Margin Buying Intensity and Future Stock Returns: Evidence from China’s A-share Market

This study examines whether margin buying contains information about subsequent stock returns in China’s A-share market. Using a monthly data set from January 2022 through June 2026, we measure margin buying intensity as monthly margin purchases divided by monthly trading value. Because the 12–2 momentum control requires pre-sample return history, including signals from early 2022 would require monthly returns from 2021. Since the final return series begins in January 2022, the fully controlled baseline specification starts in January 2023. Firm- and month-fixed-effects estimates show that a one-standard-deviation increase in margin buying intensity predicts 0.49, 0.83, and 0.96 percentage points higher cumulative returns over the next one, three, and six months, respectively. Fama–MacBeth regressions with Newey–West corrections remain significantly positive at all three horizons, and quintile sorts produce equal-weighted high-minus-low spreads of 0.97%, 1.88%, and 3.66%. The findings survive winsorization of future returns, a no-look-ahead book-to-market control, alternative financing measures, alternative momentum controls, exclusions of the September–October 2024 market episode, and screens for financial firms and ST-designated stock-months. Additional checks show that the short- and medium-horizon relation survives stringent industry-time controls and alternative normalizations of financing demand, although the six-month estimate is more sensitive under these specifications. Fully interacted tests indicate a substantial positive shift in predictive strength from 2024 onward. A symmetric comparison shows that short-selling intensity predicts lower future returns in the pooled sample. Overall, leveraged long and short trading contain directionally different predictive information, while the strength of the margin-buying signal varies materially over time.