ESG ratings are plagued by persistent methodological fragmentation and extremely low cross-agency correlation, especially for the social (S) and governance (G) pillars. While prior work has linked employee welfare to equity returns, it has failed to deliver a simple, replicable, and greenwashing-resistant way to measure how organizational friction shapes long-term firm risk and performance. This paper proposes an exploratory framework based on a simple heuristic: firms are organizational systems where employee effort and capital inputs combine to produce value added, and long‑term sustainability depends on how much of that value is returned to workers relative to what is lost to internal friction. We introduce two core constructs: the **Theoretical Organizational Signal‑to‑Noise Ratio (SNRₓ₇₄₎ₑₘ) **, an unobservable latent state variable capturing the net value returned to employees (formally defined in the Appendix) ; and its empirical proxies. Specifically, we construct an NLP‑based sentiment proxy (SNR₍₋) for validation (Section 3), and an accounting‑based proxy — **Organizational Friction Momentum (OFM) ** — defined as the residual from annual cross‑sectional regressions of the change in SG&A‑to‑value‑added ratio on standard profitability and investment variables (Section 4). First, we use Glassdoor employee reviews from 10 U. S. mega‑cap firms between 2008 and 2021 to validate the SNR₍₋ construct, showing that a sustained decline in SNR₍₋ reliably predicts higher 12‑month ahead tail risk. We then test Residual OFM on a broad cross‑section of U. S. equities from 2021 to 2025, and find its predictive power is concentrated almost entirely in human‑capital‑intensive (HCI) industries: Technology, Healthcare, and Business Services. In these sectors, the equal‑weighted long‑short return is 0. 985% per month (11. 8% annualized), with a Fama‑French five‑factor alpha of 1. 194% per month (t=2. 05) and a Fama‑MacBeth coefficient of -0. 3705 (t=-2. 67). For all other sectors, the predictive power is statistically and economically indistinguishable from zero. We also find the effect is strongly regime‑dependent: it is significant in 2024, a period of broad market breadth, but vanishes entirely in 2023, a regime of extreme liquidity concentration. Over the full 33‑month test window (2022–2025), the unconditional long‑short return for the full sample is 0. 570% per month (6. 84% annualized), with a Fama‑MacBeth coefficient of -0. 1224 (t=-1. 82). We want to be explicit: these results are preliminary and exploratory. Our sample suffers from meaningful survivorship bias (it includes only going‑concern firms) and covers a short time span, so we make no claims that OFM is a robust, unconditional alpha anomaly. The core contribution of this paper is a transparent, replicable, accounting‑based proxy for organizational friction—one of the least well‑measured components of the ESG’s S pillar—along with preliminary evidence that its predictive power is concentrated where human capital matters most.
guoyong chen (Thu,) studied this question.