P/E ratio: price-to-earnings formula, limits and traps
In short
The P/E ratio (price-to-earnings) divides a share’s price by its earnings per share: at $40 with $2 of EPS the P/E is 20, so you pay $20 for every dollar of annual profit. Trailing P/E uses the last 12 months of earnings, forward P/E uses estimates. The inverse is the earnings yield: a P/E of 20 means 5%. A low P/E can be a trap, because what matters is how much earnings will grow and for how long.
In December 1972 the Nifty Fifty, around fifty large, fast-growing American companies, traded at an average of 41.9 times earnings, against 18.9 for the S&P 500. Polaroid reached 94.8. Philip Morris stopped at 24.
In 1998 Jeremy Siegel, a professor at the Wharton School, ran the numbers with hindsight: given what the stocks returned through August 1998, what P/E did each one deserve in 1972? For Philip Morris the answer was 68.5, so anyone who paid 24 times earnings got a bargain. Polaroid had cost almost eight times what was justified, and it was one of four in the group that lost money over twenty-six years.
The group’s highest multiple turned out expensive and one well below the average turned out cheap, but neither number told you so on its own. The same holds for the stocks on your screen: the P/E is the price of one dollar of today’s earnings, and what that price is worth depends on how many dollars arrive tomorrow, for how long and with how much certainty.
What is the P/E ratio and how do you calculate it?
The P/E ratio, short for price-to-earnings, divides a share’s price by its earnings per share (EPS) over the last twelve months. A hypothetical stock at $40 with $2 of EPS has a P/E of 20: for each dollar of annual profit the market asks $20. You get the same number by dividing market capitalization by net income. FINRA, the self-regulatory body for US brokers, describes it in exactly those terms, as what you pay for a dollar of earnings.
At a P/E of 20, the price equals twenty times one year’s earnings. Your investment return depends on dividends and the price when you sell. The P/E also inherits every flaw of its denominator, meaning changing share counts, buybacks and one-off items, so it pays to know where the EPS comes from, as our guide to earnings per share and dilution explains. With negative earnings the P/E loses its meaning, and many platforms leave it blank.
Trailing vs forward P/E: what is the difference?
Trailing P/E uses earnings already reported, usually the sum of the last four quarters. Forward P/E uses analysts’ estimates for the next twelve months or the current fiscal year. At the same $40 price, expected EPS of $4 gives a forward P/E of 10. It looks like a bargain, but the forecast is doing the work: if actual EPS comes in at only $1, the P/E on that figure becomes 40.
| Annual EPS | P/E | Earnings yield |
|---|---|---|
| $1 | 40 | 2.5% |
| $2 | 20 | 5% |
| $4 | 10 | 10% |
Before comparing two P/Es, check three things. That they use the same earnings period. That both are GAAP or both are adjusted, because a higher adjusted EPS produces a lower P/E. And that price and earnings carry a date: today’s price divided by an estimate from three months ago may combine current pricing with a forecast that has not incorporated later results. Check whether the estimate has been updated.
What is earnings yield and why compare it with interest rates?
Flip the ratio and you get the earnings yield: earnings per share divided by price. At a P/E of 20 it is 5%, at 40 it is 2.5%, at 10 it is 10%. It is the share of the price the company earns in a year, whether it pays it out as a dividend or reinvests it.
Earnings yield makes the P/E comparable with a bond. A fixed-rate government bond pays its stated coupon until maturity; a stock offers an uncertain earnings yield that can grow. The more government bonds pay, the more growth a stock with a low earnings yield has to promise to hold up in the comparison. It is one of the channels through which interest rates and inflation move stocks.
P/E and growth: when is 20 times earnings expensive?
A question before the answer. Two stocks cost $40 and earn $2 per share, so both have a P/E of 20. The first grows earnings 5% a year, the second 20%. Five years from now, with the price unchanged, what P/E will each have on today’s price? Make a rough guess before you read the numbers.
After five years the first earns $2.55 per share and its P/E on today’s price falls to 15.7; the second earns $4.98 and falls to 8.0. After ten years they are at 12.3 and 3.2. Today both stocks cost the same multiple, and whoever buys the second pays far less for the earnings they will receive.
This is where the PEG ratio comes from: P/E divided by expected annual earnings growth in percentage points. That is 20 ÷ 5 = 4 for the first stock and 20 ÷ 20 = 1 for the second. It is a useful shortcut for ranking stocks in the same industry, with two limits. Growth is an estimate, and the PEG doesn’t say how many years it will last. Growth of 20% that stops after two years and then runs at 5% takes EPS from $2 to $5.43 in fifteen years; a steady 12% takes it to $10.95. Siegel reached Philip Morris’s 68.5 by looking at twenty-six years of returns, decades of growth that a single year’s PEG cannot see.
Why can a low P/E be a trap?
Cyclical earnings at their peak
In cyclical industries, such as memory chips, steel, autos and commodities, earnings rise and fall with the cycle. A low P/E can reflect unusually high earnings, so compare the denominator with earnings over a full cycle. Micron, one of the large memory makers, closed fiscal 2018 with diluted EPS of $11.51 and fiscal 2019 with $5.51: earnings more than halved in a year. Anyone who calculated the P/E on 2018 earnings was dividing the price by a number about to be cut in half, and at the same price the P/E on 2019 earnings would have been more than double.
In the hypothetical example in the figure, year 3 shows a P/E of 4.5. The following year earnings fall 80% and the price 38%. The fix is to calculate the P/E on average earnings over a full cycle: with average EPS of $5, the year-3 P/E would have been 9, twice the visible one.
A one-off gain in the denominator
Two hypothetical stocks at $40 both report EPS of $2, so a P/E of 20. For the second, though, $1 comes from selling a building, which won’t happen again. Without that gain its EPS is $1 and its P/E is 40, double the first. Write the adjustment next to the reported figure and keep the original in view.
A shrinking business
A P/E of 6 looks like a giveaway. If earnings fall 10% a year, though, after five years they are 59% of today’s and the P/E on the current price has risen to 10.2. Sometimes the market prices a decline correctly, and a low multiple is the fair price for a company that is worth a bit less every year.
The debt the P/E doesn’t see
The P/E looks only at shareholders’ capital. Two companies with the same P/E of 15, one debt-free and one with debt equal to its market value, don’t cost the same: whoever buys the second takes on creditors who get paid first and maturities that need refinancing. To compare different financial structures, investors use enterprise value (market capitalization plus net debt) divided by operating income, and our guide to company debt and maturities shows where to look.
How do you sanity-check a valuation with three scenarios?
The most honest way to use the P/E is to ask what growth you are paying for. Take the $40 stock with $2 of EPS and a five-year horizon. If you want an 8% annual return and think the market will pay 15 times earnings in five years, year-five EPS has to reach $3.92, which means growing 14.4% a year. If the company grew 6% a year over the last five years, today’s price assumes an acceleration you should be able to explain with facts, such as new products or expanding operating margins.
Then build three scenarios, each with a growth rate and an exit P/E. In the bear case EPS grows 5% and the multiple drops to 12: in five years the stock is worth $30.63, −5.2% a year. In the base case growth is 12% and the P/E 16: $56.39, +7.1% a year. In the bull case growth is 20% and the P/E 20: $99.53, +20.0% a year.
The base case is not a forecast. It shows how much the result depends on the assumptions: there are 25 points of annual return between the bear and bull cases, and almost 11 come from the exit multiple alone. Hold the exit P/E fixed and change only growth, and the gap shrinks to 14 points. The price the market will pay in five years is the most fragile assumption on the sheet, and the one worth keeping conservative.
Reading P/E as years of earnings is an accounting ratio, not an investment payback period: earnings may be retained and the exit price changes. Debt already affects profit through interest, but the multiple does not directly display repayment dates, collateral or available credit. Read the debt note alongside it.
Frequently asked questions
What is a good P/E ratio?
There is no value that is good for everyone: it depends on expected growth, how long it lasts, the industry and debt. For historical scale, in December 1972 the S&P 500 traded at 18.9 times earnings and the Nifty Fifty at 41.9. The most useful comparison is with the same stock’s history and with similar companies, on the same earnings basis.
Does a high P/E mean a stock is expensive?
Not if earnings grow enough and for long enough. A $40 stock with $2 of EPS growing 20% a year has a P/E of 20 today and 8.0 on today’s price five years out. The risk is paying for growth that never arrives: Polaroid traded at 94.8 times earnings in 1972.
What does a negative or missing P/E mean?
A negative P/E means the earnings figure used is negative. A missing P/E can also reflect zero earnings or unavailable data. Check the provider’s definition. When the company is loss-making, investors may compare enterprise value with revenue and examine cash consumption while assessing when earnings could turn positive.
What is the PEG ratio?
It is the P/E divided by expected annual earnings growth in percentage points: a P/E of 20 with 20% growth gives a PEG of 1, with 5% growth a PEG of 4. It helps compare stocks with different growth rates, but it doesn’t say how many years the growth will last.
Is trailing or forward P/E better?
Read both on a consistent earnings basis. Trailing P/E uses reported earnings, while forward P/E uses estimates that can change. If forward P/E is much lower, check whether the estimates imply earnings growth, exclude one-off charges or use different adjustments.
