Frontiers in Business, Economics and Management ISSN: 2766-824X | Vol. 7, No. 1, 2023 227 Research on the Influence of Financial Technology on The Profitability of Small and Medium‐sized Commercial Banks Yao Yang1, * 1School of Economics, Hefei University of Technology, Heifei ,230601, China *Corresponding author’s e-mail: y000cdcy@163.com Abstract: This paper analyzes the overall profitability of city commercial banks and agricultural commercial banks, analyzes how financial technology has an impact from the existing theory and impact mechanism, and establishes panel data in the empirical part for analysis, it is concluded that financial technology can promote the profitability of small and medium-sized commercial banks, but there is heterogeneity between urban and rural commercial banks, finally, the paper gives some suggestions on the healthy upgrading and transformation of small and medium-sized banks and how to regulate the use of financial technology by regulatory authorities. Keywords: Financial technology, Commercial banks, Profitability, PKU inclusive financial index. 1. Introduction Several of the hottest technologies of recent years, such as AI, Big Data, cloud computing, blockchain, digital payments and robotics consultants, are rapidly reshaping traditional banking. The profit structure of our country's commercial banks is relatively single, non-interest, Capital adequacy ratio and other factors are still restricting their development, and the emergence of network finance has also brought greater difficulties to the commercial banks, interest income and non- interest income growth slowed, and some companies have even regressed. In the face of increasingly fierce competition, China's commercial banks in the face of increasingly fierce competitive pressure, how to increase investment in the financial sector, to seek opportunities for online finance, to realize the organic combination of technology with financial products, services and business, and to promote its strategic transformation and business deepening has become an urgent problem to be solved. In this paper, through the mechanism of small and medium- sized commercial banks and data analysis, to small and medium-sized commercial banks to provide a reference for the use of financial technology. And through empirical analysis of the impact of small and medium-sized commercial banks in the development of financial technology, and put forward recommendations in the supervision, in order to better promote the transformation of such small and medium- sized banks, to guide the healthy and orderly development of the banking industry and promote the ability to serve the real economy. 2. Literature Review Delong and Geyoung's research shows that while online finance can improve U.S. profits, it can change the deposit structure of banks[1]. Li Dazhi and Xu Yihan analyzed that compared with the traditional low-interest demand deposit, the low-liquidity time deposit, the online financial products have higher liquidity and profitability, and have a certain impact on commercial banks, and a certain impact on the level of interest rates[2]. Zhan Minghua and others from a macro- analysis of the impact of bank credit channels in the face of Internet finance[3]. Research by Li, spigt & Swinkels shows that financial technology and traditional banking are complementary[4]. Wang believes that financial technology will bring other problems, such as increasing price competition and interest rate liberalization, which will reduce banks' profits and increase their risks[5]. Tang's empirical research shows that with the development of financial technology, the bank's credit risk control ability has been further improved[6],Yao Ting,based on her analysis of fintech, believes that fintech can improve its profitability by reducing transaction costs between commercial banks and customers[7]. Financial Science and technology has just started in our country. At present, the academic research on it is still at the theoretical level. 3. Analysis of Changes in NIM and ROA 3.1. Change in net interest margin In the third quarter of this year, the net interest margin of Chengshang Bank was 1.89% , which is lower than the average net interest margin of 2.07% of commercial banks, in the same period, the net interest margin of state-owned big banks, joint-stock banks and agricultural and commercial banks was 2.03% , 2.15% and 2.26% respectively. From the 2018 data, the city commercial bank net interest margin and commercial banks as a continuous narrowing trend. Some of the 16 a-share city commercial banks reported a net interest margin of 1.82% to 2.29% in the third quarter, with Qingdao Bank the lowest and Ningbo Bank the highest. 228 Figure 1. 2015-2021 the first three quarters of commercial banks NIM Change Chart 3.2. Changes in the profitability of assets In the first three quarters of this year, Chengshang's average return on assets was 0.60 per cent, lower than the 0.82 per cent average ROA of commercial banks and the lowest of the six categories of commercial banks. The figure was up 0.05 percentage points from the same period last year and in line with the average for commercial banks. Considering the average return on assets of 16 a-share listed city commercial banks in the first three quarters of about 0.85% , it can be calculated that the ROA of other city commercial banks will be less than 0.60% during the same period. For example, Hebei Bank has an annualized ROA of about 0.59% , Guilin Bank has an annualized ROA of about 0.43% , and Fudian Bank has an annualized ROA of about 0.22% . Figure 2. 2015-2021 the first three quarters of commercial banks ROA Change Chart 4. How Fintech Affects Commercial Banks' Profits The profitability of commercial banks is reflected in the aspects of customer-oriented marketing, product service, compliance risk control and Operation Management. The impact mechanism is shown in figure 3. Figure 3. Impact mechanism diagram 5. Empirical Analysis 5.1. Research hypothesis This paper holds that the long tail theory can help small and medium-sized commercial banks to tap a large potential customer base in the development of financial technology, open up the market share of the long tail, and thus increase the profitability of banks, and from the theory of financial 0 0.5 1 1.5 2 2.5 3 2015 2016 2017 2018 2019 2020 2021Q3 NIM of commercial banks NIM of city commercial bank 0 0.2 0.4 0.6 0.8 1 1.2 2015 2016 2017 2018 2019 2020 2021Q3 ROA of commercial banks ROA of city commercial bank ROA of agricultural and commercial banks 229 function, it can reduce the operating cost, improve the efficient management of information and enhance the efficiency of risk control. Hypothesis 1: the application of financial technology has a positive impact on the profitability of small and medium- sized commercial banks. Hypothesis 2: the application of fintech has heterogeneity to the promotion of its profitability. 5.2. Model design In this paper, the small and medium-sized commercial banks in 2015-2020 through the collection of data panel data for regression analysis, the model is as follows: i 0 1 2 3 4 5= L N S IZ Eitt t it it it i t itR O E F T C A R N P L C I R u v              In the model, T stands for time, I stands for commercial banks in each region, Roe stands for listed banks' profits, FT stands for their use of financial technology, CAR stands for Capital adequacy ratio, NPL stands for non-performing loans, CIR represents the cost income ratio, LNSIZE represents the natural logarithm of total assets, εit represents the error term, and UI and VT represent the individual and time-fixed effects, respectively. 5.3. Data sources Taking 2015 as the starting point of empirical analysis, the data span is 2015-2020. In order to ensure the integrity of the data and research, this article selected 43 commercial banks data, a total of 258, selected in addition to the listed 16 city commercial banks, there are also other urban commercial banks and rural commercial banks in the various regions. The bank data comes from wind database and is processed by SPSS Software. 5.4. Descriptive analysis As shown in table 1, the Roe (%) mean and standard deviation were 13.258 and 4.222, respectively, and the minimum and maximum were 1.55 and 31.71, respectively, with significant differences among small and medium-sized commercial banks. The maximum and minimum values of the total index were 431.93 and 193.29, respectively, with a mean of 302.695 and a standard deviation of 56.943. In the control variables, the average non-performing loan ratio is 1.525, the standard deviation is 0.498, which indicates that the asset quality of small and medium-sized commercial banks is very good, and the dispersion is very small. The asset size and standard deviation were 8.135 and 1.45, respectively. The results show that the asset scale of small and medium-sized commercial banks is very different, and the difference between urban and rural areas is also very big, which will have a certain impact on the profitability of the company. The mean Capital adequacy ratio (%) was 13.551 with a standard deviation of 1.65. This suggests that different banks have different risk defences. Table 1. Variable correlation coefficient table Variable Obs. Max Min Mean Std.Dev Variance ROE 231 31.71 1.55 13.258 4.222 17.828 ROA 231 2.58 0.12 0.896 0.277 0.077 Non-performing loan ratio 231 3.71 0.42 1.525 0.498 0.248 Capital adequacy ratio 231 18.94 7.15 13.551 1.65 2.724 Core Capital adequacy ratio 231 16.08 6.03 10.422 1.872 3.504 Size of capital (logarithm) 231 11.284 3.467 8.135 1.45 2.103 Cost-to-income ratio 231 77.96 18.93 30.497 7.099 50.395 Reserve coverage 231 541.23 114.05 249.991 83.763 7016.304 Total Index 231 431.93 193.29 302.695 56.943 3242.481 Depth 231 488.68 125.25 301.894 75.04 5630.976 Breadth 231 397 160.98 280.473 58.045 3369.21 Digit 231 462.23 295.07 377.547 42.873 1838.081 Then, the correlation between the explanatory variables and the control variables was tested. The results are shown in table 2. Table 2. Correlation analysis of control variables Non- performing loan ratio Capital adequacy ratio Core Capital adequacy ratio Size of capital (logarithm) Cost-to-income ratio Reserve coverage Non-performing loan ratio 1.000 -0.146 -0.013 -0.077 0.117 -0.596 Capital adequacy ratio -0.146 1.000 0.657 -0.159 -0.183 0.233 Core Capital adequacy ratio -0.013 0.657 1.000 -0.491 0.098 0.197 Size of capital (logarithm) -0.077 -0.159 -0.491 1.000 -0.615 -0.152 Cost-to-income ratio 0.117 -0.183 0.098 -0.615 1.000 0.050 Reserve coverage -0.596 0.233 0.197 -0.152 0.050 1.000 230 This was followed by a multicollinearity analysis of the control variables, as shown in table 4.4, resulting in a tolerance of 0.492, greater than 0.4, and a VIF of 2.156, much less than 10, and therefore considered multicollinearity. You can create the following panel model. Table 3. Multicollinearity test Variance expansion factor test Tolerance VIF Non-performing loan ratio 0.608 1.644 Capital adequacy ratio 0.486 2.056 Core Capital adequacy ratio 0.404 2.473 Size of capital 0.539 1.854 Size of capital (logarithm) 0.288 3.475 Cost-to-income ratio 0.529 1.889 Reserve coverage 0.589 1.698 Mean 0.492 2.156 5.5. Empirical test First, the mixed effect, the fixed effect and the random effect are used to analyze the regression model. When the first two models were compared, a P value of 0.000 was found, suggesting that the original hypothesis should be rejected, with the assumption that an individual effect was present, and therefore no mixed regression was used. In the comparison of POOL and RE, the LM test results also negated the original assumption that mixed regression was not applicable for the analysis. In the Hausman test for FE and RE, P is 0.000, strongly rejecting the original hypothesis, and therefore the FE model should be used. So select time-individual bidirectional fixed effect model. Table 4. Regression model tests Test method Statistics P value Results F-test 11.042 0 FE modle Breusch-Pagan-test 187.899 0 RE modle Hausman-test 22.754 0.004 FE modle Note: * * * , * * , * represent 1% , 5% , 10% significance levels respectively Table 5. Diagnostic tests of the static panel data model Variable FE Time-fixed effect model test Time individual fixed effects model test ROE ROE ROE Total Index 0.034*** 0.032*** 0.067** (4.918) (3.399) (2.329) Non- performing loan ti -3.075*** -3.942*** -2.375*** (-5.129) (-7.168) (-3.022) Core Capital adequacy ratio 0.1 -0.112 0.314* (0.596) (-0.641) (1.786) Capital adequacy ratio -0.136 -0.502*** -0.305** (-0.824) (-2.515) (-2.079) Cost-to-income ratio -0.239*** -0.091** -0.278*** (-5.466) (-2.193) (-3.632) Reserve coverage 0.004 0.004 0.004 (1.162) (1.098) (0.920) Size of capital (logarithm) -0.406 -1.294*** 3.13 (-0.288) -4.864 (1.022) Constant 38.776*** 48.558*** -19.803 (3.368) (10.089) (-0.642) R-squared 0.751 0.765 0.789 Note: * * * , * * , * represent 1% , 5% , 10% significance levels respectively The regression results show that the Pratt & Whitney index coefficient of the core explanatory variable is 0.067 when the 231 application time and the individual are bidirectionally fixed, it shows that the application of financial technology has a significant promoting effect on the profits of our listed commercial banks. The NPL coefficient of non-performing loan ratio is -2.375, which indicates that banks' non- performing assets will reduce ROE and affect their profits The LNSIZE factor of asset size is 3.13, which indicates that large commercial banks have a better ability to increase profits, because the larger the asset size of a bank, the more obvious its size effect. In the overall CAR sample, the Capital adequacy ratio coefficients were negative for both urban and commercial banks, but positive for rural and rural areas. This suggests that, because of the heterogeneity of banks across regions and regions, excessive Capital adequacy ratio can lead to a reduction in bank liquidity. The regression analysis of sub-samples shows that the total index of the core explanatory variable remains positive, which indicates that financial technology has a significant positive effect on corporate earnings, but there are some differences among the indexes. The application of fintech in city commercial banks is not comprehensive, and the application of fintech in bank-related fields is gradually improved, and there is no perfect fintech layout, because of its small size and good understanding of its regional businesses, the bank's use of financial technology has paid off, with the use of basic technologies such as big data and cloud computing, in addition to the line of regional advantage, can better serve the local small and medium-sized enterprises, thus the profitability of a larger degree of improvement. 5.6. Test for heterogeneity The total sample is divided into urban and rural sub- samples, urban mainly includes urban commercial banks, while rural mainly includes rural commercial banks, rural credit cooperatives and so on. Table 6. Test for heterogeneity Variable Time individual fixed effects model test(urban sub-sample) Time individual fixed effects model test(rural sub-sample) ROE ROE Total Index 0.043*** 0.043*** (5.060) (5.060) Non-performing loan ratio -3.804*** -5.587*** (-8.27) (-4.984) Capital adequacy ratio -0.016 0.014 (1.764) (0.049) Cost-to-income ratio -0.087*** -0.447*** (-3.161) (-5.704) Size of capital (logarithm) 0.611* -0.398 (1.764) (-0.921) Constant 28.776*** 51.960*** (6.532) (7.411) R-squared 0.619 0.520 Note: * * * , * * , * represent 1% , 5% , 10% significance levels respectively 6. Conclusions The banking industry's increasingly sophisticated use of fintech, and its increased investment, has boosted corporate profits to some extent. Financial technology has greatly improved commercial banks' marketing, Operation Management, product service, compliance risk control and so on, which has a positive effect on banks' profits, there is heterogeneity. There are some problems in this study, such as lack of data and insufficient disclosure of indicators, so whether the conclusions of this paper can be applied to commercial banks needs to be further explored. At present, the use of financial technology by major commercial banks can not be Quantitative analysis because it is not possible to obtain specific input from major commercial banks on financial technology. Based on the research conclusion, this paper puts forward the following suggestions: (1) strengthen the cooperation with the main financial technology companies, strengthen the training of financial technology personnel, and perfect the financial technology risk control system. (2) improving the relevant legal system and strengthening the risk control of financial institutions. References [1] Gayle Delong,Robert Deyoung (2007) Learning by Observing: Information Spillovers in the Execution and Valuation of Commercial Bank M&As. The Journal of Finance, 62(1) : 181- 126. [2] Li, D., Xu, Y. (2014) Deep thinking on the impact of internet financing on banks. International Finance, 5: 64-67. [3] Zhan, M., Liu, Z., Xu, Y., et al. (2018) Money and credit: who forecasts output more accurately? A comparative study of machine learning algorithms . Financial Forum, 23(06) : 38-52. [4] Li, spigt & Swinkels. (2017) The impact of fintech start-ups on the share prices of existing retail banks. Finance Sunac, 1: 26. [5] Wang Ke, Wu Qing Li Ji. Financial Technology and commercial bank risk-taking: an empirical analysis based on China's banking industry. Management modernization, 2017, 37(06) : 100-104 [6] Tang, Y. (2021) How does the development of financial technology in commercial banks affect credit business? Evidence based on text mining of annual reports of listed banks. Finance and Economics, 2: 38-44. [7] Yao Ting, Song liangrong. Research on the impact of financial technology on economic capital of commercial bank credit risk . Science and technology management research, 2021, 41(01): 104-110.