EV/EBITDA Sector-Relative Value on Hong Kong Stocks: Structural Discounts That Never Close

We ran a sector-relative EV/EBITDA screen on 21 years of Hong Kong stock data from HKSE. The result: 0.87% CAGR vs 0.49% for the Hang Seng Index. On paper that's +0.38% excess. But the Sharpe ratio is -0.097 and the max drawdown is -57.59%. Marginally beating the Hang Seng doesn't save you when risk-adjusted returns are negative. Hong Kong's persistent discounts aren't anomalies waiting to be corrected. They're structural, driven by factors the signal can't distinguish from temporary mispricing.

Contents

1. Method
2. The Screen
3. Sector-Relative EV/EBITDA Screen for HKSE (SQL)
4. What We Found
5. 21 invested years. Barely beats Hang Seng (+0.38%). Negative Sharpe.
6. Year-by-year returns
7. 2008: -57.6% catastrophic loss
8. The structural discount problem
9. 2011: -34.5% from the European crisis
10. 2019-2021: three consecutive damaging years
11. 2024: +33.9%, the best year
12. Backtest Methodology
13. Limitations
14. Conclusion

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

Method

Financial data sourced from key_metrics_ttm for EV/EBITDA and market cap, financial_ratios_ttm for debt metrics, profile for sector classification. 45-day lag on all financial statements to prevent look-ahead bias.

The Screen

Sector-Relative EV/EBITDA Screen for HKSE (SQL)

WITH universe AS (
    SELECT k.symbol, p.companyName, p.exchange, p.sector,
           k.evToEBITDATTM AS ev_ebitda, k.returnOnEquityTTM AS roe,
           fr.debtToEquityRatioTTM AS de, k.marketCap
    FROM key_metrics_ttm k
    JOIN financial_ratios_ttm fr ON k.symbol = fr.symbol
    JOIN profile p ON k.symbol = p.symbol
    WHERE k.evToEBITDATTM BETWEEN 0.5 AND 25
      AND k.returnOnEquityTTM > 0.08
      AND (fr.debtToEquityRatioTTM IS NULL OR (fr.debtToEquityRatioTTM >= 0 AND fr.debtToEquityRatioTTM < 2.0))
      AND k.marketCap > 2000000000
      AND p.sector IS NOT NULL
      AND p.exchange IN ('HKSE')
),
sector_medians AS (
    SELECT exchange, sector,
           PERCENTILE_CONT(0.5) WITHIN GROUP (ORDER BY ev_ebitda) AS median_ev_ebitda,
           COUNT(*) AS n_sector_stocks
    FROM universe GROUP BY exchange, sector HAVING COUNT(*) >= 5
)
SELECT u.symbol, u.companyName, u.exchange, u.sector,
       ROUND(u.ev_ebitda, 2) AS ev_ebitda_ttm,
       ROUND(sm.median_ev_ebitda, 2) AS sector_median_ev_ebitda,
       ROUND(u.ev_ebitda / sm.median_ev_ebitda, 3) AS ev_ratio_to_sector,
       ROUND((1 - u.ev_ebitda / sm.median_ev_ebitda) * 100, 1) AS discount_pct,
       ROUND(u.roe * 100, 1) AS roe_pct,
       ROUND(u.de, 2) AS debt_to_equity,
       ROUND(u.marketCap / 1e9, 2) AS mktcap_b
FROM universe u JOIN sector_medians sm ON u.exchange = sm.exchange AND u.sector = sm.sector
WHERE u.ev_ebitda / sm.median_ev_ebitda < 0.70
ORDER BY u.ev_ebitda / sm.median_ev_ebitda ASC LIMIT 30

Run this screen on Ceta Research

What We Found

21 invested years. Barely beats Hang Seng (+0.38%). Negative Sharpe.

The strategy marginally beats the Hang Seng on raw CAGR. But a negative Sharpe ratio means the strategy earned less than the risk-free rate per unit of risk over 21 invested years. You would have done better in cash on a risk-adjusted basis. Beating a market that itself barely generated positive returns isn't a meaningful win.

The 4 cash periods at the start (2000-2003) happened to avoid the early bear market. When the strategy first invested in 2004, it walked into a reasonable setup. But from 2008 onward, the structural problems in HKSE-listed companies made the signal consistently unreliable.

Year-by-year returns

2008: -57.6% catastrophic loss

The global financial crisis hit HKSE-listed companies harder than most markets. Many Hong Kong-listed firms had significant mainland China exposure through real estate and banking. When Lehman collapsed and credit markets froze, those exposures became liabilities. The -57.6% single-year loss is catastrophic and impossible to recover from in any normal compounding timeline.

The structural discount problem

Several factors create persistent HKSE discounts that the signal reads as opportunity but which don't mean-revert:

Mainland Chinese listings. A large proportion of HKSE-listed companies are mainland Chinese businesses (H-shares and Red Chips). These trade at persistent discounts to their A-share or global peers for reasons that are structural: lower governance standards, related-party risk, state ownership constraints, and limited mechanisms for minority shareholder value realization.

Property developer leverage. Companies like Evergrande, Country Garden, and their peers traded at low EV/EBITDA multiples relative to sector medians for years before 2021. The signal would have included them as "cheap vs sector." The cheap multiples reflected genuine default risk, not temporary sentiment pessimism.

Political risk premium. Post-2019, Hong Kong's special administrative status became less certain. The 2019 protest movement, the 2020 National Security Law, and subsequent pressure on the independent financial market created a risk premium that rational investors applied to HKSE listings. That premium compresses multiples but doesn't revert because the underlying risk is real.

2011: -34.5% from the European crisis

The 2011 European debt crisis triggered capital flight from Hong Kong. HKSE-listed companies had significant cross-border exposure to European credit markets. The -34.5% return (vs SPY +2.5%, -37% excess) shows how quickly HK can amplify global risk events.

2019-2021: three consecutive damaging years

• 2019: +11.3% vs SPY +32.3%. The protest movement affected commercial real estate, retail, and financial services.
• 2020: -10.3% vs SPY +15.6%. COVID + National Security Law + regulatory uncertainty.
• 2021: +6.9% vs SPY +31.3%. Evergrande default contagion + China tech regulatory crackdown affected dozens of HKSE-listed names.

Each year brought a different macro event. The sector-relative signal couldn't anticipate or protect against any of them.

2024: +33.9%, the best year

Chinese government stimulus announcements drove HKSE sharply higher in 2024. The +8.6% excess return was the second-best year in the backtest. This is the pattern: occasional large positive years driven by policy announcements, surrounded by sustained underperformance. The signal can't predict when Beijing will announce stimulus.

Backtest Methodology

Full methodology: backtests/METHODOLOGY.md

Limitations

Negative Sharpe ratio. Over 21 invested years, the strategy produced risk-adjusted returns below the risk-free rate. Beating the Hang Seng by 0.38% annually doesn't fix this. The volatility and drawdown are too severe relative to the return.

Structural discount persistence. The core assumption of sector-relative strategies, that discounts are temporary anomalies that mean-revert, is violated by HKSE's structural pricing dynamics.

MaxDD -57.59%. The second-worst max drawdown of 13 exchanges. Recovery from -57.6% requires a +134% return to break even.

Data completeness. FMP's HKSE data coverage improves post-2004. The 4 cash years at the start may partly reflect data availability rather than genuine signal failure. The invested track record starts from 2004.

Conclusion

Sector-relative EV/EBITDA on HKSE produced 0.87% CAGR over 21 invested years, beating the Hang Seng Index's 0.49% by a thin +0.38% annually. That's technically positive excess, but the negative Sharpe ratio and -57.59% max drawdown make this a risk-adjusted failure.

The problem isn't the signal construction or the quality filters. It's the nature of HKSE discounts. Property developer leverage, mainland listing risk premiums, political uncertainty, and governance concerns create discounts that look like value opportunities but are reflections of structural risk. The signal can't distinguish between "temporarily cheap" and "cheap for a reason that won't change."

Marginally beating an index that itself barely generates returns isn't a useful outcome. We're publishing this result because reporting only the successes would be misleading.

Data: Ceta Research (FMP financial data warehouse). Returns in HKD (local currency). Benchmark S&P 500 in USD. Past performance does not guarantee future results. See full methodology at github.com/ceta-research/backtests.