Digital Factors of Commercial Bank Liquidity: An SVAR Analysis (In Case of Uzbekistan)

This study evaluates the possibilities of digital analysis and econometric modelling for ensuring the liquidity of commercial banks in Uzbekistan. Using monthly data of the Central Bank of the Republic of Uzbekistan for January 2017 – December 2025 (108 observations), four recursively identified Structural Vector Autoregression (SVAR) models are estimated, in which the dependent liquidity indicators are the ratio of highly liquid assets to total assets (HQLA/TA), the liquidity coverage ratio (LCR), the net stable funding ratio (NSFR) and the instant liquidity ratio (ILR), while inflation, real broad money, the money market rate, remote banking users and real POS turnover act as macroeconomic and digital factors. Stationarity, Johansen cointegration, lag selection and Granger causality tests are followed by impulse response analysis, forecast error variance decomposition, a stress test and a signal-based early warning system. Real money supply is the main external driver of liquidity, explaining 7–19% of its variability. The rapid expansion of remote banking has a negative cumulative effect on liquidity in all four models, Granger-causes the LCR at the 10% level and explains 5.1% of LCR variability, whereas POS turnover has no significant effect. Even under the worst historical 12-month scenario all ratios remain above the regulatory minimums, but the LCR buffer shrinks to only 3.2 percentage points.