Cross-Border Dual-Listed Equities & AH Premium Puzzle

The Law of One Price is a foundational axiom in finance, asserting that two identical assets should trade at the same price. This law is systematically violated in the Chinese equity market, where companies simultaneously list "A-shares" on mainland exchanges and "H-shares" in Hong Kong. Despite identical dividend entitlements, A-shares historically trade at a massive, volatile premium to H-shares, known as the AH Premium Puzzle.

Overview

The Law of One Price is a foundational axiom in finance, asserting that two identical assets should trade at the same price. This law is systematically violated in the Chinese equity market, where companies simultaneously list "A-shares" on mainland exchanges (Shanghai/Shenzhen) and "H-shares" in Hong Kong. Despite identical dividend entitlements, A-shares historically trade at a massive, volatile premium to H-shares, known as the AH Premium Puzzle.

Key Concepts

A-Share Market — Mainland Chinese market characterized by high liquidity, retail-driven speculation, and high turnover rates.
H-Share Market — Hong Kong market dominated by global institutional investors focused on strict fundamental valuation.
Liquidity Premium — The compensation investors demand for the cost and risk of illiquidity. Modeled by metrics like Amihud Measure and Pastor-Stambaugh reversals.
Limits to Arbitrage — Structural barriers that prevent arbitrageurs from forcing prices to converge, such as agency frictions, idiosyncratic risk, and short-sale constraints.
Capital Outflow Controls — Strict government restrictions preventing domestic retail investors from transferring capital offshore to buy cheaper H-shares. This traps massive liquidity onshore, driving the AH premium.

Formulas

The Liquidity-Adjusted CAPM (LCAPM)

E(R_i) = R_f + E(c_i) + \lambda\beta_{net}^i

Where $R_f$ is the risk-free rate, $E(c_i)$ is the expected illiquidity cost, and $\lambda$ is the market price of risk.

Cointegration Spread

e_t = P_t^A - \beta P_t^H

Used to model the long-term equilibrium spread between non-stationary price series.

Vector Error Correction Models (VECM)

\Delta P_t^A = \alpha_A(P_{t-1}^A - \beta P_{t-1}^H) + \sum \gamma_{A,i}\Delta P_{t-i}^A + \sum \delta_{A,i}\Delta P_{t-i}^H + \varepsilon_t^A

\Delta P_t^H = \alpha_H(P_{t-1}^A - \beta P_{t-1}^H) + \sum \gamma_{H,i}\Delta P_{t-i}^A + \sum \delta_{H,i}\Delta P_{t-i}^H + \varepsilon_t^H

Models long-term equilibrium and short-term dynamics simultaneously, where $\alpha$ represents the speed of adjustment.

Key Takeaways

Structural Frictions Override Theory: Calculating a theoretical "fair value" based on identical cash flows is insufficient if institutional frictions prevent capital from forcing convergence.
Microstructure Asymmetry: The AH premium is driven by divergent market demographics (retail vs. institutional), capital controls, and asymmetric tax regimes.
Quantitative Trading Strategy: Quants use statistical arbitrage techniques like VECM and Machine Learning models to harvest alpha from mean-reverting properties of the AH spread, factoring in threshold cointegration due to transaction costs.