Price-to-Book Across 18 Exchanges: Only 4 Beat SPY

We ran the same P/B screen across 18 stock exchanges, 25 years of data, and compared every result to both the S&P 500 and each market's local benchmark. Four exchanges beat SPY. Fourteen didn't. But with local benchmarks, the picture changes: 12 of 18 exchanges beat their own market. That's the honest result, and it's worth understanding exactly where the value premium survived and where it didn't.

Contents

1. Method
2. The Full Results
3. Local Benchmark Context: A Different Picture
4. Finding 1: Where the Premium Survived
5. Finding 2: The Emerging Market Divide
6. Finding 3: Why the UK Failed
7. Finding 4: The Hong Kong Catastrophe
8. Finding 5: The Currency Caveat
9. Data Quality and Exclusions
10. Takeaway
11. Related Analysis

Data: FMP financial data warehouse, 2000–2025. Updated March 2026.

Method

Data source: Ceta Research (FMP financial data warehouse) Period: 2000–2025 (25 annual periods) Strategy: Top 30 stocks by lowest P/B ratio, annual rebalance, equal weight Quality filters: ROE > 8% Market cap filter: Per-exchange minimum (not a uniform $1B USD) Returns: Local currency (not converted to USD) Benchmark: SPY (7.64% CAGR, -34.9% max drawdown over the same period)

The same screen ran on each exchange. Same filters, same position count, same rebalance frequency. Differences in outcomes reflect genuine differences in how the P/B premium behaves across markets, not methodological inconsistency.

Norway (OSL) is excluded from the "beat SPY" ranking because it had 32% cash periods, meaning the screen regularly found fewer than 30 qualifying stocks. A portfolio that's one-third cash nearly a third of the time isn't a clean P/B test.

The Full Results

Returns in local currency compared to SPY (USD). Norway excluded due to >30% cash periods. See "Local Benchmark Context" below for how these results compare to each market's own index.

Local Benchmark Context: A Different Picture

The table above uses SPY as a universal benchmark, which enables cross-market comparison. But SPY is a USD-denominated US equity index. For non-US markets, comparing local currency returns to SPY conflates strategy alpha with market selection and currency effects.

With local benchmarks, the picture changes dramatically:

• Sweden: +9.47% vs OMX Stockholm 30 (2.95% CAGR). The strategy beat its local market by 9.47 percentage points annually, not just the 4.79% vs SPY shown above.
• Canada: +4.90% vs TSX Composite (4.44% CAGR). The alpha vs the local market is stronger than vs SPY.
• Brazil: +7.61% vs Bovespa (8.44% CAGR). Still the highest alpha, now measured vs the local Brazilian index.
• Taiwan: +4.07% vs TAIEX (3.91% CAGR). Significantly positive vs local market.
• Japan: +3.54% vs Nikkei 225 (2.95% CAGR). Positive vs local market despite being negative vs SPY.
• Switzerland: +2.47% vs SMI (1.90% CAGR). Beats its local market.
• Germany: +1.68% vs DAX (4.45% CAGR). Positive vs DAX despite being negative vs SPY.
• Thailand: +2.30% vs SET Index (4.16% CAGR). Beats local market.
• Korea: +2.08% vs KOSPI (3.32% CAGR). Positive vs local market.
• Singapore: +2.23% vs Straits Times (1.64% CAGR). Beats local market.
• China: +1.84% vs SSE Composite (3.54% CAGR). Marginally positive vs local market.

Still negative vs local benchmarks:

• India: -5.58% vs Sensex (11.40% CAGR). The Sensex significantly outperformed, and the P/B screen missed it.
• Hong Kong: -4.23% vs Hang Seng (0.49% CAGR). Still underperforms its already-weak local market.
• UK: -0.17% vs FTSE 100 (0.86% CAGR). Essentially matched the weak UK market rather than underperforming by 7pp vs SPY.
• South Africa: -1.78% vs SPY (no local benchmark data available).
• Malaysia: -4.97% vs SPY (no local benchmark data available).

The honest summary: 12 of 18 exchanges beat their local market. The "only 4 beat SPY" headline is a US-centric comparison that understates how well the strategy works in its local market context. Sweden's +9.47% annual excess vs OMX30 is a much stronger story than +4.79% vs SPY. Germany, Korea, Japan, Switzerland, Thailand, Singapore, and China all show positive alpha vs their local markets despite being negative vs SPY.

The meaningful failures are India (underperforms Sensex by 5.58%), Hong Kong (underperforms Hang Seng), and the US (essentially flat vs SPY).

Finding 1: Where the Premium Survived

The four exchanges that beat SPY share a pattern. Brazil and Canada are resource-heavy markets where physical assets dominate balance sheets. Sweden is an industrial economy with a long tradition of capital-intensive manufacturing. Taiwan is electronics manufacturing, where book value meaningfully reflects the factories and equipment behind the products.

In each case, book value is a real number. Factories, mines, timber tracts, semiconductor fabs. When a company trades below its book value on these exchanges, it's actually cheap relative to the hard assets backing it.

Sweden stands out even within this group. Its Sharpe ratio of 0.473 is the best of all 18 exchanges, better than Japan (0.285), Canada (0.29), and well above the US (0.228). A 0.473 Sharpe means Sweden generated strong returns with relatively low volatility, not just raw gains. The down capture vs OMX Stockholm 30 was 46.79%. When the local index fell, the portfolio fell only about half as much. That asymmetry, strong up capture and modest down capture, explains why it compounds so well over time.

Canada's 9.33% CAGR on zero cash periods is clean. The screen found qualifying stocks in every period, energy and materials companies with depressed valuations that eventually recovered. The TSX Composite returned 4.44% over the same period, making Canada's +4.90% annual excess one of the strongest vs a local benchmark.

Brazil's 16.05% CAGR leads the table by a wide margin, but comes with the important caveat that these are BRL returns. Brazilian inflation and interest rates mean the nominal CAGR looks different from a USD-converted perspective. The Bovespa returned 8.44% CAGR over the same period. The strategy beat it by 7.61 percentage points annually, the highest alpha vs a local benchmark in our study.

Finding 2: The Emerging Market Divide

The gap between Brazil (16.05%) and India (5.82%) or China (5.39%) is striking, given that all three are often grouped as "high-growth EM" markets.

Brazil had a specific combination that worked: large commodity and banking sectors, where P/B is a meaningful valuation anchor; genuine book value accountability (banks have to mark their loans, miners have real assets); and a market that regularly produces deep value situations as political and currency crises cause indiscriminate selling. Each crisis created fresh P/B opportunities that eventually resolved.

India and China didn't behave the same way. India's -65.0% max drawdown tells one part of the story. The other part is the Sensex: 11.40% CAGR over 25 years. The P/B screen returned 5.82%, underperforming the local index by 5.58 percentage points annually. India's market surged over this period, driven by technology services, financials, and consumption. The P/B screen concentrates in state-owned enterprises and legacy industrials that systematically trailed the growth sectors driving the Sensex higher.

China's 5.39% CAGR with a -59.9% max drawdown reflects a different problem. Chinese book values are often questionable. State-owned enterprises regularly trade below stated book value for rational reasons: the book value is either inflated, not available to minority shareholders, or both. The P/B signal in China is noisy in ways it isn't in, say, Sweden or Canada. That said, the strategy beat the SSE Composite (3.54% CAGR) by 1.84 percentage points annually, showing marginal alpha vs the local market.

Hong Kong is in a category of its own and gets its own section below.

Finding 3: Why the UK Failed

The UK result looks different depending on your benchmark. The LSE produced 0.69% CAGR, a -6.95% gap versus SPY. But the FTSE 100 itself returned only 0.86% CAGR over the same period. The P/B screen essentially matched the local market (-0.17% underperformance), rather than catastrophically failing as the SPY comparison suggests.

This is the market where Fama and French's value premium was most extensively replicated internationally in the 1990s. The premium didn't so much fail as the entire UK market stagnated.

The explanation is structural. The UK economy shifted decisively toward services over this period: banking, insurance, professional services, consumer brands. In a service-heavy economy, book value is increasingly a poor measure of corporate worth. Brands, customer relationships, technology platforms, and intellectual property don't appear on the balance sheet. A bank trading at 0.8x book after a credit event genuinely is cheap. A media company trading at 0.9x book might just have no assets worth measuring.

The UK P/B screen in 2000 would have found financial companies, retailers, and manufacturers. The financial companies were crushed in 2008 and never fully recovered the period's losses. The retailers ran into structural decline from Amazon and digital commerce. The manufacturers competed against cheaper European and Asian producers.

Brexit added to the stagnation. From 2016 onward, UK equities traded at persistent discounts to European peers. The P/B screen found cheap companies, but they didn't recover. The FTSE 100's 0.86% CAGR tells you the whole market struggled. The P/B screen was no worse than passive exposure.

Finding 4: The Hong Kong Catastrophe

HKSE produced -3.74% CAGR and a -89.6% max drawdown. No other exchange in this study came close to that drawdown figure.

This isn't just a bad run. It's a compounding failure across multiple distinct phases. The P/B portfolio was heavy in property developers and financial companies, both sectors with high book values and volatile earnings. Hong Kong property stocks peaked around 2007, crashed through 2008, partially recovered, then entered a second prolonged decline from 2019 onward driven by the political situation and regulatory changes affecting cross-border business.

A -89.6% max drawdown means the portfolio, at its worst point, was down 89.6% from peak. That's near-total capital destruction from peak to trough. The strategy didn't just fail in the 2008 financial crisis. It accumulated losses through multiple distinct downturns without adequate recovery periods between them.

The underlying problem is similar to China but more severe: HKSE book values were heavily concentrated in property and finance, sectors where book value matters but where book value itself was inflated during the bubble years. Buying "cheap" property developers at 0.7x book in 2007 meant buying companies whose book values were about to be written down substantially.

The currency is also relevant. HKD is pegged to USD, so there's no currency caveat here. These are effectively USD-equivalent returns. The negative 3.74% CAGR vs the Hang Seng's 0.49% CAGR means the strategy underperformed even Hong Kong's already-weak market.

Finding 5: The Currency Caveat

All returns in this study are in local currency. This matters more for some exchanges than others.

For markets like Japan (JPY) and Switzerland (CHF), local currency returns are a reasonable proxy for what a local investor would experience. Currency effects over a 25-year period can be large, but they average out more than short-term comparisons suggest.

For Brazil (BRL), the 16.05% CAGR should be treated with particular care. Brazil experienced significant inflation across this period, and BRL weakened substantially against USD. An international investor converting BRL returns to USD would show a materially different result. The alpha vs. the Brazilian benchmark remains meaningful; the absolute number less so for a non-Brazilian investor.

For markets like India and Thailand, local currency returns overstate what a USD-denominated investor would experience, as both INR and THB weakened against USD over the period.

The comparison benchmark (SPY) is in USD. So the excess return figures need this caveat: for most non-US exchanges, the "excess return vs SPY" comparison isn't apples-to-apples on a currency-adjusted basis.

Data Quality and Exclusions

Norway excluded: 32% cash periods. The screen couldn't find enough qualifying stocks in many periods, meaning the strategy wasn't being tested cleanly.

Australia (ASX) excluded from analysis: Adjusted close data quality issues with split handling.

Taiwan (24% cash, 27.9 avg stocks): Included but with limited qualifying periods. The 24% cash rate means roughly 6 out of 25 periods had fewer than 30 stocks.

Singapore (16% cash, 12.7 avg stocks): Included. The low average stock count (12.7) indicates thin qualification. Concentrated portfolio by design.

All other exchanges had clean data and full qualification periods.

Takeaway

The P/B premium is real but conditional. It survives in markets where book value measures something. Canada's mines, Sweden's factories, Brazil's banks, Taiwan's fabs. It struggles in markets where intangible value has displaced hard assets, which is most of the developed world's service-heavy economies.

The benchmark matters. Four exchanges beat SPY. Twelve beat their local market. This isn't a contradiction; it reflects that many local markets underperformed the US over this period. Sweden's +9.47% excess vs OMX Stockholm 30 is a much stronger result than the +4.79% vs SPY comparison suggests. Japan, Germany, Korea, Thailand, Switzerland, Singapore, and China all show positive alpha vs their local markets despite being negative vs SPY.

The UK result is less of a failure in context. The FTSE 100 returned 0.86% CAGR. The P/B screen essentially matched it (0.69%). Fama and French found the value premium there decades ago. The premium didn't so much disappear as the entire UK market stagnated through Brexit, financial crisis recovery, and a services-economy transition.

Hong Kong remains a cautionary example of how a P/B strategy can compound losses when book values themselves are overstated and recovery cycles are cut short by macroeconomic and political disruption. Even vs the weak Hang Seng (0.49% CAGR), the strategy underperformed.

India is the meaningful surprise. The Sensex returned 11.40% CAGR, one of the best-performing major markets in the world. The P/B screen captured only 5.82%, underperforming by 5.58 percentage points annually. The strategy systematically missed India's bull market.

If you're going to run a global P/B strategy, the lesson from 18 exchanges is: check what's in the book, and check your benchmark. In some markets, book value is real and the strategy generates alpha vs the local index. In others, it's noise, or the entire market has moved in a direction the strategy can't follow.

Related Analysis

• P/B Value Strategy: US Markets Deep Dive
• P/B Value Strategy: Sweden Best Sharpe of 18 exchanges (0.473). +9.47% vs OMX30.
• P/B Value Strategy: Canada Resource stocks, +4.90% vs TSX Composite, +38pp in 2022.
• P/B Value Strategy: Brazil Highest alpha (+7.61% vs Bovespa). Returns in BRL.

Data: Ceta Research (FMP financial data warehouse, 2000–2025) Returns in local currency. All results gross of transaction costs. Past performance does not guarantee future results. This is educational content, not investment advice.