P/E Ratio Calculator (Price-to-Earnings)
P/E (Price-to-Earnings) ratio calculator. Trailing + forward P/E with earnings yield + Fed Model spread vs Treasury yield. The world's most-used stock valuation multiple.
P/E Ratio Calculator
How to use the P/E ratio calculator
Enter share price + TTM EPS
Share price: current market price from your broker, Yahoo Finance, or Google Finance. TTM EPS (Trailing Twelve Months Earnings Per Share): sum of the last four reported quarterly EPS figures. Use diluted EPS, net of preferred dividends, and reported (GAAP or IFRS) earnings rather than an adjusted figure for the cleanest comparison. A loss gives no P/E — the tool says so and still shows the (negative) earnings yield.
Enter forward 12-month EPS estimate
Forward EPS = the consensus analyst estimate for the next 12 months, or your own (future revenue × expected margin ÷ shares). It is optional: leave it blank and the tool shows the trailing figures only. A forward P/E is only as good as the estimate under it.
Set 10-year Treasury yield for Fed Model context
The "Fed Model" compares earnings yield (E/P) to the 10-year Treasury yield. If E/P > Treasury yield, equities look cheap against bonds by this yardstick. The default 5.09% is the US Treasury 10-year par yield for 16 October 2026; change it to today's figure. For ASEAN stocks use the matching-tenor government yield in the same currency (SGS, MGS, Indonesian government bonds). The comparison is a heuristic with a known flaw — it sets a nominal bond yield against an earnings yield whose earnings rise with inflation — so read it as context.
Read trailing + forward P/E
Trailing P/E: based on past earnings — concrete but backward-looking. Forward P/E: based on consensus future earnings — forward-looking but only as good as the estimates. A trailing P/E of 35 with forward P/E of 20 means analysts expect 75% EPS growth — sanity-check this. The interpretation line places the trailing P/E (or the forward one, if there is no trailing) in a band and sets it beside the S&P 500's long-run average of 16.2 from Shiller's data.
Read Fed Model verdict + use as input to PEG
The Fed Model spread tells you whether earnings yield beats the risk-free rate. Positive spread = equities attractive vs bonds; negative = bonds win. To go further: take the P/E here and plug it into the PEG ratio calculator which divides P/E by the expected annual EPS growth rate in percent.
P/E ratio — price over earnings, and what it is quietly forecasting
The price-to-earnings ratio is the number most market commentary reaches for first ("the S&P is trading at N times earnings…"). The math is simple: P/E = share price / earnings per share. A stock at $150 with $7.50 in TTM earnings has a trailing P/E of 20 — meaning investors are paying $20 today for each $1 of last year's earnings. The reciprocal (E/P, called earnings yield) at 5% is comparable to a bond yield: it tells you the immediate income-equivalent return you're getting per dollar invested, if earnings stay flat forever.
Trailing vs forward — the most important distinction
Two flavours dominate practice. Trailing P/E uses last 12 months' earnings — concrete, audited, backward-looking. Good for stable mature companies; misleading for cyclicals (peak earnings produce low P/E that looks "cheap" but isn't) and for fast-growers (last year's EPS understates today's reality). Forward P/E uses analysts' consensus estimate for next 12 months — forward-looking but only as accurate as the estimates. When trailing P/E is much higher than forward P/E, analysts expect strong earnings growth; when forward is higher than trailing, expect a slowdown. Always check both. The Shiller CAPE (cyclically adjusted P/E) divides the real price by ten years of inflation-adjusted average earnings, which smooths the business cycle; it is usually quoted for whole indexes. In Robert Shiller's own S&P data the CAPE averaged 17.8 from 1881 to 2026 and stood at 40.7 on 2 October 2026.
A high multiple is a forecast in disguise. The question is never whether a P/E is high, but whether the earnings growth it assumes will arrive.
Sector context — why one number isn't enough
P/E means little without a comparison. Computed from Robert Shiller's monthly S&P data, the index's trailing P/E averaged 16.2 from January 1871 to June 2026 (median 15.1); on 16 October 2026 FactSet put the S&P 500 at 26.0 times trailing and 19.0 times forward earnings. Industries differ far more than that. Aswath Damodaran's January 2026 US data (aggregate market value ÷ aggregate net income): software 49.13, general retail 43.04, general utilities 21.67, integrated oil and gas 15.70, money-center banks 14.99. A P/E of 30 is ordinary for software and would be extraordinary for an oil major. Always compare against (a) sector average, (b) the stock's own 5-10 year historical range, (c) growth expectations (this is where PEG comes in — see PEG Ratio Calculator). A low P/E in isolation isn't "cheap" — it often signals real problems: cyclical peak earnings, structural decline, governance concerns.
ASEAN markets — check the date on any benchmark
An index P/E is a moving number, and an undated one is worse than none. One dated example: on 16 October 2026 the holdings of the SPDR Straits Times Index ETF, which tracks Singapore's STI, had a weighted-average P/E of 16.50, and three banks — DBS, OCBC and UOB — made up 58.2% of the fund (State Street Global Advisors). An index that heavy in banks should be compared with bank multiples, not with a technology-heavy index such as the S&P 500. For any market, take the figure from the index provider or an ETF tracking it, note the date, and compare sector by sector rather than headline against headline.
Trailing against forward, earnings yield as the reciprocal, and Shiller's CAPE
P/E = price / earnings per share, and it exists only for positive earnings. A loss does not make a negative P/E; it makes no P/E.
Reciprocal = earnings yield (E/P). A P/E of 20 = 5% earnings yield, directly comparable to bond yields.
Trailing P/E uses TTM (past 12 months) EPS. Forward P/E uses next-12-month consensus. Always check both.
S&P trailing P/E, Shiller data Jan 1871 – Jun 2026: mean 16.2, median 15.1. FactSet, 16 Oct 2026: 26.0 trailing, 19.0 forward.
Shiller CAPE uses 10-year inflation-adjusted EPS. Mean 17.8 over 1881–2026; 40.7 on 2 Oct 2026 (Shiller data).
Industries differ widely (Damodaran, US, Jan 2026): software 49.13, utilities 21.67, integrated oil 15.70, money-center banks 14.99.
Fed model: forward earnings yield against the 10-year Treasury — the comparison the Fed's July 1997 Monetary Policy Report drew.
For loss-making firms (negative EPS), P/E is undefined. Use EV/Sales, EV/EBITDA, or P/B instead.
P/E is before growth adjustment. For growth-adjusted: PEG = P/E ÷ expected growth in percent — popularised by Peter Lynch.
Cyclical P/E inversion: at a trough EPS collapses and P/E soars — the S&P's hit 123.7 in May 2009 (Shiller data). At a peak the reverse makes a stock look cheap.
Frequently asked questions
-
Both, for different purposes. Trailing P/E is concrete — last 12 months' actual reported earnings. Use it for stable mature firms and as a check on optimistic forward estimates. Forward P/E is what investors are paying for expected earnings; use it for growth stocks and recovering cyclicals. The gap between them tells you what analysts expect: forward < trailing = expected growth; forward > trailing = expected decline. On 16 October 2026 FactSet put the S&P 500 at 26.0 trailing and 19.0 forward — a gap that assumes strong earnings growth.
-
Sometimes — but more often it's a value trap warning. Reasons a stock might trade at a "cheap" P/E: (a) cyclical peak earnings — energy producers at oil-price highs, semiconductor firms at supply-cycle peaks, (b) structural decline — fixed-line telcos, legacy print media, (c) governance/management concerns — accounting scandals, dilutive insider behaviour, (d) regulatory/litigation overhang — opioid manufacturers, tobacco. A genuinely cheap stock will look cheap on multiple metrics simultaneously (low P/E + low P/B + high free cash flow yield + sustainable dividend). Single-metric "cheap" is rarely the bargain it looks like.
-
Three structural reasons. (1) Higher expected earnings growth — software scales at a lower marginal cost than industrial production. (2) More predictable revenue — recurring subscription revenue is steadier than cyclical industrial revenue. (3) Optionality — expansion paths not yet in current earnings. The risk: if growth disappoints, the multiple falls as well as the earnings. A P/E going from 35 to 20 is a 43% price decline before any change in earnings.
-
A rough yardstick that compares the S&P 500's forward earnings yield (E/P) with the 10-year Treasury yield. The name comes from the Federal Reserve's July 1997 Monetary Policy Report to Congress, which observed that the ten-year Treasury yield exceeded the ratio of twelve-month-ahead earnings to prices by the largest amount since 1991; the comparison was named for the Fed by market commentators rather than adopted by it as a model. Its known flaw: a bond yield is nominal and fixed, while earnings grow with inflation, and equities carry more risk than Treasuries. Dated example: on 16 October 2026 the S&P 500's forward P/E was 19.0 (FactSet), an earnings yield of 5.26%, against a 10-year Treasury par yield of 5.09% (US Treasury) — a spread of about 0.17 points, which the model would call roughly neutral.
-
P/E is undefined for a company with zero or negative EPS — dividing a price by a loss gives a number that ranks a deeper loss as "cheaper". The tool still shows the earnings yield, which is meaningful when negative. Losses are common for early-stage growth companies, biotech in long research phases and recent listings. For these, use EV/Sales, EV/EBITDA (if EBITDA is positive), price-to-book, or price to free cash flow. Some practitioners use an "adjusted EPS" that adds back non-cash charges to reach a positive figure — treat that with care, because it can hide a real loss.
-
For the cleanest comparison, reported (GAAP or IFRS) EPS — the audited number. Analyst consensus and company guidance usually use an adjusted figure that excludes stock-based compensation, restructuring and one-off charges. The gap can be large: a software company can be loss-making on GAAP and profitable on its adjusted measure, mostly because stock-based compensation is excluded — and many analysts regard that compensation as a real cost. Whichever you choose, use the same basis for every company and every year you compare.
-
PEG = P/E ÷ expected annual earnings growth, with growth in percent. It adjusts P/E for growth, making cross-company comparisons more meaningful: a stock with P/E 30 and 25% growth has a PEG of 1.2; one with P/E 15 and 5% growth has a PEG of 3.0 — dearer per unit of growth despite the lower P/E. Rules of thumb popularised by Peter Lynch treat a P/E roughly equal to the growth rate (PEG near 1) as fair; they are heuristics, not a standard. For scale, Damodaran's January 2026 US data put the PEG of the whole market at 1.90. Use our PEG ratio calculator to compute it.
-
Compare dated figures, sector by sector. On 16 October 2026 the holdings of the SPDR STI ETF (Singapore) had a weighted-average P/E of 16.50 (State Street Global Advisors), while FactSet put the S&P 500 at 26.0 times trailing earnings. Much of that gap is composition: DBS, OCBC and UOB were 58.2% of the STI fund, and banks trade on lower multiples than software almost everywhere — Damodaran's January 2026 US data has money-center banks at 14.99 and software at 49.13. Before reading a market as cheap, compare its banks with banks and its technology companies with technology companies, and match the P/E to expected growth.
-
No. Price, EPS, Treasury yield — every input stays in your browser. The P/E + earnings-yield + Fed-model computation runs as client-side JavaScript. Open DevTools → Network when you click Calculate and you'll see no request carrying your figures.
-
Foundational: Graham & Dodd, Security Analysis (1934). Valuation by multiples: Damodaran A., Investment Valuation, and his data page (pages.stern.nyu.edu/~adamodar/), which publishes P/E, PEG and growth by industry each January. History and CAPE: Shiller R., Irrational Exuberance, and his S&P data file at shillerdata.com. Current index valuations: FactSet's weekly Earnings Insight.
Method & sources
How it computes
Price divided by earnings per share, on trailing twelve-month EPS and on a forward estimate, defined only for positive EPS; a loss returns no P/E and the reason. Earnings yield = EPS ÷ price for any EPS, a loss being a negative yield. Implied growth = forward ÷ trailing EPS − 1. Fed-model spread = forward earnings yield − 10-year Treasury yield, in percentage points (RecaCorpFin.peRatios).
What this tool implements
- Trailing and forward are shown side by side and never blended. Trailing is fact and stale; forward is timely and an estimate. A single "P/E" that hides which one it used is the most common way this ratio misleads.
- A non-positive EPS returns no ratio rather than a negative one, and the page says why. A negative P/E is not a cheap stock; it is an undefined ratio. The earnings yield is still shown, because a negative yield is meaningful. A price of zero or less is refused.
- ⚠️ The Fed model comparison is a heuristic with a poor forecasting record, not a valuation method. It compares a real (earnings) yield with a nominal (Treasury) one, which is its central and well-known flaw. It is shown because it is widely quoted, and labelled because it is widely misused. Its origin is the comparison in the Federal Reserve's July 1997 Monetary Policy Report.
- ⚠️ P/E is not comparable across sectors or across accounting regimes. The page quotes Damodaran's January 2026 US industry P/Es (aggregate market value ÷ aggregate net income) to show the spread, dated.
- PEG = P/E ÷ expected annual EPS growth in percent — the definition Damodaran's industry table uses (money-center banks 14.17 ÷ 13.74 = 1.03).
- Corrected 2026-10-19: "S&P 500 current ~22" (26.0 trailing, FactSet 16 Oct 2026); "CAPE current ~30+" (40.7, 2 Oct 2026); "CAPE available only for indexes"; "STI current ~12" (16.50); KLCI, SET, IDX, VN-Index, HSI, Nikkei and sector ranges removed as unsourced; 10-year default 4.5% "mid-2026" → 5.09% dated 16 Oct 2026; "Peter Lynch's famous heuristic" softened to a heuristic; price 0 classed "Very low" with Infinity yield; loss hid the earnings yield; blanks became defaults.
Sources
- Graham B, Dodd DL. Security Analysis, 1934 — the earnings-multiple framing that this ratio still rests on.
- Asness CS. Fight the Fed Model. Journal of Portfolio Management 2003;30(1):11-24 — the standard critique of the equity-yield-versus-bond-yield comparison this tool displays, and the reason it is presented with a caveat rather than as a verdict.
- Board of Governors of the Federal Reserve System. Monetary Policy Report to the Congress, July 1997, Section 2 — the ten-year Treasury yield against twelve-month-ahead earnings ÷ price: https://www.federalreserve.gov/boarddocs/hh/1997/july/ReportSecti…
- Shiller RJ. U.S. stock market data (ie_data.xls), S&P Composite price, earnings and CAPE, monthly from 1871, file saved 5 Oct 2026: https://shillerdata.com/
- FactSet. Earnings Insight, 16 October 2026 — S&P 500 forward 12-month P/E 19.0 (5-yr avg 19.8, 10-yr avg 19.1), trailing 26.0: https://advantage.factset.com/hubfs/Website/Resources%20Section/R…
- U.S. Department of the Treasury. Daily Treasury Par Yield Curve Rates, 2026 — 10-year 5.09% on 16 Oct 2026: https://home.treasury.gov/resource-center/data-chart-center/inter…
- Damodaran A. Price Earnings Ratios by Sector (US), data as of January 2026, NYU Stern: https://pages.stern.nyu.edu/~adamodar/New_Home_Page/datafile/peda…
- State Street Global Advisors. SPDR Straits Times Index ETF (ES3) — fund characteristics and top holdings as of 16 Oct 2026: https://www.ssga.com/sg/en/individual/etfs/spdr-straits-times-ind…
What can make this go out of date
- Nothing is fetched at runtime: price, both EPS figures and the Treasury yield are entered by the user. The default Treasury yield (5.09%) is the 16 Oct 2026 US 10-year par yield and goes stale daily; the page dates it.
- The benchmarks in the prose are dated snapshots — FactSet (weekly), Treasury (daily), Shiller (monthly), Damodaran (each January), SSGA (daily) — and need re-checking at each review; PE_BENCHMARKS in corpfin-kernel.js holds the ones the result line quotes.
Pick up where you left off
Stored only in this browser — never sent to our servers.