An innovative Bayesian multiple indicator-multiple cause analysis of bank productivity
用贝叶斯蒙特卡洛方法改进银行效率与生产率增长的测算模型，分析2008-2015年欧盟15国银行数据，发现效率差异大、生产率下降主要源于技术退步，且小银行效率低但生产率增长更强。
Abstract Using Bayesian Monte Carlo methods, we augment a stochastic distance function measure of bank efficiency and productivity growth with indicators of financial stability, profitability, and capitalization. Our novel Multiple Indicator-Multiple Cause (MIMIC)-style model provides more precise estimates of policy-relevant parameters, including bank efficiency and productivity growth. Analyzing EU-15 banks from 2008 to 2015, we find significant disparities in efficiency, revealing a ‘two-speed’ banking sector. Productivity growth has declined, driven primarily by technological regress rather than managerial inefficiencies. Small and peripheral banks exhibit lower efficiency than larger, core-EU banks, though productivity growth appears stronger among smaller institutions. We show that greater technical efficiency is associated with higher profitability, capitalization, and financial stability, as well as reduced earnings volatility.