How a No-Short-Selling Model Produces Non-Normal Returns

A new academic paper posted on arXiv under identifier 2505.18723 examines Deviations from Normality in a financial model that prohibits short-selling. The research challenges a foundational assumption in quantitative finance: that asset returns follow a normal distribution. For traders and risk managers, the paper exposes a structural blind spot in how standard pricing and portfolio models behave when short-selling is restricted.

The Mechanism Behind Non-Normal Returns

The paper's title points to a specific dynamic. When short-selling is banned, the usual arbitrage forces that push prices toward fundamental values are weakened. Negative information cannot be expressed through short positions. Without those forces, return distributions become skewed or fat-tailed in ways that standard Value-at-Risk models and Black-Scholes pricing fail to capture. The available summary does not provide specific numerical findings. The implication is clear nonetheless: any strategy relying on normality assumptions may systematically misprice risk in markets where short-selling is constrained.

The Practical Impact on Stock Selection and Risk Models

For traders building watchlists, the paper reinforces a critical point. Markets with short-selling bans – whether regulatory, such as the 2008 U.S. ban on financial stocks, or structural, such as certain emerging markets – require different modeling approaches. AlphaScala's stock market analysis already accounts for liquidity and volatility regimes. This research suggests that distributional assumptions should also be stress-tested against the short-selling environment. A stock that appears cheap under a normal distribution may actually carry higher tail risk if short-sellers cannot correct mispricing.

The paper also connects to broader debates about market efficiency. Without short-selling, negative information is slower to enter prices. That delay can create persistent deviations from fair value. Quantitative models built on normality will systematically underestimate this effect. For traders using mean-reversion strategies, this is a critical blind spot. The model's results suggest that the probability of extreme negative moves is higher than standard models predict when short-selling is restricted.

Many exchange-traded funds and smart-beta strategies implicitly assume normality in their risk models. When short-selling is restricted – for example, in markets with bans on naked short sales or in China's A-share market – these funds may underestimate drawdown risk. The paper provides a mathematical framework to quantify that error, even if the summary does not reveal specific parameters.

The Next Catalyst for the Framework

The paper is a theoretical contribution. Its practical test will come when quantitative funds begin incorporating its framework into backtesting environments. A replication study using real market data would be the next concrete catalyst. Historical episodes like the 2008 short-selling ban or the 2020 COVID-era restrictions in certain European markets provide natural laboratories. Until such a study appears, the research serves as a reminder to question normality assumptions in any portfolio that faces short-selling constraints.

For a related discussion on how market structure affects trading decisions, see AlphaScala's analysis of Alphabet's $84.75B raise and its implications for capital allocation under constraints.