【作者】 徐娜；

【导师】 李英；

【作者基本信息】 东北大学 ， 金融学， 2015， 硕士

【摘要】 近年来,随着我国利率市场化和金融深化改革的加速推进,商业银行的经营环境愈发复杂。政府放宽银行业的准入机制,民营银行设立热潮来袭,以云计算与大数据等为代表的新一代互联网技术迅速崛起,第三方支付、阿里贷、移动支付等互联网金融蓬勃发展。这些变化都给商业银行的传统经营模式与盈利水平带来了巨大的挑战。在这样竞争激烈的复杂环境中,我国商业银行可持续发展的关键就是要提高自身的竞争力,而效率是银行业竞争力的集中体现。因此,研究我国商业银行的效率及其影响因素对我国银行业今后的发展有重要的理论意义和现实意义。本文主要运用DEA-Malmquist指数模型和Panel Data模型,对我国商业银行的全要素生产率及其影响因素进行实证研究,在此基础上提出相应的政策建议。具体内容如下。(1)回顾和梳理商业银行效率和全要素生产率的相关理论;阐述全要素生产率的测算方法,包括参数法和非参数法;介绍DEA模型与基于DEA模型的Malmquist指数方法。(2)基于我国14家商业银行2007-2013年间的数据,采用DEA-Malmquist指数模型,分别从银行业整体视角、所有制视角、个体视角对全要素生产率及其分解进行实证分析。实证结果表明,考察期间内商业银行的平均全要素生产率指数大于1,是得到改进的,并且改进主要得益于技术进步;股份制商业银行的全要素生产率要高于国有商业银行。(3)对影响我国商业银行全要素生产率的因素进行实证分析。具体过程是,运用测算出的全要素生产率指数作为因变量,以GDP增长率、M2增长率、互联网金融资产规模增长率、市场集中度、创新能力、资产收益率和资产费用率等7个因素作为自变量,建立固定效应面板模型,对其进行分析。实证结果表明,市场集中度和资产费用率对全要素生产率呈显著的负向影响,其他5个因素对其呈显著的正向影响。(4)提出提高我国商业银行全要素生产率的政策建议,包括提高成本控制能力,降低银行的资产费用率;促进银行业技术进步,提高银行的创新能力;促进多元化发展,拓展银行的业务模式。更多还原

【Abstract】 In recent years,with in-depth advancing of China’s interest rate marketization and financial deepening reform,the operating environment of commercial banks become more and more complex.The government relaxes entry mechanisms of banking,more and more private banks begin to set up.With the rapid rise of new generation of Internet technology such as cloud computing and big data,Internet finance such as the third party payment,Ali-loan,mobile payments is developing rapidly.All of these have brought the huge challenge to commercial bank’s traditional business model and profitability.In such complex and fiercely competitive environment,the key to the sustainable development of Chinese commercial banks is to improve their competitiveness.And efficiency is the concentrated reflection of banking competitiveness.Therefore,it have important theoretical meaning and realistic meaning to the future development of Chinese banking industry to research the efficiency of commercial banks and its influencing factors.In this paper,we use DEA-Malinquist index model and Panel Data model to do the empirical research on total factor productivity and its influencing factors of Chinese commercial banks.Put forward some corresponding policy recommendations on this basis.More details are as follows.(1)Review and summarize the related theory of efficiency and total factor productivity;expound the measurement of total factor productivity,including parametric and nonparametric method;introduce the DEA model and Malmquist index method based on DEA model.(2)Based on 14 commercial banks in China from 2007 to 2013,use DEA-Malmquist index model to do the empirical analysis about total factor productivity and its decomposition,respectively from three perspective which are the banking sector as a whole,ownership and each individual bank.The empirical results show that the total factor productivity index on average of the commercial banks during the period of investigation is greater than 1.It is improved and the improvement thanks to the technological progress.Total factor productivity of the joint-stock commercial banks exceeds the state-owned commercial banks.(3)Do the empirical analysis about influencing factors of total factor productivity of Chinese commercial banks.Specific process is to use the total factor productivity index measured before to be as dependent variable,and use seven factors to be as independent variables,including GDP growth rate,M2 growth rate,Internet finance’s assets growth rate,market concentration,innovation,return on assets and cost on assets.Establish a fixed effects regression model and do the empirical analysis.The empirical results show that the market concentration and cost on assets have significantly negative influence on the total factor productivity,on which five other factors have significantly positive influence.(4)Give some advice to improve total factor productivity of Chinese commercial banks,including improve the ability of cost control,reduce the banks’cost on assets;promote technological progress of the banking,improve the innovation ability of the bank;promote the diversified development,expand the bank’s business model.更多还原