Cyclical Sector Timing: 15 Markets, Same Signal, Radically Different Outcomes

India: +4.55% annual excess CAGR. China: -17.97% annual excess. Same signal. Same sectors. Same selection method. 24 years of data.

Contents

1. The Strategy
2. Full 15-Exchange Results
3. The India vs China Contrast
4. Taiwan and Korea: Signal Fires Too Rarely
5. What Drives Success vs Failure
6. Dedicated Markets vs Comparison-Only
7. Takeaway

That contrast is the central finding of our global cyclical timing study. The revenue growth expansion signal (invest when ≥50% of cyclical stocks show positive year-over-year revenue growth, hold cash otherwise) produces outcomes that vary dramatically by market structure, not by signal quality.

The Strategy

The method is identical across all 15 exchanges:

• Universe: Basic Materials, Industrials, Energy, Consumer Cyclical stocks
• Signal: ≥50% of universe with positive YoY FY revenue growth → INVEST. Below 50% → CASH
• Selection: Top 30 by ROE among qualifying stocks (positive revenue growth AND ROE ≥ 5%)
• Rebalancing: Annual (July), 45-day data lag, equal weight, max 30 stocks
• Period: 2001–2024

Full methodology: backtests/METHODOLOGY.md US flagship blog (detailed methodology + SQL): blog.tradingstudio.finance/cyclical-sector-timing-us-backtest

Full 15-Exchange Results

SPY benchmark CAGR: 8.89%. All returns in local currency vs SPY USD benchmark.

The India vs China Contrast

India and China are the extremes. The gap is 22.5 percentage points of excess CAGR annually.

Why India works:

India's cyclical economy is driven by domestic demand and government capital expenditure. Infrastructure buildout, manufacturing capacity, and urbanization create multi-year revenue growth cycles in Basic Materials (cement, steel) and Industrials (engineering, capital goods). When the signal fires on BSE/NSE, it reflects real productive expansion, not financial leverage.

The 2008 result is the cleanest evidence: India cyclicals +24.3%, SPY -26.1%. India's capex cycle was domestically funded, insulated from US credit markets. Revenue growth in cement and construction continued regardless.

Why China fails:

China returned -9.08% CAGR over 24 years with a -90.9% max drawdown. The signal fired correctly in terms of revenue growth. Chinese cyclical companies did expand. But two structural problems destroyed returns:

First, Chinese markets have extreme boom-bust cycles unrelated to corporate fundamentals. Policy shifts, regulatory crackdowns, and liquidity injections create returns that are driven by government action, not by the companies the signal identified.

Second, the -90.9% max drawdown reflects the 2007-2008 cycle where Chinese cyclical stocks fell 90%+ from peak. No revenue-based signal can protect against that scale of market-structural collapse.

With 29% cash periods, the signal was also frequently off. When invested in Chinese cyclicals, the portfolio was fully exposed to these structural risks.

Taiwan and Korea: Signal Fires Too Rarely

Taiwan and Korea both show meaningful CAGR underperformance but the mechanism is different from China.

Taiwan (42% cash): Taiwan's cyclical sector is dominated by semiconductor-adjacent materials and tech-related industrials. The revenue cycles in these sectors don't follow traditional economic expansion patterns. They follow chip demand cycles, which are unpredictable and don't correlate with broad cyclical revenue growth signals. The signal stays off 42% of the time because the economic structure doesn't match the signal's underlying model.

Korea (38% cash): Similar dynamics. Korean industrials include large conglomerates (chaebol) with complex cross-holdings, and the energy/materials sectors are smaller relative to the overall economy. The signal can't get a clean read on broad cyclical expansion.

In both cases, the strategy isn't wrong. It's that the sectors covered don't capture what drives those economies. The cyclical timing model is well-matched to resource-heavy and industrial-export economies. It's poorly matched to tech-manufacturing economies where semiconductor cycles dominate.

What Drives Success vs Failure

The markets where cyclical timing works best share common characteristics:

Industrial export economies with transparent revenue cycles. Germany (Sharpe 0.411), Switzerland (Sharpe 0.415), India (CAGR +4.55%), Sweden (+0.78%) all have cyclical sectors where revenue growth clearly signals expansion or contraction. The businesses are capital goods, precision machinery, cement, steel. Orders come in, revenues grow.

Commodity economies with identifiable boom-bust cycles. Canada (down capture 28.4%) and Australia (-29.01% max drawdown) reflect how commodity-linked cyclicals provide good downside protection, even if overall returns disappoint. The signal correctly times commodity cycles, but it can't generate excess over SPY because SPY's tech weight outperforms in the periods when cyclicals are range-bound.

Policy-driven or tech-cycle economies are poor fits. China (policy-driven), Taiwan/Korea (semiconductor cycles), Hong Kong (financial/real estate weighted) all underperform. The revenue growth signal was designed for traditional economic cycle detection. It doesn't translate.

Dedicated Markets vs Comparison-Only

Dedicated markets (8 blogs): India, Sweden, Germany, US, UK, Switzerland, Australia, Canada. These have sufficient data quality, meaningful average stock counts, and interpretable signal dynamics.

Comparison-only (7 markets): Brazil, Japan, South Africa, Korea, Taiwan, Hong Kong, China. These markets show extreme results (China -90.9% DD), thin data (South Africa 25% cash with AvgStk 15.7), or structural mismatches (Taiwan/Korea semiconductor cycles) that make standalone investment recommendations inappropriate.

Takeaway

The revenue growth expansion signal works in industrial economies with transparent capex cycles. It doesn't work in policy-driven or semiconductor-cycle markets.

The best case for cyclical sector timing is India: 24 years of sustained domestic industrial expansion, confirmed by revenue data, selected by ROE quality screen. The worst case is China: policy-driven markets where corporate fundamentals disconnect from stock prices.

Everywhere else falls on a spectrum. Germany and Switzerland get you near-market returns with better risk characteristics. Canada gives you excellent downside protection at the cost of lagging compounding.

The signal is right about the economic cycle. Whether that translates to excess returns depends entirely on the market.

Part of a Series: US | UK | Switzerland | Sweden | India | Germany | Canada | Australia

Data: Ceta Research (FMP financial data warehouse). 15 exchanges, 2001-2024, annual rebalance (July), 45-day data lag. All returns in local currency vs SPY USD benchmark. Past performance doesn't guarantee future results. This is educational content, not investment advice.