A two-sided reform, a one-sided market: margin buying, turnover, and return predictability in China

This is the authors' abstract. We don't add key points for this paper.

China’s margin-trading and short-selling reform was designed to support two-sided trading, yet the resulting market is dominated by leveraged purchases. We ask whether this economically large financing channel contains incremental information about future returns beyond turnover. Across cross-sectional regressions, portfolio sorts, and out-of-sample forecasts, observed margin flow and the margin-financed share do not provide stable incremental return predictability once broader trading activity is considered. The event evidence instead shows that eligibility was directed towards stocks whose return and trading dynamics were already changing, precluding a causal interpretation of post-admission patterns. The gap between scale and predictive content is informative: the results are consistent with margin finance scaling up trading motives already reflected in turnover rather than creating a distinct, measurable information channel. The results also show why trading scale, conditional price information, and policy identification should be evaluated separately when assessing market reforms.