One Composite Screen, 17 Exchanges: Where Value-Momentum Works Around the World

We ran a single value-momentum composite on 17 stock exchanges spanning 25 years. Same filters, same ranking, same rebalancing everywhere. Each exchange is measured against its own local index, not a foreign benchmark. 14 of the 17 beat their local market. India led at 15.69% CAGR, but the more striking finding is how consistently this strategy generated alpha across different market structures.

Contents

1. The Screen
2. Full Results: All 17 Exchanges
3. The Top Tier: High Alpha vs Local Benchmarks
4. Marginal Results
5. Underperformers: Three Exchanges That Lost to Their Local Index
6. Key Insights
7. Currency Warning
8. Takeaway
9. Part of a Series

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

The Screen

P/E between 0 and 20. ROE above 10%. D/E below 1.0. Stocks that pass get ranked by a composite percentile score combining value and 12-month momentum. Top 30, equal weight, rebalanced semi-annually. Entry uses next-day close (MOC execution model). Transaction costs are size-tiered (0.1-0.5% one-way). The approach follows the Asness, Moskowitz, and Pedersen (2013) finding that value and momentum, combined, produce more stable alpha than either factor alone.

Full methodology and US-specific analysis: Value-Momentum US.

Full Results: All 17 Exchanges

Each exchange is compared to its local index in the same currency. This gives a more honest measure of alpha than a cross-currency SPY comparison. Exchanges without a local index in the dataset (JNB, MIL, KLS) fall back to SPY.

* No local index in FMP dataset. SPY used as cross-currency benchmark. ** Norway benchmark (^OSEAX) only has data from 2013. Results cover 13 years, not 25. Not directly comparable.

The Top Tier: High Alpha vs Local Benchmarks

Hong Kong (HKSE): +8.03% vs Hang Seng, 9.34% CAGR. The largest excess return of any exchange. Down capture of 68.33% against the Hang Seng means real crisis protection in local terms. The -62.75% max drawdown is severe in absolute terms, but the Hang Seng itself was worse. Beta of 1.063 against the local index tells a different story than the old SPY comparison: the portfolio tracks the local market closely but with strong alpha on top. Full analysis: Value-Momentum Hong Kong.

Sweden (STO): +6.75% vs OMX30, 9.67% CAGR. The second-largest alpha when measured against local benchmarks. Down capture of 58.29% shows genuine downside protection that wasn't visible in the old SPY comparison. Sweden's industrial base produces exactly the kind of companies this screen targets. Sortino of 0.697 confirms the return distribution is favorably skewed. Full analysis: Value-Momentum Sweden.

UK (LSE): +5.90% vs FTSE 100, 6.98% CAGR. The UK is the biggest surprise. Against SPY, it looked like a -1.21% failure. Against the FTSE 100, it delivered +5.90% excess with 155.77% up capture and 60.78% down capture. The explanation: the FTSE 100 has been one of the weakest developed-market indices over 25 years. Value-momentum stocks within the UK comfortably beat the UK index, even if they couldn't match the S&P 500 from across the Atlantic.

Switzerland (SIX): +5.78% vs SMI, 7.76% CAGR. The highest Sharpe (0.485) of any exchange. Down capture of 56.55% against the SMI is strong. Like the UK, Switzerland was a near-miss against SPY but a clear winner against its own market. The SMI is concentrated in a few mega-caps (Nestle, Novartis, Roche). The broader value-momentum portfolio diversifies away from that concentration.

India (NSE): +4.34% vs Sensex, 15.69% CAGR. Still the highest CAGR by a wide margin. The alpha dropped from +7.99% (vs SPY) to +4.34% (vs Sensex) because India's market itself has been strong. But 4.34% annual excess over a strong local index for 21 years is a genuinely impressive result. Down capture of 61.76% vs the Sensex provides meaningful protection. Full analysis: Value-Momentum India.

Canada (TSX): +4.80% vs TSX Composite, 9.17% CAGR. Zero cash periods and 40.17% down capture make Canada the most consistent non-US result. The portfolio absorbed less than half the TSX Composite's losses while participating 111% in its gains. Full analysis: Value-Momentum Canada.

South Africa (JNB): +3.85% vs SPY, 11.61% CAGR. Best drawdown protection: -28.44% max drawdown, 1.65% down capture. The SPY benchmark is cross-currency (ZAR vs USD), so the low down capture partly reflects currency divergence. Still, the absolute drawdown number is the shallowest of any exchange. Full analysis: Value-Momentum South Africa.

China (SHZ+SHH): +3.54% vs SSE Composite, 7.24% CAGR. Positive alpha, but with -60.76% max drawdown. The SSE Composite is volatile enough that even modest stock selection beats the index. Down capture of 100.15% means no crisis protection.

US (NYSE+NASDAQ+AMEX): +3.41% vs S&P 500, 11.18% CAGR. The deepest universe (0% cash, 29.6 avg stocks) and the highest Sharpe among the alpha-positive group at 0.438. Down capture of 77.02% is solid. Full analysis: Value-Momentum US.

Germany (XETRA): +2.50% vs DAX, 7.57% CAGR. Down capture of 48.83% against the DAX is among the lowest. The DAX returned about 5% CAGR, so 2.50% excess is meaningful in relative terms. Full analysis: Value-Momentum Germany.

Marginal Results

Italy (MIL): +1.39% vs SPY, 9.15% CAGR. Positive excess, but the 25% cash rate and 15 average stocks per period make this a concentrated bet. The SPY benchmark makes the excess hard to interpret.

Japan (JPX): +0.89% vs Nikkei, 3.83% CAGR. Barely positive excess. -65.57% max drawdown is the worst of all 17 exchanges. Japan's deflation cycle punished value screens, but the strategy at least kept pace with the Nikkei. The 2023-2024 surge (+46.68%, +37.43%) suggests the environment may be shifting.

Taiwan (TAI+TWO): +0.53% vs TAIEX, 4.31% CAGR. Essentially flat vs the local index. Taiwan's tech-heavy market doesn't produce enough qualifying value stocks. 22% cash periods.

Underperformers: Three Exchanges That Lost to Their Local Index

Korea is the closest to neutral. The -0.56% shortfall might reverse with a few years of different data. The structural issue is that chaebols (Samsung, Hyundai, SK) carry higher leverage than the D/E < 1.0 filter allows, so the screen misses the companies that drive the KOSPI.

Thailand is the clearest failure. A negative Sharpe means you'd have been better off in risk-free assets.

Norway has a structural mismatch: 57% cash means the screen couldn't find enough qualifying stocks more than half the time. The -1.12% excess should be treated with caution since benchmark data only starts in 2013.

Key Insights

14 of 17 beat their local benchmark. That's an 82% hit rate, much higher than the 41% hit rate when everything was measured against SPY. The old framing made it look like the strategy failed in most markets. In reality, it underperformed the local index in only three countries.

Local benchmarks reveal the real alpha. The UK looked like a -1.21% failure vs SPY. Against the FTSE 100, it's +5.90%. Hong Kong went from +1.57% vs SPY to +8.03% vs Hang Seng. The strategy generates alpha within markets. The question is whether the market itself keeps up with the S&P 500.

Down capture separates the top tier. Canada (40.17%), Germany (48.83%), Switzerland (56.55%), and Sweden (58.29%) all absorbed less than 60% of their local index's drawdowns. South Africa's 1.65% is the lowest but uses a cross-currency benchmark.

Asia-Pacific: alpha exists but drawdowns are severe. Japan (+0.89%), Taiwan (+0.53%), and China (+3.54%) all generated positive excess vs their local indices. But max drawdowns of -65.57%, -46.67%, and -60.76% make the ride brutal. Korea and Thailand are the only true failures.

The D/E filter is both feature and bug. It creates South Africa's remarkable downside protection by accidentally eliminating commodity companies. But it also kills Norway, Korea, and Italy by removing too many names from structurally leveraged markets.

Currency Warning

All returns are in local currency. An investor buying Indian stocks in INR would face currency conversion. India's 15.69% CAGR in INR terms won't translate directly to 15.69% in USD. The same applies to every non-US exchange. The excess returns shown here measure the strategy's effectiveness in its home market vs its home index, not what a US-based investor would actually receive.

Takeaway

Value-momentum composites work in most markets. 14 of 17 exchanges beat their local benchmark. The strategy generates genuine within-market alpha, across developed and emerging markets, across different sector compositions and market structures.

The standout finding: Hong Kong's +8.03% excess and Sweden's +6.75% are the largest alpha figures when measured against local indices. India combines the highest absolute CAGR (15.69%) with strong alpha (+4.34% vs Sensex). For downside protection, Canada's 40% down capture and South Africa's -28.44% max drawdown are in a class of their own.

Three out of 17 lost to their local index (Korea, Thailand, Norway). That's an 18% failure rate, concentrated in markets with structural screening mismatches.

Part of a Series

• Value-Momentum: US Stocks (11.18% CAGR)
• Value-Momentum: India (15.69% CAGR)
• Value-Momentum: Sweden (9.67% CAGR)
• Value-Momentum: South Africa (11.61% CAGR)
• Value-Momentum: Hong Kong (9.34% CAGR)
• Value-Momentum: Canada (9.17% CAGR)
• Value-Momentum: Germany (7.57% CAGR)

Data: FMP warehouse via Ceta Research. 2000-2025 backtest, semi-annual rebalancing, next-day close execution. Full methodology in the US deep-dive.