Confirmation and Matching in Post-Trade Operations: Employee Perceptions of Settlement Risk Reduction at a Global Custodian

This study, undertaken during a six-week internship in the post-trade operations function of State Street Corporation, Bengaluru, examines the role of trade confirmation and matching in reducing settlement risk, and situates employee-level perception evidence within the broader regulatory and academic literature on post-trade operational risk. A structured five-point Likert questionnaire covering ten confirmation- and matching-related statements was administered to 150 employees drawn from Trade Operations, Middle Office, Back Office, Custody Services, Risk Management and other operational functions. A one-sample t-test confirms that the composite perception score (M = 3.96, SD = 0.29) is significantly above the neutral midpoint (t = 40.146, p < 0.001), indicating strong overall employee confidence that confirmation and matching reduce settlement risk. However, internal-consistency reliability of the ten-item scale was low (Cronbach's α = −0.046), and neither Pearson correlation nor multiple regression (R² = 0.015, F(6,143) = 0.371, p = 0.896) found significant relationships among individual perception items, indicating that the items function as distinct judgements rather than a single unified attitude. One-way ANOVA found no significant difference in perception across departments (F = 0.45, p = 0.812) or years of experience (F = 0.749, p = 0.525), while the difference between male and female respondents approached but did not reach significance (t = 1.847, p = 0.067). The findings reveal a consistent "principle versus practice" gap: employees strongly endorse the conceptual importance of confirmation and matching but rate current matching timeliness, discrepancy resolution and system reliability more moderately. The study recommends prioritising matching turnaround time, strengthening exception-resolution workflows, continuing automation investment, and redesigning future perception instruments as validated, multi-dimensional scales.