ARTICLE

https://doi.org/10.1057/s41599-024-02630-4 OPEN

Financial inclusion, education, and employment:

empirical evidence from 101 countries
Xiaoling Song1, Jiaqi Li2 & Xueke Wu1 ✉

This study analysed the data of 101 countries from the World Bank’s Financial Inclusion and
Financial Availability Survey databases for the years 2011, 2014, 2017, and 2021. It con-
1234567890():,;

structed a new financial inclusion evaluation index system using digital empowerment and
established a dual fixed-effect panel data regression model to examine the impact of financial
inclusion on employment. The study also considered the moderating effect of education on
financial inclusion affecting employment. The results indicate that financial inclusion is
beneficial to improving employment levels. The level of education development is propor-
tional to the level of employment, and it has a regulatory effect of promoting the financial
inclusion’s positive impact on employment level. The heterogeneity analysis demonstrates
that the effect of financial inclusion on improving employment levels is greater in female
groups. The promotion of financial inclusion is the largest in low- and middle-income
countries, followed by high-income countries, indicating that financial inclusion must have a
certain economic foundation to fully play its role, and it has more obvious benefits for
relatively poor and weak areas than for high-income areas. The conclusions of this study have
implications for improving employment and augmenting financial inclusion.

1 Beijing Language and Culture University, Beijing, China. 2 Beijing Foreign Study University, Beijing, China. ✉email: [email protected]

HUMANITIES AND SOCIAL SCIENCES COMMUNICATIONS | (2024)11:172 | https://doi.org/10.1057/s41599-024-02630-4 1

ARTICLE HUMANITIES AND SOCIAL SCIENCES COMMUNICATIONS | https://doi.org/10.1057/s41599-024-02630-4

C
Introduction
ompared with the characteristics of traditional finance, expectations, family’s economic ability, education level, and one’s
which involves ‘preferring the rich to the poor’, financial own ability. At the macro level, the provision of jobs by enter-
inclusion strives to provide financial products and services prises and public institutions, government support, and economic
that can cover all of society and affordably and sustainably meet situation also affect social employment. As an important factor
the needs of individuals and enterprises in transactions, savings, affecting people’s livelihoods and financial markets, financial
credit, and other aspects. Digital financial services using mobile inclusion has attracted many scholars to study its impact on
phones and other devices cover more than 80 countries world- employment. The positive role of financial inclusion in employ-
wide, providing those most likely to be excluded: poor, rural, ment is mainly reflected in directly and indirectly increasing
female or young individuals (Allen et al. 2016) with more employment and optimising the structure. First, the literature
humanised financial services and developing the economy at the shows that the significance of financial inclusion in optimising
macro level by supporting small and micro enterprises, assisting employment structure is the depth, breadth, and digitisation of
agriculture, and stimulating investment, among others (Fu 2020). financial inclusion in order from high to low, and savings and
Financial inclusion must be improved in many aspects, and formal accounts have the most obvious effect on enhancing
there remains room for improving the coverage of financial household financial flexibility (Sakyi-Nyarko et al. 2022). In
inclusion. According to the World Bank, nearly one-third of addition, financial inclusion can positively influence employ-
adults will still be unbanked by 2021. Simultaneously, there is a ment’s sustainability, with data from Belt and Road countries
gender gap in the use of financial inclusion, with many women in showing that this effect is more significant in high- and middle-
remote and poor areas having significantly lower levels of elec- income countries (Zhechen and Guosheng 2021). The digitali-
tronic device usage and participation in formal economic activ- sation brought about by financial inclusion can also promote
ities than men (Kim 2022). Asfaw et al. (2009) also found that equity in employment opportunities. Second, entrepreneurship,
gender gap in finance is more obvious when using scarce as a form of employment, is also significantly affected by financial
resources and is exacerbated by poverty, which is more severe in support. Evidence shows that the development of financial
low- and middle-income countries. Financial inclusion essentially inclusion is positively correlated with the level of rural entre-
aims to address the sustainable development and exclusion issues preneurship (Romero-Castro et al. 2023). Finally, some studies
of finance. The rational economic man hypothesis in traditional start from the unemployment rate and prove that the degree of
finance has inherent flaws, and it can be said that inclusive financial inclusion development is inversely proportional to the
finance is a supplement to the gaps and imperfections in the unemployment rate, and this relationship is more significant in
current financial system of a country (Renzhia and Baek 2020). countries with higher incomes (Mais et al. 2020).
The continuous development of financial inclusion requires a Other scholars have researched the relationship between the
stable macroeconomic environment, which cannot be separated level of educational development and employment. With the
from regulatory support (Johnson 2013). However, the develop- increasing emphasis on employment-oriented education, many
ment level of financial inclusion is seriously unbalanced, and the scholars have conducted research on education’s impact on
quality of financial services and level of market supervision are employment level and quality. Education investment represents
uneven. Therefore, it is vital to explore the uneven development the strength of education support, which is of great significance
of financial inclusion in the world, the temporal changes in the for many issues, such as the improvement of social quality and
degree of development of financial inclusion, and the world’s sustainable production. One study found that, in the context of
adaptation to the digitalisation of financial inclusion. the transfer of employment to the public sector, the increased
Recently, the downward economic pressure brought about by accessibility of higher education brought about by education
the global pandemic and the inconvenience of daily travel have expansion doubled the probability of entering paid employment,
made unemployment rates rise sharply, which greatly affected the with an 80% increase in hourly wages. The main reason for this is
employment of the self-employed and manual workers, and also that the public sector has higher salaries and higher educational
highlighted the urgency and necessity of governments to improve requirements (Ravi 2021). However, the increase in labour pro-
the quality of employment. Adequate liquidity, capital sources, ductivity can only temporarily increase the employment level
and flexible working modes are necessary conditions for enter- during the period of wage stickiness; after the gradual increase in
prises to cope with the impact of the rapid development of sci- wages, employment will decline to original levels (Blanchard and
ence, technology, and economic turbulence. In addition to the Katz 1997). Therefore, in the long run, investment in education
stability of enterprises, the economic security and employment of cannot improve employment by improving society’s overall
residents are also closely related. The development of inclusive production efficiency.
finance provides more financing channels for micro, small, and Overall, abundant research results exist on the influence of
medium-sized enterprises and provides financial services for the financial inclusion on employment levels, but a majority of those
whole life cycle (Lai et al. 2022). The smooth financing channels studies are based on the measurement of the development degree
of enterprises and institutions, especially small and micro enter- of financial inclusion, with banks as the main focus, and its
prises over short periods, the availability of start-up funds, and impact on employment. There are relatively few empirical studies
micro-credit are particularly important for guaranteeing indivi- on the comprehensive evaluation system of traditional and new
dual employment and improving employment levels. Can finan- digital finance under digital empowerment and how it affects
cial inclusion guarantee job provision by expanding financing employment. There are heterogeneity studies in the literature,
channels? Can entrepreneurship be promoted by directly pro- most of which explore the reasons for the differences in using
viding financial services to individuals? In practice, it is important financial services by men and women. Gender is rarely used as the
to pay attention to the employment and entrepreneurship effects basis for heterogeneity classification to study the distinctions in
of financial inclusion. the external impact of financial inclusion in different gender
Existing research can be divided into two parts: financial groups. With greater analysis of gender differences mediating
inclusion and employment levels, and education development financial inclusion’s impact on employment, the differences in
levels and employment levels. Some scholars have studied the financial inclusion’s benefits for different genders can be further
relationship between financial inclusion and employment. At the studied, thereby enriching research on gender equality in
micro level, employment is related to an individual’s career employment. Studies on the relationship between education and

2 HUMANITIES AND SOCIAL SCIENCES COMMUNICATIONS | (2024)11:172 | https://doi.org/10.1057/s41599-024-02630-4

HUMANITIES AND SOCIAL SCIENCES COMMUNICATIONS | https://doi.org/10.1057/s41599-024-02630-4 ARTICLE

employment are also supported by many classical theories; through traditional business logic, making entrepreneurship and
however, few studies examine the moderating effect of education employment no longer confined to offline enterprises and
on employment. Although many studies focus on entrepreneurial institutions. The online intelligent payment system serves as the
or vocational education, the general level of education develop- cradle of many enterprises relying on this transaction mode,
ment is not the focus; studying its moderating effect has reference which significantly contributes to employment and entrepreneur-
significance for enhancing financial inclusion’s employment ship (Rotatori et al. 2021; Senyo et al. 2023).
promotion effect. Thus, studying the role of investment in edu- Second, at the micro level, financial inclusion has reshaped all
cation in terms of financial inclusion and job security should be walks of life. Financial inclusion demonstrates the characteristics
encouraged. of digital transactions, intelligent products, and humanised
The marginal contributions of this paper are as follows. (1) The services, which encourage enterprises to introduce and develop
research scope is expanded to about 100 countries, which makes new technologies, and even reshape business models. Moreover,
it more objective and robust; it also conforms to the concept of the expansion of financing channels makes enterprises more
financial inclusion covering a wide range of people. In the tem- willing to invest in new technologies and equipment. The rise of
poral dimension, the latest (2021) data of the World Bank are finance marginalises labour’s role in the process of income
introduced, and the survey data of 2011, 2014, 2017, and 2021 are generation and sharing, resulting in large-scale employment
selected. (2) Under the background of digital empowerment, a stagnation (Lin 2016). However, labour and technology are
financial inclusion evaluation index system has been established inseparable and interdependent, and such inseparability can
that considers both traditional finance and new digital finance improve productivity, labour value, and demand (Autor 2015).
formats to assess the development level of financial inclusion The improvement in production efficiency brought about by
more objectively and comprehensively in various countries. (3) technological innovation actively impacts overall employment in
Education is added as a moderating variable to study the role of a society. Overall, therefore, financial inclusion is more likely to
financial inclusion in promoting employment. (4) We use the stimulate demand for labour and promote employment.
degree of economic development as the basis for heterogeneity In the labour supply market, financial inclusion can also
classification regression and explore the differences in employ- increase employment by influencing individuals, families, and the
ment circumstances between men and women worldwide. job-search environment. Digital savings and access to digital credit
are important channels for inclusive financial services to affect
entrepreneurship (Koomson et al. 2023). Financial inclusion
Theoretical analysis and research hypotheses enables marginalised groups and regions to obtain more
The impact of financial inclusion on employment levels. The convenient financial support and financial services, thereby
improvement in employment level requires a simultaneous reducing household financing constraints and easing expected
increase in labour demand and supply. Under certain conditions, economic pressure. It increases individual entrepreneurial beha-
the development of financial inclusion can increase the labour viour, creating more entrepreneurs (Evans and Jovanovic 1989).
force and expand the demand for the labour force by affecting In summary, the following hypothesis is proposed:
credit, enterprises, workers, and other aspects, so as to improve H1: The financial inclusion index is directly proportional to
the employment level. employment levels.
The factors that affect labour demand can be divided into
macro- and micro-factors. First, at the macro level, according to
the financial deepening theory (Shaw 1973), financial inclusion The moderating effect of educational development level
has the characteristic of relaxing interest rate control compared Education and employment. Education plays an important role in
with traditional finance. A steady and falling loan interest rate transferring talent to employment positions and professional
effectively stimulates the increase in loan demand, broadens training for the labour force. Education’s regulatory mechanism
financing channels, and reduces financing resistance, thus within financial inclusion’s influence on employment starts with
increasing corporate financing and investment and driving the education’s influence on employment. According to human capital
investment behaviour of residents, as a result of which the theory, education can equip the labour force with more profes-
number of enterprises increases. The expansion of scale is also sional knowledge reserves and higher professional quality, reduce
conducive to the high-quality development of the real economy, the risk of unemployment (Xing et al. 2018), and improve the
thereby creating additional job opportunities and promoting a employment level. It is noteworthy that the Wiles test and the
higher employment level. However, residents’ judgement of sheepskin effect theory suggest that educational qualifications can
economic conditions will be more optimistic because of financial only play a screening role in the job market, that is, obtaining a
inclusion, thus enhancing their marginal propensity to consume, higher degree is a qualification for higher salaries (Miller and
stimulating the economy, expanding production, and expanding Volker 1984). However, in the era of rapid technological iteration,
employment (Bo et al. 2022). From a financial perspective, a emerging new industries and job functions can improve the utili-
relaxed interest rate control policy leads to an increase in the real sation rate of talent (Yang 2022). Therefore, from this perspective,
interest rate, an increase in capital costs, and an improvement in the improvement of the national education level brought about by
the labour force as a production factor favoured by enterprises, educational investment can improve the matching degree of edu-
thus increasing the demand for labour (Boustanifar 2014). cation and employment, thus promoting employment. Moreover,
Improved financial availability and coverage will attract interna- many studies have demonstrated the positive effect of education
tional market players into a country, creating more jobs and investment on labour employment (Ahn and Winters 2023).
business opportunities (Parameshwara and Raghurama 2011). To conclude, the following hypothesis is also proposed:
In addition, the technological and information innovation H2: The level of educational development is directly propor-
brought about by financial inclusion has made transactions more tional to the level of employment.
digitised and convenient, broadened financing channels, and
provided more start-up and operating funds for small and Financial inclusion, education, and employment. Correlation
medium enterprises (SMEs, which refer to relatively small analysis shows that the level of educational development has a
enterprises in terms of scale, capital, manpower, turnover, etc. weak correlation with employment and the financial inclusion
in this context). More importantly, digitalisation and media break index. According to the above theoretical analysis, the impact of

HUMANITIES AND SOCIAL SCIENCES COMMUNICATIONS | (2024)11:172 | https://doi.org/10.1057/s41599-024-02630-4 3

ARTICLE HUMANITIES AND SOCIAL SCIENCES COMMUNICATIONS | https://doi.org/10.1057/s41599-024-02630-4

financial inclusion on employment occurs more by affecting the weight. Next, the financial inclusion dimension index was syn-
economy and enterprises than indirectly affecting employment thesised, and four financial inclusion indexes were used to syn-
through education. Therefore, we speculate that education’s reg- thesise the final total financial inclusion index using the same
ulatory effect is more obvious. calculation method. The calculation formulas were as follows:
The mechanism by which education regulates financial inclusion Step 1: Calculate the coefficient of variation ∂rs of index s in
to promote employment can be analysed in terms of families and dimension r.
individuals. On the one hand, many studies mentioned financial σ
education level’s influence on individuals’ use of financial services ∂rs ¼ rs ðr ¼ 1; 2; ¼ ; n; s ¼ 1; 2 ¼ mÞ ð1Þ
xrs
and participation in financial activities. For example, financial
literacy has a positive and significant impact on farmers’ awareness Step 2: Calculate the weight wrs of index s in dimension r.
of digital credit, and the improvement in education level can increase ∂
residents’ efficiency in using the conveniences created by financial wrs ¼ s rs ðr ¼ 1; 2; ¼ ; n; s ¼ 1; 2 ¼ mÞ ð2Þ
∑1 ∂rs
inclusion (Sarfo et al. 2023). As mentioned above, the development
of financial inclusion can increase the demand for the labour force in Step 3: Calculate the index Frs of index s in dimension r.
various ways, and the progress in education can help more job

A  mrs
seekers in the labour force secure positions that match their abilities, Frs ¼ wrs *
 ðr ¼ 1; 2; ¼ ; n; s ¼ 1; 2 ¼ mÞ ð3Þ
increase the efficiency of labour force utilisation, and consequently Drs  drs
enhance the employment level. Conversely to the above, when the Step 4: Calculate the financial inclusion index IFIr in dimension
education level is higher, the overall quality of the labour force is r.
higher, knowledge levels are higher, and adaptability is more robust. pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Fr12 þ F2 þ ¼ þ F2
It is therefore reasonable to infer that higher education development
IFIr ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
r2 rs
ðr ¼ 1; 2; ¼ ; n; s ¼ 1; 2 ¼ mÞ
levels can enhance financial inclusion in employment. Moreover, wr1 þ w2r1 þ ¼ þ w2rs
2
studies have found that education can reduce the gender gap in
financial inclusion in low-income countries (Sharif et al. 2022), ð4Þ
thereby improving the level of financial inclusion in society. If the Step 5: Use IFIr to repeat the above calculation steps to gain IFI.
hypothesis that financial inclusion promotes employment is valid, Here σrs is the standard deviation, xrs is the mean, A represents
increased education levels can amplify its employment promotion the specific value, Drs is the maximum, drs is the minimum, which
effect by enhancing the development of financial inclusion. are all of the s index in dimension r.
Therefore, the following hypothesis is proposed: The inclusive benefit index system selected in this study is
H3: Educational development level plays a positive moderating shown in Table 1.
role in the influence of financial inclusion on employment level. Formula (1) involved the degree of use dimension selection, the
proportion of savings users in financial institutions, the number
of loan users of commercial banks among 100,000 people, the
Research design
proportion of GDP of outstanding loans of commercial banks, the
Data selection. In this study, the panel data of 101 countries in
proportion of GDP of outstanding deposits of commercial banks,
the years 2011, 2014, 2017, and 2021 were selected as research
and the number of deposit accounts of commercial banks among
samples; consequently, there were 384 valid data in the basic
100,000 people. Formula (2) functioned to benefit the specific
panel regression. Next, different indicator systems were used to
groups: the proportion of financial institution account owners
calculate the financial inclusion index each year. As a result of the
among low- and middle-income groups, the proportion of debit
lack of data for different countries in different years, the number
card owners among low- and middle-income groups, the
of countries in 2011, 2014, 2017, and 2021 was 96, 99, 100, and
proportion of credit card owners among low- and middle-
92, respectively. All the data in this study were selected from the
income groups, and the proportion of borrowers among low- and
World Bank’s database, in which the construction of the financial
middle-income groups. Formula (3) involved quality: compre-
inclusion index primarily included ‘World Bank Financial
hensive level of coverage, scope, and availability of credit
Inclusion’, ‘Financial Access Survey’, ‘Doing Business’, ‘Education
information represented by credit information depth index,
Statistics’, and other databases.
comprehensive level of scope and availability, rights and
participation of creditors in the bankruptcy liquidation process
Variable settings measured by creditor participation index, performance of
Explained variable: level of employment. Some scholars chose to use regulatory authorities in the credit market measured by credit
the unemployment rate to explore the relationship between financial supervision performance, coverage of credit institutions, and
inclusion and employment level (Mais et al. 2020). However, con- popularity of credit services measured by credit institution
sidering that financial inclusion aims to benefit all social groups, coverage index. Formula (4) involved availability: number of
especially disadvantaged and marginalised groups, this study did not ATMs per 1000 square metres, number of ATMs per 100,000
choose the unemployment rate, but the employment level repre- adults, number of insurance companies per 100,000 adults, and
sented by the proportion of the number of employed and the number of commercial bank branches per 100,000 adults.
working-age population, including the impact of financial inclusion When using different indicator systems to compute the IFI for
on social employment to the maximum extent. 2021, more consideration was given to digitalisation, including the
popularity index of using mobile phones or Internet bill payment.
Core explanatory variable: financial inclusion index. Based on Indicators reflecting respondents’ subjective feelings about financial
Sarma’s (2012) construction of the inclusive financial index and services were added, such as the percentage of respondents who cited
referring to the index composition and classification of the G20 medical expenses for a major illness as a major source of stress in
Inclusive Financial System, this study selected the four dimen- their personal financial situation and the percentage of respondents
sions of financial service usage, financial service availability, who cited savings as a major source of emergency funds.
financial service quality, and benefit to specific groups. In each
dimension, multiple secondary indicators were selected. The Moderating variable: educational development level. The propor-
variation coefficient of each index was used to calculate the tion of educational financial expenditure has always been an

4 HUMANITIES AND SOCIAL SCIENCES COMMUNICATIONS | (2024)11:172 | https://doi.org/10.1057/s41599-024-02630-4

HUMANITIES AND SOCIAL SCIENCES COMMUNICATIONS | https://doi.org/10.1057/s41599-024-02630-4 ARTICLE

Table 1 Financial inclusion index system.

Degree of use Percentage of users over 15 who deposited at financial institutions in the past year
Outstanding loans with commercial banks (% of GDP)
Outstanding deposits with commercial banks (% of GDP)
Average number of the mobile money transactions in active mobile money accounts
Percentage of respondents over 15 with a credit card
Number of commercial bank borrowers per 100,000 adults
Retail non-cash transactions per 100,000 adults
Number of commercial bank deposits per 100,000 adults
Function to benefit the special group Proportion of SMEs with outstanding loans or lines of credit
Proportion of low- and middle-income users over 15 who have accounts with financial institutions
Percentage of low- and middle-income users over 15 who used a mobile phone or Internet to pay bills in the past
year
Percentage of low- and middle-income users over 15 who have borrowed from financial institution in the past
year
Proportion of low- and middle-income users over 15 whose most concerned financial issues is medical treatment
of serious illnesses
Quality Credit information depth index
Creditor participation index
Credit supervision performance
Credit institution coverage index
Availability Number of credit cards per 100,000 adults
Number of mobile agent outlets per 100,000 adults
Number of ATMs per 1,000 square metres
Number of ATMs per 100,000 adults
Number of insurance companies per 100,000 adults

important indicator of the educational development level. According (6) Level of openness to the outside world: The degree of
to the United Nations Education Scientific and Cultural Organiza- openness of the economy can promote an increase in
tion, by 2020, the United States will spend 959.5 billion U.S. dollars employment quality and human capital level (Acharya
on education, ranking first in the world, while China will spend USD 2017). The impact of import and export trade includes the
259 billion, ranking second only to the United States. Taking the direct impact of overseas markets on domestic producers
availability of data into account, this study used the ratio of gov- and the indirect impact on domestic supply by influencing
ernment financial expenditure on education to total GDP to mea- domestic demand. Therefore, the ratio of import and export
sure the degree of education development and education support. trade to GDP reflects the level of economic openness.
(7) Foreign investment: Considering that foreign investment
Control variables. In addition to financial inclusion and educa- affects the provision of jobs by acting on foreign-owned
tional factors, the aspects that affect a country’s employment enterprises (Riccardo et al. 2022), foreign investment was
situation also include the original economy, industry, consump- included.
tion level, finance, and openness of the region. The application of variables and definitions in this study is
(1) Economic level: The economic level affects employment summarised in Table 2.
mainly by influencing demand; therefore, adding the
natural logarithm of per capita GDP controls the level of Model Selection. This article tests the individual effects of the
economic development. model with Prob > F = 0.0000. Therefore, the test results indicate
(2) Urbanisation level: The study found that the urbanisation that the fixed effects model is superior to the mixed OLS model.
level can promote informal employment of the rural labour Subsequently, the Hausman test was conducted, and the p-value
force by expanding production scale and implementing was 0.0000. Therefore, the original hypothesis was rejected and a
infrastructure construction related to rural transformation fixed effects model was chosen for regression. Finally, to further
(Mingxing et al. 2023). Therefore, the impact of the reduce the problems caused by missing variables, we choose
urbanisation process on employment can be controlled by bidirectional fixed effects. To implement models, the coefficients
adding the urbanisation rate. of variables will indicate relationships among them and t values
(3) Industrial added value: Low industrial added value will will inform us of the reliability of the results.
squeeze the space of the highly paid labour force and cause
unemployment, so industrial added value was added to Model construction
reflect the industrial development of a country. Basic regression model. To verify H1: The financial inclusion
(4) Household consumption level: The demand represented by index was proportional to the employment level, and a model was
residents’ consumption level indirectly affects the demand constructed (1).
for productivity and the labour force; therefore, the Employmentit ¼ α0 þ β1  IFIit þ β2  GDPperit þ β3  Urbanizationit þ
consumption potential of a country was measured by β4  Industryit þ β5  Consumptionit þ β6  Governmentit þ β7  Openupit þ β8
adding the household consumption level.  Foreigninvestit þ εit
(5) Fiscal expenditure: Government consumption expenditure
will have an impact on employment and employment quality; ð5Þ
therefore, we took the ratio of fiscal expenditure to GDP into Regulatory effect model. Based on model (1), the adjusting vari-
consideration to reflect the scale of the government. able education development level (Education) and its interaction

HUMANITIES AND SOCIAL SCIENCES COMMUNICATIONS | (2024)11:172 | https://doi.org/10.1057/s41599-024-02630-4 5

ARTICLE HUMANITIES AND SOCIAL SCIENCES COMMUNICATIONS | https://doi.org/10.1057/s41599-024-02630-4

Table 2 Study variables.

Type Name Symbol Definition

Independent variable Financial inclusion IFI Financial Inclusion Index
Adjusting variable Education Education Government expenditure on education /GDP
Dependent variable Employment Employment Employed population/working-age population
Control variable Economic development GDPper Natural ratio of real GDP per capita
Urbanisation rate Urbanisation Urban population/total population
Industrial added value Industry Industrial value added /GDP
Resident consumption level Consumption Final consumption expenditure /GDP of household and non-profit
institutions serving households
Fiscal expenditure Government General government final consumption expenditure /GDP
Degree of openness Openup (Import value + Export value)/GDP
Foreign investment Foreigninvest Net foreign direct investment inflows /GDP
Heterogeneity classification Gender Maleemploy Male employment level
variables Femaleemploy Female employment level
Income level Incomegroup
Variables for robustness test Unemployment rate Unemployment Unemployment/(employed + unemployed)

with the financial inclusion index (Education*IFI) were added to standard deviations of the control variables, the natural logarithm
construct model (2): of the per capita GDP was the largest, at 1.478, indicating a large
gap in economic level among the countries.
Employmentit ¼ α0 þ β1  IFIit þ β2  Educationit þ β3  Educationit  IFIit
This study constructed an independent financial inclusion
þβ4  GDPper þ β5  Urbanizationit þ β6  Industryit þ β7  Consumptionit þ β8 
index for 2011, 2014, 2017, 2021, and conducted an analysis and
Governmentit þ β9  Openupit þ β10  Foreigninvestit þ εit
comparison. Due to the absence of some data for some countries,
ð6Þ all the data of 17 countries, including Guatemala, Haiti, Jamaica,
and Tunisia, were finally removed, and only the IFI values of the
If the coefficient symbols of the IFI and the interaction item existing countries in the past four years were retained. The results
Education*IFI were the same and both were significant, it would are displayed in Table 4.
indicate that the education development level had strengthened It is obvious from the results that over time, the average value
the role of financial inclusion on employment. If the symbols of the IFI first increased and then decreased. A possible reason for
were opposite and both were significant, it would suggest that the this is that the measurement index system selected for each year
level of educational development weakened the role of financial was different. For an evaluation system that is becoming
inclusion on employment. increasingly digital and focuses on users’ actual experience, many
countries may not have sufficient time and resources to adapt to
Regression results analysis the adjustment. In addition, the impact of the COVID-19
Descriptive statistics. To observe the data characteristics of the pandemic in 2019 had an impact on the financial system of
variables selected in this study, the natural logarithm of per capita some countries, especially poor and weak countries and regions,
GDP was first taken to limit its order of magnitude to about 0–10, resulting in a low financial inclusion index. However, from the
the other variables were 0–1, and the dimension difference of standard deviation point of view, there was a downward trend,
each variable was small. Next, descriptive statistics were carried which indicates that the development of financial inclusion in the
out for the main variables. Since some data were missing and the world was synchronised to a certain extent. This is also in line
sample size differed, the interpolation method was used to deal with the original intention of financial inclusion to eliminate the
with the missing values. As some countries lacked four years of gap between rich and poor and extreme poverty, and to narrow
data on education (Education), fiscal expenditure (Government), this gap worldwide.
resident consumption level (Consumption), and degree of open-
ness (Openup), it was not possible to utilise all of them. The
results are summarised in Table 3, demonstrating the sample size, Basic regression analysis. Controlling for other conditions to
mean, standard deviation, and extreme values of the IFI, verify H1, the financial inclusion index was proportional to the
employment level, gender employment level, education, economic employment level in regression model (1). As shown in column
level, and other control variables. As can be observed from Table (1) of Table 5, when no control variables were added, IFI and
3, the mean value of the financial inclusion index was 0.619, the employment level (Employment) were regressed: the coefficient of
lowest level was 0.453, the most developed level was 0.840, and IFI was 0.184 and significant at the 1% level, which initially
the standard deviation was 0.107, thereby indicating that the verified H1. The control variables were again added to the
development of financial inclusion from 2011 to 2021 was still regression. In column (2) of Table 5, the coefficient of IFI was
unbalanced among countries. Second, the mean level of 0.119 and significant at 5% level, which again verified H1. With
Employment was 0.568, indicating that the employment status of the other control conditions unchanged, the degree of develop-
the working-age population still had substantial room for ment of financial inclusion was proportional to the employment
improvement compared with the global average level, and the level of the country. Regarding the control variables, except for
standard deviation of 0.113 indicated that the employment level the natural logarithm of per capita GDP, the remaining variables
was uneven, with a large gap between the lowest level of 0.308 and were negatively correlated with employment level, which was not
the maximum level of 0.852. On average, male employment completely consistent with the references. Possible reasons
(Maleemploy) was substantially higher than female employment include: (1) the year and region of the selected data source were
(Femaleemploy), and the standard deviation of female employ- different; (2) the economic factors represented by the control
ment was larger. The extreme value of education (education) had variables were inversely proportional to the employment level.
a large gap, but its degree of dispersion was low. Among the Specifically, urbanisation, as represented by the urbanisation rate,

6 HUMANITIES AND SOCIAL SCIENCES COMMUNICATIONS | (2024)11:172 | https://doi.org/10.1057/s41599-024-02630-4

HUMANITIES AND SOCIAL SCIENCES COMMUNICATIONS | https://doi.org/10.1057/s41599-024-02630-4 ARTICLE

Table 3 Descriptive statistics. financial inclusion. There may be missing estimator variables,
leading to the endogeneity of the model. The natural logarithm of
mobile cell subscriptions per 100 people (Mobile) was used as an
Variable Sample size Mean Standard Min. Max.
instrumental variable. Column (1) in Table 6 shows the results for
deviation
the first stage. The number of mobile subscriptions was inversely
IFI 384 0.619 0.107 0.453 0.840 proportional to IFI; the F value was 10.49 and the significance
Employment 384 0.568 0.113 0.308 0.852
level was 5%, indicating a strong correlation. Column (2) shows
Maleemploy 384 0.666 0.105 0.408 0.903
Femaleemploy 384 0.470 0.153 0.074 0.830
the results of the second stage. IFI was directly proportional to
Education 321 0.144 0.047 0.051 0.316 employment levels (Employment), significant at the 5% level,
GDPper 384 8.841 1.478 5.905 11.725 which is consistent with the original conclusion. In addition, the
Government 373 0.160 0.051 0.024 0.293 exogeneity test of the independent variable was carried out, and
Urbanisation 384 0.625 0.221 0.157 1.000 the P value was 0.001, rejecting the null hypothesis, indicating the
Industry 384 0.257 0.096 0.028 0.727 necessity of dealing with endogeneity. In the unidentifiable test,
Consumption 373 0.634 0.151 0.237 1.188 the P value was 0.020, rejecting the null hypothesis and passing
Foreigninvest 384 0.056 0.160 −0.417 2.228 the unidentifiable test.
Openup 373 0.856 0.554 0.226 3.791

Extended analysis: regulatory effects. We next verified H2 and

H3: controlling for other conditions, the level of education
Table 4 IFI descriptive statistics. development promotes the positive effect of financial inclusion on
the level of employment. Regression model (2) shows the results
Year Number of Mean Standard Min. Max. in column (3) of Table 6. The results illustrate that the regression
countries deviation coefficient of Education was 0.500, significant at the 1% level, in
2011 84 0.340 0.054 0.812 0.212 line with H2, indicating that the increase in education input
2014 84 0.390 0.148 0.587 0.127 effectively promotes the improvement of the employment level.
2017 84 0.326 0.040 0.713 0.176 Second, the regression coefficient of the interaction term Educa-
2021 84 0.191 0.026 0.522 0.101 tion*IFI was 1.445, significant at the 1% level, verifying H3; that
is, the more developed the education level, the more the devel-
opment of financial inclusion in the region promotes an increase
may exacerbate resource inequality, employment opportunity in the employment level.
inequality, and competition for housing and transportation
resources between urban and rural areas, thus exacerbating social Panel regression analysis of instrumental variables extended
problems, including unemployment (Pilehvar 2021). Greater analysis: heterogeneity regression
economic openness can crowd out local markets and create jobs. The impact of financial inclusion on different gender
employment levels. Gender differences in the labour market and
Robustness test workplace have long been a heavily debated social topic. Women
Change core variable. To test the robustness of the model, first, a suffer from lack of health, education, work opportunities, and
robustness test was carried out by replacing the core variables. control over their own lives and choices (Kabeer 1999). As a
Employment, the variable explained in model (1), was replaced by result, they have limited digital finance opportunities. Gender
unemployment, and the results are shown in column (4) of Table gaps persist in many developing countries and are not only
5. The results indicate that the regression coefficient of IFI was unfair, but also reduce economic performance (Klasen 2018).
−0.115 and significant at the 5% level, showing that the devel- Klasen also shows that lowering gender gaps in education leads to
oped degree of financial inclusion was inversely proportional to higher economic performance. There have also been some
the unemployment rate. This reflects that the development of reviews focusing on the factors causing the financial exclusion of
financial inclusion is conducive to employment to a large extent, women (Saluja et al. 2023). Based on these findings, this study
which is consistent with the original conclusion. The regression speculates that because financial inclusion serves vulnerable and
results of the control variables were in line with the original marginalised groups, the positive influence of financial inclusion
conclusion, which preliminarily verifies the robustness. on employment in female groups is greater than that in male
groups. Therefore, employment in model (1) was replaced with
Removing extreme values. Second, this study adopted tail-tail male employment level (Maleemploy) and female employment
processing for the data of the maximum 10% and minimum 10% level (Femaleemploy) for regression, and the results are shown in
of IFI and conducted regression analysis again to further test the columns (1) and (2) of Table 7. As shown in the table, the IFI of
robustness of model (1). Column (3) of Table 5 displays the both men and women in employment was 5%, and the regression
results. Consistent with previous results, the regression coefficient coefficient of women was greater than that of men, which verifies
of IFI was 0.117 and significant at the 5% level. The results show the conjecture. This shows that the employment improvement
that the developed degree of financial inclusion was proportional effect of financial inclusion was more obvious among women.
to the employment level, which is consistent with the original
conclusion, and the regression coefficient sign and significance of The influence of financial inclusion on the employment level in
the control variables were in accordance with the original model, different economically developed regions. The digitalisation of
further verifying the model’s robustness. financial inclusion is steadily intensifying, relying increasingly on
the popularisation and application of electronic products and
Endogenous processing: Instrumental variable method. Since it technology. These require a better credit management mechanism
is impossible to rule out the reverse causality between employ- and market supervision capacity and a stronger resistance to the
ment and the development of financial inclusion, countries with risk of non-performing loans. More economically developed
high employment levels have a high probability of a better eco- regions are often better able to meet the infrastructure require-
nomic situation, which itself is beneficial to the development of ments of financial inclusion. By contrast, financial inclusion may

HUMANITIES AND SOCIAL SCIENCES COMMUNICATIONS | (2024)11:172 | https://doi.org/10.1057/s41599-024-02630-4 7

ARTICLE HUMANITIES AND SOCIAL SCIENCES COMMUNICATIONS | https://doi.org/10.1057/s41599-024-02630-4

Table 5 Basic regression and robustness test.

Variables (1) (2) (3) (4)

Employment Unemployment
IFI 0.184*** (0.057) 0.119** (0.055) 0.117** (0.056) −0.115** (0.057)
GDPper 0.014 (0.009) 0.014* (0.009) −0.028*** (0.010)
Government −0.153* (0.079) −0.157** (0.079) 0.147* (0.084)
Urbanisation −0.362*** (0.109) −0.366*** (0.109) 0.232** (0.112)
Industry −0.155*** (0.054) −0.156*** (0.054) 0.064 (0.055)
Consumption −0.166*** (0.039) −0.168*** (0.039) 0.114*** (0.040)
Foreigninvest −0.007 (0.010) −0.007 (0.010) −0.027*** (0.010)
Openup −0.005 (0.013) −0.004 (0.013) −0.042*** (0.013)
Year, Country fixed fixed fixed fixed
R^2 0.111 0.216 0.216 0.235

*p < 0.1; **p < 0.05; ***p < 0.01.

Table 6 Instrumental variables and moderating effects.

Variable (1) (2) (3)

IFI Employment
IFI 1.611** (0.748) 0.202*** (0.065)
Education 0.500*** (0.149)
Education*IFI 1.445 *** (0.381)
Mobile −0.023** (0.010)
GDPper 0.036*** (0.011) −0.037 (0.031) 0.016 (0.010)
Government 0.051 (0.097) −0.297* (0.174) −0.163* (0.085)
Urbanisation 0.551** (0.098) −1.176*** (0.385) −0.330*** (0.118)
Industry −0.190*** (0.065) 0.072 (0.178) −0.168*** (0.063)
Consumption −0.092* (0.049) −0.047 (0.106) −0.173*** (0.047)
Foreigninvest −0.011 (0.012) 0.007 (0.022) −0.008 (0.010)
Openup 0.019 (0.015) −0.032 (0.029) −0.001 (0.129)
Year, Country fixed fixed fixed
R^2 0.245 0.225

*p < 0.1; **p < 0.05; ***p < 0.01.

be more effective in poor areas because it benefits specific groups and an overall decrease in 2021. Second, the development level of
and aims to narrow the gap between rich and poor. To explore financial inclusion can significantly promote the improvement of
the heterogeneity of the effect of economic development on employment levels. Both tail reduction and replacement of the
financial inclusion, based on the survey results of the World explained variables verify the model’s robustness. Third, the
Bank, this study divided the countries into four groups: low higher the level of educational development, the higher the level
income, low middle income, high middle income, and high of employment, and the stronger the function of financial
income, and placed them into model (1) for regression. As shown inclusion in promoting employment. Fourth, the employment
in Table 7, the regression coefficients of IFI were all positive; promotion effect of financial inclusion is more obvious among
highest for low-middle-income countries, followed by low- women. In middle- and low-income countries and above, the
income countries, high-middle-income countries, and high- effect of financial inclusion on employment will decrease with the
income countries. The coefficient in the low-income group was increase in income levels, which shows that financial inclusion
no longer significant, but significant at higher levels in the other has certain requirements on the economic foundation at the
groups. According to the comprehensive results, the effect of present stage, and there is a clear bias in benefiting poor and weak
financial inclusion on employment level was positive at all areas.
income levels, but it put forward certain requirements for the
degree of economic development. However, the benefits were
greater in lower-income countries than in higher-income Policy suggestions
countries. The government and financial institutions. First, policy should
focus on the coordinated development of financial inclusion
among regions. A large standard deviation in the financial
Conclusions and further discussion inclusion index indicates that its development in different
Research results. By studying the data of more than 100 countries countries is unbalanced. To develop financial inclusion, low-
and constructing the financial inclusion index and panel model, income countries should, on the premise of conforming to their
this study concludes the following. First, the global development national conditions, expand the coverage of financial infra-
of financial inclusion is seriously unbalanced, and the compre- structure, enhance residents’ willingness to use inclusive financial
hensive level needs to be improved; on average, the utilisation rate services, and formulate financial services and policies suitable for
of surplus labour force still has large room for improvement; the implementation in remote areas. Second, it accelerates the digital
financial inclusion index showed a downward trend year by year, transformation of financial inclusion, and improves the quality

8 HUMANITIES AND SOCIAL SCIENCES COMMUNICATIONS | (2024)11:172 | https://doi.org/10.1057/s41599-024-02630-4

HUMANITIES AND SOCIAL SCIENCES COMMUNICATIONS | https://doi.org/10.1057/s41599-024-02630-4 ARTICLE

and effectiveness of inclusive financial services. When separately

−0.690*** (0.150)

0.294*** (0.091)
(6) High income

0.141*** (0.045)
calculating the financial inclusion index for the four years in

−0.006 (0.012)
0.321** (0.127)
0.022 (0.018)

0.027 (0.019)
question using descriptive statistics, the overall financial inclusion

0.011 (0.122)
index declined after more digital indicators and user experience
evaluation indicators were added in 2021. The digital wave of

0.336
fixed
financial inclusion is endowed with increasing significance, and
financial institutions—as the main entities undertaking the
development of financial inclusion—should promote mobile
device lending, trading products and services, and complete
digital transformation as soon as possible. While providing ser-
vices, institutions should also pay attention to the user’s experi-

−0.078*** (0.026)
−0.570*** (0.127)
−1.625*** (0.185)

ence feedback and carry out regular assessment and rectification.

0.550*** (0.154)
−0.046 (0.028)

−0.070 (0.057)
(5) High middle

−0.229 (0.182)

Third, the development of financial inclusion and establishment

0.154 (0.301)

of a more direct channel for financial inclusion to help employ-

ment should be established. Based on the conclusion of the basic
income

0.379
regression analysis, financial institutions and governments should
fixed

continue to increase the coverage of financial infrastructure and

use of financial services, and cater to specific groups of people to
maximise financial inclusion. The government can use inclusive
policies to broaden financing channels for more vulnerable
groups and help reduce their concerns about employment and
−0.144*** (0.041)

entrepreneurship. Attention should be paid to controlling foreign

0.383*** (0.140)
0.634** (0.280)
−0.195* (0.106)
(4) Low middle

1.100*** (0.312)

0.089* (0.052)

capital’s influence on domestic manufacturers, and efforts should

0.009 (0.264)
0.065 (0.302)

be made to solve the problems of housing shortages and the

protection of rights of rural residents during urbanisation to
income

reduce its negative impact on employment levels. Moreover, some

0.461
fixed

scholars concluded that financial technology is greater related to

small micro-finance institutes (Hussain and Rasheed 2023).
Micro finance, as well established-tools for financial inclusion,
can serve as a source of inspiration. The government can intro-
duce more favourable policies for microcredit, and at the same
−0.260** (0.096)

time, financial institutions can launch innovative targeted pro-

−0.449* (0.261)
−0.343* (0.176)

−0.030 (0.078)
−0.023 (0.253)
(3) Low income

−0.363 (0.212)
0.738 (0.476)

0.643 (0.410)

ducts like microcredit, to benefit groups that are excluded by

Employment

formal channels of finance. Fourth, in view of the results of the

adjustment effect analysis, the education development level can
0.663

enhance the effect of financial inclusion in promoting the level of

fixed

employment. Therefore, the government can consider increasing

the proportion of education investment, developing local educa-
tion to higher levels, strengthening the coordination between
education investment and financial inclusion, and playing a more
aggressive role in promoting education development. Countries
−0.197*** (−0.046)
−0.214*** (−0.062)
−0.315** (−0.122)
−0.172* (−0.092)

with a low financial inclusion index should increase the propor-

0.139** (−0.062)

−0.019 (−0.014)
−0.01 (−0.011)

tion of financial literacy training in education and enhance the

Female employ

0.011 (−0.01)

breadth and depth of the use of financial inclusion to promote

employment. Meanwhile, the promotion of micro finance will
also enhance residents’ understanding of financial products and
0.193
fixed
(2)

the operation of financial institutions, thereby enhancing their

financial literacy. More innovative products will also spawn some
new positions for society. This is also one aspect of the role of
microfinance in employment through education.
−0.140*** (−0.044)
−0.159*** (−0.059)

Society as whole and individual residents. First, descriptive sta-

−0.438*** (−0.118)
−0.183** (−0.089)

−0.003 (−0.011)

tistics and heterogeneity regression results show that, compared

0.025** (−0.01)
0.132** (−0.06)

0.022 (−0.014)

with men, women are at a disadvantage both regarding the

Male employ
Table 7 Heterogeneity analysis.

employment level itself and financial inclusion’s positive impact

in promoting employment level—they do not enjoy the same
0.266

benefits brought by financial inclusion as men, featuring low

fixed
(1)

availability and usage of financial products and services. Society

*p < 0.1; **p < 0.05; ***p < 0.01.

as a whole should pay due attention to this issue. For example, the
government should formulate policies and regulations to protect
women’s rights in employment opportunities. Second, the het-
erogeneity regression analysis shows that financial inclusion’s
Year, Country
Consumption
Foreigninvest
Urbanisation
Government

promotion effect on employment in low-income countries is

Variable

Industry

Openup
GDPper

insufficiently significant. For low-income persons, special atten-

tion is required regarding the improvement of infrastructure,
R^2
IFI

such as networks, mobile terminals, and digital media. For low-

HUMANITIES AND SOCIAL SCIENCES COMMUNICATIONS | (2024)11:172 | https://doi.org/10.1057/s41599-024-02630-4 9

ARTICLE HUMANITIES AND SOCIAL SCIENCES COMMUNICATIONS | https://doi.org/10.1057/s41599-024-02630-4

income groups, more attention should be paid to improving Miller PW, Volker PA (1984) The screening hypothesis: an application of the wiles
infrastructure conditions such as networks, mobile terminals, and test. Econ Inq 22:121–127. https://doi.org/10.1111/j.1465-7295.1984.tb00671.x
digital media. Low-income countries can strengthen orderly Mingxing C, Xinrong H, Jiafan C et al (2023) Urbanization and vulnerable
employment: Empirical evidence from 163 countries in 1991–2019. Cities
exchanges and cooperation with the outside world to promote 135:104208
their own financial inclusion. Parameshwara P, Raghurama A (2011) Role of banks in financial inclusion. Indian
This study still has some limitations, such as not including all J Appl Res 3:336–338. https://doi.org/10.15373/2249555X/JUNE2013/112
countries in the analysis due to missing data. Due to limitations in Pilehvar AA (2021) Spatial-geographical analysis of urbanization in Iran. Palgrave
data availability, the moderating effect of education was not explored Commun 8:1–12
Ravi S (2021) The returns to higher education and public employment. World Dev
in heterogeneity classification regression for gender. If more research 144:105471
can be added in the future the study will be more constructive. Renzhia N, Baek YJ (2020) Can financial inclusion be an effective mitigation
measure? evidence from panel data analysis of the environmental Kuznets
curve[J/OL]. Finance Research Letters, 37. https://doi.org/10.1016/j.frl.2020.
101725
Data availability Riccardo C, Roberto G, Michael S (2022) Does foreign investment hurt job creation
The data that support the findings are openly available in the at home? The geography of outward FDI and employment in the USA. J Econ
World Bank’s Financial Inclusion and Financial Availability Geogr 22:53–79
Survey database. Other related materials used or analysed during Romero-Castro N, López-Cabarcos MA, Piñeiro-Chousa J (2023) Finance, tech-
the study are available from the corresponding author on rea- nology, and values: A configurational approach to the analysis of rural
entrepreneurship. Technol Forecasting Soc Change 190. https://doi.org/10.
sonable request. 1016/j.techfore.2023.122444
Rotatori D, Lee EJ, Sleeva S (2021) The evolution of the workforce during the
Received: 16 August 2023; Accepted: 4 January 2024; fourth Industrial Revolution. Hum Resour Dev Int 24:92–103. https://doi.org/
10.1080/13678868.2020.1767453
Sakyi-Nyarko C, Ahmad AH, Green CJ (2022) The gender-differential effect of
financial inclusion on household financial resilience. J Dev Stud 58:692–712.
https://doi.org/10.1080/00220388.2021.2013467
Saluja OB, Singh P, Kumar H (2023) Barriers and interventions on the way to
empower women through financial inclusion: a 2 decades systematic review
References (2000–2020). Humanit Soc Sci Commun 10:148. https://doi.org/10.1057/
Acharya RC (2017) Impact of trade on Canada’s employment, skill and wage
s41599-023-01640-y
structure. World Econ 40:849–882. https://doi.org/10.1111/twec.12360
Sarfo Y, Musshoff O, Weber R (2023) Farmers’ awareness of digital credit: Does
Ahn K, Winters JV (2023) Does education enhance entrepreneurship? Small Bus
financial literacy matter? J Int Dev:1–9. https://doi.org/10.1002/jid.3774
Econ 61:717–743. https://doi.org/10.1007/s11187-022-00701-x
Sarma M (2012) Index of financial inclusion A measure of financial sector inclu-
Allen F, Demirguc-Kunt A, Klapper L et al. (2016) The foundations of financial
siveness. No. 1207, Competence Centre on Money, Trade, Finance and
inclusion: Understanding ownership and use of formal accounts. J Financ
Development, Hochschule für Technik und Wirtschaft, Berlin
Intermed 27:1–30. https://doi.org/10.1016/j.jfi.2015.12.003
Senyo PK, Gozman D, Karanasios S et al. (2023) Moving away from trading on the
Asfaw A, Lamanna F, Klasen S (2009) Gender gap in parents’ financing strategy for
margins: Economic empowerment of informal businesses through FinTech.
hospitalization of their children: evidence from India. Health Econ
Inf Syst J 33:154–184. https://doi.org/10.1111/isj.12403
19(3):265–279. https://doi.org/10.1002/hec.1468
Sharif SP, Naghavi N, Hassam Waheed E et al (2022) The role of education in
Autor DH (2015) Why are there still so many jobs? The history and future of
filling the gender gap in financial inclusion in low-income economies. Int J
workplace automation. J Econ Perspex 29:3–30. https://doi.org/10.1257/jep.
Emerg Markets 18:1–36
29.3.3
Shaw ES (1973) Financial deepening in economic development. Oxford University
Blanchard O, Katz LF (1997) What we know and do not know about the natural
Press, New York
rate of unemployment. J Econ Perspect 11:51–72. https://doi.org/10.1257/jep.
Xing C, Yang P, Li Z (2018) The medium-run effect of China’s higher education
11.1.51
expansion on the unemployment of college graduates. China Econ Rev
Bo Y, Fang M, Weihua D et al (2022) Digital financial inclusion and rural
51:181–193. https://doi.org/10.1016/j.chieco.2017.05.005
household subsistence consumption in China. Econ Anal Policy 76:627–642.
Yang CH (2022) How artificial intelligence technology affects productivity and
https://doi.org/10.1016/J.EAP.2022.09.007
employment: Firm-level evidence from Taiwan. Res Policy 51. https://doi.org/
Boustanifar H (2014) Finance and employment: evidence from U.S. banking
10.1016/j.respol.2022.104536
reforms. J Bank Fin 46:343–354
Zhechen G, Guosheng H (2021) Digital financial inclusion and sustainable
Evans DS, Jovanovic B (1989) An estimated model of entrepreneurial choice under
employment: Evidence from countries along the belt and road. Borsa Istanb
liquidity constraints. J Pol Econ 97:808–827. https://doi.org/10.1086/261629
Rev 21:307–316
Fu J (2020) Ability or opportunity to act: What shapes financial well-being? [J/OL]
World Dev 128. https://doi.org/10.1016/j.chieco.2021.101628
Hussain S, Rasheed A (2023) Financial inclusion based on financial technology and
risky behaviour of micro-finance institutes: evidence from South Asian Acknowledgements
micro-finance banks. Digit Policy Regul Gov 25(5):480–489. https://doi.org/ This research project was supported by the National Social Science Foundation Project
10.1108/DPRG-12-2022-0149 ‘Multi-dimensional research on the comparative evaluation and enhancement mechan-
Johnson S (2013) From microfinance to inclusive financial markets: the challenge ism of financial inclusion under digital empowerment’ (Project No. 21BJL087). We
of social regulation. Ox Dev Stud 41:35–52 would like to thank Editage (www.editage.cn) for their English language editing.
Kabeer N (1999) Resources, agency, achievements: reflections on the measurement
of women’s empowerment. Dev Change 30(3):435–464
Kim K (2022) Assessing the impact of mobile money on improving the financial
Author contributions
XS conceived the overall research idea and designed the research method and was a
inclusion of Nairobi women. J Gend Stud 31:306–322. https://doi.org/10.
major contributor in writing the manuscript. JL analysed and interpreted data on
1080/09589236.2021.1884536
inclusive finance and technological innovation in various countries and built an inclusive
Klasen S (2018) The impact of gender inequality on economic performance in
financial indicator system. XW modified and optimized the model. All authors read and
developing countries. Annu Rev Resour Econ 10(1). https://doi.org/10.1146/
approved the final manuscript.
annurev-resource-100517-023429
Koomson I, Martey E, Etwire PM (2023) Mobile money and entrepreneurship in
East Africa: the mediating roles of digital savings and access to digital credit.
Competing interests
Inf Technol People 36:996–1019. https://doi.org/10.1108/ITP-11-2021-0906
The authors declare no competing interests.
Lai JT, Xie J, Cao S et al. (2022) Digital financial inclusion and illegal fundraising in
China. Appl Econ 54(48):5575–5590
Lin KH (2016) The rise of finance and firm employment dynamics. Organ Sci Ethical approval
27:972–988. https://doi.org/10.1287/orsc.2016.1073 This article does not contain any studies with human participants performed by any of
Mais S, Claudia G, Sarkisyan A (2020) Cross-country variation in financial the authors
inclusion: a global perspective. Eur J Fin 26:4–5

10 HUMANITIES AND SOCIAL SCIENCES COMMUNICATIONS | (2024)11:172 | https://doi.org/10.1057/s41599-024-02630-4

HUMANITIES AND SOCIAL SCIENCES COMMUNICATIONS | https://doi.org/10.1057/s41599-024-02630-4 ARTICLE

Informed consent Open Access This article is licensed under a Creative Commons
This article does not contain any studies with human participants performed by any of Attribution 4.0 International License, which permits use, sharing,
the authors. adaptation, distribution and reproduction in any medium or format, as long as you give
appropriate credit to the original author(s) and the source, provide a link to the Creative
Commons license, and indicate if changes were made. The images or other third party
Additional information
Supplementary information The online version contains supplementary material material in this article are included in the article’s Creative Commons license, unless
available at https://doi.org/10.1057/s41599-024-02630-4. indicated otherwise in a credit line to the material. If material is not included in the
article’s Creative Commons license and your intended use is not permitted by statutory
Correspondence and requests for materials should be addressed to Xueke Wu. regulation or exceeds the permitted use, you will need to obtain permission directly from
the copyright holder. To view a copy of this license, visit http://creativecommons.org/
Reprints and permission information is available at http://www.nature.com/reprints licenses/by/4.0/.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in
published maps and institutional affiliations. © The Author(s) 2024

HUMANITIES AND SOCIAL SCIENCES COMMUNICATIONS | (2024)11:172 | https://doi.org/10.1057/s41599-024-02630-4 11