|
|
|
| Leverage, Information and Liquidity: Asymmetric Effects of Margin Trading and Short Selling on Stock Liquidity |
| WANG Yongqin, LI Zhuochu, XIA Mengjia
|
| School of Economics, Fudan University;Economics Department, University of Pennsylvania |
|
|
|
Abstract Market liquidity and price discovery constitute two central dimensions of financial market efficiency (O’Hara, 2003), with adequate liquidity being a necessary condition for efficient price formation. In equity markets, investors hold heterogeneous beliefs: optimistic investors tend to buy, whereas pessimistic investors tend to sell. Liquidity facilitates the aggregation of these heterogeneous opinions by reducing trading frictions. More fundamentally, liquidity is intrinsically linked to the endogenous determination of marginal price setters, namely, which investors’ beliefs ultimately get reflected in market prices.
Leverage reshapes this equilibrium by altering marginal investors. Optimistic investors above the marginal type can long through margin financing, while pessimistic investors below the marginal type can short to express negative views. When markets are sufficiently liquid and long-short participation is balanced, both bullish and bearish beliefs can be incorporated into prices in a relatively symmetric manner, enhancing price efficiency. In contrast, under limited liquidity, information may still enter prices but often in an asymmetric and distorted manner. When optimistic investors dominate trading activity, leveraged buying amplifies one-sided price pressure and weakens the disciplining role of opposing beliefs.
This mechanism becomes particularly salient during economic expansions, when market-wide optimism prevails. In upswings, optimistic investors increase leverage to scale up purchases, amplifying price appreciation. Rising asset prices, in turn, relax collateral constraints and support higher leverage, reinforcing the influence of optimistic beliefs on prices. This positive feedback loop can exacerbate liquidity risk: when investors’ beliefs converge and trading is dominated by liquidity demand rather than supply, the scarcity of sellers may cause market liquidity to deteriorate.
Motivated by these considerations, this paper addresses three interrelated questions. First, how does market liquidity behave when asset prices are primarily driven by optimistic beliefs? Second, do margin trading and short selling exert symmetric effects on liquidity? Third, how do these effects evolve over market cycles, and through which information channels do they operate?
In an idealized setting with balanced long-short participation, margin trading and short selling together constitute a symmetric belief-expression mechanism: margin trading allows optimistic investors to lever up long positions, while short selling enables pessimistic investors to express bearish views. Under such conditions, heterogeneous beliefs are more fully incorporated into prices, improving both pricing efficiency and liquidity. In practice, however, margin trading and short selling are highly asymmetric. Margin financing volumes vastly exceed short-selling activity, resulting in a “one-sided leverage” market structure. In such an environment, optimistic beliefs can be disproportionately amplified and belief convergence may be intensified, while liquidity pressure may rise during market upswings, undermining overall market efficiency.
China’s introduction of margin trading and short selling in 2010 provides a unique quasi-natural experiment to examine these mechanisms. Using daily data on A-share stocks from 2009 to 2015, we employ a difference-in-differences (DID) framework to identify the causal effects of margin trading and short selling on stock-level liquidity. We find that, in the long run, the introduction of margin trading and short selling reduces the liquidity of eligible stocks. Decomposing the effects reveals pronounced asymmetry: margin financing significantly impairs liquidity, whereas short selling improves liquidity. Because margin financing overwhelmingly dominates trading activity, its negative effect drives the aggregate outcome. Moreover, the adverse liquidity effect is substantially stronger during bullish periods, consistent with asymmetric trading constraints that amplify liquidity imbalances in such periods. Mechanism analyses further show that margin trading, particularly margin financing, reduces stock price information efficiency and suppresses information production. The suppression of pessimistic information is especially pronounced during bullish periods. In addition, firms with higher default risk exhibit stronger incentives for information production and rely more heavily on private information to mitigate potential losses, highlighting substantial heterogeneity in firms’ informational responses.
This study contributes to the literature along three dimensions. First, at the theoretical level, existing research has largely progressed along two separate lines: one emphasizing the role of leverage and collateral constraints in asset pricing, the other examining how heterogeneous investor beliefs affect market efficiency. These two lines overlook their intrinsic linkages. This paper bridges these strands by integrating the collateral theory (Geanakoplos, 2010) with the heterogeneous-belief asset pricing model (Hong and Stein, 2003), providing a unified framework that explains how collateral constraints regulate belief expression in prices and generate asymmetric liquidity effects over the business cycle. Second, at the empirical level, the paper exploits China’s institutional setting and the 2010 margin trading reform to implement a clean identification strategy, avoiding cross-country confounding factors related to institutional and cultural heterogeneity. This approach provides systematic evidence on the structural relationship among leverage, information, and liquidity, and documents their cyclical asymmetries. Third, in terms of policy relevance, the findings deepen our understanding of market microstructure and offer implications for macroprudential regulation. Structural constraints in China’s capital market, particularly the imbalance between long and short mechanisms and limited securities lending supply, remain binding, and the “one-sided leverage” structure persists. More broadly, the leverage-information-liquidity framework developed in this paper provides a useful lens for analyzing liquidity and risk transmission in other leveraged financial markets, including bonds, foreign exchange, and derivatives.
|
|
Received: 23 November 2023
Published: 01 April 2026
|
|
|
|
|
|
|
|
