start learning →
beginner10 min read

The P/E ratio and valuation basics: P/E, forward P/E, EV/EBITDA and their limits

The price-to-earnings ratio tells you how many years of current profit the market is paying for a share. Here is how to calculate it, how trailing, forward and cyclically adjusted versions differ, why EV/EBITDA exists, and where all of them mislead.

key takeaways
  1. The P/E ratio is the share price divided by earnings per share, or market value divided by net income; it shows how much investors pay for each unit of annual profit.
  2. Trailing P/E uses the last 12 months of reported earnings; forward P/E uses analysts’ forecasts, which are often revised and are frequently optimistic.
  3. EV/EBITDA compares the whole business, including debt, with operating earnings before interest, tax, depreciation and amortisation, so it is less distorted by differences in debt.
  4. One-off gains and losses, cyclical peaks and accounting choices can make a P/E look cheap or expensive for reasons that have nothing to do with value.
  5. Valuation ratios describe how a share is priced; they do not tell you when its price will change.

The price-to-earnings ratio (P/E) is a share’s price divided by its earnings per share over a year. A P/E of 20 means investors are paying 20 units of currency for every 1 unit of annual profit the company makes per share. It is the most quoted valuation number in markets because it is simple, but it is only a starting point: it depends on which earnings you use, it is distorted by debt, one-offs and the economic cycle, and it tells you nothing about timing.

This guide explains how to calculate P/E and its main variants, how to read it, the other ratios professionals use alongside it, especially EV/EBITDA, and the most common ways these numbers mislead. If you are new to shares, start with how stock markets work.

Calculating the P/E ratio

There are two equivalent ways:

  • Per share: share price ÷ earnings per share (EPS).
  • Whole company: market capitalisation ÷ net income attributable to shareholders.

Both give the same answer, because market capitalisation is price × shares and EPS is net income ÷ shares.

Worked example: the basic calculation

Illustrative numbers. A company’s share price is 60.00. Over the last 12 months it reported net income of 600 million with 200 million shares outstanding.

  • EPS = 600 ÷ 200 = 3.00.
  • P/E = 60 ÷ 3 = 20.
  • Market capitalisation = 60 × 200 million = 12,000 million. Check: 12,000 ÷ 600 = 20.

Earnings yield

Turn the P/E upside down and you get the earnings yield: EPS ÷ price. A P/E of 20 is an earnings yield of 1 ÷ 20 = 5%. Earnings yield is handy for comparing shares with bonds and cash, since it is expressed as a percentage like a yield. It is not the cash an investor receives, though; most companies pay out only part of their earnings as dividends, and earnings are an accounting figure, not cash flow.

Which earnings? Trailing, forward and adjusted

The same share can have several quite different P/E ratios depending on the earnings figure used.

Trailing P/E

Uses the last 12 months of reported earnings, often called TTM (trailing twelve months). It is based on actual results, but it looks backwards. If earnings are about to rise or fall sharply, the trailing P/E is out of date.

Forward P/E

Uses analysts’ forecasts for the next 12 months or the next financial year. Using the same company with a share price of 60, if analysts expect EPS of 3.60 next year, the forward P/E is 60 ÷ 3.6 ≈ 16.7, against a trailing P/E of 20.

Forward P/E reflects expectations, which is what share prices trade on. But forecasts are often revised, sometimes heavily, and consensus estimates have tended to be optimistic early in the year and trimmed as results approach. Our guides to earnings season and reading an earnings report explain how actual results are compared with these estimates.

Reported versus adjusted earnings

Companies report earnings under accounting standards (IFRS or US GAAP), and many also publish "adjusted" or "underlying" figures that strip out items they consider unusual, such as restructuring costs, impairments or share-based pay. Adjusted EPS is often higher than reported EPS. Neither is automatically right; the question is whether the excluded costs really are one-offs. A company that has "one-off" restructuring charges every year has an ongoing cost.

Worked example: a one-off gain flatters the P/E

Illustrative numbers. A company with a market value of 8,000 million reports net income of 700 million, which includes a 300 million gain from selling a division.

  • Reported P/E: 8,000 ÷ 700 ≈ 11.4. It looks cheap.
  • Underlying net income: 700 − 300 = 400 million. Underlying P/E: 8,000 ÷ 400 = 20.

The business is valued at 20 times its ongoing profit, not 11. Screens that sort shares by P/E will rank this company as a bargain unless the one-off is removed. A one-off loss does the reverse, making the P/E look high for a year.

Reading a P/E: what is high or low?

A P/E has meaning only in context. Ways to frame it:

  • Against the company’s own history. Is the share priced higher or lower than its usual range, and has anything changed to justify that?
  • Against peers in the same industry. Banks, utilities and software companies typically sit at very different multiples because their growth and risk differ. Comparing across sectors is rarely useful; see sector rotation explained for how sectors differ.
  • Against the market. Index-level P/E ratios are published for the S&P 500 and other indices.
  • Against interest rates. When bond yields are higher, investors can earn more without owning shares, which tends to pull valuation multiples down. Our bonds and yields guide explains the link.

A high P/E usually means the market expects strong earnings growth, sees the earnings as unusually stable, or both. A low P/E usually means it expects slow growth, sees high risk, or thinks current earnings are at a peak. The growth vs value stocks guide develops this.

The PEG ratio

The PEG ratio divides P/E by the expected annual earnings growth rate in percent. A P/E of 30 with 15% expected growth gives a PEG of 2; a P/E of 12 with 4% growth gives 3. It is a rough way to compare companies growing at different speeds, but the growth figure is a forecast and often wrong.

The cyclically adjusted P/E (CAPE)

For cyclical companies and whole markets, one year of earnings can be misleading. The cyclically adjusted P/E, popularised by economist Robert Shiller, divides price by the average of ten years of inflation-adjusted earnings. It smooths out booms and recessions. It is used mainly for broad indices over long periods and has a poor record as a short-term timing signal.

Where the P/E breaks down

No earnings, no P/E

A company with a loss has a negative P/E, which is meaningless, and one with tiny earnings has an enormous P/E. Young growth companies are often valued on sales instead.

Cyclical peaks and troughs

For cyclical businesses such as miners, carmakers or chipmakers, earnings swing hugely through the cycle. The P/E often looks lowest at the top of the cycle, when earnings are at a peak and about to fall, and highest near the bottom, when earnings are depressed. Buying cyclicals because the P/E is low can mean buying at the worst point.

Debt is invisible

P/E looks only at the equity. Two companies with the same P/E can carry very different amounts of debt, which changes their risk completely. This is the main reason EV/EBITDA exists.

Accounting choices

Depreciation methods, the treatment of research spending, acquisitions and tax rates all affect net income. Companies in different countries or using different standards may not be directly comparable.

Share count changes

Earnings per share can rise because of a shrinking share count rather than growing profit. Our guide to share buybacks shows how a buyback can lift EPS while net income falls.

Enterprise value and EV/EBITDA

Enterprise value

Enterprise value (EV) is a measure of the value of the whole business, not just the shares. A simplified version:

  • EV = market capitalisation + total debt − cash.

The logic: if you bought the entire company, you would pay for the shares, take on the debt, and get the cash. Fuller versions also add items such as preferred shares, minority interests and, under some conventions, lease liabilities.

EBITDA

EBITDA is earnings before interest, tax, depreciation and amortisation. It approximates operating profit before financing choices (interest), tax regimes and non-cash accounting charges (depreciation and amortisation). Because it is before interest, it belongs to all providers of capital, lenders and shareholders alike, which matches EV.

Worked example: same P/E story, different debt

Illustrative numbers. Two companies each have EBITDA of 1,000 million, depreciation of 200 million, a 25% tax rate and a market capitalisation of 10,000 million. Company A has 5,000 million of debt at 5% interest; Company B has no debt. Cash is ignored for simplicity.

Company A:

  • Operating profit (EBIT): 1,000 − 200 = 800 million.
  • Interest: 5,000 × 5% = 250 million.
  • Pre-tax profit: 550 million. Net income after 25% tax: 412.5 million.
  • P/E: 10,000 ÷ 412.5 ≈ 24.2.
  • EV: 10,000 + 5,000 = 15,000 million. EV/EBITDA: 15,000 ÷ 1,000 = 15.

Company B:

  • EBIT: 800 million, no interest.
  • Net income after tax: 600 million.
  • P/E: 10,000 ÷ 600 ≈ 16.7.
  • EV: 10,000 million. EV/EBITDA: 10.

The two businesses produce identical operating profits. The market is putting the same price on their shares, but buying Company A means taking on 5,000 million of debt too. EV/EBITDA shows clearly that the whole of Company A costs 15 times operating earnings against 10 for Company B. And Company A’s shareholders carry more risk: if EBITDA falls, its interest bill does not.

Limits of EV/EBITDA

EBITDA ignores the cost of replacing equipment (depreciation is a real cost for capital-heavy businesses), ignores tax and interest, which have to be paid in cash, and is not defined under accounting standards, so companies can calculate it differently. It is useful for comparing businesses with different debt, but it overstates the cash available to owners of capital-intensive companies.

Other common ratios

  • Price-to-book (P/B): share price ÷ net assets per share. Useful for banks and insurers, whose assets are mostly financial and carried near market value. Less useful for software or brand-led companies whose main assets are not on the balance sheet at full value.
  • Price-to-sales (P/S) and EV/sales: useful for companies with little or no profit, but they ignore margins entirely. A company with a 2% margin and one with a 30% margin can have the same sales.
  • Dividend yield: annual dividend ÷ share price. Relevant to income investors and to dividend adjustments on CFDs, but a very high yield can be a sign the market expects a cut.
  • Free cash flow yield: free cash flow ÷ market value. Cash-based, so harder to flatter with accounting choices, though it can swing with investment cycles.

No single ratio is sufficient. Analysts typically look at several together and at how they have changed over time.

Why valuation is not a trading signal

Valuation ratios tell you how the market is pricing a business today. They do not tell you when that will change. Expensive shares can get more expensive for years if earnings keep growing or sentiment stays strong; cheap shares can stay cheap or get cheaper. Many short-term moves in share and index prices are driven by earnings surprises, interest rate expectations, flows and risk appetite rather than by valuation.

For traders using share or index CFDs, valuation is background context: it helps explain why some shares react violently to small earnings misses (high expectations built into a high P/E) and why rate changes move certain sectors more. It does not replace a plan for entries, exits and risk. Size each position from the money you would lose if your stop is hit, using the position size calculator and the method in the position sizing guide. Share CFDs are capped at 5:1 leverage for retail clients under ESMA rules, financing is charged nightly, and most retail CFD accounts lose money; see why most retail CFD traders lose.

Common mistakes

  • Comparing P/Es across very different industries. A bank and a software company are not on the same scale.
  • Mixing trailing and forward figures. Make sure two P/Es being compared use the same earnings basis.
  • Ignoring one-offs. Check what is in net income before trusting a low P/E.
  • Buying cyclicals because the P/E is low. It can be lowest at the earnings peak.
  • Ignoring debt. Use EV-based ratios when debt levels differ.
  • Trusting forecasts blindly. Forward P/E is only as good as the estimates behind it.
  • Using valuation as a timing tool. Cheap and expensive can persist for long periods.

Frequently asked questions

What is a good P/E ratio?

There is no single good number. A P/E only makes sense compared with the company’s own history, its industry peers, the wider market and the level of interest rates. A P/E of 25 may be normal for a fast-growing software company and very high for a utility.

What is the difference between trailing and forward P/E?

Trailing P/E uses the last 12 months of reported earnings. Forward P/E uses analysts’ forecasts for the next 12 months or financial year. If earnings are expected to grow, the forward P/E is lower than the trailing one; if they are expected to fall, it is higher.

Why use EV/EBITDA instead of P/E?

EV/EBITDA values the whole business, including its debt, against operating earnings before interest, tax and non-cash charges. That makes companies with different amounts of debt easier to compare. P/E looks only at the shares and net income, which are both affected by how much the company has borrowed.

Can a company have a negative P/E?

Mathematically yes, if it made a loss, but a negative P/E has no useful meaning. Loss-making companies are usually compared on price-to-sales, EV/sales or measures of cash burn instead.

What is the Shiller CAPE ratio?

The cyclically adjusted price-to-earnings ratio divides price by the average of the past ten years of inflation-adjusted earnings. It is used mostly for whole stock markets, to smooth out the effect of booms and recessions on one year’s earnings. It has been a poor guide to short-term market moves.

Does a low P/E mean a share is cheap?

Not necessarily. The P/E may be low because earnings include a one-off gain, because earnings are at a cyclical peak and about to fall, or because the business is in decline. A low P/E is a reason to investigate, not a conclusion.

Not investment advice. This is general information and education, written without knowledge of your circumstances. It is not a recommendation to buy or sell any instrument. Past performance is not a reliable indicator of future results. Trading leveraged products carries a high risk of losing money. Risk warning