A groundbreaking study by researchers at the University of Illinois Chicago, the University of Louisville, and the University of Kansas is poised to reshape our understanding of short selling and its implications for family-owned businesses. The research, soon to be published in the prestigious Journal of Finance, systematically challenges a widely held belief that family firms are uniquely vulnerable to information leaks that manifest as unusually heavy short selling before disappointing earnings announcements. This new analysis, leveraging a more comprehensive dataset and rigorous methodology, suggests that the predictive power of short selling and its sensitivity to negative news are consistent across both family and non-family enterprises.
The genesis of this debate lies in a seminal 2011 study by Ronald Anderson, David Reeb, and Wanli Zhao, also published in the Journal of Finance. That influential paper posited that short selling could indeed predict future returns in family firms, but not in their non-family counterparts. Furthermore, it reported a startling disparity: nearly seventeen times more abnormal short selling activity was observed before negative earnings announcements in family firms compared to non-family firms. The authors of that study speculated that internal disputes within controlling families might provide short sellers with privileged access to non-public information, thereby creating an information asymmetry that benefited them in family-owned companies. This assertion, supported by the empirical evidence presented at the time, gained significant traction within financial academia and industry circles, influencing perceptions of risk and information flow in family businesses.
However, the inherent nature of short selling—its reliance on astute market analysis and the identification of overlooked corporate weaknesses—suggests that such skills should be equally applicable, if not more so, in the broader universe of publicly traded companies, including those not characterized by family ownership. It is this very principle that prompted Ferhat Akbas, Constantin Mihas Chair and Professor of Finance at the University of Illinois Chicago; Minjie Huang, Associate Professor of Finance at the University of Louisville; and Felix Meschke, Associate Professor of Finance at the University of Kansas School of Business, to re-examine the foundational findings.
Revisiting the Data: A More Comprehensive Approach to Short Selling
The current study by Akbas, Huang, and Meschke builds directly upon the groundwork laid by Anderson, Reeb, and Zhao. By gaining access to the same family-firm data utilized in the earlier research, the new team was able to conduct a comparative analysis using a significantly more expansive record of short-selling activities. This extended dataset, covering the same firms and the same time period as the original study, provided a richer and more nuanced view of the marketplace.
Crucially, their findings diverge sharply from the earlier conclusions. The revised analysis demonstrates that short selling effectively predicts future returns in both family and non-family firms. Moreover, when examining the period leading up to earnings announcements, the researchers found no discernible evidence that pre-announcement short selling activity in family firms was any more sensitive to negative earnings news than in non-family firms. This direct empirical challenge suggests that the unique vulnerabilities attributed to family firms in the context of short selling may have been overstated.
Short Selling as a Predictor of Future Returns: A Unified Phenomenon
To investigate the predictive power of short selling on future stock performance, Akbas, Huang, and Meschke employed a daily ranking methodology. They categorized both family and non-family firms separately, assessing the percentage of each stock’s daily trading volume that originated from short sales. Subsequently, they compared the subsequent returns of stocks that experienced relatively light short selling with those that were heavily shorted.
The results of this portfolio analysis were compelling and consistent across firm types. Lightly shorted stocks, irrespective of their family ownership status, consistently delivered higher returns. After meticulously adjusting for established market factors such as market risk, firm size, value characteristics, and momentum, the researchers calculated a daily return gap. For family firms, this gap was an estimated 6.5 basis points (0.065%) in favor of lightly shorted stocks. For non-family firms, the corresponding gap was 4.9 basis points (0.049%). While both these gaps were statistically significant, indicating a genuine market effect, the difference between the two gaps was not statistically significant. This finding strongly implies that short sellers possess valuable insights into future returns that are applicable to both family and non-family businesses, without a clear advantage in one over the other.
Further reinforcing this conclusion, the researchers also conducted regression analyses. These statistical models revealed that short selling exhibited a stronger relationship with subsequent returns in non-family firms than in family firms. This regression outcome directly contradicts the notion that short sellers possess a greater informational edge in family-owned companies. Neither the portfolio performance metrics nor the regression analyses provided support for the hypothesis that family firms are uniquely susceptible to informed short selling.
Short Selling in the Shadow of Earnings Announcements: No Differential Sensitivity
The period immediately preceding corporate earnings announcements presents a critical juncture for market participants, and for short sellers in particular, it can be an opportunity to capitalize on anticipated negative news. If short sellers foresee disappointing financial results, they are likely to increase their short selling activity before such information becomes public. To test this hypothesis, Akbas, Huang, and Meschke quantified "abnormal short selling." This metric was calculated by comparing the average daily short sales in the 30 days leading up to an earnings announcement with the firm’s annual average short sales outside of these pre-announcement periods. The core question was whether firms experiencing larger negative earnings surprises would also exhibit more significant abnormal short selling activity prior to the announcement.
The findings here were equally striking in their divergence from previous assumptions. The researchers found no statistically significant difference in how pre-announcement short selling responded to earnings surprises in family versus non-family firms. In fact, the point estimates from their analysis suggested the opposite trend: the response of short selling to negative surprises was larger in non-family firms. This response was significant at the 10% level for negative surprises, and overall, the relationship across all types of surprises was statistically insignificant for family firms. This empirical outcome directly refutes the idea that family firms are uniquely targeted or more vulnerable to informed short selling ahead of negative news.
The Crucial Role of Comprehensive Short-Selling Data
A significant factor contributing to the differing conclusions between this new study and the earlier work lies in the data utilized. Akbas, Huang, and Meschke leveraged short-selling data sourced from exchanges and various other trading venues, enabling them to construct a more comprehensive picture of short-selling activities. While both studies examined the same firms and time periods, the current research’s samples for the return and earnings analyses contained a greater number of observations, enhancing the statistical power and reliability of their findings.
To validate the coverage and accuracy of their data, the researchers conducted a rigorous assessment by comparing their dataset with the Wharton Research Data Services (WRDS) Short Volume database. The WRDS data, which began in August 2009, partially overlaps with the period examined in the new study, extending through December 2010. During this overlapping period, the researchers found that their data captured an impressive 99% of the security-days for which WRDS reported short-sale volume. Moreover, on matched observations, the correlation coefficients were exceptionally high: 0.99 for short volume and 0.90 for the short-selling ratio. Importantly, portfolio sorts constructed using the WRDS data replicated the non-family result identified in their primary analysis, demonstrating that heavily shorted stocks underperformed, a finding consistent with their broader conclusions.
While this comparison provides strong confidence in the coverage of their data, particularly for the later years of the study period, it cannot definitively establish coverage for the earlier years examined. Furthermore, the researchers acknowledge that they cannot pinpoint the exact source of the discrepancies between their findings and those of Anderson, Reeb, and Zhao, as they did not have direct access to the original short-selling records used in the 2011 study. Nevertheless, the breadth and validation of their data offer a robust foundation for their conclusions.
Evidence from More Recent Eras: Consistency in Predictability
To ascertain whether the observed patterns held true in more contemporary market conditions, Akbas, Huang, and Meschke extended their analysis to more recent data. They employed two distinct definitions of a family firm to ensure a comprehensive examination. The first definition was based on founding-family ownership, with data extending through December 2025. The second definition focused on family ties among top executives, utilizing data up to December 2019.
Across both definitions of family firms, the results remained consistent with their earlier findings. Short selling consistently predicted future returns in non-family firms. While the evidence for family firms in these more recent periods was weaker, this observation aligns with and further supports their earlier regression results, which indicated a less pronounced relationship between short selling and future returns in family firms compared to non-family firms. This suggests a consistent pattern of short-selling efficacy across different firm structures over time.
Implications for Corporate Governance, Investment Strategies, and Regulatory Oversight
The findings of Akbas, Huang, and Meschke carry significant implications for various stakeholders within the financial ecosystem.
For Boards of Directors and Corporate Advisors: The earlier study had suggested that unusually heavy short selling preceding an earnings announcement might signal a particular vulnerability to information leaks in family firms. This new research directly challenges that assertion. Boards and their advisors should recognize that while unusual short selling activity still warrants scrutiny regarding "who knew what and when," the heightened vulnerability previously attributed to family firms is not empirically supported. The sensitivity of pre-announcement short selling to bad earnings news appears to be a uniform market phenomenon, rather than a specific risk associated with family-owned enterprises.
For Institutional Investors: The study underscores the persistent informational value embedded within short-selling activity for both family and non-family firms. Institutional investors can continue to leverage short-selling data as a valuable indicator of future stock performance. The finding that stocks with heavy short selling subsequently underperform those with light short selling, across both categories of firms, reinforces the notion that short sellers’ ability to identify companies likely to experience difficulties is not confined to family-owned companies. This suggests a broader applicability of short-selling analytics in investment decision-making.
For Regulators: The research provides a clear message to regulatory bodies: informed short selling does not appear to be an exclusive "family-firm problem." The study’s findings indicate that short sellers possess information about future returns in both family and non-family firms with comparable efficacy. Consequently, the researchers’ results do not support the implementation of additional or targeted oversight of trading activities specifically within family firms. Existing regulatory frameworks, designed to ensure market integrity and transparency for all publicly traded companies, appear to be sufficiently robust to address informed short selling across the entire market spectrum.
In essence, this comprehensive re-evaluation of short selling and its relationship with family firms offers a more unified and evidence-based perspective. By moving beyond the specific focus on family firms as uniquely vulnerable, the study provides a more generalized understanding of market dynamics, benefiting investors, corporate leaders, and regulators alike. The rigorous methodology and expansive data employed by Akbas, Huang, and Meschke offer a compelling counterpoint to established assumptions, paving the way for a more nuanced and accurate interpretation of short-selling behavior in the modern financial landscape.
