Market Composition and Data-Quality Constraints in Global Listed-Securities Reference Files: Evidence from a Pre-2020 Cross-Market Dataset

Abstract

This study examines the empirical usability of pre-2020 global listed securities reference data by assessing market composition, venue concentration, and identifier completeness. Using a source file of 237 raw records and 56 fields, the analysis retains 147 observations created before 2020 and applies descriptive frequency analysis, cross-tabulation, robust summary statistics, and field-completeness checks. The findings show that the retained sample is dominated by exchange-traded funds and common equity, which together account for nearly the entire dataset. Listings are geographically diverse but concentrated in a limited number of countries, particularly GB, AU, HK, US, and NL, while trading activity is similarly clustered across major venues and currencies. Security-level identifiers, including ISIN, SEDOL, issuer name, security description, and trading currency, are complete or near-complete, supporting reproducible descriptive and cross-market analysis. However, materially weaker LEI and GICS coverage limits the dataset’s suitability for issuer-level linkage and sector-based empirical designs without supplementary enrichment. The study contributes by reframing securities reference data as research infrastructure rather than passive administrative metadata and by proposing a Reference-Data Usability Assessment Framework (RDUAF) to guide pre-analysis evaluation of dataset coverage, completeness, and research suitability. Overall, the evidence suggests that the examined file is well-suited to security-level market-structure profiling prior to the COVID-19 period but requires additional data integration to support more advanced firm-level, sectoral, or causal finance research.