<p>This study empirically examines “liquidity dependence” in the Japanese banking system. Liquidity dependence describes a tendency whereby Quantitative Easing (QE) expands banks’ demandable liabilities, but these expansions do not reverse when policy shifts to Quantitative Tightening (QT), leading to heightened reliance on central bank liquidity. Acharya and Rajan (J Financ 79:2943, 2024) and Acharya et al. (Liquidity dependence and the waxing and waning of central bank balance sheets. NBER Working Paper 31050, 2024) document this phenomenon in the United States, where it is considered a contributing factor to the liquidity stresses observed in September 2019 and March 2023. Although similar patterns have been reported in Europe, empirical evidence for Japan remains limited. This study fills this gap and further contributes by evaluating more than two decades of Japan’s QE and QT history through the lens of liquidity dependence. Based on time-series analyses of macroeconomic data and panel analyses of bank-level microdata, we find that QE expands demandable deposits, whereas QT fails to unwind them and instead compresses time deposits, revealing asymmetric adjustments consistent with liquidity dependence. Moreover, by incorporating bank heterogeneity and comparing our findings with U.S. evidence, we show that the degree of asymmetry varies across banks and across countries. These results suggest that prolonged QE has structural implications for financial stability and for the design of QT.</p>