Submitted:
18 November 2026
Posted:
19 November 2026
You are already at the latest version
Abstract
This article investigates whether increases and decreases in the accounting equi-ty-to-assets ratio are associated asymmetrically with subsequent non-performing loans in Vietnamese commercial banks. Using a panel of 27 listed or trading-registered banks and 270 bank-year observations in the principal lagged specification, we estimate two-way fixed-effects models with bank-clustered standard errors. Annual changes in the equi-ty-to-assets ratio are decomposed into positive and negative signed components. The full-sample one-year-lagged model yields a positive coefficient for capital-ratio increases (0.2070) and a negative coefficient for decreases (−0.1501); a cluster-adjusted Wald test of slope equality gives F(1,26) = 3.9446 (p = 0.0577). A post-estimation 1% winsorization of current and lagged NPL retains all 27 banks and yields F(1,26) = 4.4311 (p = 0.0451); a sep-arate unmodified-data sensitivity specification excluding NVB gives F(1,25) = 5.8171 (p = 0.0235). The contemporaneous specification shows no comparable evidence of asymmetry (p = 0.6823). The findings highlight the importance of adjustment timing, signed capital changes, extreme-tail sensitivity, and influential restructuring episodes for interpreting the capital–credit-risk relationship. The analysis offers implications for capital planning, bank supervision and the transparent interpretation of small-cluster panel evidence.
Keywords:
equity-to-assets ratio
; non-performing loans
; asymmetric adjustment
; banking risk
; Vietnam
Copyright: This open access article is published under a Creative Commons CC BY 4.0 license, which permit the free download, distribution, and reuse, provided that the author and preprint are cited in any reuse.