The Impact of Bank Geographical Distribution on Credit Efficiency: An Analysis of Intermediary Effect Based on Credit Behavior
-
摘要: 银行物理网点分布对信贷效率的影响一直是学术界关注的重要问题,本文基于金融地理的视角,从理论和实证的角度分析了银行物理网点分布对银行信贷行为的影响,进而利用中介效应模型分析了二者关系对信贷效率的影响。研究发现:出于降低交易成本和控制信贷风险的考虑,银行距企业及其总部的地理越远,银行信贷集中度越高,中长期贷款比例越低; 同时,物理网点地理分布与信贷效率之间存在通过银行信贷行为的中介效应,但不同的信贷行为对信贷效率影响的方向不同,即银行提高信贷集中度的行为加剧了地理环境对银行信贷效率的负向影响,而降低中长期贷款比例的行为则可削弱地理环境对银行信贷效率的负向影响。因此,为优化信贷行为,提高信贷效率,应充分认识到地理环境在金融活动中的作用,从降低信息摩擦的角度优化银行地理分布。Abstract: The impact of banks' geographical distribution on credit efficiency has always been an important topicamong researchers.Froom the perspective of financial geography, this paper analyzes the influence of geographical distribution of bank physical outlets on bank credit behavior from the theoretical and empirical level, and applies the mediation effect model to analyze the impact of their relationship between on credit efficiency. The study finds that, for the sake of reducing transaction costs and controlling credit risk, the farther the bank physical outlets is from the enterprise and from its headquarters, the higher the concentration of bank credit is, and the lower the proportion of medium-and long-term loans are. At the same time, there is amediation effect between the geographical distribution of physical outlets and credit efficiency through bank credit behavior, but different credit behaviors have different impact on credit efficiency. That is, the increase of banks' concentration of credit aggravates the negative effect of geographical environment on the bank credit efficiency, while the reduction of the proportion of medium-and long-term loans can weaken the negative impact of geographical environment on bank credit efficiency. Therefore, to optimize credit behavior and improve credit efficiency, we should fully realize the role of geographical environment in financial activities and optimize the geographic distribution of bank physical outlets from the perspective of reducing information friction.
-
Key words:
- bank physical outlets /
- credit behavior /
- credit efficiency /
- mediation effect .
-
表 1 变量说明
名称 含义 定义 被解释变量 Dir 银行信贷流向指标 以各银行最大十家贷款客户贷款占总贷款比例(贷款集中度)表示 Mat 银行期限结构指标 以各银行中长期贷款占总贷款比例表示 解释变量 Ope 物理网点运营距离指标 以各银行各省分支网点密度加权和表示,权重为该银行在某省分支机构数占其全部分支机构数的比例 Fun 物理网点功能距离指标 以各银行分支机构所在省与其总部所在省距离的加权和表示,权重为该银行在某省分支机构数占其全部分支机构数的比例 控制变量 Gyg 银行股权性质指标 以各银行国有股权(国家持股+国有法人持股)比例表示 Cap 银行规模指标 以各银行资产规模表示 Mar 银行金融环境指标 以各银行总部所在地的市场化指数表示 表 2 变量描述性统计
名称 观测数 平均值 标准差 最小值 最大值 Dir 153 3.38 1.60 1.62 12.95 Mat 153 44.94 11.94 12 74 Ope 153 44.21 32.44 1.84 187.13 Fun 153 4.66 1.41 0.61 5.89 Cap 153 5.08 5.74 0.062 24.10 Gyg 153 46.54 25.47 1.30 100 Mar 153 8.29 1.06 6.53 9.78 表 3 信贷投向对银行物理网点运营距离和功能距离回归结果
因变量 自变量 系数 标准误 t值 p值 Dir Ope 0.014 0.005 2.869 0.005 Fun 0.402 0.188 2.140 0.035 Cap 0.026 0.041 0.641 0.523 Gyg 0.018 0.009 1.915 0.058 Mar -0.307 0.123 -2.494 0.014 _cons 3.040 0.827 3.676 0.000 F检验:固定效应对混合效应 H0:个体效应不存在 F(16,114)=43.20 p值:0.00 Hausman检验:固定效应对随机效应 H0:应为随机效应 chi2(6)=12.81 p值:0.046 表 4 信贷期限结构对银行物理网点运营距离和功能距离回归结果
因变量 自变量 系数 标准误 t值 p值 Mat Ope -0.096 0.055 -1.749 0.083 Fun -6.594 2.235 -2.950 0.004 Cap 0.710 0.455 1.560 0.122 Gyg 0.397 0.104 3.809 0.000 Mar 1.807 1.380 1.309 0.093 _cons 73.940 9.280 7.968 0.000 F检验:固定效应对混合效应 H0:个体效应不存在 F(16,114)=11.69 p值:0.00 Hausman检验:固定效应对随机效应 H0:应为随机效应 chi2(6)=36.04 p值:0.00 表 5 信贷效率对银行物理网点地理分布回归结果
因变量 自变量 系数 标准误 t值 p值 Eff Ope 0.015 0.003 4.769 0.000 Fun 0.328 0.126 2.607 0.010 Cap -0.001 0.026 -0.049 0.961 Gyg 0.014 0.006 2.357 0.020 Mar 0.099 0.078 1.279 0.203 _cons -1.318 0.522 -2.525 0.013 F检验:固定效应对混合效应 H0:个体效应不存在 F(16,114)=5.50 p值:0.00 固定效应对随机效应检验 H0:应为随机效应 chi2(6)=15.96 p值:0.014 表 6 信贷行为中介效应回归结果
因变量 自变量 系数 标准误 t值 p值 Eff Ope 0.016 0.003 5.212 0.000 Fun 0.383 0.121 3.159 0.002 Dir 0.168 0.056 3.018 0.003 Mat 0.015 0.005 3.088 0.003 Cap -0.017 0.024 -0.698 0.487 Gyg 0.005 0.006 0.831 0.408 Mar 0.179 0.073 2.447 0.016 _cons -2.962 0.600 -4.935 0.000 F检验:固定效应对混合效应 H0:个体效应不存在 F(16,112)=7.57 p值:0.00 固定效应对随机效应检验 H0:应为随机效应 chi2(7)=35.56 p值:0.00 表 7 信贷行为中介效应分析
自变量 直接效应 中介变量Dir 中介效应 大小 方向 性质 c a b a×b a×b×c Ope 0.016*** 0.014*** 0.168*** 0.002 45 >0 互补性中介效应 Fun 0.383*** 0.402** 0.067 54 >0 互补性中介效应 自变量 直接效应 中介变量Mat 中介效应 大小 方向 性质 c a b a×b a×b×c Ope 0.016*** -0.096* 0.015*** -0.001 44 <0 竞争性中介效应 Fun 0.383*** -6.594*** -0.104 31 <0 竞争性中介效应 注:a、b、c含义如图 4,*、**、***分别代表系数在10%、5%和1%置信水平下显著。 -
[1] 殷孟波. 金融业全面开放下的商业银行信贷行为研究[M]. 成都: 西南财经大学出版社, 2012: 326-328. [2] CHRISTOPHERS B. The territorial fix: Price, power and profit in the geographies of markets[J]. Progress in human geography, 2014(6): 754-770. [3] O'BRIENR. Global financial integration: the end of geography[M]: London: Royal Institute of International Affairs, 1992: 56-61. [4] DEYOUNGR, GLENNON D, NIGRO P. Borrower-lender distance, credit scoring, and loan performance: Evidence from informational-opaque small business borrowers[J]. Journal of financial intermediation, 2008(1): 113-143. [5] PETERSEN MA, RAJAN RG. Does distance still matter? The information revolution in small business lending[J]. The journal of finance, 2002(6): 2533-2570. [6] KNYAZEVA A, KNYAZEVA D. Does being your bank's neighbor matter?[J]. Journal of banking & finance, 2012(4): 1194-1209. [7] RAJAN U, SERU A, VIG V. Statistical default models and incentives[J]. The American economic review, 2010(2): 506-510. [8] RAJAN U, SERU A, VIG V. The failure of models that predict failure: Distance, incentives, and defaults[J]. Journal of financial economics, 2015(2): 237-260. [9] FILOMENI S, UDELL GF, ZAZZARO A. Hardening Soft Information: How Far Has Technology Taken Us?[J]. Csef working papers, 2016(2): 342-350. [10] AGARWAL S, HAUSWALD R. Distance and private information in lending[J]. The review of financial studies, 2010(7): 2757-2788. [11] BARTOLI F, FERRI G, MURRO P, et al. SME financing and the choice of lending technology in Italy: Complementarity or substitutability?[J]. Journal of banking & finance, 2013 (12): 5476-5485. [12] AGHION P, TIROLE J. Formal and real authority in organizations[J]. Journal of political economy, 1997(1): 1-29. [13] STEIN JC. Information production and capital allocation: Decentralized versus hierarchical firms[J]. The journal of finance, 2002(5): 1891-1921. [14] ALESSANDEINI P, PRESBITERO AF, ZAZZARO A. Banks, Distances and Firms' Financing Constraints[J]. Review of finance, 2009(2): 261-307. [15] SKRASTINS J, VIG V. How organizational hierarchy affects information production. IMFS working paper series[J], 2015(1): 92-93. [16] COTUGNO M, STEFANELLI V. Bank size, functional distance and loss given default rate of bank loans. International Journal of financial research[J], 2011(1): 165-180. [17] 陶锋, 胡军, 李诗田, 等. 金融地理结构如何影响企业生产率?——兼论金融供给侧结构性改革[J]. 经济研究, 2017(9): 55-71. doi: 10.3969/j.issn.1005-913X.2017.09.022 [18] 祝继高, 饶品贵, 鲍明明. 股权结构, 信贷行为与银行绩效——基于我国城市商业银行数据的实证研究[J]. 金融研究, 2012(7): 6-18. doi: https://www.cnki.com.cn/Article/CJFDTOTAL-JRYJ201207006.htm [19] 赵尚梅, 史宏梅, 杜华东. 地方政府在城市商业银行的大股东掏空行为——从地方政府融资平台贷款视角的研究[J]. 管理评论, 2013 (12): 32-41. doi: https://www.cnki.com.cn/Article/CJFDTOTAL-ZWGD201312005.htm [20] 王秀丽, 鲍明明, 张龙天. 金融发展、信贷行为与信贷效率——基于我国城市商业银行的实证研究[J]. 金融研究, 2014(7): 94-108. doi: https://www.cnki.com.cn/Article/CJFDTOTAL-JRYJ201407007.htm [21] 王小鲁, 余静文, 樊纲. 中国分省份市场化指数报告(2018)[M]. 北京: 社会科学文献出版社, 2019: 45-70. [22] HSIAO C. Analysis of panel data[M]. London: Cambridge University Press, 2014: 78-90. [23] 宋昌耀, 李涛, 李国平. 地理距离对中国民营企业银行贷款的影响[J]. 地理学报, 2021(8): 1835-1847. doi: https://www.cnki.com.cn/Article/CJFDTOTAL-DLXB202108003.htm