of such growth. That would also require the fortitude to resist the urge to buy something because it is going up.

The real warning here is the PE ratios look backwards. Sometimes that means they underestimate the performance of a company in the next year, such as going from 1999 to 2000. A PE ratio of 20 in 1999 would have made Oracle a steal. Sometimes that means they overestimate the performance of a company, such as going from 2000 to 2001. The PE ratio of 26 in 2000 was misleading. Without some research into the business itself, just looking at the PE ratio will tell you nothing.

All things considered, just putting in a bit of money every year and ignoring everything else is a good strategy. In fact, this is the market index strategy that Benjamin Graham recommends for people who don’t want to read through detailed reports from every company in their portfolio.