CSRIndeX: A Python tool for computing normative CSR benchmarks from SEC filings

Measuring corporate social responsibility (CSR) remains empirically challenging due to the lack of standardised, comparable data on firms’ social and environmental expenditures. While corporate disclosures increasingly reference sustainability and responsibility, actual CSR spending is rarely reported in a consistent, machine-readable format. This paper introduces CSRIndeX, an open-source Python software that addresses this gap by combining disclosure-based CSR measurement with normative spending benchmarks derived from firm profitability. CSRIndeX automatically retrieves Form 10-K filings from the United States Securities and Exchange Commission (SEC), extracts CSR-related narrative disclosures, and computes coverage-based disclosure breadth (Disclosure Quality, DQ) and thematic balance metrics. In parallel, the software retrieves firm-level net income from the SEC XBRL Company Facts API. It calculates user-defined normative CSR spending targets as a proportion of fiscal-year profits. By explicitly separating descriptive disclosure metrics from normative benchmarks, CSRIndeX provides a transparent and reproducible framework for empirical analysis of CSR, policy evaluation, and comparative research at the firm level.