Frontiers in Business, Economics and Management 
ISSN: 2766-824X | Vol. 13, No. 3, 2024 

 

338 

Digital	Inclusive	Finance,	Household	Leverage	Ratio	
and	Economic	Growth	
Fucong Bi1, a 

1Shanghai University, Shanghai, China 
abi_15632892805_bi@163.com 

 

Abstract: In this paper, from 2011 to 2020 part of the provincial panel data in China as a research sample, using the system 
GMM model discusses the digital inclusive finance, residents leverage and the relationship between the economic growth. 
Through the research found: digital universal financial and economic growth between a "U" type of nonlinear relationship, 
residents between leverage and economic growth presents a "inverted U" nonlinear relationship. In the background of the rapid 
development of digital informatization, it is necessary to strengthen digital financial supervision, strictly control the leverage of 
the household sector and stabilize the expectations of the real estate market, coordinate the development of digital finance in the 
region, and promote steady economic growth. 

Keywords: Digital inclusive finance; Resident leverage ratio; Economic growth. 
 

1. Foreword 
According to data released by the People's Bank of China, 

by the end of 2023, the balance of inclusive micro loans was 
29.4 trillion yuan, up 23.5% year on year, or 5.61 trillion yuan, 
up 1.03 trillion yuan year on year. Inclusive finance has 
played an indispensable role in solving the "short, small, 
urgent and frequent" loan needs, supporting the sustainable 
development of small and micro economies, helping poverty 
alleviation and realizing the rural revitalization strategy. 

2. Literature Review 
First of all, there are mainly three different theories on the 

influence of household leverage ratio on economic growth, 
namely, positive effect theory, negative effect theory and 
threshold theory. Liu Ziwei, Li Xuelian, and Xue Lei (2021) 
believe that an appropriate increase in household debt can 
promote economic development without increasing the 
internal vulnerability of the economy. According to the PEST 
analysis method, Cheng Xuejun and Li Xinhe (2021) 
concluded that excessive debt overdraws the consumption 
potential and has adverse effects on economic development.  

Second, the existing literature disagrees on the role of 
digital finance in promoting economic development. Yang 
Gang and Zhang Hengyi (2022) believe that digital inclusive 
finance and regional innovation can significantly promote 
economic growth. The research results of Li Tao (2016) show 
that with the deepening of China's economic and financial 
development, the development of digital inclusive finance 
will make its marginal cost greater than its marginal income, 
which will have a negative impact on China's economic 
development. 

According to the study of past literature, there is no 
consensus on these two issues. Therefore, it is of great 
significance to further explore the internal relationship 
between these three to provide decision-making reference for 
promoting stable economic growth. 

3. Research Hypothesis 
Based on the above analysis of the existing literature, the 

following assumptions are made: 
Hypothesis 1: There is a "inverted" non-linear relationship 

between the U-shaped leverage ratio and economic growth; 
Hypothesis 2: There is a "U" -shaped nonlinear relationship 

between digital inclusive finance and economic growth. 

4. Empirical Study 
(1) Description of the samples and the variables 
Based on the availability of real data, the empirical analysis 

sample used in this paper includes interprovincial panel data 
from 29 provinces from 2011 to 2020. Table 1 shows the 
names, symbolic representations, economic implications, and 
data sources of each regressor. 

In addition, instrumental variables are cited to control the 
possible endogeneity problems in economometric models. 
Referring to the research, this paper uses mobile phone 
popularity (Lin Lin West, 2022) and the first-order lag item of 
digital financial inclusion index as instrumental variables.  

(2) Estimation method 
This paper uses the systematic generalized moment 

estimation method (system GMM) proposed by Arellano & 
Bover et al. in 1995, as follows: 
y , =αy , +βX , +μ +ε ,  
As shown in the formula, the variable y ,  represents the 

time delay of y ,  , and X ,   represents the core explanatory 
variables. β is the coefficient to be estimated for each variable. 
Based on previous literature (Ma Yong and Chen Yulu, 2017), 
in order to avoid the endogeneity between variables and 
between variables and residues, a two-step system 
generalized moment method is used to estimate the model. 

In addition, based on the existing research results, this 
paper conducts two tests on the estimation results of GMM: 
first, Hansen test, which is mainly used to test the validity of 
the instrumental variables; second, second-order sequence 
correlation test (AR (2) test), which is mainly used to test 
whether the residual items have a second-order sequence 
correlation. 

(3) Empirical analysis and robust test 
As shown in Table 2, models (1), (2) and (3) are the 

regression results before the addition of the instrumental 



 

339 

variable (mobile phone penetration), and models (4), (5) and 
(6) are the regression results after the addition of the 
instrumental variables. It can be seen that the coefficients of 

the core explanatory variables before and after the addition of 
the instrumental variables are consistent. 

 
Table 1. Regression variables and data sources 

Variable name symbol Economic meaning data sources 

Economic growth level lnGDP 
Amount of gross real national income 

The log of the (100 million yuan) 
National Bureau of 

Statistics of China（NBS） 
Residential leverage ratio debt Household sector loans / GDP NBS, Provincial Bureau of 

Statistics, the People's Bank 
of China 

The square side of the resident 
leverage ratio 

debt2 
Household sector loans / GDP 

square 
The development level of digital 

inclusive finance 
fina Digital Financial inclusion General Index 

Institute of Digital Finance, 
Peking University 

Corporate leverage ratio TDR Non-financial corporate equity capital / assets NBS and the Wind database 

Housing prices HP 
Average sales price of commercial housing 

(yuan / m 2) 
NBS 

Housing sales HS 
Commercial housing sales area (ten thousand 

square meters) 
NBS 

Financial development level fd Financial sector value added value / GDP Wind database 
Degree of industrialization ind Industrial added value / GDP Wind database 

urbanization level urban 
Urban population / permanent resident 

population at the end of the year 
NBS 

per capita disposable income dip 
Per capita disposable income of all provinces 

and cities (yuan) 
NBS 

Per capita available consumption 
expenditure 

hcp 
Per capita consumption expenditure of all 

provinces and cities (yuan) 
NBS 

 
Through the analysis of the empirical results, the following 

conclusions can be drawn: 
Digital inclusive finance and economic growth show a "U" 

-shaped non-linear relationship, that is, with the improvement 
of the former level, the latter shows a trend of first decline and 
then increase. Hypothesis 1 is established; 

The non-linear relationship between the leverage level of 
Chinese residents and economic growth is established. That 
is to say, there is a resident leverage level. Before this level, 
it has a positive impact on economic development. 
Hypothesis 2 is established. 

 
Table 2. Systematic GMM estimation results 

 mobel(1) mobel(2) mobel(3) mobel(4) mobel(5) mobel(6) 

_cons 
2789.2*** 

(7.56) 
2813.2*** 

(7.63) 
2695.0*** 

(7.27) 
2652.9*** 

(6.84) 
2683.8*** 

(6.92) 
2548.8*** 

(6.55) 

L.lnGDPP 
0.986*** 
(218.24) 

0.985*** 
(219.94) 

0.984*** 
(211.50) 

0.986*** 
(217.50) 

0.985*** 
(219.31) 

0.984*** 
(211.34) 

fina 
-1.076** 
(-2.73) 

 
-1.290** 
(-2.98) 

-1.125** 
(-2.84) 

 
-1.316** 
(-3.04) 

fina2  
-0.00257** 

(-2.95) 
  

-0.00265** 
(-3.03) 

 

debt   
16.55* 
(2.09) 

  
16.69* 
(2.11) 

debt2   
-17.32* 
(-2.08) 

  
-17.89* 
(-2.15) 

mob    
1.683 
(1.14) 

1.580 
(1.07) 

1.847 
(1.23) 

AR(2) statistics    
-0.30 

（0.765） 
-0.52 

（0.605） 
0.44 

（0.657） 

Hansen statistics    
9.23 

（0.161） 
8.59 

（0.284） 
11.62 

（0.169） 
Notes: (1) *, * * and * * * indicate that the regression coefficients for each variable are significant at the confidence levels of 
10%, 5% and 1%, respectively; (2) L.lnGDPP Is the first-order lag term of the variable lnGDPP; (3) the second line of the 

variable coefficient contains the corresponding Z-statistic; (4) the first line of the Hansen test and AR (2) test is the statistics of 
the corresponding test, and the second row is the corresponding P-value.similarly hereinafter; (5) data for control variables has 

been omitted from this table. 
 

As shown in Table 2, the model set in this paper has passed 
the Hansen test and AR (2) test, from which it can be judged 
that the selection of instrumental variables is appropriate in 

the model estimation process, and it also shows that the model 
estimation results will not be affected by the second order 
sequence correlation, which verifies the effectiveness of the 



 

340 

results. 

5. Study Conclusions and Policy 
Recommendations 

(I) Study conclusions 
First of all, it can be seen from the benchmark regression 

analysis that the leverage effect of Chinese residents has 
played a certain role in promoting the development of China's 
economy. However, in the long run, the debt ratio of Chinese 
residents and economic growth show an "inverted U-shaped" 
relationship. Secondly, from the perspective of benchmark 
regression, China's digital finance still has a certain inhibitory 
effect on economic growth, but in the long run, digital 
inclusive finance and residents' debt level show a "U" shaped 
nonlinear relationship. 

(2) Policy recommendations 
As a major way to stimulate domestic demand, 

consumption plays a decisive role in promoting internal 
circulation. The leverage of households can promote the 
expansion of consumption and promote the economy from 
external demand-driven growth to consumption-driven 
growth, but at the same time, the risk of household debt 
should not be ignored. Therefore, based on the above research 
conclusions, the author tries to provide reference opinions for 
the policy formulation of digital inclusive finance in China: 

First, strengthen the supervision and management of digital 
finance to prevent major financial system risks. Secondly, 

strengthen the control of the increasing leverage of household 
enterprises, and deploy the risk reduction work as early as 
possible. Finally, maintain the policy orientation of "housing 
not speculation" to stabilize the expectations of the property 
market. 

Reference Documentation 
[1] Liu Ziwei, Li Xuelian, Xue Lei. Resident debt growth, financial 

risk and debt structure adjustment under the high leverage rate 
[J]. Finance and Economics Science, 2021 (09): 14-28. 

[2] Cheng Xuejun, Li Xinhe. Suggestions on the development, risk 
and response of household leverage ratio: Based on the 
research perspective of legal finance [J]. Southwestern Finance, 
2021 (10): 20-31. 

[3] Yang Gang, Zhang Hengyi. Digital financial inclusion, 
Regional Innovation, and Economic Growth [J]. Statistics and 
Decision-making, 2022,38 (02): 155-158. 

[4] Li Tao, Xu Xiang, Sun Shuo. Financial Inclusion and 
Economic Growth [J]. Financial Research, 2016 (04): 32-43. 

[5] Lin Lin West, Xiao Yubo. Digital finance, technological 
innovation and regional Economic growth [J]. Journal of 
Lanzhou University (Social Science Edition), 2022,50 (02): 47-
59. 

[6] Ma Yong, Chen Yulu. Financial leverage, leverage volatility 
and economic growth [J]. Economic Research, 2017,52 (06): 
31-45.