Gender and Financial Inclusion: Analysis of financial inclusion of women in the SADC region

Full text

Gender and financial inclusion
Analysis of financial inclusion of women in the SADC region
Prepared by FinMark Trust
August 2016
Policy research paper No. 01/2016
Authors
Ashenafi Beyene Fanta
Kingstone Mutsonziwa
1
1
For inquiries please contact Kingstone Mutsonziwa at kingstonem@finmark.org.za

Acknowledgement
We are grateful for the useful comments from Professor Daniel Makina from the
Department of Finance, Risk Management and Banking of UNISA who availed himself to
review the paper. We also extend gratitude to the CEO of FinMark Trust, Dr Prega
Ramsamy for insightful comments on earlier versions of the paper.

I
Executive summary
Gender gap prevails even in countries with the highest financial inclusion: The gender
gap in bank account ownership is highest in Botswana, Swaziland, and Mauritius, while
South Africa is the only country with a positive gender gap, i.e., women being more
financially included than men. This is may be mainly driven by women receiving social
grants through SASSA card.
The gap in account usage is wider than account ownership: The gender gap is wider
when usage rather than access is used to gauge financial inclusion. For instance, in Malawi
the gap in bank account usage is 19 percent while the gap in access to bank account is only
8 percent. Not surprisingly, despite the higher proportion of females being banked in South
Africa, it is actually males that have more used accounts than females. Most bank accounts
owned by females in South Africa are either dormant or mailbox accounts. The gender gap
in access to bank credit is bigger even in countries where the level of access is the highest.
More females use someone else’s account: Usage of someone else’s account is the
highest among females, and this is the case especially in Zambia, Tanzania and Swaziland.
Furthermore, more females cite lack of money as the primary reason for not having a bank
account. Females also cite remoteness of bank branches and lack of understanding about
how banks operate as reasons for not having a bank account.
Gender affects financial inclusion: Gender affects financial inclusion even after controlling
for individual characteristics such as household size, age, education, place of residence,
marital status, employment status, income, and level of education, implying that financial
services are biased against females.
Promote financial literacy through financial education: Financial education programs
targeted at females will enable them to develop a reasonable understanding about the
language used by banks, benefits of owning a bank account, and how to apply for it. Such
programs should also enable females to develop skills in household financial management
that leads to their empowerment and increased involvement in household financial
decisions.

II
Introducing agency banking and mobile money in rural areas: Agency banking and
mobile money will help females in rural areas that are excluded from owning a bank
account due to remoteness of bank branches.
Strengthening informal financial service providers to expand the outreach of financial
services to females in rural areas: Accessibility to informal financial service providersand
their ability to design products that suit the needs of individuals makes them ideal for many
females in the region. However, informal operators may not have the technical know-how
of managing financial services and may also lack resources to satisfy the needs of their
clients. Building the financial and managerial capacity of the informal financiers may allow
majority of rural females to get quality financial services at a cheaper cost.
Mitigating risks in the informal sector: While the informal sector supports female access
to financial services, there are always the risks of exploitation. Safety nets need to be
provided through appropriate consumer protection measures.
Income generating capability to improve financial inclusion and address gender
disparity: Both the descriptive and economic analyses point to the fact that financial
inclusion is strongly linked with income generating capability. With women facing wider
exclusion it is important to also address the issue of gender equality in economic activities.

III
Contents
1. Introduction .................................................................................................................................... 1
2. Financial inclusion of women in the SADC region: an overview ..................................................... 4
3. Data and methodology ................................................................................................................... 5
4. Analysis and results ......................................................................................................................... 6
4.1. Preliminary analysis ................................................................................................................ 6
4.1.1. Account ownership by gender across countries ............................................................ 6
4.1.2. Bank account usage status by gender ............................................................................ 9
4.1.3. Borrowing by gender .................................................................................................... 11
4.1.4. Savings by gender ........................................................................................................ 12
4.2. Gender and financial inclusion: Econometric analysis ......................................................... 14
4.2.1. Descriptive statistics and Chi-Square test results. ....................................................... 14
4.2.2. Econometric model results ........................................................................................... 14
4.3. Barriers to financial inclusion of women .............................................................................. 20
4.3.1. More females use someone else’s account .................................................................. 20
4.3.2. Barriers to account ownership by gender .................................................................... 21
References ............................................................................................................................................ 25
Appendix A ........................................................................................................................................... 27
Appendix B ............................................................................................................................................ 28
List of Tables
Table 1: Sample size and year survey was conducted .............................................................. 6
Table 2: Percentage of adult population making ................................................................... 21
Table 3: Summary statistics of independent variables by gender ......................................... 28
Table 4: Logistic regression output: gender and account ownership ................................... 29
Table 5: Logistic regression output: gender and access to credit ......................................... 30
Table 6: Logistic regression output: gender and access to savings ...................................... 31

IV
List of Figures
Figure 1: SADC access strand by gender (including South Africa) ......................................... 4
Figure 2: SADC access strand by gender (excluding South Africa) ........................................ 5
Figure 3:Bank account ownership by gender in each country ............................................... 7
Figure 4:Account ownership at a formal financial institution by gender in each country ...... 8
Figure 5: Informal account ownership by gender in each country ......................................... 8
Figure 6: Proportion of financially excluded population by gender in each country .............. 9
Figure 7:Bank account status in each country ..................................................................... 10
Figure 8: Bank account status by gender in each country .................................................... 11
Figure 9: Credit Strands by gender for each country ........................................................... 12
Figure 10: Saving Strands by gender for each country ......................................................... 13
Figure 11: Usage of someone else’s account by gender in each country ............................. 20

1
1. Introduction
Financial inclusion, i.e., access to and uptake of affordable financial services, allows
individuals to store value in a safe place, access credit, and through insurance products
manage risks. Without inclusive financial systems, poor individuals and small enterprises
need to rely on their own limited savings and earnings to invest in their education, become
entrepreneurs, or take advantage of promising growth opportunities (Beck and Honohan,
2008). Financial inclusion is an important tool for eradicating poverty and narrowing
income inequality and as such it is an integral part of inclusive development and a building
block for poverty reduction strategy (Chibba, 2009). The role of financial inclusion in
poverty alleviation is supported by empirical evidence. For instance, Burgess, Pande and
Wong (2005) reported that state-led branch expansion into rural unbanked locations and
the enforcement of directed bank lending in India led to reduction in poverty through
increased bank borrowing among the poor, in particular among low caste and tribal groups.
Although access to finance has not been directly spelt either in the Millennium
Development Goals (MDGs) or in the new Sustainable Development Goals (SDGs), access
to financial services is an important direct or indirect contributor to the achievement of
most of the goals (Claessens and Feijen, 2007). For instance, in the case of education and
health, one important effect of access to financial services is through the income effect:
better access to financial services improves incomes and therefore the possibility of
obtaining health and education services. It also contributes to the fourth SDG goal of
gender equality because allowing women direct access to financial services might improve
their possibilities to become entrepreneurs, thus increasing their individual incomes, their
chances to become more independent, and their participation in family and community
decision making. Improving financial inclusion has thus received attention in a number of
national governments. According to Demirgüç-Kunt et al. (2015), out of 143 economies, 67
percent have a mandate to promote financial inclusion and more than 50 countries have set
formal targets and ambitious goals for financial inclusion. International organisations,
including the G20 and the World Bank, are also beginning to formulate strategies to
promote financial inclusion.

2
Demirgüç-Kunt et al. (2015) reported that 62 percent of adults worldwide have an account
at a bank or another type of financial institution or with a mobile money provider. They also
highlighted the existence of marked disparity in financial inclusion across regions. For
instance, 94 percent of adults in the OECD countries have an account compared to only 34
percent in sub-Saharan Africa (SSA). Besides, there is a gender gap in financial inclusion
which is highest in developing countries, with account penetration being lower among
women. While the global gender gap
2
is 7 percent, it is 9 percent in SSA (30 percent for
women compared to 39 percent for men). Based on the latest FinScope data for 12
countries, the gender gap in the SADC region is 5 percent (60 percent for men and 55
percent for women) which is slightly below the SSA average.
Exclusion of women from financial services has been reported by a number of studies that
have found that women are more excluded than men both at firm and individual levels.
Studies report that female-owned firms face more financial constraints than male-owned
businesses (see for instance, Presbitero et al., 2014; Henderson et al., 2015; and Beck et al.,
2011). Using firm level data from countries in the Caribbean, Presbitero et al. (2014)
reported that women-led businesses are more likely to be financially constrained than
other comparable firms. Similarly, Henderson et al. (2015) noted that men are more
favourably treated when it comes to access to credit lines than women in the US, and
women and minority applicants are concerned that they receive even less favourable
treatment from lenders that is unrelated to their creditworthiness. This has been further
strengthened by Beck et al. (2011) in a European study in which they reported that female
borrowers are less likely to secure a loan when the loan officer is male. They also reported
that female borrowers assigned to opposite-sex officers get loans with unfavourable terms
such as higher interest rates and shorter maturities. More recently, Demirgüç-Kunt et al.
(2015) confirmed the existence of a gender gap in financial inclusion even after controlling
for a host of individual characteristics including income, education, employment status,
rural residency and age.
The gender gap is worrisome because exclusion of women from economic activities means
that their important contribution to economic development will be missed. Furthermore,
exclusion deprives women of human rights which should allow them to have equal
2
Gender gap refers to the difference in the level of financial inclusion between men and women.

3
opportunity to participate in social and economic activities. Two important arguments, i.e.,
the human rights argument and the capabilities argument, put forward by Beneria et al.
(2015) can be used to establish the importance of ensuring gender equality and hence
eliminating the gender gap. According to the human rights argument, women should enjoy
equal access to financial services so that they have equal participation in social and
economic activities. Women constitute 50 percent of the world population and their
exclusion would be detrimental to equitable economic growth. A gender gap therefore not
only affects women but also the whole nation by derailing economic growth. For instance,
Knowles et al. (2002) reported that educational gender gap is an impediment to economic
growth. Earlier, Klasen (2002) reported that gender inequality in education directly affects
economic growth by lowering the average level of human capital. Recently, Klasen and
Lammana (2009) reported that economies in the Middle East and North Africa lose
economic growth opportunities due to the gender gap in education suggesting that barriers
to female employment are not only disadvantageous to women, but also appear to reduce
economic growth in both developed and developing countries.
The capabilities argument emphasises the abilities of women in enhancing household
welfare and hence reducing poverty. Studies confirm that women manage resources better
than men. For instance, Pitt and Khandker (1998) reported that microcredit has a larger
effect on the behaviour of poor households in Bangladesh when women are the program
participants. The effect was captured in the form of a higher gain in annual household
consumption expenditure for women compared to men. Similarly, a recent study in India by
Swamy (2014) reported that women with access to microcredit experience a higher income
growth than men (8.40 percent for women against 3.97 percent for men). They also report
that women use the resources in a manner that improves family well-being and contribute
to significant increase in savings levels of the households. Similarly, women's credit is found
to have an impact on the health of both boys and girls while credit provided to men do not
have a similar effect on children’s health (Pitt et al., 2003). This is attributed to the fact that
women often make more optimal household spending decisions affecting children’s
welfare (Rawlings and Rubio, 2005). The financial inclusion of women also leads to their
empowerment. As reported by Pitt et al. (2006), access to credit leads to women taking a
greater role in household decision making, having greater access to financial and economic

4
resources, having greater social networks, having greater bargaining power vis-a -vis their
husbands, and having greater freedom of mobility.
Against the foregoing benefits of women access to financial services, this policy research
paper examines the significance of the gender gap in financial inclusion in the SADC region
with a view to providing policy prescriptions. Unlike previous studies that mainly focused on
examining the state of access to formal and informal financial services, the paper subjects
the FinScope data in the public domain to econometric analysis in order to study the
relationship between gender and financial inclusion in the SADC region.
The rest of the paper is organised as follows: The next section presents an overview of
financial inclusion in the SADC region emphasising the cross-country comparisons. Section
three describes the data and presents the econometric model used in examining the
relationship between gender and financial inclusion. Section four presents the analysis and
discussions which are divided into preliminary analysis and econometric results. Section
five includes the conclusion followed by policy recommendations in section six.
2. Financial inclusion of women in the SADC region: an overview
The comparison of the Access Strand for males and females in the SADC region shows that
females are less banked than males with a 6 percent gap. As shown in Figure 1,
proportionally lesser women have accounts at other formal financial institutions. However,
informal inclusion is slightly higher among females narrowing the overall level of gender
gap in financial inclusion to 5 percentage points.
Figure 1: SADC Access Strand by gender (including South Africa)
Source: FinScope
The above picture significantly changes when one excludes South Africa where women are
more financially included than men. The Regional gender gap doubles for bank account

5
ownership while account ownership at other formal and informal providers remains almost
unchanged. The gap in financial exclusion also doubles from 5 percent to 10 percent. A
similar experiment using other countries with a relatively higher level of financial inclusion
in the Region (such as Mauritius and Botswana) did not change the picture. A probe into the
nature of financial inclusion in South Africa shows that more women are banked due to the
SASSA MasterCard
3
ownership.
Figure 2: SADC Access Strand by gender (excluding South Africa)
Source: FinScope
3. Data and methodology
Data for the study were obtained from nationally representative FinScope Consumer
Surveys conducted in different years in Botswana, the Democratic Republic of Congo
(DRC), Malawi, Mauritius, Mozambique, South Africa, Swaziland, Tanzania, Zambia, and
Zimbabwe. Although FinScope surveys were available for Namibia and Lesotho for 2011,
the two countries were excluded because the data are old. Table 1 reports the FinScope
survey, size and year for each country.
3
SASSA MasterCard was introduced by the South African Social Security Agency (SASSA) and given to social
grant recipients in South Africa. The card allows grant recipients to cash out their money at ATMs of any bank
and also swipe the card at shops. In 2015, 10.5million South Africans (representing 28 percent of adult
population) had SASSA MasterCards, comprising 42 percent of adult females and 13 percent of adult males.
Further, 63 percent of SASSA card owners also have a bank account in their own name while 37 percent have
only SASSA card. Of the total adult population, those who have a bank account in their own name (including
those who also have SASSA card) constitute 64 percent. The total banked population (which also includes
those who have only a SASSA MasterCard) is 77 percent which includes those who have a bank account in
their own name (64 percent) and those who have only a SASSA card (13 percent). The gender split of those
who have only a SASSA card is 19 percent of the total banked females compared to 7 percent of banked
males. This clearly shows the importance of SASSA card in closing gender gap in financial inclusion in South
Africa.

6
Table 1: Sample size and year survey was conducted
No
Country
Year of survey
Sample size
1
Botswana
2014
1,503
2
Democratic Republic of Congo
2015
5,000
3
Malawi
2014
3,005
4
Mauritius
2014
4,000
5
Mozambique
2014
3,905
6
South Africa
2015
5,000
7
Swaziland
2014
3,440
8
Tanzania
2013
7,987
9
Zambia
2014
8,479
10
Zimbabwe
2014
4,000
Total
46,319
We also used Findex 2014 microdata to analyse the differences in account status (i.e.,
dormant, mailbox, and used) between gender groups.
4. Analysis and results
Preliminary analysis was conducted using descriptive statistics to compare access to
account, savings and credit across gender groups. A more robust analysis was conducted
through inferential statistics using three econometric models (for more details see
Appendix A). The models were used to capture the effect of gender on access to account,
savings, and credit while keeping unchanged factors such as income, household size, age,
place of residence, marital status, level of education, and employment status. The paper
utilised the FinScope data to look at overall access and also segment it into individual
elements that include access to banks, access to formal financial institutions (i.e. banks and
non-bank formal financial institutions), and access to informal finance.
4.1. Preliminary analysis
4.1.1. Account ownership by gender across countries
Account ownership was analysed across gender groups by first looking at bank account
ownership as this represents the most common pathway of access. As depicted in Figure 3,
females have lesser access than males, except in South Africa where it is vice-versa. The
gender gap varies across countries with Botswana (14%), Swaziland (14%), and Mauritius
(11%) having the largest gap and the Democratic Republic of Congo (2%) having the

7
smallest gap. South Africa is the only country in the region where more females are banked
than males. This is may be mainly driven by social grant payments that go to relatively
more females than males.
Figure 3: Bank account ownership by gender in each country
Source: FinScope
A further probe into access was made by including non-bank financial institutions such as
insurance companies, microfinance institutions, saving and credit associations and the like.
As shown in Figure 4, the gender gap narrows in most countries except in Tanzania,
Zambia, and the Democratic Republic of Congo where the gap has actually widened
implying that access of females to non-bank formal institutions is even worse than access
to bank accounts in these countries. The gender gap narrows even in South Africa (albeit
marginally) where females have better access than males. The gap persists in Malawi and
Mozambique. In Botswana and Swaziland, where the gap in bank account ownership is the
largest, it narrows down significantly, implying that the non-bank formal institutions are
more accessible than banks to females in these countries.
91
74
62
57
31
29
26
24
18
13
80
80
48
43
23
21
22
15
10
11
Maurtius
South Africa
Swaziland
Botswana
Malawi
Zambia
Zimbabwe
Mozambique
Tanzania
DRC Female Male

8
Figure 4: Account ownership at a formal financial institution by gender in each country
Source: FinScope
The gender gap is also evident in the access to informal accounts. However, the gap is
opposite with regard to formal account ownership. As depicted in Figure 5, females have
more access to informal accounts than males except in Zimbabwe and the Democratic
Republic of Congo where more males than females access informal accounts, and in
Botswana where access to informal accounts is the same between males and females. The
largest gender gap in access to informal accounts is observed in Swaziland followed by
Zambia which is consistent with the fact that the gender gap for formal account ownership
is among the highest for the two countries. The fact that more females access informal
accounts shows that informal providers serve as alternative routes to females that are
excluded from the formal sector.
Figure 5: Informal account ownership by gender in each country
Source: FinScope
93
81
71
69
66
63
43
39
38
28
83
86
65
61
65
51
33
32
30
19
Maurtius
South Africa
Botswana
Swaziland
Zimbabwe
Tanzania
Zambia
DRC
Malawi
Mozambique Female Male
18
15
14
13
13
5
8
3
8
1
24
18
17
16
12
12
8
4
4
3
Zambia
Mozambique
Tanzania
Malawi
DRC
Swaziland
Botswana
South Africa
Zimbabwe
Maurtius Female Male
1

9
The overall picture of gender disparity in financial inclusion can be illustrated by looking at
the proportion of people that are financially excluded. As depicted in Figure 6, females are
more excluded than males in all countries except in South Africa. The largest gender gap is
present in Tanzania (9%) followed by Mauritius (8%) and the Democratic Republic of Congo
(8%). The lowest gap is in Swaziland mainly due to the fact that the large gap in formal
access is offset by the gap in informal access. South Africa is an outlier because fewer
females than males are financially excluded.
Figure 6: Proportion of financially excluded population by gender in each country
Source: FinScope
4.1.2. Bank account usage status by gender
The foregoing sections focused on account ownership paying less attention to the extent of
usage of the accounts. The importance of usage as a better measure of financial inclusion
was highlighted by Rami (2009) who observed that government programmes in India have
led to inclusion of large numbers of low-income households without a proportionate
increase in usage. Similarly, a recent report by the UNCDF (2016) revealed that a large
majority of bank account owners in developing countries seldom use the accounts to store
value, to transact, and to access credit and most of the bank accounts are either dormant or
used as a mailbox. Analysis of account status
4
was conducted using Findex 2014 microdata
4
According to UNCDF (2016), accounts are classified into three categories as used, mailbox, and dormant
accounts. Used accounts are those that experience at least three withdrawals and/or deposits in a month
while mailbox accounts are those that experience withdrawals or deposits at most twice in a month. Dormant
accounts are those that did not have any withdrawal/deposit in a month.
57
48
49
39
23
26
21
26
6
16
63
56
54
43
32
31
27
27
14
10
Mozambique
DRC
Malawi
Zambia
Tanzania
Zimbabwe
Botswana
Swaziland
Maurtius
South Africa Female Male

10
for eight
5
SADC countries included in the survey. As shown in Figure 7, despite South Africa
and Mauritius being at the top of the access rank in the Region, it is actually Botswana that
leads in terms of percentage of used accounts. With 3 percent used accounts and 87
percent having dormant accounts, Zimbabwe is at the bottom of the usage ranking in the
Region.
Figure 7: Bank account status in each country
Source: Global Findex 2014
The gender gap in account usage is by far wider than access to a bank account. As
presented in Figure 8, Malawi has the highest gender gap in bank account usage (19%)
which is more than twice the gap in bank account ownership (8%). The lowest gender gap
in account usage is observed in Botswana (1%). Used accounts are skewed towards males
while mailbox and dormant accounts are skewed towards females. The gender gap is
present even in Mauritius (the country at the top of financial inclusion ranking in the region)
where 33% of male bank account holders have a used account compared to only 21% of
females. Most dormant accounts belong to females in the countries except Botswana
where males have more dormant accounts. In South Africa, a lower proportion of females
have used accounts and a relatively higher proportion of females have either a mailbox or
5
Swaziland and Zambia were not included in the analysis due to unavailability of Global Findex data for these
countries.
28 32 27 23 29 23 17
3
47 42 45 44 36
35
30
10
25 26 28 33 35 42
53
87
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
South Africa Botswana Mauritius DRC Malawi Zambia Tanzania Zimbabwe
Used Mailbox Dormant

11
dormant account which may be due to more females receiving social grants through a bank
account.
Figure 8: Bank account status by gender in each country
Source: Global Findex 2014
4.1.3. Borrowing by gender
Analysis of access to credit by gender reveals that the gap between males and females is
apparent in all countries. As shown in Figure 9, females have lower access to bank credit
than males except in the Democratic Republic of Congo and Malawi where there is no
perceptible difference between females and males. A large gender gap is observed in
Mauritius (12%) followed by Botswana (8%). A relatively low gender gap is observed in
Tanzania (1%), Zambia (2%), and Zimbabwe (2%). The gap prevails even when we factor in
access to credit from other formal institutions. For instance, in Mauritius, Mozambique and
Zambia, the gender gap in accessing formal credit is the same as the gap in accessing bank
credit, which means that non-bank financial institutions do not present any better
advantages to females than banks. However, in South Africa and Tanzania, the gap
disappears when credit from non-bank financial institutions is considered, implying that
more females access credit from non-bank financial institutions in the two countries. In
32
31
26
19
37
18
33
21
30
26
21
13
27
20
5
1
41
45
48
37
34
39
45
45
45
49
30
31
34
36
11
8
27
24
26
44
29
43
22
34
24
25
49
57
40
44
84
91
M
F
M
F
M
F
M
F
M
F
M
F
M
F
M
F
Botsw DRC Mala Maur SA Tanz Zamb Zimb
Active Mailbox Dormant

12
contrast, in Malawi, Swaziland, and Zimbabwe the gap widens further when credit from
non-bank financial institutions is factored in, which means that those institutions are less
accessible to females than banks in these countries. With more females than males
accessing informal credit the reality in the informal credit market is opposite to the one in
the formal market. The only exceptions are Mozambique, Tanzania, and Zambia where
more males than females have access to informal credit. In South Africa, males and females
have equal access to informal credit. The overall picture of access to credit shows that
females are more excluded than males in Botswana (7%), Mauritius (7%), Mozambique
(4%), Tanzania (7%), and Zambia (7%). There is no gender gap in access to credit in South
Africa and Zimbabwe. Interestingly, Swaziland, the Democratic Republic of Congo, and
Malawi have more males excluded than females. Lower exclusion of females is mainly
driven by more females accessing credit from the informal sector in these countries.
Figure 9: Credit Strands by gender for each country
Source: FinScope
4.1.4. Savings by gender
The other dimension of examining gender disparity in financial inclusion is by looking at
access to savings. Not surprisingly, the trend in gender gap persists in the saving market as
well. As depicted in Figure 10, the gender gap is highest in Botswana (11%), Swaziland

13
(11%), and Mauritius (10%). The gap is narrower in the Democratic Republic of Congo (2%),
South Africa (3%), and Mozambique (4%). The gap narrows in Botswana and Tanzania
when we consider saving at both bank and non-bank financial institutions, implying that
more females than males have access to non-bank financial institutions as a channel for
saving. In contrast, the non-bank financial institutions seem to be used more by males than
females in Mauritius, South Africa, Swaziland, and Zimbabwe where the gap widened
further. Not surprisingly, females dominate the informal saving market in the countries
except in Botswana, the Democratic Republic of Congo, and Zambia where more males
than females use informal mechanisms for saving. Analysis of the exclusion trend across
the countries shows that females are more excluded than males in all the countries. The
largest gender gap in exclusion is observed in the Democratic Republic of Congo, Mauritius,
and Zambia. Overall, our preliminary analysis of access to saving revealed that more
females are excluded from the saving market and this remains unchanged even after
accounting for informal saving channels in which more females are likely to participate. The
fact that females are excluded from the saving market in all the countries is consistent with
our report on barriers to account ownership in which more females cite lack of money as
the primary reason.
Figure 10: Saving Strands by gender for each country
17
14
38
27
66
56
8
6
17
8
15
7
17
9
13
7
9
5
36
25
10
6
11
16
6
2
3
3
3
2
3
2
3
5
12
8
1
1
12
9
10
14
17
16
4
6
50
40
27
28
37
34
69
72
24
29
26
27
19
29
63
65
35
41
25
36
39
51
53
62
45
57
11
14
51
56
64
67
34
37
M
F
M
F
M
F
M
F
M
F
M
F
M
F
M
F
M
F
M
F
South
Africa Bots Maurt DRC M ala Zamb Tanz Zimb Mozam Swazi
Bank Other formal Informal Excluded

14
Source: FinScope
4.2. Gender and financial inclusion: Econometric analysis
In this section, the results of the econometric analysis are presented which is divided into
descriptive statistics and the econometric model results.
4.2.1. Descriptive statistics and Chi-Square test results.
The descriptive statistics confirmed differences between males and females in terms of
access to an account, borrowing, and saving. As reported in Table 3 in Appendix B, the
gender gap in account ownership is significant. When we disaggregate the elements of
access to account ownership it becomes clear that females own fewer bank accounts than
men and the difference is significant. Similarly, formal account ownership is biased towards
men. Comparison of access to borrowing shows that females have equal access to
borrowing. However, females’ access to credit both at a bank or non-bank financial
institution is less than that of men and this is offset by more females accessing credit from
the informal sector. The gender gap is also apparent from differences in access to saving
between males and females. Females have less access to saving than males. They also have
lower access to formal saving and this is not offset by a higher access to the informal
market. In general, comparison of mean values of variables of financial inclusion for males
and females show that females are excluded from the formal market and hence resort to
the informal market as an alternative.
Differences also prevail in the mean value of the predictor variables that include household
size, income, age, marital status, employment status, educational status, and place of
residence. Most females live in households that are larger in size, earn less income, are
younger, are divorced and are unmarried. Most of them are unemployed and have primary
schooling.
4.2.2. Econometric model results
The results of the econometric analysis are presented in this section where the effect of
gender on financial inclusion was analysed while keeping unchanged age, income, marital
status, employment status, place of residence, and level of education. The econometric
model is more robust in telling the true story about the effect of gender on financial

15
inclusion because it determines the relationship between the two while keeping all other
factors that might affect financial inclusion unchanged.
Gender and financial inclusion: gender and account ownership
Our first analysis involved determining the effect of gender on account ownership where
we have the variable ‘have account’ capturing those who have an account at any institution
(i.e. bank, non-bank formal financial institution or informal financial service providers). As
reported in Table 4 in Appendix B, gender does not have an effect on account ownership.
Account ownership is rather affected by age, income, marital status, employment status,
place of residence, and level of education. Age is found to have a positive effect on access.
Income has a significant positive effect on access, and this is consistent with the fact that
mostly the poor are financially excluded. Of the marital status categories, married people
are found to have greater chance of access while no similar effect is observed for those who
are single or divorced. The fact that marriage increases the chance of account ownership
may be explained on the ground that married people have increased financial responsibility
that leads to a higher demand for financial services. Employment is the other important
factor that increases chance of account ownership. Employed people have 1.48 times more
chance of account ownership than unemployed people. A positive and significant
coefficient for urban suggests that account ownership is skewed towards urban dwellers,
and at the flip side it suggests that most rural people are financially excluded. The
coefficients for the level of education are all positive and increase as we move from primary
schooling to post-secondary schooling. This implies that although those who have some
schooling have higher chance of account ownership than those with no schooling, increases
in the level of educational status significantly increases the chance of account ownership.
Analysis of bank account ownership and gender shows that gender enters the model
negatively and significantly which implies that females have a lower chance of account
ownership to a bank account. Bank account ownership is determined by age, income,
marital status, employment status, place of residence, and the level of education. Income
increases the chance of having an account at a bank. Analyses of categories of marital
status show that married and singles have a higher chance of bank account ownership. The
fact that rural people are excluded from the banking sector is revealed through a significant

16
positive coefficient for the variable ‘urban’. The effect of level of education is clear from
positive and significant coefficients that increase as one moves from primary to post-
secondary education implying that increased level of education increases the chance of
having a bank account.
Using formal account ownership (i.e., account at a bank and non-bank financial institution),
we gain a different insight into determinants of access. Formal account ownership, like
bank account ownership, is biased against females. Females are also excluded from the
non-bank financial institutions as can be understood from a negative and significant
coefficient for gender. This implies that factors that impede females’ access to a bank
account also hold when it comes to access to accounts at non-bank financial institutions.
Both married and single females have a higher chance of having a formal account. Those
living in urban areas also have a higher chance of having a formal account which shows that
even the non-bank financial institutions are beyond the reach of people living rural areas.
The importance of education for having formal accounts is evident from significant
coefficients for the three levels of education that increases as we move upwards.
Comparison of the second and third models, i.e., having a bank account and having a
formal account, shows that variables except income that affect ownership of a bank
account also affect ownership of account at non-bank financial institutions suggesting that
these institutions can serve the low income groups of society had it not been for their
location in urban centres.
We also analysed the relationship between gender and informal account ownership, and
the econometric results are insightful. Gender has a statistically significant positive
relationship with access to informal accounts suggesting that females are more likely to use
informal financial service providers. This is consistent with our descriptive analysis which
shows that most females have informal access. Despite a significant relationship with
formal account ownership, age does not have a significant relationship with informal
access. This suggests that the informal sector is accessible to the young as much as it is to
the old. Not surprisingly, increase in income is negatively related to chance of having an
informal account suggesting that the informal financial sector is accessed by low income
groups. While marriage does not have a significant relationship with informal access,
divorced and single females have less chance of having informal access. While employment

17
does not have an effect on informal access, unemployment increases it. A negative and
significant coefficient for the variable ‘Urban’ means living in urban areas decreases the
chance of informal access. While primary schooling increases the chance of informal access,
attaining secondary and post-secondary schooling decreases it. This implies that those who
have primary schooling have both formal and informal access. In general, those that have
informal access can be described as females, low income groups, unemployed, rural
dwellers, those having primary schooling or no schooling at all.
Gender and financial inclusion: gender and access to credit
As reported in Table 5 in Appendix B, gender is negatively and significantly related to
access to credit which suggests that females have lesser access to credit than males. Age,
income, employment status, place of residence, and level of education are the other
important predictors of overall access. Increase in age is related to increased chance of
access to credit and no non-linearity is observed suggesting that people have an increased
chance of accessing credit even at older age. Increase in income is related to higher chance
of accessing credit, employment does not boost the chance of access to credit, while
unemployment decreases it. As is the case with account ownership, access to credit is also
skewed to people living in urban areas. The role of education is evident from the positive
and significant coefficients for the three levels of education, and the chance of accessing
credit increases with increase in the level of education.
As shown in Table 5 in Appendix B, the coefficient for the variable ‘gender’ is negative and
significant implying that females have a lower chance of accessing bank credit. Marital
status affects access to credit in such a manner that the divorced have a lower likelihood of
access to bank credit. Employment is a strong predictor of access to bank credit with the
unemployed having a lower chance of access and the employed having higher chance of
access. Employed people are twice more likely to have access to a bank credit than the
unemployed. Bank credit is available to those living in urban areas suggesting that people
living in rural areas are excluded from bank credit too. The level of education predicts
access to bank credit in such a manner that having primary schooling is not related to
access to bank credit while having secondary or post-secondary schooling increases it.

18
We also look at formal credit to see if non-bank financial institutions offer a different
opportunity to females. As reported in Table 5 in Appendix B, the coefficient for gender is
negatively and statistically significant suggesting that females are equally excluded from
the non-bank credit market. We did not find any statistically significant relationship
between marital status and access to formal credit. Employment status is a strong
predictor of access in such a manner that while unemployment decreases the chance of
access, employment increases it. While those living in urban areas have a higher chance of
accessing formal credit, people in rural areas are less likely to access formal credit. Having
primary schooling is found to have no effect on the chance of accessing formal credit but
having secondary or post-secondary schooling is related to a higher chance of access to
formal credit. Having a post-secondary education increases the chance of access to formal
credit by more than four times. In general, the likelihood of accessing formal credit
decreases for females, low income earners, those living in rural areas, the unemployed, and
those with no schooling.
The model for informal credit shows that access to informal credit is gender neutral. In
other words, there is no significant disparity between females and males in accessing
informal credit. It is also clear from the outputs that age and marital status are not related
to informal access. Not surprisingly, income is negatively and significantly related to
informal access suggesting that increase in income decreases the chance of accessing
informal credit. This is consistent with the outputs of the previous models in which we have
shown income to be an important predictor of access to both bank and non-bank credit.
The role of level of education on the likelihood of access to informal credit is such that only
post-secondary schooling decreases the chance of access while both primary and
secondary schooling have no significant relationship. This suggests that people with a
primary and secondary schooling still rely on the informal market to satisfy their demand
for credit.
Gender and financial inclusion: gender and access to saving
We also look at the relationship between gender and access to saving to see if the disparity
in access between males and females is significant while controlling for other factors. As
presented in Table 6 in Appendix B, gender is negatively and significantly related to access

19
to saving suggesting that females have a lower chance of access to saving. Age and access
to saving are linearly related implying that an increase in age increases ones chance of
access to saving without any threshold. Income is the other important predictor of access
but no similar effect is observed for marital status variables. Employment increases one’s
chance of access to saving while no evidence could be generated for a decrease in the
chance of saving associated with unemployment. Living in urban areas increases the
likelihood of access. Education is found to be an important predictor of access to saving
with the likelihood of saving increasing as the level of education increases.
In the second model, we analyse the role of gender on access to saving at a bank where we
found that gender is negatively and significantly related to bank saving suggesting that
females have a lower chance of saving at a bank. Age increases the chance of saving at a
bank. As expected, income is among the factors that strongly predict access to saving at a
bank. Among the categories for marital status, we find that being single is related to a
lower chance of saving at a bank. As is the case with other dimensions of access, employed
people have a better chance of saving at a bank. People living in urban areas are more likely
to save at a bank than those living in rural areas. Educational level affects saving at a bank
in such a manner that the likelihood of saving at a bank increases for those having
secondary schooling and above. Primary schooling is found to have no relationship with
saving at a bank.
Females also have a smaller chance of saving at non-bank formal institutions evidenced by
a negative and significant coefficient. This suggests that forces that hinder females’ access
to saving at a bank also serve as a barrier to their access to non-bank formal saving. Age is
found to have a positive effect on access to formal saving. Access to formal saving is largely
determined by income. Marriage is not related to access to formal saving but singles and
those divorced are found to have a lower chance of accessing formal saving. While
unemployment decreases the chance of accessing formal saving, employment increases it
significantly. Only secondary and post-secondary schooling are related to higher chance of
accessing formal saving.
Analysis of predictors of informal saving shows that gender is positively and significantly
related to the chance of accessing informal saving suggesting that females often resort to

20
informal providers to satisfy their demand for saving. Surprisingly, increase in household
size is related to a higher chance of saving informally. Increase in personal income is
inversely related to the chance of accessing informal saving. While no relationship is found
between unemployment and informal saving, employment decreases the likelihood of
informal saving. Not surprisingly, living in urban areas decreases the chance of saving
informally which suggests that such form of saving is typically for rural people. The level of
education is related to access to informal saving in such a manner that those having
primary schooling are more likely to access informal saving while secondary schooling and
post-secondary schooling are related to a lower chance of accessing informal saving.
4.3. Barriers to financial inclusion of women
4.3.1. More females use someone else’s account
A reason for females having lower bank account ownership is because some of them use
somebody else’s account. As presented in Figure 11, usage of somebody else’s account
varies across countries with Zambia (25%) at the top and South Africa (4%) at the bottom.
Usage of somebody else’s account is the highest in countries with the lowest bank account
penetration and vice-versa, with the exception of Mauritius and the Democratic Republic of
Congo. Proportionately, more females use someone else’s account in all the countries
except in Malawi and the Democratic Republic of Congo where the split is even. Usage of
someone else’s account is particularly common among women in Zambia, Tanzania, and
Swaziland.
Figure 11: Usage of someone else’s account by gender in each country
Source: FinScope
5
7
7
5
7
4
1 1
1
20
16
8
10
7
6
3 4
1
Zambia Tanzania Mauritius Swaziland Malawi Botswana South Africa Zimbabwe DRC
Women
Men

21
4.3.2. Barriers to account ownership by gender
We also analysed the top three barriers to account ownership and compared them across
gender groups. Lack of money is the top most important reason for not having a bank
account in all the countries. In fact, more females than males cite this as the primary reason
for not having a bank account. For instance, 88 percent of females cite lack of money as the
top most important reason for not having a bank account in Botswana compared to only 56
percent of males. Other factors that act as a barrier against account ownership of females
are described below.
Women are less involved in household financial decision making
Female involvement in household financial decisions is the other important factor
explaining lower level of financial inclusion. As shown in Table 2, proportionately fewer
females are involved in household financial decisions which lead to lower level of demand
for financial services. This was reported as a barrier by females in Mauritius and Zambia.
Table 2: Percentage of adult population making financial decisions alone
Country
Male
Female
Botswana
38.30
32.50
DRC
39.2%
21.5%
Malawi
39.0%
24.3%
Mauritius
32.0%
20.0%
Mozambique
26.0%
24.0%
Source: FinScope
Females have low financial literacy
Lower level of financial literacy among females is the other serious barrier to account
ownership in many countries in the region. The literacy related barriers are mostly either
attitudinal or awareness related. Awareness related barriers include lack of understanding
about benefits of having a bank account, how banks work, the financial language they use
or where and how to apply for a bank account. It also involves attitude related problems
including females having a feeling that bank accounts are not for people like them.
Financial literacy related problems inhibit female ownership of bank accounts in Botswana,
Democratic Republic of Congo, Malawi, Mauritius, Mozambique, Swaziland, Tanzania and
Zimbabwe.

22
More females are not able to maintain minimum balance for opening a bank account
The minimum balance required to open a bank account can act as a barrier especially when
it is relatively high compared to average income of people in lower income categories. This
can also explain lower bank account ownership among females. In Mauritius, Zambia,
Tanzania, and Zimbabwe, females report inability to maintain the minimum balance as a
barrier.
More females cite remoteness of a bank branch
Bank branch penetration often varies from country-to-country and proportionately more
branches are located in urban centres leading to higher level of exclusion of people in the
rural areas. Females in rural areas often cite remoteness of bank branches as a barrier, and
this is the case in Zambia and Tanzania.
More females prefer informal providers to banks
Relatively more females report that they can obtain financial services from elsewhere in the
community. This implies that women have a higher propensity towards using informal
financial services which is consistent with our analysis in an earlier section of the report
that, more females than males are informally served.
Conclusions, recommendations and policy considerations
Access to financial services such as savings, credit and insurance enables individuals to
store value in a safe place, receive and transfer value, and manage liquidity and risk.
However, more females are excluded from financial services and this leads to their
exclusion from social and economic activities. Financial exclusion of females means that
their potential contribution to economic growth is lost. This paper documents the gender
gap in financial inclusion in the SADC region in access to accounts, savings and credit using
FinScope Consumer survey data for ten countries in the region.
Our preliminary analysis shows that the gender gap in bank account ownership is the
highest in Botswana, Swaziland, and Mauritius while South Africa is the only country with a
positive gender gap, i.e. females having more access than males mainly driven by most
females receiving social grants through a SASSA card. However, the gap in the region
narrows significantly when we consider account ownership at non-bank formal financial
institutions implying that these institutions serve as alternative routes when banks are

23
inaccessible. Females have more access to informal finance accounts except in Zimbabwe
where males have more access and in Botswana where both males and females have equal
access.
Most of the preliminary findings were confirmed through comparison of mean values that
shows a significant gender gap in account ownership, saving and credit. Significant
differences were observed between males and females in income, age, employment status,
and level of education.
The results of our econometric analysis shows that gender affects financial inclusion even
after controlling for individual characteristics suggesting existence of gender biases in the
region. Income, level of employment, place of residence and level of education were found
to have a strong effect on financial inclusion. Employed people have better access to bank
accounts, credit and savings irrespective of their gender, level of education and income
which implies that employment per se increases account ownership due to people’s ability
to earn a regular income. Secondary and post-secondary schooling are strongly linked to
better access to financial services compared to primary schooling. Based on the foregoing
conclusions, we forward the following recommendations which in our view would help in
closing the gender gap in financial inclusion in the Region.
 Promoting financial literacy through financial education: Financial education
programs targeted at females will enable them to develop a reasonable understanding
about the language used by banks, benefits of owning a bank account, and how to apply
for it. Such programs should also enable females to develop skills in household financial
management that leads to their empowerment and increased involvement in household
financial decisions.
 Introducing agency banking in rural areas: Agency banking will help females in the
rural area that are excluded from owning a bank account due to remoteness of bank
branches.
 Strengthening informal financial service providers to expand outreach of financial
services to females in rural areas: Accessibility of informal financial service providers
and their ability to design products that suits the needs of individuals makes them ideal
for many females in the region. However, informal operators may not have the technical

24
know-how of managing financial services and may also lack resources to satisfy the
needs of all their clients. Building the financial and managerial capacity of the informal
financiers may allow majority of rural females to obtain quality financial services at a
cheaper cost.
 Mitigating risks in the informal sector: While the informal sector supports female
access to financial services, there are always the risks of exploitation. Safety nets need
to be provided through appropriate consumer protection measures.
 Income generating capability to improve financial inclusion and address gender
disparity: Both the descriptive and econometric analyses point to the fact that financial
inclusion is strongly linked with income generating capability. With women facing wider
exclusion it is important to also address the issue of gender equality in economic
activities.

25
References
Beck, T., & Honohan, P. (2008). Finance for all? Policies and pitfalls in expanding access
(Vol. 41792). World Bank Publications.
Beck, T., Behr, P., & Madestam, A. (2011). Sex and credit: Is there a gender bias in lending?
European Banking Center Discussion Paper, (2011-027).
Benería, L., Berik, G., & Floro, M. (2015). Gender, development and globalization:
economics as if all people mattered. Routledge.
Burgess, R., Pande, R., & Wong, G. (2005). Banking for the poor: Evidence from India.
Journal of the European Economic Association, 3(2‐3), 268-278.
Chibba, M. (2009). Financial inclusion, poverty reduction and the millennium development
goals. European Journal of Development Research, 21(2), 213-230.
Claessens, S., & Feijen, E. (2007). Financial sector development and the millennium
development goals (No. 89). World Bank Publications.
Demirgüç-Kunt, A., Klapper, L. F., & Singer, D. (2013). Financial inclusion and legal
discrimination against women: evidence from developing countries. World Bank Policy
Research Working Paper, (6416).
Demirgüç-Kunt, A., Klapper, L. F., Singer, D., & Van Oudheusden, P. (2015). The Global
Findex Database 2014: measuring financial inclusion around the world. World Bank Policy
Research Working Paper, (7255).
Henderson, L., Herring, C., Horton, H. D., & Thomas, M. (2015). Credit Where Credit is Due?
Race, Gender, and Discrimination in the Credit Scores of Business Startups. The Review of
Black Political Economy, 42(4), 459-479.
Klapper, L., & Singer, D. (2015). The Role of Informal Financial Services in Africa. Journal of
African Economies, 24(suppl 1), i12-i31.
Klasen, S. (2002). Low schooling for girls, slower growth for all? Cross‐country evidence on
the effect of gender inequality in education on economic development. The World Bank
Economic Review, 16(3), 345-373.

26
Klasen, S., & Lamanna, F. (2009). The impact of gender inequality in education and
employment on economic growth: new evidence for a panel of countries. Feminist
economics, 15(3), 91-132.
Klasen, S., & Lamanna, F. (2009). The impact of gender inequality in education and
employment on economic growth: new evidence for a panel of countries. Feminist
economics, 15(3), 91-132.
Lopez-Claros, A & Zahidi, S.(2005) Women’s Empowerment: Measuring the Global Gender
Gap, World Economic Forum.
Muravyev, A., Talavera, O., & Schäfer, D. (2009). Entrepreneurs' gender and financial
constraints: Evidence from international data. Journal of Comparative Economics, 37(2),
270-286.
OECD (2012) Women’s economic empowerment, The OECD DAC Network on Gender
Equality (GENDERNET), Organization for Economic Cooperation and Development.
Pitt, M. M., & Khandker, S. R. (1998). The Impact of Group‐Based Credit Programs on Poor
Households in Bangladesh: Does the Gender of Participants Matter? Journal of political
economy, 106(5), 958-996.
Pitt, M. M., Khandker, S. R., & Cartwright, J. (2006). Empowering women with micro
finance: Evidence from Bangladesh. Economic Development and Cultural Change, 54(4),
791-831.
Pitt, M. M., Khandker, S. R., Chowdhury, O. H., & Millimet, D. L. (2003). Credit programs for
the poor and the health status of children in rural Bangladesh*. International Economic
Review, 44(1), 87-118.
Presbitero, A. F., Rabellotti, R., & Piras, C. (2014). Barking up the Wrong Tree? Measuring
Gender Gaps in Firm’s Access to Finance. The Journal of Development Studies, 50(10),
1430-1444.

27
Ramji, M. (2009). Financial inclusion in Gulbarga: Finding usage in access. Institute for
Financial Management and Research, Chennai, India. http://www. centre-for-microfinance.
org/wp-content/uploads/attachments/csy/1901/26_Ramji_Financial, 20.
Rawlings, L. B., & Rubio, G. M. (2005). Evaluating the impact of conditional cash transfer
programs. The World Bank Research Observer, 20(1), 29-55.
Seguino, S. (2000). Accounting for gender in Asian economic growth. Feminist Economics,
6(3), 27-58.
Swamy, V. (2014). Financial inclusion, gender dimension, and economic impact on poor
households. World Development, 56, 1-15.
UN (2014) Gender equality and sustainable development, World Survey on the Role of
Women in Development 2014, United Nations.
UNCDF (2016) Lost in the mail: Why bank account access is not translating into usage,
accessed on May 31,2016, URL:
http://www.uncdf.org/sites/default/files//Documents/insight_note_04.pdf
Appendix A
The binary logistic regression model used in which the probability of financial inclusion is
described by the following function:
   
 or   

 …………………………………………….(1)
Where
is the probability the ith person is financially included.
 is the value of the unobserved variable for the ith person.
The logistic regression model assumes that  is linearly related to the predictors
           ………………………………………(2)
Where

28
 is the jth predictor for the ith person that include gender, household size, income, age, place of residence,
level of education, marital status, and employment status.

 is the jth coefficient
 is the number of predictors
The dependent variable in each model is dichotomous taking the values 1 or 0. In the first
model, respondents who have a bank account are assigned the value 1 and 0 otherwise. In
the second model, those who save at a financial institution are assigned 1 and 0 otherwise.
In the third model, those who accessed credit are assigned 1 and 0 otherwise.
Appendix B
Table 3: Summary statistics of independent variables by gender
The variables Household size and age are found to be non-normally distributed but with
equal variance for men and women. Therefore, we used Mann-Whitney U Test, a non-
parametric test, to check difference in the mean value of the variables for the two groups.
For categorical variables, we used Chi-square test and statistical significance is reported
using asterisks in the last column of the table.
Variable
Description
Female
Male
Significance
Have Account
Have an account
0.66
0.7
****
Account@Bank
Have a bank account
0.33
0.4
***
Account@Formal
Have account at a bank or non-
bank financial institution
0.52
0.59
***
Account@Informal
Have account at informal financial
service provider
0.14
0.11
***
Borrowed
Borrowed from somewhere
0.36
0.36
-
Bank Credit
Borrowed from a bank
0.08
0.12
***
Formal Credit
Borrowed from a bank or non-
bank financial institution
0.16
0.2
***
Informal Credit
Borrowed from an informal
financial service provider
0.18
0.14
***
Saved
Saved somewhere
0.52
0.56
***
Saved@Bank
Saved at a bank
0.19
0.24
***
Saved@Formal
Saved at a bank or non-bank
financial institutions
0.24
0.31
***

29
Saved@Informal
Saved at an informal financial
service provider
0.26
0.24
***
Household size
Household size
4.64
4.48
***
Income
Income of the respondent
1.39
1.6
***
Age
Age of the respondent
38.04
39.78
***
Divorced
A person is divorced
0.07
0.03
***
Married
A person is married
0.63
0.67
***
Single
A person never married
0.28
0.37
***
Unemployed
A person is jobless
0.63
0.55
***
Employed
A person is employed
0.34
0.42
***
Urban
A person lives in urban area
0.43
0.4
***
Primary
A person has primary education
0.4
0.34
***
Secondary
A person has secondary education
0.4
0.46
***
Post-secondary
A person has post-secondary
education
0.08
0.12
***
Note: *** significant at 1% level,
Table 4: Logistic regression output: gender and account ownership
Variables
Have account
Have bank
account
Have formal
account
Have informal
account
Gender
-0.01
-0.151***
-0.192***
0.266***
HHSize
-0.005
.000
-0.005
0.0050
Age
0.064***
0.069***
0.063***
0.0100
Age squared
-0.001***
-0.001***
-0.001***
0.0000
Personal Monthly Income
0.624***
0.974***
0.000
-0.355***
Divorced
-0.039
0.008
0.099
-0.23**
Married
0.175***
0.196***
0.136***
0.016
Single
-0.036
0.205***
0.211***
-0.384***
Unemployed
-0.112
-0.186**
-0.202**
0.226*
Employed
0.39***
0.557***
0.355***
0.1
Urban
0.770***
0.725***
1.055***
-0.375***
Primary schooling
0.416***
0.422***
0.485***
0.1200*
Secondary schooling
1.039***
1.301***
1.328***
-0.1400**
Post-secondary schooling
2.131***
2.759***
2.723***
-1.4700***
Constant
-2.006***
-5.564***
-2.929***
-2.085***
Country fixed effect
YES
YES
YES
YES
N
23,825
23,825
23,825
23,825

30
Variables
Have account
Have bank
account
Have formal
account
Have informal
account
Cox & Snell R2
0.184
0.281
0.286
0.057
Negelkerke R2
0.253
0.407
0.381
0.098
Note: *** significant at 1% level, ** significant at 5% level, *** significant at 10% level.
Binary logistic regression estimation coefficients are reported.
Table 5: Logistic regression output: gender and access to credit
Variables
Have
credit
Have bank
credit
Have formal
credit
Have informal
credit
Gender
-0.132**
-0.234**
-0.215***
0.05
HHSize
-0.002
0.017
0.025
0.000
Age
0.04***
0.089***
0.085***
0.005
Age squared
0.000
-0.001***
-0.001***
0.000
Personal monthly income
0.368***
1.247
0.856***
-0.123*
Divorced
-0.035
-0.36***
-0.222
0.03
Married
0.083
0.217
0.101
-0.024
Single
-0.089
0.125
-0.013
-0.12
Unemployed
-0.289***
-0.442*
-0.541***
-0.058
Employed
0.027
0.733***
0.372***
-0.383***
Urban
0.183***
0.607***
0.313***
0.002
Primary schooling
0.286**
0.034
0.268
0.151
Secondary schooling
0.374***
0.527**
0.640***
0.111
Post-secondary schooling
0.786***
1.445***
1.547***
-0.625***
Country fixed effect
YES
YES
YES
YES
Constant
-1.848
-8.683***
-5.733***
-.660**
Country fixed effect
YES
YES
YES
YES
N
11,569
13,142
13,142
11,569
Cox & Snell R2
0.373
0.104
0.123
0.264
Negelkerke R2
0.530
0.354
0.300
0.409
Note: *** significant at 1% level, ** significant at 5% level, *** significant at 10% level.
Binary logistic regression estimation coefficients are reported.

31
Table 6: Logistic regression output: gender and access to savings
Variables
Have a
saving
Saving at a
bank
Have formal
saving
Have informal
saving
Gender
-0.082*
-0.215***
-0.266***
0.118***
HHSize
-0.003
0.012
0.001
0.017**
Age
0.042***
0.05***
0.06***
-0.001
Age Squared
0.0000
0.0000
0.0000
0.0000
Personal monthly income
0.547***
0.978***
0.999***
-0.068**
Divorced
0.014
-0.17
-0.271*
0.108
Married
0.058
-0.014
-0.088
0.023
Single
-0.113
-0.198*
-0.199**
-0.057
Unemployed
-0.003
-0.034
-0.241**
0.086
Employed
0.399***
0.582***
0.554***
-0.174**
Urban
0.229***
0.491***
0.464***
-0.149***
Primary schooling
0.338***
0.16
0.168
0.259***
Secondary schooling
0.593***
0.889***
0.936***
-0.227***
Post-secondary schooling
0.940***
1.759***
1.961***
-0.55***
Country fixed effect
YES
YES
YES
YES
Constant
-2.33***
-6.134***
-5.201***
-1.047***
Country fixed effect
YES
YES
YES
YES
N
12,196
13,142
13,142
12,196
Cox & Snell R2
.250
0.179
0.247
0.065
Negelkerke R2
.336
0.327
0.391
0.089
Note: *** significant at 1% level, ** significant at 5% level, *** significant at 10% level.
Binary logistic regression estimation coefficients are reported.