Purpose Over the past decade, financial crises have intensified, largely due to globalization and the integration of financial systems. The 2007–08 Global Financial Crisis revealed the crucial role of “too-big-to-fail” institutions in exacerbating systemic risks, leading regulators to identify Systemically Important Financial Institutions (SIFIs). However, the “too-big-to-fail” approach has limitations, as it overlooks the complexities of financial interconnections, such as common exposures and indirect relationships between institutions. Using India's 2018–19 NBFC crisis as a case, this article examines: (1) whether firm-level traits such as size, leverage and funding capture systemic importance and (2) whether network centrality measures (PageRank, Betweenness, Eigenvector, Closeness) better predict institutional vulnerability during crises. Design/methodology/approach This article explores three risk theories – “too-big-to-fail,” “too-connected-to-fail” and “too-central-to-fail” – in assessing systemic risk within financial networks. A complex network approach, focusing on PageRank and other centrality measures like eigenvector centrality, eccentricity centrality, closeness centrality and betweenness centrality, is used to analyze and compare the predictive power of these theories in identifying vulnerable institutions, with the predictive power of firm-level characteristics. Findings By analyzing various centrality scores such as PageRank, Betweenness, Eigenvector and Closeness, the study finds that institutions most central in the financial network suffered the greatest losses during the NBFC crisis. However, eccentricity scores did not significantly predict systemic risk. Contrary to traditional assumptions, asset size alone was not a significant predictor of systemic risk, while short-term funding reliance and non-interest income emerged as key factors. The study supports the argument that smaller institutions, like NBFCs, may face greater vulnerabilities during financial crises, challenging the notion that size alone dictates systemic importance. Originality/value This study extends systemic risk research by integrating the “too-big-to-fail,” “too-interconnected-to-fail” and “too-central-to-fail” perspectives in the Indian context, demonstrating that size alone is a weak predictor of systemic vulnerability compared to network-based centrality measures. By applying complex network analytics to the NBFC crisis, the article provides regulators with dynamic, early-warning tools that capture contagion channels beyond firm size, making systemic risk monitoring more precise and policy-relevant.
Chaturvedi et al. (Fri,) studied this question.