Is the medium the message? Social disclosure channels and firm risk

Battle Count: 78

The paper is moderately relevant to quantitative trading. It provides insights into how disclosure channel information affects idiosyncratic risk, which is a key input for risk models and portfolio construction. The finding that first-time SEC disclosure increases idiosyncratic risk could inform event-driven strategies or risk-adjusted position sizing. However, the paper does not propose trading signals, backtest strategies, or directly address alpha generation. Its primary contribution is to risk understanding rather than return prediction.

The econometric methodology (panel OLS regression, Fama-French factor model, propensity score matching) is standard and well-documented. However, implementation requires access to specialized commercial data (Datamaran), multiple financial databases (Compustat, CRSP, IBES, MSCI), and careful construction of first-time/continued disclosure variables across three channels and three social topics. The interaction term analysis adds moderate complexity. No code is provided.

The study relies on Datamaran (a commercial machine-learning-based text analysis dataset) for sustainability disclosure data, which is not publicly available. Other data sources (Compustat, CRSP, IBES, MSCI ESG, Kenneth French's factor data) are standard but require subscriptions. The regression specifications are fully detailed, but the Datamaran data dependency significantly limits reproducibility. No code or replication package is mentioned.