The impact of uncertain financial risk on the operation efficiency of banks

Li Huizhi; Yu Xianghua

doi:10.1016/j.heliyon.2023.e21378

. 2023 Nov 13;9(12):e21378. doi: 10.1016/j.heliyon.2023.e21378

The impact of uncertain financial risk on the operation efficiency of banks

Li Huizhi ^a,^b, Yu Xianghua ^c,^d,^∗

PMCID: PMC10703606 PMID: 38076192

Abstract

This paper introduces the latest research on the application of the new dynamic network DEA method in the measurement of the operating efficiency of the banking industry, and measures the operating efficiency of 15 listed banks in China by using the directional distance function (DDF) and the new dynamic network DEA model combined with the Malmquist index. To effectively evaluate the impact of uncertain financial risks on the operating efficiency of banks, we respectively selected the measurement indicators of total assets, number of employees, deposits, and net profit to calculate the operating efficiency of banks in the process of medium and long-term investment. According to the actual situation of bank operation, the process of bank operation is divided into the input stage of capital and manpower, the profit stage of capital operation, and operation stage, and a two-stage DEA network model is constructed. By measuring the overall operating efficiency of China's 15 banks from 2017 to 2022, the results show that the new dynamic network DEA model can better distinguish the efficiency of China's banks from the single-stage DEA model. In the new dynamic network model, the total factor productivity of most banks in China is reduced to varying degrees, which is greatly affected by pure technical efficiency.

Keywords: New dynamic network, DEA, Malmquist index, Total factor productivity, Bank operating efficiency

1. Introduction Foreword

Banks play a significant role in people's daily life and provide consumers with a range of financial services. The serving objectives of banks range from individuals to corporations.Even the government has a close relationship with the banks.Individuals get the reward from deposition and investment and meet their needs of consumption by takin loans from the banks.The banks also provide Help enterprises to borrow money to finance investment projects, investment to obtain income, insurance to protect property, and bills to pay for goods. To help the government finance public programs, protect public property, and pay for public expenditures. Therefore, the research on the operating efficiency of banks plays a vital role in the stability of financial markets and the healthy development of the economy.

In 1987, Charnes and others [1]first put forward the Data Envelopment Analysis (DEA). Their first model is named the CCR model. From the perspective of production function, it is used to study the "production departments" with multiple production inputs and outputs. However, the single-stage DEA model can not show the production process within the DMU, which leads to the inefficiency of the evaluated DMU. Because of this, Färe and Grosskopf [2]established a Malmquist productivity change index to examine productivity changes in two adjacent periods, and defined it as the Malmquist total factor productivity index. Kao [3]established a two-stage network DEA model. As an input to the second phase. Different from the previous studies which regard the whole production process and the two sub-processes as independent, the tandem relationship between the two sub-processes is considered when measuring the efficiency, and the overall efficiency is the product of the efficiencies of the two sub-processes.Therefore, the efficiency calculated by the two-stage DEA method is more meaningful than that calculated by the independent two-stage DEA method.

The new dynamic network DEA model is widely used in the evaluation of bank efficiency at present. Based on the analysis of the structure and characteristics of the banking business, Han and Su [4]regard banks as a series of two-stage network production structures with intermediate inputs and intermediate outputs. The first stage is liability business and intermediary business, and the second stage is asset business. Based on this network structure, a comprehensive network DEA model with intermediate inputs and intermediate outputs is established to measure the overall efficiency of banks, sub-stage efficiency, correlation index, and projection of each bank. Lu et al. [5]measured the technical efficiency and pure technical efficiency of major commercial banks based on the network DEA method from the new perspective of savings intermediate variables. Taking savings as an intermediate variable, Xiang and Zhao [6]divide the operation process of commercial banks into two stages: fundraising and fund operation. Based On the idea of network DEA and weight balance cross-efficiency, this paper uses the entropy method to deal with the aggregation problem of cross-efficiency, and constructs a cross-efficiency evaluation model for each subsystem and chain system of commercial banks. Evaluate the comprehensive efficiency and stage efficiency of listed commercial banks. Wang et al. [7]constructed a relational network DEA model for a chain system under the assumption of variable returns to scale, and measured the pure technical efficiency of the whole system and each sub-process to examine the effectiveness of decision-making unit links Huang et al. [8]Combined with the relational network DEAmodel under the condition of constant returns to scale, the scale efficiency of chain-like DMUs is further measured.Zhao [9]divided its business process into two stages: fundraising and fund operation, and used an improved principal component analysis method to screen indicators and build an index system. The cross-efficiency evaluation model of weight balance is used to solve the problem of cross-efficiency, and the entropy method is used to deal with the aggregation problem of cross-efficiency, and the cross-efficiency evaluation model of each subsystem and chain system of commercial banks is constructed. Wang et al. [10]used the DEA-Malmquist index method to analyze 28 sub-sectors of China's the manufacturing industry from 2003 to 2008, calculated the total factor energy efficiency index of manufacturing industry considering environmental effects, and used Tobit model to study its influencing factors.Wu et al. [11] established the DEA-Malmquist efficiency index with expected output and undesired output on agricultural carbon emissions of 31 provinces (cities, districts) in China from 2000 to 2011 to study the agricultural economic accounting. Carbon emissions, based on the systematic measurement of agricultural carbon emissions, measured the changing trend of agricultural carbon emission efficiency, and analyzed the provincial differences and changing trends of the agricultural Malmquist carbon emission efficiency index and its decomposition index. Wang and Ma [12]Based on the logistics industry efficiency of 30 provincial regions in China from 1997 to 2009, using the Malmquist-luenberger (ML) productivity index method to measure the efficiency of the logistics industry including unexpected output, and using the three-stage DEA model to study the external operating environment of logistics The impact of conditions on the efficiency of China's logistics industry.Zhang Hewei [13] researched the national and provincial dairy farming models (free-range, small-scale, medium-scale,and large-scale) by combining the DEA-Malmquist productivity index method with the provincial panel data of dairy cattle breeding in China from 2004 to 2011. Conduct a comparative analysis to observe the relationship between the dairy farming model and the total factor productivity (TFP) of raw milk production. Chen et al. [14] established the DEA-Malmquist index model through the panel data of 281 township health centers in Hunan Province from 2000 to 2008, to study the changes in total factor productivity and its decomposition items in township health centers. Li et al. [15] used the original data of China's major coastal ports and used the dynamic DEA-Malmquist total factor productivity index as a theoretical tool to evaluate the dynamic efficiency of the major coastal ports on both sides of the Taiwan Strait. competitive situation.Sun et al. [16] quantified and calculated the efficiency of financial support for provincial economic development through the provincial data of China from 1998 to 2010 subdivided it into three major industries, and used the DEA-Malmquist index method to study whether the efficiency of economic development is related to stable and sustainable economic development. Gong et al. [17] established the super-efficiency DEA model and the Melmqulst index model based on the raw data of tourism in 13 cities in Jiangsu from 2001 to 2011 to measure static efficiency, and dynamic efficiency. Liu et al. [18] used the non-parametric DEA-Malmquist index method to quantitatively analyze the contribution rate of my country's scientific and technological progress. Feng et al. [19] used the DEA method to evaluate the overall development status of the basic data of scientific research in my country's universities from 2000 to 2009, and used the Malmquist index to analyze the dynamic changes of scientific research efficiency in universities. Huang and Zhang [20] used the Malmquist index decomposition method of the DEA model to analyze the strategic emerging industrial technology by using the inter-provincial panel data of 28 provincial administrative regions and three major regions (East, Central, and West) in China from 2005 to 2012. The innovative TFP growth is decomposed into changes in technological progress, pure technical efficiency and knowledge innovation efficiency, and its dynamic changes in TFP, reasons for TFP growth changes, inter-provincial and regional differentiation characteristics, and regional differences in the source of TFP growth momentum are analyzed.

In conclusion, the research method of this study is based on the research of previous scholars and we conducted a deeper EDA refinement stage, which can show the research object more clearly. And make more accurate conclusions and recommendations.

2. Research methods and design

2.1. Research methods

The Data Envelopment Analysis (DEA) is expressed as the ratio of output to input. It attempts to maximize the efficiency of service units by comparing the efficiency of a particular unit with the performance of a group of similar units providing the same service. In this process, some units that achieve 100 % efficiency are called relatively efficient units, while other units that score less than 100 % efficiency are called inefficient units. It is suitable for the performance evaluation of multi-objective decision-making units with multi-input and multi-output. Because there is no need to assume or estimate the production function and formula in advance, it is also applicable to the input or output indicators which are difficult to price, and avoids the impact of the subjective factors of decision-makers, so many mature DEA models have emerged, and gradually become one of the most important methods of relative efficiency evaluation.

Network DEA model is a DEA method to study the efficiency of network decision-making units with multi-stage input and output of a network structure. The purpose of establishing network DEA model is to make full use of the input and output information of each stage. If the traditional DEA method regards the evaluation system as a "black box", the network DEA model is to "open the black box" to evaluate the efficiency.

Malmquist index can be used to analyze the change in productivity and the effects of technical efficiency and technical progress on the change of productivity when the data are panel data with multi-period observations. The commonly used Malmquist total factor productivity (TFP, TFP) index analysis, is used to analyze the change in productivity in two periods (the ratio of productivity in the latter period to that in the former period). F Färe R et al. (1992) first used the DEA method to calculate the Malmquist index, and decomposed the Malmquist index into two changes: one is the technical efficiency change of the evaluated DMU in two periods. The second is the technological change (TC), which reflects the frontier shift in DEA analysis. It can be used to process panel data and measure the dynamic change of total factor productivity of decision-making units with multiple inputs and outputs in different periods.

2.2. Study design

The three-stage network DEA model designed for the bank operation process division is shown in Fig. 1.

2.2.1. Establishment of model

Now, suppose that the evaluation has n DMUs, each bank has m kinds of input variables and s kinds of output variables, X_ijis the ith input of the jth bank, and Y_rjis the rth output of the jth bank. Thus, the input of the jth bank can be expressed as_j= (X_1j, X_2j, …, x^T; And the output is expressed as Y_j= (Y_1j, Y_2j, …, y_sj)^T, say V = (ν₁, ν₂, …, ν_m)^Tis the coefficient vector of input, and the coefficient vector of output Y is u = (u₁, u₂, …, u_s)^T, the function formula is:

\begin{array}{l} E_{k} = \max \sum_{s}^{r = 1} u_{r} Y_{r k} / \sum_{m}^{i = 1} v_{i} X_{i k} \\ s . t . \sum_{s}^{r = 1} u_{r} Y_{j k} / \sum_{m}^{i = 1} v_{i} X_{i k} ⩽ 1, j = 1,2, \dots, n \\ u_{r}, v_{i} ⩾ ε, r = 1,2, \dots, s; i = 1,2, \dots, m \end{array}

ε is a small non-Archimedean number. Each DMU applies m inputs to produce s outputs, and E_kis the relative efficiency of the DMU_k. When E (TV41) = 1, DEA is effective, and when E (TV41) < 1, DEA is ineffective. Suppose a system consists of a series of two subprocesses. Using m inputs_ikand s outputs Y_rk, the intermediate product is Z_px, which is the output of stage 1 and the input of stage 2. Traditional two-stage DEA studies use the above model to measure the overall efficiency, and the following models measure the efficiency of stages 1 and 2, respectively.

\begin{array}{l} E_{k}^{1} = \max \sum_{q}^{p = 1} w_{p} Z_{p k} / \sum_{m}^{i = 1} v_{i} X_{i k} \\ s . t . \sum_{q}^{p = 1} w_{p} Z_{p j} / \sum_{m}^{i = 1} v_{i} X_{i k} ⩽ 1, j = 1,2, \dots, n \\ w_{p}, v_{i} ⩾ ε, p = 1,2, \dots, q; i = 1,2, \dots, m \end{array}

Because the efficiencies of the two models are calculated independently, another model is used to build the relevance of the entire process:

\begin{array}{l} E_{k}^{2} = \max \sum_{3}^{r = 1} u_{r} Y_{r k} / \sum_{q}^{p = 1} w_{p} Z_{p k} \\ s . t . \sum_{s}^{r = 1} u_{r} Y_{r j} / \sum_{q}^{p = 1} w_{p} Z_{p j} ⩽ 1, j = 1,2, \dots, n \\ u_{r}, w_{p} ⩾ ε, r = 1,2, \dots, q; i = 1,2, \dots, m \end{array}

The following model is transformed into a linear equivalent model:

\begin{array}{l} E_{k} = \max \sum_{s}^{r = 1} u_{r} Y_{r i} / \sum_{m}^{i = 1} v_{j} X_{i k} \\ s . t . \sum_{s}^{r = 1} u_{r} Y_{r j} / \sum_{m}^{i = 1} v_{i} X_{i j} ⩽ 1, j = 1,2, \dots, n \\ \sum_{s}^{r = 1} w_{p} Z_{p j} / \sum_{m}^{i = 1} v_{i} X_{i j} ⩽ 1, j = 1,2, \dots, n \\ \sum_{s}^{r = 1} u_{r} Y_{i j} / \sum_{q}^{p = 1} w_{p} Z_{p j} ⩽ 1, j = 1,2, \dots, n \\ u_{r}, v_{i}, w_{p} ⩾ ε, r = 1,2, \dots, s; i = 1,2, \dots, m; p = 1,2, \dots, q \end{array}

Because the optimal multiplier solved by the following model may not be unique, the overall efficiency values $E_{k} = E_{k}^{1} \cdot E_{k}^{2}$ It's not unique. So to solve this problem, we need to find a set of multipliers to produce the maximum efficiency value, and then calculate the overall efficiency score according to the following model. The model is:

\begin{array}{l} E_{k} = \max \sum_{r = 1}^{s} u_{r} Y_{r k} \\ s . t . \sum_{i = 1}^{m} v_{i} X_{i j} = 1, j = 1,2, \dots, n \\ \sum_{r = 1}^{s} u_{r} Y_{r j} - \sum_{i = 1}^{m} v_{i} X_{i j} ⩽ 0, j = 1,2, \dots, n \\ \sum_{p = 1}^{q} w_{p} z_{p j} - \sum_{i = 1}^{m} v_{i} X_{i j} ⩽ 0, j = 1,2, \dots, n \\ \begin{array}{c} \sum_{r = 1}^{s} u_{r} Y_{r j} - \sum_{p = 1}^{q} W_{p} Z_{p j} ⩽ 0, j = 1,2, \dots, n \\ u_{r}, v_{i}, w_{p} ⩾ ε, r = 1,2, \dots, s; i = 1,2, \dots, m; p = 1,2, \dots, q \end{array} \end{array}

According to the following model, $E_{k}^{1}$ Efficiency of the second stage: $E_{k}^{2} = E_{k} / E_{k}^{1}$ . Or by replacing the objective function of the following model with $\sum_{r = 1}^{S} U_{r} Y_{r k}$ . The first constraint is replaced by $\sum_{p = 1}^{q} W_{p} Z_{p j} = 1$ , the Phase 1 efficiency value $E_{k}^{1} = E_{k} / E_{k}^{2}$ .

\begin{array}{l} E_{k}^{1} = \max \sum_{p = 1}^{q} w_{p} Z_{p k} \\ s . t . \sum_{i = 1}^{m} v_{i} X_{i k} = 1, j = 1,2, \dots, n \\ \sum_{r = 1}^{s} u_{r} Y_{r j} - E_{k} \sum_{i = 1}^{m} v_{i} X_{i j} ⩽ 0, j = 1,2, \dots, n \\ \sum_{p = 1}^{q} w_{p} z_{p j} - \sum_{i = 1}^{m} v_{i} X_{i j} ⩽ 0, j = 1,2, \dots, n \\ \begin{array}{c} \sum_{r = 1}^{s} u_{r} Y_{r j} - \sum_{p = 1}^{q} W_{p} Z_{p j} ⩽ 0, j = 1,2, \dots, n \\ u_{r}, v_{i}, w_{p} ⩾ ε, r = 1,2, \dots, s; i = 1,2, \dots, m; p = 1,2, \dots, q \end{array} \end{array}

The DEA model can not measure the change in efficiency value in different periods. This problem can be solved by combining the Malmquist index theory with the DEA method, that is, the DEA-Malmquist model.

Assume that (X^t, y^t) represents the input and output of the t-th period, (X^t+1, y^t+1) represents the input and output of the t + 1-th period, $D_{c}^{t} (x^{t}, y^{t})$ 、 $D_{c}^{t + 1} (x^{t + 1}, y^{t + 1})$ Is the output distance function of the corresponding period, and the subscript C is the constant return to scale. Malmquist index formula:

t f p = M^{t + 1} (x^{t + 1}, y^{t + 1}, x^{t}, y^{t}) = {[\frac{D_{c}^{t} (x^{t + 1}, y^{t + 1})}{D_{c}^{t} (x^{t}, y^{t})} \times \frac{D_{c}^{t + 1} (x^{t + 1}, y^{t + 1})}{D_{c}^{t + 1} (x^{t}, y^{t})}]}^{\frac{1}{2}}

The Malmquist index is decomposed into the technical efficiency change index (Eff-ch) and the technical progress index (Tech), and the specific formula is as follows:

E f f c h = \frac{D_{c}^{t + 1} (x^{t + 1}, y^{t + 1})}{D_{c}^{t + 1} (x^{t}, y^{t})}

T e c h = {[\frac{D_{c}^{t} (x^{t + 1}, y^{t + 1})}{D_{c}^{t + 1} (x^{t + 1}, y^{t + 1})} \times \frac{D_{c}^{t} (x^{t}, y^{t})}{D_{c}^{t + 1} (x^{t}, y^{t})}]}^{\frac{1}{2}}

$t f p = M^{t + 1} (x^{t + 1}, y^{t + 1}, x^{t}, y^{t}) = E f f c h \times T e c h$ When the returns to scale are variable, the technical efficiency change index is decomposed into the pure technical efficiency index (pech) and the scale efficiency index (sech).

p e c h = \frac{D_{v}^{t + 1} (x^{t + 1}, y^{t + 1})}{D_{v}^{t + 1} (x^{t}, y^{t})}

sech = {[\frac{D_{v}^{t + 1} (x^{t + 1}, y^{t + 1}) / D_{c}^{t + 1} (x^{t + 1}, y^{t + 1})}{D_{v}^{t} (x^{t}, y^{t}) / D_{c}^{t} (x^{t}, y^{t})} \times \frac{D_{v}^{t + 1} (x^{t + 1}, y^{t + 1}) / D_{c}^{t + 1} (x^{t + 1}, y^{t + 1})}{D_{v}^{t} (x^{t}, y^{t}) / D_{c}^{t} (x^{t}, y^{t})}]}^{\frac{1}{2}}

$D_{v}^{t + 1} (x^{t + 1}, y^{t + 1})$ When the returns to scale are variable, the output distance function is based on the technical conditions of the t + 1 period, and the subscript V is variable returns to scale. Then the technical efficiency change index is: effch = pech × sech. Therefore, total factor productivity is the efficiency of bank operation. The formula is:

tfp = Effch \times Tech = pech \times sech \times tech

3. Empirical analysis

This paper takes the data from 2017 to 2022 as the research period, and selects 15 listed banks as the research object. The data comes from the China Statistical Yearbook (2017–2022) and the annual report of the bank's official website. Through the comparative analysis of pure technical efficiency, scale efficiency,and comprehensive efficiency of 15 banks, the source factors affecting operational efficiency are analyzed, and reasonable suggestions are put forward. The comprehensive efficiency value is derived by MaxDEA software, and the operating efficiency of 15 banks is shown in Table 1-Operating efficiency tables of the 15 banks.

Table 1.

Operating efficiency tables of the 15 banks.

Overall production and operation stage operation comprehensive efficiency						Comprehensive Efficiency of Bank Operation at the Stage of Capital and Manpower Input					Comprehensive Efficiency of Bank Operation in the Profitable Stage of Capital Operation
Bank	2017–2018	2018–2019	2019–2020	2020–2021	2021–2022	2017–2018	2018–2019	2019–2020	2020–2021	2021–2022	2017–2018	2018–2019	2019–2020	2020–2021	2021–2022
Everbright Bank	1.02074	1.05938	0.94149	1.10974	0.98938	1.07676	1.10310	1.07833	1.01353	1.00293	0.95469	0.95625	0.91000	1.12528	0.98752
Industrial and Commercial Bank	0.97871	0.96504	0.91542	1.04515	0.94115	0.99989	1.00683	1.09210	0.99687	1.00308	0.98641	0.97202	0.84085	1.04790	0.92357
Huaxia Bank	0.98540	0.93511	0.86653	1.05003	1.03372	0.98463	1.01530	0.94230	0.96793	1.06231	1.01150	0.94946	0.92190	1.06048	0.94622
China Construction Bank	0.99967	0.96153	0.91879	1.03302	0.99199	1.02464	1.00316	1.01648	1.01083	0.97756	1.00361	0.98107	0.90533	1.02337	0.96813
Bank of Communications	1.02658	1.04687	0.95906	1.09456	0.97856	1.10405	1.00517	1.00727	1.03556	1.01100	0.93786	1.02082	0.96001	1.09375	0.99691
Minsheng Bank	0.97637	1.07259	0.61944	0.98280	0.94345	1.05403	1.05618	1.00276	1.00380	1.01350	0.93068	0.95902	0.61940	0.98039	0.95675
Agricultural Bank	0.97696	0.95499	0.95863	0.94048	0.91612	1.00614	0.99012	1.08757	0.92399	0.98250	0.97949	0.96701	1.00308	0.96396	0.93244
Ping An Bank	1.01147	1.07947	0.93302	1.12573	1.15403	1.00750	1.03125	1.00039	0.99311	1.03438	1.00566	1.02081	0.94121	1.12344	1.13907
Pudong Development Bank	0.99594	1.03562	0.86191	0.88862	0.93156	1.02911	1.03895	0.97520	1.05308	1.04630	0.97230	0.95565	0.87173	0.84779	0.92088
Bank of Shanghai	1.26081	0.86526	0.93744	1.03851	0.97640	1.16051	0.87942	1.05498	1.16157	1.07564	1.22696	0.86916	0.93379	1.02223	0.97568
Industrial Bank	1.02859	1.08208	0.95087	1.16169	0.98879	1.03297	1.09102	1.01227	0.99853	0.98389	0.99068	0.95826	0.95466	1.24374	0.97517
Bank of China	0.95222	0.97991	0.94789	1.01222	1.00214	1.01356	1.00792	0.99686	0.98366	1.02908	0.95460	0.98719	0.95198	1.03124	0.93823
China Zheshang Bank	1.15766	1.15722	0.8325	1.21812	1.1100	1.02458	1.02542	1.02900	1.01886	1.08305	1.06419	1.11630	0.83042	1.09383	0.97093
China Merchants Bank	0.97508	1.03491	0.83576	0.93437	0.94198	1.08167	1.14105	1.01820	0.96263	2.50007	0.93351	0.98327	0.82278	0.95586	0.83309
CITIC Bank	1.02164	1.01987	0.95153	1.14441	1.02647	1.01491	1.02989	1.03689	1.04581	0.97614	1.00553	0.97596	0.93626	1.15312	1.02316
Mean value	1.02452	1.01666	0.89535	1.05196	0.99505	1.04100	1.02832	1.02337	1.01132	1.11876	0.99718	0.97815	0.89356	1.05109	0.96585

Open in a new tab

Through the analysis of the data in Table 1, Table 2, Table 3.It can be seen that the average value of the comprehensive efficiency of the bank's overall production and operation stage shows a state of fluctuation, and the average value of the comprehensive efficiency of the bank's overall production and operation stage has the same fluctuation trend as that of the profit stage of capital operation. From 2017 to 2022, the average operating efficiency of banks shows a downward trend year by year, with a slight increase from 2021 to 2022. From 2017 to 2022, the comprehensive efficiency of China Merchants Bank in 2021–2022 is the highest (2. 50007) among the 15 banks in the capital and manpower input stage in Table 1. Among the three indicators of pure technical efficiency, pure technical change, and scale change rate, the scale change rate has the greatest impact on the comprehensive operating efficiency of China Merchants Bank (1. 83247). From 2017 to 2022, the 15 banks in Table 1 had the lowest comprehensive operating efficiency (0. 87942) in 2018–2019, while Table 3 shows that the pure technical efficiency, pure technical change,and scale change rate have the greatest impact on the comprehensive operating efficiency of China Merchants Bank (0. 84185).

Table 2.

2021-2022 Bank EDA Index table.

t(Period)	Bank	MI(t-1, t)	PEC(t-1, t)	PTC(t-1, t)	SCH(t-1, t)
2021–2022	Everbright Bank	1.00293	0.82202	1.29986	0.93862
2021–2022	Industrial and Commercial Bank	1.00308	1.25331	0.92918	0.86134
2021–2022	Huaxia Bank	1.06231	0.87331	1.10837	1.09748
2021–2022	China Construction Bank	0.97756	0.96803	1.06248	0.95046
2021–2022	Bank of Communications	1.01100	0.97483	1.09768	0.94481
2021–2022	Minsheng Bank	1.01350	0.75554	1.32415	1.01304
2021–2022	Agricultural Bank	0.98250	0.97265	1.01433	0.99585
2021–2022	Ping An Bank	1.03438	0.72528	1.50170	0.94970
2021–2022	Pudong Development Bank	1.04630	0.93203	1.19399	0.94021
2021–2022	Bank of Shanghai	1.07564	1.11626	0.88769	1.08551
2021–2022	Industrial Bank	0.98389	0.88370	1.15575	0.96332
2021–2022	Bank of China	1.02908	0.99460	1.04166	0.99328
2021–2022	China Zheshang Bank	1.08305	0.96423	1.14485	0.98111
2021–2022	China Merchants Bank	2.50007	1.36653	0.99837	1.83247
2021–2022	CITIC Bank	0.97614	0.83571	1.19886	0.97427

Open in a new tab

Table 3.

2018-2019 Bank EDA Index table.

t(Period)	Bank	MI(t-1, t)	PEC(t-1, t)	PTC(t-1, t)	SCH(t-1, t)
2018–2019	Everbright Bank	1.10310	1.07148	1.04902	0.98140
2018–2019	Industrial and Commercial Bank	1.00683	0.98253	1.21306	0.84474
2018–2019	Huaxia Bank	1.01530	0.94869	1.02297	1.04616
2018–2019	China Construction Bank	1.00316	0.99891	1.09105	0.92043
2018–2019	Bank of Communications	1.00517	0.95627	1.16526	0.90205
2018–2019	Minsheng Bank	1.05618	1.04834	1.05459	0.95531
2018–2019	Agricultural Bank	0.99012	1.01011	1.01399	0.96668
2018–2019	Ping An Bank	1.03125	1.01752	1.05296	0.96250
2018–2019	Pudong Development Bank	1.03895	1.06851	1.05258	0.92375
2018–2019	Bank of Shanghai	0.87942	0.84185	0.94212	1.10879
2018–2019	Industrial Bank	1.09102	1.07685	1.05260	0.96252
2018–2019	Bank of China	1.00792	1.00292	1.00492	1.00005
2018–2019	China Zheshang Bank	1.02542	1.00538	1.04983	0.97151
2018–2019	China Merchants Bank	1.14105	1.00880	0.96625	1.17059
2018–2019	CITIC Bank	1.02989	1.04691	1.05122	0.93580

Open in a new tab

Through the analysis of the data in Table 1，4 and 5.It can be seen that from 2017 to 2022, the comprehensive operating efficiency of Shanghai Bank is the highest in 2017–2018 (1. 22696), while from Table 4, it can be seen that Shanghai Bank has the highest operating efficiency. Among the three indicators of pure technical efficiency, pure technical change, and scale change rate, the pure technical efficiency (1. 15751) has the greatest impact on the overall efficiency of China Merchants Bank. From 2017 to 2022, the comprehensive efficiency of Minsheng Bank in 2019–2020 is the lowest (0. 61940), while from Table 5, among the three indicators of pure technical efficiency, pure technical change, and scale change rate, the pure technical efficiency (0. 69411) has the greatest impact on Minsheng Bank's comprehensive efficiency.

Table 4.

2017-2018 Bank EDA Index table.

t(Period)	Bank	MI(t-1, t)	PEC(t-1, t)	PTC(t-1, t)	SCH(t-1, t)
2017–2018	Everbright Bank	0.95469	0.93815	1.00072	1.01689
2017–2018	Industrial and Commercial Bank	0.98641	0.98540	1.05441	0.94936
2017–2018	Huaxia Bank	1.01150	0.99161	1.01505	1.00493
2017–2018	China Construction Bank	1.00361	1.01670	1.09571	0.90089
2017–2018	Bank of Communications	0.93786	0.94409	1.03688	0.95806
2017–2018	Minsheng Bank	0.93068	0.93403	0.99163	1.00482
2017–2018	Agricultural Bank	0.97949	0.98496	0.99854	0.99590
2017–2018	Ping An Bank	1.00566	0.99071	1.00956	1.00546
2017–2018	Pudong Development Bank	0.97230	0.92789	1.04801	0.99984
2017–2018	Bank of Shanghai	1.22696	1.15751	1.02400	1.03515
2017–2018	Industrial Bank	0.99068	0.96560	1.02462	1.00130
2017–2018	Bank of China	0.95460	0.95820	1.02758	0.96949
2017–2018	China Zheshang Bank	1.06419	1.08923	1.14652	0.85215
2017–2018	China Merchants Bank	0.93351	0.99655	0.83555	1.12109
2017–2018	CITIC Bank	1.00553	0.96132	1.04651	0.99949

Open in a new tab

Table 5.

2019-2020 Bank EDA Index table.

t(Period)	Bank	MI(t-1, t)	PEC(t-1, t)	PTC(t-1, t)	SCH(t-1, t)
2019–2020	Everbright Bank	0.91000	0.96693	0.90719	1.03740
2019–2020	Industrial and Commercial Bank	0.84085	0.96021	0.90203	0.97081
2019–2020	Huaxia Bank	0.92190	0.98874	0.93552	0.99666
2019–2020	China Construction Bank	0.90533	0.98025	0.98573	0.93693
2019–2020	Bank of Communications	0.96001	1.11891	0.86907	0.98724
2019–2020	Minsheng Bank	0.61940	0.69411	0.91911	0.97090
2019–2020	Agricultural Bank	1.00308	1.42300	0.98361	0.71665
2019–2020	Ping An Bank	0.94121	1.00863	0.92483	1.00899
2019–2020	Pudong Development Bank	0.87173	0.97224	0.90788	0.98759
2019–2020	Bank of Shanghai	0.93379	1.11199	0.81863	1.02579
2019–2020	Industrial Bank	0.95466	1.08588	0.86892	1.01177
2019–2020	Bank of China	0.95198	1.07450	0.90601	0.97787
2019–2020	China Zheshang Bank	0.83042	0.90185	0.87873	1.04786
2019–2020	China Merchants Bank	0.82278	0.93654	0.88649	0.99101
2019–2020	CITIC Bank	0.93626	1.08818	0.89353	0.96290

Open in a new tab

Through the analysis of the data in Table 1，6 and 7. It can be seen that from 2017 to 2022, the comprehensive operating efficiency of Shanghai Bank is the highest in 2017–2018 (1. 26081), while from Table 6, it can be seen that Shanghai Bank has the highest operating efficiency. Among the three indicators of pure technical efficiency, pure technical change, and scale change rate, the pure technical efficiency (1. 15751) has the greatest impact on the overall efficiency of China Merchants Bank. From 2017 to 2022, the comprehensive efficiency of Minsheng Bank in 2019–2020 is the lowest (0. 61944), while from Table 7, the pure technical efficiency (0. 67485) has the greatest impact on the comprehensive efficiency of China Merchants Bank in the three indicators of pure technical efficiency, pure technical change, and scale change rate.

Table 6.

2018 operating efficiency compare.

t(Period)	Bank	MI(t-1, t)	PEC(t-1, t)	PTC(t-1, t)	SCH(t-1, t)
2018	Everbright Bank	1.02074	0.96091	1.02759	1.03373
2018	Industrial and Commercial Bank	0.97871	0.99826	1.04080	0.94198
2018	Huaxia Bank	0.98540	0.96439	1.00629	1.01540
2018	China Construction Bank	0.99967	1.01670	1.06670	0.92176
2018	Bank of Communications	1.02658	0.99928	1.08070	0.95059
2018	Minsheng Bank	0.97637	0.85958	1.13423	1.00144
2018	Agricultural Bank	0.97696	0.97109	1.00829	0.99778
2018	Ping An Bank	1.01147	0.98521	1.01386	1.0126
2018	Pudong Development Bank	0.99594	0.86857	1.14118	1.00477
2018	Bank of Shanghai	1.26081	1.15751	1.03476	1.05265
2018	Industrial Bank	1.02859	0.93766	1.12940	0.97128
2018	Bank of China	0.95222	0.94394	1.01425	0.99460
2018	China Zheshang Bank	1.15766	1.05919	1.16779	0.93592
2018	China Merchants Bank	0.97508	1.04680	0.84561	1.10153
2018	CITIC Bank	1.02164	0.93380	1.05910	1.03301

Open in a new tab

Table 7.

2020operating efficiency compare.

t(Period)	Bank	MI(t-1, t)	PEC(t-1, t)	PTC(t-1, t)	SCH(t-1, t)
2020	Everbright Bank	0.94149	0.96217	0.92543	1.05735
2020	Industrial and Commercial Bank	0.91542	1	0.94012	0.97372
2020	Huaxia Bank	0.86653	0.94444	0.92534	0.99152
2020	China Construction Bank	0.91879	0.97512	0.98564	0.95596
2020	Bank of Communications	0.95906	1.11891	0.86907	0.98626
2020	Minsheng Bank	0.61944	0.67485	0.92542	0.99186
2020	Agricultural Bank	0.95863	1.29301	0.98361	0.75374
2020	Ping An Bank	0.93302	0.97029	0.92539	1.03911
2020	Pudong Development Bank	0.86191	0.96552	0.88333	1.01059
2020	Bank of Shanghai	0.93744	1.11199	0.82453	1.02243
2020	Industrial Bank	0.95087	1.12287	0.86892	0.97455
2020	Bank of China	0.94789	1.02811	0.94021	0.98060
2020	China Zheshang Bank	0.83251	0.96759	0.83980	1.02451
2020	China Merchants Bank	0.83576	0.96764	0.79333	1.08871
2020	CITIC Bank	0.95153	1.09308	0.88953	0.97861

Open in a new tab

Overall, the comprehensive efficiency of Minsheng Bank in 2019–2020 shows the lowest value (0. 61944) in the stage of capital and manpower input, the stage of capital operation profit, and the stage of global production and operation. The pure technical efficiency index (67485) is the most important factor affecting comprehensive efficiency.

4. conclusion

Based on the single-stage DEA model, this paper establishes a two-stage new dynamic network DEA-Malmquist model, which examines the operating efficiency of 15 banks in China from 2017 to 2022 based on considering the internal operating process of banks, and makes an empirical analysis of the operating efficiency of 15 banks. The results show that: (1) From 2017 to 2022, the average value of bank operating efficiency shows a fluctuating state, from 2017 to 2021, the average value of bank operating efficiency shows a downward trend year by year, until 2022, there is a slight increase. (2) Pure technical efficiency is the main factor affecting the operational efficiency of banks, and its role in promoting and hindering the operational efficiency of banks is greater than that of scale efficiency; (3) There are some differences in the change range and influencing factors of the operating efficiency of different banks.Therefore, first, establish a complete database and build a private domain traffic pool. The traditional silo operation mode of banks has simple logic and single form, relatively speaking, it cuts off the integrity of bank data assets, and it is difficult to achieve the scale efficiency of data. Opening up the correlation between bank data and constructing the overall data map will provide a better future for banks. We will lay a solid foundation for all business operations, marketing activities, and customer management. Second, keep pace with the times and use scientific methods to reduce the burden on employees. The use of data kanban facilitates automatic data retrieval and analysis by bank staff, and liberates labor costs. It facilitates global multi-dimensional visual management of bank managers, and provides a reliable basis for bid management, index management, operation management, and forecasting decision-making. Third, refine the customer classification and reach out accurately. Take customer thinking as the starting point and end point of operation work, establish emotional links with customers, and tap potential customers. Fourth, is product thinking. In this day and age, the lines between product and customer service are increasingly blurred. Product positioning that meets customer expectations and enhances the core competitiveness of the entire industry. Fifth, grasp the scene and build an ecology. Banks need to have warm scenarios to acquire customers, not only financial scenarios with high frequency and rigid demand, but also non-financial scenarios with high frequency and rigid demand.Fifth, in the future, our banks will actively promote the goal of "reducing costs, increasing revenue, and improving efficiency" in the application of artificial intelligence, big data, cloud computing, blockchain, and other financial technologies, hoping to achieve digital "curve overtaking". Digital transformation is an inevitable choice for small and medium-sized commercial banks to conform to the trend of The Times and seize development opportunities.

Data availability statement

The original contributions presented in the study are included in the article or

supplementary material, further inquiries can be directed to the corresponding

Author/s.

CRediT authorship contribution statement

Li Huizhi: Writing – review & editing, Writing – original draft, Supervision, Resources, Methodology, Formal analysis. Yu Xianghua: Software, Investigation.

Declaration of competing interest

The author, Li Huizhi is the first author and the corresponding author of Yu Xiang Huawei.All authors disclosed no relevant relationships.

Footnotes

^{Appendix A}

Supplementary data to this article can be found online at https://doi.org/10.1016/j.heliyon.2023.e21378.

Contributor Information

Li Huizhi, Email: 247000705@qq.com.

Yu Xianghua, Email: weall2000@126.com.

Appendix A. Supplementary data

The following is the Supplementary data to this article.

Multimedia component 1

mmc1.docx^{(33.7KB, docx)}

References

1.Charnes A., Cooper W.W., Rhodes E. Measuring the efficiency of decision-making units. Eur. J. Oper. Res. 1978;2(6):429–444. [Google Scholar]
2.Färe R., Grosskopf S. Kluwer Academic Publishers; Boston: 1996. Intertemporal Production Frontiers: with Dynamic DEA [M] p. 656. 656. [Google Scholar]
3.Kao C. Efficiency decomposition in network data envelopment analysis: a relational model. Eur. J. Oper. Res. 2009;192:949–962. [Google Scholar]
4.Han Song, Su Xiong. Research on the structural efficiency of Chinese commercial banks: based on complex network DEA model. China Management Science. 2016;24(8):9. [Google Scholar]
5.Lu Feng, Liu Weiqi, Shi Jinfeng. Research on the efficiency of China's commercial banks — — based on the network DEA method from the new perspective of savings. China Soft Science. 2012;(2):11. [Google Scholar]
6.Xiang Xiaodong, Zhao Ziliao. Research on the efficiency evaluation of China's commercial banks based on network DEA cross-efficiency model. Industrial Technology and Economy. 2017;(2):9. [Google Scholar]
7.Yang J.W., Wang M.Q., Li D., et al. Evaluating the efficiency of commercial banks by using the DEA model of shared input correlation network. Accounting Monthly (II) 2016;(10):4. [Google Scholar]
8.Huang Yi, Ge Hong, Feng Yingjun. Relational network DEA model based on chain system: a case study of commercial banks in China. Systems Engineering Theory and Practice. 2009;(5):9. [Google Scholar]
9.Zhao Ziliao. Fuzhou University; 2018. Research on the Efficiency Evaluation of China's Commercial Banks Based on Network DEA Cross-Efficiency Model [D] [Google Scholar]
10.Wang Shanshan, Qu Xiaoe. Research on the change of total factor energy efficiency of China's manufacturing industry based on environmental effects. China Population, Resources and Environment. 2011;21(8):130–137. [Google Scholar]
11.Wu Xianrong, Zhang Junbiao, Tian Yun, et al. China's provincial agricultural carbon emissions: calculation, efficiency changes and influencing factors——based on DEA-malmquist index decomposition method and the Tobit model application. Resour. Sci. 2014;36(1):10. [Google Scholar]
12.Wang Weiguo, Ma Yueyue. Systems Engineering; 2012. Efficiency of China's Regional Logistics Industry——a Malmquist-luenberger index method based on a three-stage DEA model[J] [Google Scholar]
13.Zhang Fei, Longbao Wei. Research on my country's dairy farming scale and raw milk production efficiency——based on the DEA-malmquist method. Agricultural Modernization Research. 2013;34(4):5. [Google Scholar]
14.Chen Xiaoling, Liu Ying, Wang Xiaowan, et al. Dynamic efficiency analysis of township health centers in Hunan Province from 2000 to 2008 based on the DEA-Malmquist index model. Chin. J. Evidence-Based Med. 2012;12(5):4. [Google Scholar]
15.Li Lanbing, Liu Jun, Li Chunhui. Dynamic efficiency evaluation of main coastal ports in three regions of Taiwan Strait—based on DEA-malmquist total factor productivity index. Soft Sci. 2011;25(5):5. [Google Scholar]
16.Sun Aijun, Jiang Yu, Fang Xianming. Comparison of financial support for economic development efficiency——analysis based on DEA-malmquist index method. J. Cent. Univ. Finance Econ. 2011;(11):6. [Google Scholar]
17.Gong Yan, Gong Yan, zheng-rong Guo, et al. China Postdoctoral Science Foundation; 2013. Research on Jiangsu Tourism Development Efficiency and Countermeasures——Based on Super Efficiency DEA and Malmquist Index Analysis[C]//China Economic Sustainable Development Postdoctoral Academic Forum. National Postdoctoral Office of the Management Committee. [Google Scholar]
18.Yu Jie, Liu Runsheng, Cao Yan, et al. Research on the contribution rate of my country's scientific and technological progress based on the DEA-Malmquist method: 1979-2004. Soft Sci. 2009;23(2):6. [Google Scholar]
19.Guangdi Feng, Chen Peipei, Tian Jinfang. Analysis of scientific research efficiency of Chinese universities based on DEA-malmquist method——an empirical study from 30 provincial panel data [J]. Modern finance and economics. Journal of Tianjin University of Finance and Economics. 2012;(9):13. [Google Scholar]
20.Huang Haixia, Zhang Zhihe. Technological innovation efficiency of China's strategic emerging industries——based on DEA-malmquist index model. Technological Economy. 2015;34(1):8. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Multimedia component 1

mmc1.docx^{(33.7KB, docx)}

Data Availability Statement

The original contributions presented in the study are included in the article or

supplementary material, further inquiries can be directed to the corresponding

Author/s.

[bib1] 1.Charnes A., Cooper W.W., Rhodes E. Measuring the efficiency of decision-making units. Eur. J. Oper. Res. 1978;2(6):429–444. [Google Scholar]

[bib2] 2.Färe R., Grosskopf S. Kluwer Academic Publishers; Boston: 1996. Intertemporal Production Frontiers: with Dynamic DEA [M] p. 656. 656. [Google Scholar]

[bib3] 3.Kao C. Efficiency decomposition in network data envelopment analysis: a relational model. Eur. J. Oper. Res. 2009;192:949–962. [Google Scholar]

[bib4] 4.Han Song, Su Xiong. Research on the structural efficiency of Chinese commercial banks: based on complex network DEA model. China Management Science. 2016;24(8):9. [Google Scholar]

[bib5] 5.Lu Feng, Liu Weiqi, Shi Jinfeng. Research on the efficiency of China's commercial banks — — based on the network DEA method from the new perspective of savings. China Soft Science. 2012;(2):11. [Google Scholar]

[bib6] 6.Xiang Xiaodong, Zhao Ziliao. Research on the efficiency evaluation of China's commercial banks based on network DEA cross-efficiency model. Industrial Technology and Economy. 2017;(2):9. [Google Scholar]

[bib7] 7.Yang J.W., Wang M.Q., Li D., et al. Evaluating the efficiency of commercial banks by using the DEA model of shared input correlation network. Accounting Monthly (II) 2016;(10):4. [Google Scholar]

[bib8] 8.Huang Yi, Ge Hong, Feng Yingjun. Relational network DEA model based on chain system: a case study of commercial banks in China. Systems Engineering Theory and Practice. 2009;(5):9. [Google Scholar]

[bib9] 9.Zhao Ziliao. Fuzhou University; 2018. Research on the Efficiency Evaluation of China's Commercial Banks Based on Network DEA Cross-Efficiency Model [D] [Google Scholar]

[bib10] 10.Wang Shanshan, Qu Xiaoe. Research on the change of total factor energy efficiency of China's manufacturing industry based on environmental effects. China Population, Resources and Environment. 2011;21(8):130–137. [Google Scholar]

[bib11] 11.Wu Xianrong, Zhang Junbiao, Tian Yun, et al. China's provincial agricultural carbon emissions: calculation, efficiency changes and influencing factors——based on DEA-malmquist index decomposition method and the Tobit model application. Resour. Sci. 2014;36(1):10. [Google Scholar]

[bib12] 12.Wang Weiguo, Ma Yueyue. Systems Engineering; 2012. Efficiency of China's Regional Logistics Industry——a Malmquist-luenberger index method based on a three-stage DEA model[J] [Google Scholar]

[bib13] 13.Zhang Fei, Longbao Wei. Research on my country's dairy farming scale and raw milk production efficiency——based on the DEA-malmquist method. Agricultural Modernization Research. 2013;34(4):5. [Google Scholar]

[bib14] 14.Chen Xiaoling, Liu Ying, Wang Xiaowan, et al. Dynamic efficiency analysis of township health centers in Hunan Province from 2000 to 2008 based on the DEA-Malmquist index model. Chin. J. Evidence-Based Med. 2012;12(5):4. [Google Scholar]

[bib15] 15.Li Lanbing, Liu Jun, Li Chunhui. Dynamic efficiency evaluation of main coastal ports in three regions of Taiwan Strait—based on DEA-malmquist total factor productivity index. Soft Sci. 2011;25(5):5. [Google Scholar]

[bib16] 16.Sun Aijun, Jiang Yu, Fang Xianming. Comparison of financial support for economic development efficiency——analysis based on DEA-malmquist index method. J. Cent. Univ. Finance Econ. 2011;(11):6. [Google Scholar]

[bib17] 17.Gong Yan, Gong Yan, zheng-rong Guo, et al. China Postdoctoral Science Foundation; 2013. Research on Jiangsu Tourism Development Efficiency and Countermeasures——Based on Super Efficiency DEA and Malmquist Index Analysis[C]//China Economic Sustainable Development Postdoctoral Academic Forum. National Postdoctoral Office of the Management Committee. [Google Scholar]

[bib18] 18.Yu Jie, Liu Runsheng, Cao Yan, et al. Research on the contribution rate of my country's scientific and technological progress based on the DEA-Malmquist method: 1979-2004. Soft Sci. 2009;23(2):6. [Google Scholar]

[bib19] 19.Guangdi Feng, Chen Peipei, Tian Jinfang. Analysis of scientific research efficiency of Chinese universities based on DEA-malmquist method——an empirical study from 30 provincial panel data [J]. Modern finance and economics. Journal of Tianjin University of Finance and Economics. 2012;(9):13. [Google Scholar]

[bib20] 20.Huang Haixia, Zhang Zhihe. Technological innovation efficiency of China's strategic emerging industries——based on DEA-malmquist index model. Technological Economy. 2015;34(1):8. [Google Scholar]

PERMALINK

The impact of uncertain financial risk on the operation efficiency of banks

Li Huizhi

Yu Xianghua

Abstract

1. Introduction Foreword

2. Research methods and design

2.1. Research methods

2.2. Study design

Fig. 1.

2.2.1. Establishment of model

3. Empirical analysis

Table 1.

Table 2.

Table 3.

Table 4.

Table 5.

Table 6.

Table 7.

4. conclusion

Data availability statement

CRediT authorship contribution statement

Declaration of competing interest

Footnotes

Contributor Information

Appendix A. Supplementary data

References

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Cite

Add to Collections

PERMALINK

The impact of uncertain financial risk on the operation efficiency of banks

Li Huizhi

Yu Xianghua

Abstract

1. Introduction Foreword

2. Research methods and design

2.1. Research methods

2.2. Study design

Fig. 1.

2.2.1. Establishment of model

3. Empirical analysis

Table 1.

Table 2.

Table 3.

Table 4.

Table 5.

Table 6.

Table 7.

4. conclusion

Data availability statement

CRediT authorship contribution statement

Declaration of competing interest

Footnotes

Contributor Information

Appendix A. Supplementary data

References

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases