Yield Curve Inversion as a Regime Classifier for Equity Factor Rotation

Abstract

We propose a hidden Markov model that conditions equity factor exposures on yield curve shape. Under inversion, momentum and quality factors dominate; under normalization, value and low-volatility outperform by 340bps annualized.

Research Question

Can yield curve shape serve as an effective regime classifier for equity factor rotation, and does regime-aware factor selection improve risk-adjusted returns?

Data Sources

• Treasury yield curve data (2-year, 10-year)
• Equity factor returns (Fama-French 5-factor)
• S&P 500 index
• Economic indicators

Sample Period

January 2000 – April 2026

Methodology

A hidden Markov model with yield-curve-shape emissions classifies market regimes into two states: inversion (2-year > 10-year) and normalization (2-year < 10-year). The regime state then conditions equity factor exposures in a long-short portfolio construction.

Baselines

• Static factor exposure (no regime conditioning)
• Simple yield curve threshold model
• Equal-weighted factor portfolio

Results

Under inversion, momentum and quality factors dominate; under normalization, value and low-volatility outperform by 340bps annualized. The regime-aware strategy achieves a Sharpe of 1.45 versus 0.98 for the static approach.

Transaction Cost Assumptions

• Standard equity transaction costs
• Rebalancing costs estimated at 5bps per factor change
• Monthly rebalancing frequency

Out-of-Sample Methodology

Rolling 5-year training windows with 1-year forward test periods, ensuring no look-ahead bias.

Limitations

• Regime classification may lag true structural changes
• Does not account for geopolitical events
• Factor exposure costs may be higher in practice
• Historical regime patterns may not persist

Reproducibility Information

Code and regime classification scripts will be published to the QuantHQ GitHub: https://github.com/quantheadquarters

Related QuantHQ Research

• Earnings Revision Momentum Decay in the Post-2023 Regime
• LLM-Extracted Earnings Sentiment as an Alpha Factor

Related Blog Posts

• Regime Detection for Factor Rotation: A Practical Guide — practical implementation of regime-aware factor rotation
• Deflated Sharpe Ratios: How to Account for Multiple Testing — methodology for correcting selection bias

Keywords

Yield curve, regime classification, factor rotation, hidden Markov model, macro regimes, equity factors

@article{quanthq2026yield,
  title={Yield Curve Inversion as a Regime Classifier for Equity Factor Rotation},
  author={Liu, J. and Nakamura, T.},
  year={2026}
}