Stock market valuations by country
Every major market ranked cheapest to most expensive by an overall value score (out of 10, 10 = cheapest) that blends CAPE, P/E, P/B, P/S and dividend yield. Tap any column to re-sort. Right now China screens as the best value and the US among the most expensive.
Latest observation: August 2026
By Dominic Roe · Data Engineer & Business Intelligence Developer
| Whole worldVT | 30.2 | 22.6 | 3.32 | 2.55 | 1.59% | 14.7% | 2.8 |
| ChinaEstimate | 15.9 | 9.9 | 1.13 | 1.32 | 1.87% | 11.4% | 8.5 |
| ItalyEstimate | 24.4 | 15.3 | 1.84 | 1.59 | 3.05% | 12.0% | 7.7 |
| United KingdomEstimate | 22.0 | 15.7 | 2.31 | 1.62 | 3.09% | 14.7% | 7.5 |
| GermanyEstimate | 25.2 | 17.9 | 1.87 | 1.28 | 1.94% | 10.4% | 6.8 |
| JapanEstimate | 36.2 | 19.1 | 1.94 | 1.66 | 3.86% | 10.2% | 6.2 |
| FranceEstimate | 25.8 | 18.4 | 2.08 | 1.50 | 2.83% | 11.3% | 6.2 |
| SpainEstimate | 27.9 | 16.0 | 2.18 | 1.81 | 2.71% | 13.6% | 5.8 |
| SwedenEstimate | 29.1 | 14.7 | 2.71 | 2.85 | 3.44% | 18.5% | 5.4 |
| NetherlandsEstimate | 34.7 | 20.7 | 2.64 | 1.95 | 4.24% | 12.7% | 5.3 |
| South KoreaEstimate | 36.3 | 15.9 | 1.97 | 1.67 | 1.25% | 12.4% | 5.0 |
| AustraliaEstimate | 22.9 | 21.4 | 2.75 | 3.43 | 2.89% | 12.8% | 4.6 |
| CanadaEstimate | 30.1 | 20.5 | 2.82 | 2.78 | 1.27% | 13.8% | 3.1 |
| IndiaEstimate | 33.6 | 22.9 | 3.14 | 2.72 | — | 13.7% | 2.8 |
| SwitzerlandEstimate | 30.8 | 25.9 | 4.45 | 2.93 | 1.73% | 17.2% | 2.1 |
| TaiwanEstimate | 44.2 | 28.1 | 3.91 | 2.49 | 1.04% | 13.9% | 2.0 |
| United States✓ Real | 42.5 | 25.1 | 5.19 | 3.64 | 1.09% | 20.6% | 1.4 |
Reading the colours: the Score and the trailing multiples (P/E, P/B, P/S, yield) are shaded green when cheap and red when dear versus the other markets. CAPE is shaded against each market’s own history — so green there means cheap for that country, not merely a low number. ROE is a quality gauge, not a price: green means more profitable, and it is shown as context only — it does not feed the Score.
Only the US CAPE is a measured series. Rows tagged Estimate use a free ETF price proxy and an anchor (see the methodology); the trailing multiples are real, current fund figures.
How the value score works
No single multiple settles whether a market is cheap, so the score combines five. For the trailing multiples we rank every market against the others — lowest P/E, P/B and P/S count as cheap, and a higher dividend yield counts as cheap too. CAPE is the exception: because it is the one measure with deep history, a market scores well on it for being cheap against its own past, not merely for carrying a structurally low multiple — which stops banks-and-energy markets from looking permanently cheap. We average those five ranks (equal-weighted) and put the result on a 1–10 scale where 10 is the cheapest. A market that is cheap on all five lands near 10; one that is dear across the board lands near 1.
Two honest caveats. The score is mostly relative — the four multiples rank these markets against each other today, not against fair value (only the CAPE component looks at a market’s own history) — and it inherits each measure’s quirks, most of all that price-to-book is distorted by sector mix (a banks-and-energy market looks cheap on book almost by construction). It is context for research, not a recommendation. CAPE for every market except the US is a clearly-labelled estimate; see the methodology.
How is the value score calculated?
Five measures are combined, equal-weighted. The four trailing multiples — P/E, P/B, P/S and dividend yield — are each ranked against the other markets, giving a 0–100 'cheapness' (a low P/E is cheap; a high dividend yield is cheap). CAPE is treated differently: because it is the one measure with deep history, it is ranked against each market's OWN past rather than against other markets — so a market scores well on CAPE when it is cheap relative to its own history, not merely because it carries a structurally low multiple. The five cheapness ranks are averaged and mapped to a score out of 10 where 10 is the cheapest. It is regenerated from the data, never hand-set.
Does a high score mean a market is a good investment?
No. The score is mostly RELATIVE — cheapest versus these peers, right now (only the CAPE component looks at a market's own history) — not an absolute verdict or a buy signal. Cheap markets are often cheap for reasons: slower growth, more cyclicality, political risk, or a heavy weighting of low-multiple sectors like banks and energy. Treat it as a starting point for research, not advice.
Why are there two colouring systems in the table?
CAPE is coloured against each market's OWN history (green means cheap for that country, not just a low number) — the most useful way to read a single market. The trailing multiples are coloured across markets (green means cheap versus the others, today), because there is no free long history for per-country P/E, P/B or P/S. The score blends both frames the same way it is built: its CAPE component is judged against the market's own history and its multiple components against the other markets.
Why is price-to-book so different between countries?
Book value depends heavily on sector mix. Markets weighted toward banks, energy and materials carry more tangible assets and trade at low price-to-book almost structurally, while technology-heavy markets carry little book value and trade high. A low P/B is not automatically cheap — read it alongside the other measures.
What does return on equity (ROE) tell me, and why isn't it in the score?
ROE — here derived as price-to-book ÷ price-to-earnings — measures how much profit a market earns on its shareholders' equity. It is a quality gauge, not a price, so it helps explain why some expensive markets deserve their premium: the US trades on high multiples partly because it earns a high return on equity (around 20%), while a cheaper-looking market such as the UK earns considerably less. Because high quality is precisely what justifies a higher valuation, folding ROE into the cheapness score would partly cancel out the very signal the score is built to measure — so we show it beside the score as context, not inside it.
Is the data real?
Only the US CAPE is a measured series (from Robert Shiller). Other countries' CAPE is an estimate built from a free ETF price proxy and an anchor — marked 'Estimate' in the Data tag. The trailing multiples are real, current fund-reported figures. Sample (demo) data, where shown, is labelled and must not be used for investment decisions.