Building CetaResearch. Data analysis for policy, research, and startups.
P/E compression
P/E compression on German stocks returns 3.32% annually vs 4.45% for the DAX. The DAX reinvests dividends, so unlike most markets in this study the comparison is like-for-like. Germany is the fair test, and the strategy loses it.
backtests
ASX cyclical timing returned 8.58% CAGR (+4.69% alpha over ASX 200) over 24 years. Australia's resources-heavy economy ties cyclical signals closely to commodity prices. 2015: +15.6% vs ASX -5.7%. 2020: +40.8% vs ASX +21.1%. 32.0% down capture vs the ASX 200.
backtests
We ran Benjamin Graham's formula (sqrt(22.5 × EPS × BVPS)) across US stocks from 2000 to 2024. The honest result: 7.01% CAGR vs 7.64% for SPY. The down capture of 68.8% is real. The structural underperformance since 2013 is also real.
backtests
We tested buying S&P 500 sectors when their aggregate P/E falls 1+ standard deviation below the 5-year average. 12.70% CAGR vs 10.65% SPY over 21 years, with 67.6% down-capture ratio. Full methodology and SQL included.
ETF
We tested low-concentration quality stocks on multiple global exchanges from 2005 to 2025. South Africa was the only exchange that beat SPY (12.09% vs 10.61%). Norway came closest with the best Sharpe ratio (0.373). Look-ahead bias from snapshot ETF data limits all results.
ETF
The S&P 500 holds 500 stocks. The top 10 account for 39%+ of the portfolio. We measured the effective number of stocks in 2,000+ ETFs using the Herfindahl-Hirschman Index. SPY behaves like a 50-stock portfolio, not 500.
ETF
Norway produced the best risk-adjusted result in our ETF concentration series. 9.45% CAGR vs 10.61% for SPY, but with a 0.373 Sharpe ratio, 0.712 Sortino, and near-identical max drawdown. The 2013-2018 run delivered six straight years of outperformance.
ETF
South Africa is the only exchange where low-concentration quality stocks beat SPY. 12.09% CAGR vs 10.61% over 2005-2025, with a 0.470 beta and +2.33% alpha. Most of the excess came from 2005-2012. The signal weakened after 2015.
piotroski
The Piotroski F-Score inverts in India. Score 8-9 returned 2.2% CAGR over 27 years on NSE. Score 0-2 returned 9.4%. The spread is -7.2%. Both trail the Sensex (10.9%). India's growth-driven market punishes the quality signal.
piotroski
The Piotroski F-Score works in Hong Kong. Score 8-9 returned 7.6% CAGR with Sharpe 0.207 over 27 years, beating Hang Seng (2.7%) by +4.87%. Score 0-2 returned 2.1% with -76.5% max drawdown. The +5.5% spread is one of the strongest in our study.
piotroski
Japan's Piotroski F-Score spread is +1.1% over 27 years. Score 8-9 beats the Nikkei 225 by +1.31%. The alpha comes from avoidance (+6.7%), not selection (-5.5%). Score 8-9 underperformed All Value (9.9%). The F-Score on JPX is a filter, not a stock picker.
piotroski
UK Score 8-9 returned 4.8% CAGR over 27 years, beating the FTSE 100 (1.3%) by +3.48%. But the unfiltered value universe returned 6.9% and Score 0-2 returned 8.3%. Spread: -3.5%. Selection alpha -2.1%, avoidance alpha -1.3%.