Unilu, Belval Campus, Luxembourg, by mrspatbile
My research covers topics in Corporate Finance, including Governance, Investor Activism, and their connection to Asset Pricing. In my empirical work, I primarily use open-source data, eventually incorporating low-cost sources as needed, scripting all research steps to improve reproducibility.
Working Papers
Activist Investors Stakes and Market Movements
JEL Classification: G14, G23, G30, G32
Keywords: large shareholders, blockholders, activist investors, activism, corporate governance, corporate finance, voice
This paper investigates two questions related to activist blockholders: i) What drives their purchase decisions? ii) Is a higher share of activist holdings value increasing? We employ information disclosed in regulatory filings (SEC Schedule 13D) to address these inquiries as follows. The period between the event date (the day the investor reaches the regulatory ownership threshold, also known as trigger date) and the day of its public disclosure (filing date) is referred to as “pre-disclosure accumulation period” or “grace period”. We find that activist investors do indeed use the grace period to adjust their stakes, taking advantage of market-wide price swings. When the market experiences an upward deviation (last quintile) from its trend within the 10 days following the trigger date, initial stakes in dollars are 19% lower. Though activist investors use the pre-disclosure accumulation period strategically, we do not find evidence that this (plausibly exogenous) variation in ownership has an impact on the overall valuation of the firm, at least within the scope of our dataset and specification.
Post-event market prices, including computed values for trend (control variable) and deviation from the trend (instrumental variable - IV).
Question
What drives purchase decisions of activists?
Does activists use the accumulation interval strategically?
Is a higher activist stake value increasing?
Methodology
Variable: average of daily deviation of market returns, from its trend (10 leading days)
OLS (ownership) + IV (abnormal returns)
Findings
Strategic use of pre-disclosure interval
OLS: 5th Q deviations =⇒ decrease ownership in 19%
IV: no evidence that ownership stake is value increasing
Is there a premium for Sustainable Development Goals?
JEL Classification: G14, G23, G30, M14
Keywords: large shareholders, blockholders, activist investors, activism, corporate governance, sustainable finance, SDGs, textual analysis, NLP
This paper investigates the relationship between company valuations after being targeted by blockholder activists when their stated investment objectives exhibit some degree of similarity to Sustainable Development Goals (SDGs) as developped by the UN. Our empirical approach employs Natural Language Processing (NLP) to establish similarity scores from the textual content of activist regulatory filings (investment objectives) and the SDGs. We find a robust positive relationship that is both economically and statistically significant. Switching from no similarity to the average similarity increases the abnormal returns around the filing date of an activist investor by approximately 2%, about one fifth of the average abnormal return. These findings imply that activist investors may play a crucial rule in the transition towards a more sustainable economy.
The plot shows two principal components of LLM embedding for the textual content of: (1) activists' stated objectives (blue) and (2) the UN SDG goals and targets (red circle). Note that out-of-the-shelf LLM embeddings are not superior to simpler similarity techniques in this context - though with fine-tuning these models are expected to excel.
Question
Does activists have objectives associated to SDGs?
If so, do the intervention is linked to higher abnormal returns?
Methodology Similarity scores:
investment objective vs. SDG (goals)
investment objective vs. SDG (goals + objectives)
Findings
Higher similarity correlated to higher abnormal returns. Note we are not claiming causality.
Unveiling Non-Core Activist Blockholder Events
JEL Classification: G14, G23, G30, G32, C81
Keywords: activist investors, blockholder activism, corporate governance, data collection, data extraction, reproducibility
Blockholder activism is characterized by large shareholders with a declared intention to exert influence in corporate decision-making. Researchers have frequently used blockholder activist events as a means to investigate a variety of topics in Corporate Finance. While regulatory filings, the primary source of data, are publicly available, the process of extracting a suitable events dataset, one that exclusively includes instances in which an external investor targets a company with the intent to exert influence on its business, lacks thorough documentation, making it unfeasible to be reliably reproduced. In this paper, we discuss key considerations for using public data to compile an activist dataset that identifies non-core activist blockholder events. Furthermore, we investigate the consequences of failing to exclude these non-core events on two common research outcomes: short-term abnormal returns and ownership stakes. This work represents an initial step toward establishing a fully reproducible process for dataset assembly, potentially reducing barriers to future research in this field.
Number of flagged records identifying non-core activist events. Each datapoint corresponds to SEC filings (non-13Ds) captured in windows predating the activist event. Figures are aggregated quarterly.
Reproducible methodology to extract data using only public sources
Reveals the effect of non-core events in abnormal returns and ownership stakes
Methodology
fully documented, reproducible, quasi open-source
OLS outcomes over flags for non-core events
Findings
Suggest that previous published results might:
understate abnormal returns (insiders, bankruptcies, mergers)
overstate ownership stakes (duplicated filings)
Essays in Blockholder Activism PhD Thesis
The three research topics above were integral components of the PhD Thesis in Finance, titled Essays in Blockholder Activism, which I defended at the University of Luxembourg. The text in the thesis reflect the version of those working papers as of the day of the PhD defense, March 7, 2024, whereas the individual links (to be included soon) will direct to the most recent versions.
Dissertation defence committee
Dr Ulf Von Lilienfeld, dissertation supervisor
Professor, Université du Luxembourg
Dr Michael Halling, Chairman
Professor, Université du Luxembourg
Dr Roman Kräussl
Professor, Bayes Business School
Dr Christos Koulovatianos
Professor, Université du Luxembourg
Dr Jan Schnitzler
Professor, Grenoble École de Management
Research Tools and Resources
Data sourced from SEC Edgar and WRDS (CRSP + Kenneth French series) via APIs for full reproducibility. Articles edited and formatted with LaTeX on Overleaf. The following plots help give context to the data used in my research.
Number of records before and after cleaning (left) and resulting spatial distribution of filers (activists) and targeted companies.