Feminist Economics
ISSN: 1354-5701 (Print) 1466-4372 (Online) Journal homepage: https://www.tandfonline.com/loi/rfec20
Gendered Paths to Asset Accumulation? Markets,
Savings, and Credit in Developing Countries
Cheryl R. Doss, Carmen Diana Deere, Abena D. Oduro, Hema Swaminathan,
Zachary Catanzarite & J. Y. Suchitra
To cite this article: Cheryl R. Doss, Carmen Diana Deere, Abena D. Oduro, Hema Swaminathan,
Zachary Catanzarite & J. Y. Suchitra (2019) Gendered Paths to Asset Accumulation?
Markets, Savings, and Credit in Developing Countries, Feminist Economics, 25:2, 36-66, DOI:
10.1080/13545701.2019.1566753
To link to this article:  https://doi.org/10.1080/13545701.2019.1566753
Published online: 04 Feb 2019.
Submit your article to this journal 
Article views: 140
View Crossmark data
Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=rfec20
Feminist Economics, 2019
Vol. 25, No. 2, 36–66, https://doi.org/10.1080/13545701.2019.1566753
GENDERED PATHS TO ASSET ACCUMULATION?
MARKETS, SAVINGS, AND CREDIT IN DEVELOPING
COUNTRIES
Cheryl R. Doss, Carmen Diana Deere, Abena D. Oduro, Hema
Swaminathan, Zachary Catanzarite, and J. Y. Suchitra
ABSTRACT
An extensive literature shows how property inheritance is biased against women
in many developing countries, yet relatively little attention has been given to
gender bias in other means of acquiring physical assets, such as the market.
Using individual-level data from Ecuador, Ghana, and Karnataka, India, this
study analyzes modes of acquisition and financing of housing, agricultural land,
other real estate, and businesses. The findings show that women acquire fewer
of their assets through the market than men, and that in asset markets, both
men and women are more likely to use their own savings than to use credit. The
study also analyzes current loans for asset acquisition and finds that, in general,
women tend to be somewhat disadvantaged in securing formal bank loans. The
results suggest that financial inclusion to promote more gender equal access to
accumulation of assets should focus on both savings and credit, with priority to
savings.
KEYWORDS
Asset markets, inheritance, Ecuador, Ghana, India
JEL Codes: E21, J16, O57
INTRODUCTION
A consensus is emerging among academics and policymakers that women’s
empowerment is associated with their ownership of property. Increasing
poor women’s property ownership may also provide them and their families
a pathway out of poverty. The evidence suggests that women are less likely
than men to own any form of immovable property. As development efforts
aim to increase women’s property ownership, it is important to step back
and examine how women accumulate property and whether the modes of
acquisition differ for women and men. This would assist in understanding
the constraints that women face in acquiring assets and developing ways to
bridge the gender asset gap.
© 2019 IAFFE
GENDERED PATHS TO ASSET ACCUMULATION
Individuals acquire assets through inheritance, state and/or community
transfers of property, and markets. Much of the focus on women’s asset
acquisition in developing countries, particularly of land, has been on the
gendered patterns of inheritance. Some attention has also been given
to gender biases in state programs of land or housing distribution or in
the privatization of community lands. Much less attention has been given
to men’s and women’s participation in asset markets, and the extent to
which such participation is financed through savings and/or credit. As
asset markets grow in importance, it is critical to understand their gender
dimensions and, particularly, whether there are differences in how men
and women finance their asset acquisitions.
Using detailed asset ownership data for individuals, we analyze how
women and men acquire their immovable property and businesses in
Ecuador, Ghana, and Karnataka, India.1 These assets constitute the bulk of
household physical or non-financial gross wealth: 90 percent in Ecuador,
82 percent in Ghana, and 93 percent in Karnataka.
The three countries represent very different patterns of urbanization
and market development and have different legal norms, all of which
have crosscutting impacts on the potential for asset accumulation by
both women and men. Ecuador, a high-middle-income country, is more
urban than either Ghana or India, both low-middle-income countries. In
Ecuador, agricultural land is relatively less important in the aggregate
wealth portfolio than in the other two countries, and the real estate market
in both rural and urban areas is fairly well developed. In contrast, in
Ghana, much of the land and housing in rural areas continues to be under
customary tenure, with kinship and community playing an important role
in access to property. In India, agricultural land is the most important asset
of the majority of the population, which is still rural. Land is more often
transferred intergenerationally than by any other means, and land markets
are not well developed. Overall, inheritance is a much more important
means of acquiring land and other immovable property in Ghana and India
than in Ecuador.
The three countries also differ with respect to their marital regimes, with
important implications for women’s property ownership. Ghana and India
are characterized by the separation of property marital regime, whereby
property acquired within marriage is considered the exclusive property of
the individual who acquired it or in whose name it is legally registered.
Ecuador, by contrast, has the more gender-inclusive partial community
property regime, whereby any property bought during marriage is the joint
property of both spouses. Not surprisingly, the gender asset gap among
couples is minimal in Ecuador compared to the other two countries where
it looms large (Deere et al. 2013).
Irrespective of the degree of development of asset markets, there are
differences in the patterns of modes of acquisition of property by gender.
37
ARTICLE
In each of the three countries, women acquire fewer of their assets through
the market than men. For both men and women who acquire immovable
property and businesses through the market, the use of savings is the most
common way of financing the acquisition. Relatively few of these assets have
been purchased using credit.
In addition to this retrospective analysis of how assets have been acquired,
we consider current loans that have been utilized for asset acquisition in
Ecuador and Ghana. Loans for this purpose are more readily available in
Ecuador than in Ghana, but in the former there is a large gender gap in
favor of men. Further, while in Ecuador the great majority of the loans
for asset acquisition are from formal sources, in Ghana, informal sources
predominate. In both countries, private bank credit favors men.
As markets for immovable property become increasingly important in
developing countries and as the opportunities for opening and expanding
businesses increase, it is critical that women be able to acquire assets on
the same terms as men. While strengthening women’s inheritance rights
and ensuring that these are honored in practice will always be important,
it is necessary to also ensure that women have access to formal financial
institutions, including savings accounts, to acquire assets through the
market.
PATTERNS OF ASSET ACQUISITION
Assets may be acquired through inheritance, the market, or state or
community transfers, but the two most common means of asset acquisition
are inheritance and purchase. A long-standing debate in the literature
on wealth accumulation in developed countries regards the extent to
which wealth is accumulated primarily through inheritance rather than self-
acquired – via savings and themarket – and the impact on wealth inequality.
This debate concludes that inherited wealth is more concentrated than the
distribution of wealth as a whole (Davies and Shorrocks 2000). Estimates
of the share of households who have ever received an inheritance vary
widely: from only 4.4 percent in the United Kingdom and 12.0 percent in
the United States, to over 40 percent in Sweden, Belgium, and Switzerland
(Semyonov and Lewin-Epstein 2013: 1138).
Similar analyses have yet to be done for developing countries, so we
do not know much regarding the relative importance of these sources of
wealth and how they might change in the process of development. The
variation in the incidence of inheritance in developed countries, however,
suggests that this process is certainly heterogeneous.
In addition, very little is known about how the broad patterns of asset
acquisition differ for men and women, although there has been substantive
research on particular aspects. Women often face discrimination in
inheritance, with sons favored over daughters in many regions of the world;
38
GENDERED PATHS TO ASSET ACCUMULATION
but women may also face constraints in acquiring assets through markets.
The extent to which these constraints affect the overall patterns of asset
acquisition for men and women has not been studied systematically (Deere
and Doss 2006).
An extensive literature focuses on the discrimination against women in
inheritance law and practice regarding immovable property, specifically
agricultural land and housing.2 This issue was recognized in the
international arena through the 1979 UN Convention on the Elimination
of All Forms of Discrimination against Women (CEDAW), and since then,
many countries have passed legislation strengthening women’s inheritance
rights, especially under intestate succession. These initiatives, however,
have had limited success in changing actual inheritance practices.
For example, the Hindu Succession Act (1956) in India was amended in
2005 to correct gender inequalities in inheritance by sons and daughters.
This law could potentially benefit many Indian women, but has faced
implementation challenges, such as poor understanding of the legislation,
communities not recognizing women as landowners, and women
themselves being hesitant to assert their rights (Sircar and Pal 2014).
In Ghana, the Wills Act of 1971 did not contain explicit provisions for
inheritance by sons and daughters. The intent of the Intestate Succession
Act of 1985 was to protect the interests of surviving spouses, and while it
provides for children to inherit a portion of the estate of the deceased
parent, there is no provision for sons and daughters to inherit equally. It
too has faced implementation challenges (Fenrich and Higgins 2001).
The increasing formalization of property rights and the growing
commoditization of land and other property have brought new challenges
for strengthening women’s property rights. State titling programs have
often resulted in land and housing being titled only in the name of
the male head of household with women losing many of their claims
(Lastarria-Cornhiel 1997). However, a growing number of countries are
now recognizing the importance of joint titling of immovable property in
the names of both spouses in state programs (Deere and León 2001; Datta
2006; Ali, Deininger, and Goldstein 2014). The formalization of property
rights is often designed explicitly to allow markets to play a greater role in
allocating immovable property, particularly over time.
It cannot be assumed, however, that asset markets will be gender
neutral, since participation in these depends on the purchasing power of
individuals. If women have access to fewer financial resources than men,
whether through savings or access to credit, they will be less likely to
participate in asset markets and acquire assets in their own name. The
ability to save is related to labor force participation and earnings. While
women’s labor force participation has been increasing in some world
regions, most notably Latin America and the Caribbean, between 1990 and
2013 it fell in Central and Eastern Europe, Central Asia, and South Asia
39
ARTICLE
(UN Women 2015). Moreover, in countries such as India, women’s labor
force participation is concentrated in nonremunerated family labor, where
they do not earn income directly. When they do earn income directly,
women’s wages are systematically lower (Elson 1999). Further, they do not
always control the income they earn.3
Women’s ability to purchase assets may also be limited by social norms.
Susana Lastarria-Cornhiel et al. (2014) note that women have less access to
land sales and rental markets not only because of financial constraints, but
also because husbands may be reluctant for their wives to purchase land
on their own. A woman who seeks to purchase property on her own may
be perceived as a bad wife or one who is planning to leave her husband.
Sellers of agricultural land may discriminate against women if women are
not socially recognized as farmers. This was the case in northern Peru in
the 1950s and 1960s, where some landowners would not sell land parcels to
women or charged them higher prices (Deere 1990).
Comparing the modes of acquisition of land among landowners from
surveys in four Latin American countries, Deere and León (2001) find
that although men are much more likely than women to have inherited
land, the most frequent way that women landowners acquired their land
was through inheritance. A much smaller share of women landowners than
men landowners purchased their land, implicitly suggesting that women
face greater challenges in land markets.
In much of Africa, women do not traditionally inherit land. They may
acquire use rights to land through their husbands or fathers, but not
ownership rights. Thus, the market may provide an important option
for women to acquire land. In Uganda, a country with relatively well-
developed land markets, an analysis in three regions of the country finds
that women are more likely to purchase land than inherit it; 30 percent of
women purchased land while 23 percent acquired it through inheritance
or gifts. A gender gap prevails nevertheless given that 43 percent of male
respondents had purchased land, and 41 percent had inherited it (Doss
et al. 2012b).4 Thus, even where land markets are active, women may face
more constraints than men in participating in this market.
Regarding housing, Greta Friedemann-Sánchez (2006), in a study
of floriculture workers in Colombia, found few gender differences in
inheritance: male and female workers were similarly likely to have inherited
a housing lot or dwelling. Among those who had purchased these assets,
however, men were significantly more likely than women to have used their
own savings. Women used a larger number of financial sources, relied more
upon loans, and worked for pay for more years to acquire their property.
Friedemann-Sánchez attributes these differences to cultural norms whereby
women have greater family obligations than do men, and therefore save
less. The main source of credit for these paid workers was an employee
credit fund, with equal access for men and women.
40
GENDERED PATHS TO ASSET ACCUMULATION
The literature on business acquisition focuses more on the use of
credit, rather than comparing the roles of inheritance and markets,
perhaps because the majority of small businesses in developing countries
are primarily forms of self-employment. A World Bank (2010) study of
entrepreneurship in Latin America found that personal savings are the
main source of funding to start a business, for both small and larger
businesses, whether owned by women or men.5 However, women are more
likely than men to use their own savings as start-up capital. Among the
reasons reported for women’s lower use of credit was their inability to
provide collateral if major household assets are registered only in the name
of the male household head.
In their review of the literature, Leora F. Klapper and Simon C. Parker
(2011) find that women entrepreneurs have more limited access to capital
for their businesses and are less likely to rely on formal financial institutions
than men. Some of the studies they analyze from Latin America have
found that the main gender difference is on the demand side, rather than
outright discrimination, since women are less likely than men to apply for
formal financing to maintain or expand their business. On the other hand,
they report on studies from a number of developing countries that have
documented the many barriers women face in acquiring formal business
loans, from low financial literacy to the inability to provide collateral or a
guarantor for a loan.
The rapid expansion of microfinance programs over the past several
decades has undoubtedly expanded women’s access to credit. The impact
of these programs, particularly group lending in Asia and Africa, on poverty
reduction and/or women’s empowerment have been amply debated.6
Although these outcomes may depend on whether loan recipients use
their loans to accumulate assets, relatively few studies have focused on
this issue, particularly from a gender perspective. For example, Carina
Van Rooyen, Ruth Stewart, and Thea De Wet’s (2012) systematic review
of impact studies of microfinance programs in Sub-Saharan Africa found
only thirteen studies of sufficient quality to measure the causal impact of
credit or credit plus savings interventions on outcomes.7 Of these, only five
measured the impact on asset accumulation. While all five of these studies
utilized samples that included both men and women, only two included
an analysis by gender of the loan recipient, and only one identified which
household member owned the asset acquired with the credit.
Neither of the two that included a gender analysis found gender
differences in asset accumulation as a result of the microfinance program.
Carolyn Barnes, Gary Gaile, and Richard Kibombo’s (2001) longitudinal
study of microcredit programs aimed at microentrepreneurs in Uganda
found that program clients significantly increased the amount spent on
acquiring business assets compared to non-clients, with no significant
difference by gender of the loan recipient. Moreover, client households
41
ARTICLE
were more likely than those of non-clients to become homeowners, and
the value of all other household durables was similarly higher for clients,
with again no difference by gender of the loan recipient. In a study of rural
Zanzibar, Tanzania, Conner Brannen (2010) found that the households
of both male and female members of village savings and loan associations
were more likely to be homeowners than those of the control group (made
up of new members). In addition, prior year expenditure on durable
assets was significantly greater among member households than the control
group, but the impact was smaller in households where the woman was the
borrower. The study, however, did not ask to whom the assets acquired
belonged. In the other high quality studies on the impact of microcredit
on asset acquisition, any gendered impact must be deduced from the type
of asset acquired or the composition of the membership, which is clearly
insufficient for a rigorous gender analysis.
Among more recent studies on Africa, Namizata Binaté Fofana
et al. (2015) focus on both household asset acquisition and women’s
empowerment in rural Côte d’Ivoire. They found that womenmicrofinance
borrowers had higher incomes and a higher value of household assets than
the control group. However, the positive effect was with respect to durable
goods that might be owned and used by all household members – a bed,
chair, radio, and hoe. They found no effect on what they call “women’s
assets,” such as clothing and jewelry.
The growing recognition that microcredit is not a panacea for either
poverty reduction or women’s empowerment has led to a greater focus
on how individuals may be savings constrained. The constraints include
banking regulations and charges regarding savings accounts (Beck,
Demirgüç-Kunt, and Soledad Martinez Peria 2008; Dupas and Robinson
2013), as well as behavioral explanations (Ashraf, Karlan, and Yin 2010; Kast
and Pomeranz 2014). Overall, gender analyses of the relationship between
savings and asset accumulation are still pretty sparse in the literature.8
This paper contributes to filling the lacunae in the literature about how
men and women acquire assets throughmarkets. Although the incidence of
purchasing immovable property is relatively low in many countries, growing
markets for immovable property provide an opportunity for women to
acquire property outside of slowly changing inheritance systems. However,
participation in asset markets requires access to savings and/or credit, and
surprisingly little is known about their relative roles in asset acquisition or
the extent to which this relationship is gendered.
DATA AND METHODOLOGY
In order to collect individual asset ownership data, the Gender Asset
Gap project in 2010 fielded nationally representative household surveys in
Ecuador and Ghana, and household surveys at the state level in Karnataka,
42
GENDERED PATHS TO ASSET ACCUMULATION
India. The surveys utilized a two-stage sampling technique, which resulted
in a sample of 2,892 households in Ecuador; 2,170 in Ghana; and 4,110 in
India.
The surveys employed two instruments, a household and an individual
questionnaire. The first instrument collected data on household
demographics, livelihoods, and an inventory of physical assets owned by
members of the household, identifying the owner of each asset. The
individual questionnaire was administered separately to a maximum of two
adult members of the household (usually the man and women considered
most knowledgeable) and collected further information on ownership of
physical assets as well the data on ownership of financial assets and debt.
Although the questionnaires were based on a similar template, they were
adapted to the local context in each country following extensive qualitative
research. All inquired about the method of acquisition of each of the major
assets, but the placement of the questions in the instruments differed.9 In
addition, for Ecuador and Ghana, the survey includes detailed information
on each outstanding loan held by the respondent and identifies the
source and purpose of each loan.10 We can identify the person or
persons responsible for repaying each loan; whether the loan was used
to acquire or enhance immovable property or businesses or to purchase
livestock, agricultural equipment, and consumer durables; and whether it
was obtained through formal or informal sources.
Our primary aim in this paper is to establish whether there are broad
patterns in how assets are acquired across countries, whether there are
any systematic gender differences in the modes of asset acquisition, and
whether this depends on the type of asset, specifically, immovable property
versus businesses. Once we establish whether there are gender differences
in participation in asset markets, we analyze whether there are systematic
differences by gender in the use of savings versus credit to finance these
purchases and by type of asset.
Our analysis relies upon descriptive statistics to establish patterns. What is
relatively unique about our data sets is that we have gender-disaggregated
data on each asset that is currently owned and how it was acquired. Our
contribution is thus to examine a new question considering multiple assets
in a comparative framework.
Data limitations do not allow us to answer the interesting question of
what distinguishes women (or men) who were able to purchase assets
from those who were not. While we collected information on the year that
the various assets were acquired (and can thus deduce the age at which
a person acquired the asset), we did not collect information on marital
status, education, locale, and employment at the time of purchase. Many
of these, particularly marital status and locale, may have changed. Since we
are dealing with retrospective data, we are hesitant to carry out an analysis
even by current rural/urban locale of residence, since all three of our
43
ARTICLE
countries, but particularly Ecuador and Ghana, have been characterized
by a high degree of internal and international migration over the past
several decades. In such contexts, even ethnicity may be fluid, precluding a
rigorous analysis of intersectionality.
An example from our qualitative research may best illustrate our
concern for why it might be inappropriate to consider a person’s current
characteristics adequate predictors of their likelihood to purchase an asset.
In urban Ecuador, we were surprised by the number of women who
had themselves purchased either the housing lot upon which they had
subsequently built a dwelling with their partner, or who owned a second
urban dwelling. The focus group discussions and interviews indicated that
some of these purchases had been facilitated by their inheritance of a land
parcel in their community of origin, which they subsequently sold. In the
survey, this second urban dwelling would accurately appear as purchased
with the use of savings, but her individual characteristics at the time of
the survey – urban, mestizo (mixed-race background), married, perhaps
a housewife – might differ greatly from those at the time of the purchase
– for example, rural, indigenous, single, and an unpaid worker on a family
farm.
PATTERNS OF ASSET OWNERSHIP
Before considering the patterns of asset acquisition, it is useful to have
a sense of the gendered patterns of ownership in these three countries.
Figure 1 presents the incidence of current asset ownership among adult
(18 years and over) men and women for the four key assets.11 A person is
considered an owner regardless of whether they own the asset individually
Figure 1 Incidence of ownership of key assets by adults, by sex and country
44
GENDERED PATHS TO ASSET ACCUMULATION
or jointly. The most commonly owned asset, by either men or women, is
the principal residence. The one exception is that in Ghana the asset most
often owned by women is a business, although many of these are very small.
In Karnataka, men are also likely to own agricultural land, and there is
a large gender gap. In Ghana, much of the agricultural land is not owned
by individuals but is held under customary tenure. Considering only land
that is private property, there is also a large gender gap. In Ecuador, which
is much more urbanized than the other two countries, only 7 percent
of individuals nationally are landowners, but there is no gender gap in
ownership rates. Relatively few people own other real estate in any of the
three countries, with the highest incidence being among men in Ghana.
Finally, businesses exhibit a different pattern than immovable property.
In both Ghana and Ecuador, a larger share of women than men own
businesses. A gender gap in favor of men persists in Karnataka.
ASSET ACQUISITION IN ECUADOR, GHANA, AND
KARNATAKA
The overall pattern of acquisition by type of asset is depicted in Figure 2. In
the case of the main residence, in both Ecuador and Ghana the majority are
acquired through the market, either through purchase or by construction
of the dwelling.12 In contrast, the majority of residences in Karnataka are
inherited. The “other” category refers primarily to government programs
or community transfers.
Figure 2 Mode of acquisition of immovable property and businesses (percent of
currently owned assets acquired through purchase, inheritance, and other means)
45
ARTICLE
In all three countries, the most common way that agricultural land
is acquired is through inheritance or gifts.13 Ecuador shows the highest
incidence of land acquired through the market. In both Ghana and
Karnataka, a relatively small share of land parcels is acquired through the
market. In Ghana, “other” forms have been important such as claiming
ownership of community (lineage) lands.
The acquisition of other real estate in Ecuador and Ghana tends to
follow the same pattern as the dwelling, with the most common mode of
acquisition being purchase and/or construction. In contrast, in Karnataka,
other real estate is more frequently inherited than purchased.
Businesses show a markedly different pattern when compared to
immovable property: in all three countries, most businesses are acquired
through purchase or founded through self-efforts, with inheritance being
of relatively minor importance. This largely reflects the importance of
self-employment in developing countries.
In broad strokes, this comparative analysis suggests that the relative
importance of inheritance in acquiring immovable property is associated
with the degree of rurality and the breadth and depth of asset markets. As
countries urbanize, inheritance of land becomes relatively less important
and the acquisition of dwellings through the market relatively more
important. However, even in Karnataka, which is the most rural of the three
sites and where inheritance dominates in the acquisition of property, the
market plays at least some role in the acquisition of the residence and other
real estate, while less so in terms of agricultural land.
In order to consider the gendered patterns of asset acquisition, we
consider the mode of acquisition by the form of ownership and by country
(Tables 1–3). We categorize four types of ownership: owned by a man
individually, by a woman individually, jointly by a couple, and “other”
joint ownership. The last category includes assets jointly owned by two or
more household members other than the main couple, as well as assets
owned jointly by a household member with someone residing outside the
household.14
The dominant form of ownership differs strikingly across the three
countries (note the number of assets owned in Tables 1–3). To a large
extent, this reflects differences in the marital property regime. In Ecuador,
most agricultural parcels, main residences, and other real estate are owned
jointly by a couple, reflecting its partial community property regime.
The exception to this pattern is businesses, where individual ownership
dominates. In Ghana and India, the separation of property marital regime
does not confer joint property rights to individuals upon marriage, thus
most property is owned individually, with a gender gap in favor of men for
all forms of property except businesses in Ghana.
Most notable, is that the patterns of acquisition vary not only across assets,
but also depending on whether the assets are owned by men, women, or
46
GENDERED PATHS TO ASSET ACCUMULATION
47
Table 1 Incidence of modes of acquisition, by form of ownership, Ecuador
% of purchasers using:
Form of Owned % purchased/ Savings
ownership assets (n) constructed % inherited % other Purchased (n) Any savings Any credit and credit
Main residence
Man 253 84.6 13.0 2.4 214 85 32.7 18.2
Woman 429 84.2 14.9 0.9 361 84.2 29.6 16.6
Couple 729 93.7 4.4 1.9 683 80.2 39.1 21.2
Other 316 85.4 13.9 0.6 270 85.9 30 17.4
Total 1727 88.5 10.0 1.5 1528 82.9 34.4 19
p 0.000 0.000 0.192 0.099 0.055 0.256
Agricultural land
Man 137 37.2 58.4 4.4 51 94.1 5.9 0
Woman 125 26.4 72.8 0.8 33 93.9 18.2 12.1
Couple 174 52.9 44.3 2.9 92 78.3 28.3 6.5
Other 78 52.6 46.2 1.3 41 80.5 7.3 0
Total 514 42.2 55.3 2.5 217 84.8 17.5 4.6
p 0.000 0.000 0.262 0.027 0.002 0.024
(Continued).
ARTICLE
48
Table 1 Continued.
% of purchasers using:
Form of Owned % purchased/ Savings
ownership assets (n) constructed % inherited % other Purchased (n) Any savings Any credit and credit
Other real estate
Man 91 58.2 38.5 3.3 53 94.3 20.8 15.1
Woman 122 51.6 45.9 2.5 63 79.4 33.3 17.5
Couple 121 86.8 13.2 0.0 105 73.3 40 13.3
Other 74 51.4 48.7 1.4 38 89.5 28.9 21.1
Total 408 63.5 35.1 1.7 259 81.5 32.8 15.8
p 0.000 0.000 0.270 0.003 0.101 0.702
Business
Man 505 90.1 2.0 8.7 455 83.1 24.2 7.7
Woman 766 85.1 3.5 12.4 652 85 20.2 5.4
Couple 220 97.3 0.9 1.8 214 72.9 37.9 11.7
Other 258 86.1 4.7 9.7 222 84.7 22.1 7.2
Total 1749 88.2 0.7 9.6 1543 82.7 24.1 7.2
p 0.000 0.037 0.000 0.000 0.000 0.002
Notes: Respondents could indicate more than one form of acquisition. Statistical significance tests refer to Pearson Chi-sq test across all forms of ownership.
GENDERED PATHS TO ASSET ACCUMULATION
49
Table 2 Incidence of modes of acquisition, by form of ownership, Ghana
% of purchasers using:
Form of Owned % purchased/ Savings
ownership assets (n) constructed % inherited % other Purchased (n) Any savings Any credit and credit
Main residence
Man 491 84.3 15.5 0.2 414 93.5 2.2 0.7
Woman 173 60.2 39.3 0.6 104 80.8 2.8 1
Couple 41 97.6 2.4 0.0 40 92.5 2.5 2.5
Other 64 14.3 85.7 0.0 10 70 0 0
Total 769 73.8 25.9 0.3 568 90.7 2.3 0.9
p 0.000 0.000 0.000 0.934 0.702
Agricultural land
Man 937 6.7 53.3 40.0 65 89.2 6.2 0
Woman 299 7.0 76.3 16.7 23 78.3 4.4 0
Couple 38 18.4 60.5 21.1 8 100 0 0
Other 93 3.2 71.0 25.8 3 66.7 0 0
Total 1367 6.9 59.7 33.4 99 86.9 5.1 0
p 0.020 0.000 0.000 0.251 0.856 NA
(Continued).
ARTICLE
50
Table 2 Continued.
% of purchasers using:
Form of Owned % purchased/ Savings
ownership assets (n) constructed % inherited % other Purchased (n) Any savings Any credit and credit
Other real estate
Man 451 78.1 18.9 3.1 353 0.22 1.55 0
Woman 237 59.1 37.6 3.4 140 0.42 2.11 0
Couple 25 76.0 24.0 0.0 19 0 4 0
Other 9 33.3 66.7 0.0 3 0 0 0
Total 722 71.2 25.8 3.1 515 0.28 1.8 0
p 0.00 0.00 0.76 0.96 0.77 NA
Business
Man 307 94.1 5.5 0.0 289 94.8 3.1 1.0
Woman 813 93.7 5.3 0.5 760 87.6 5.1 0.3
Couple 22 90.9 13.6 0.0 19 89.5 0 0
Other 20 75.0 25.0 0.0 16 87.5 6.3 0
Total 1162 93.5 5.9 0.3 1084 89.6 4.5 0.5
p 0.008 0.001 0.632 0.009 0.393 0.406
Notes: Respondents could indicate more than one form of acquisition. Statistical significance tests refer to Pearson Chi-sq test across all forms of ownership. For
agricultural land, other ownership claims to family land, clearing prime forest, encroachment, through sharecropping (some types of tenancies allow for eventual
ownership), and adjudication by the community or the state.
GENDERED PATHS TO ASSET ACCUMULATION
51
Table 3 Incidence of modes of acquisition, by form of ownership, Karnataka, India
% of purchasers using:
Form of Owned % purchased/ Savings
ownership assets (n) constructed % inherited % other Purchased (n) Any savings Any credit and credit
Main residence
Man 2013 36.8 56.7 9.4 741 82.1 54.3 38.9
Woman 738 30.5 59.4 13.7 225 79.6 45.3 28
Couple 130 45.4 43.1 17.7 59 86.4 54.2 42.4
Other 298 33.6 63.4 5.7 100 79 49 34
Total 3,179 35.4 57.4 10.4 1125 81.51 52 36.4
p 0.001 0.001 0.000 0.555 0.112 0.019
Agricultural land
Man 1867 9.3 85.8 3.8 173 85 41.6 28.9
Woman 363 11.9 82.1 5.0 43 81.4 30.2 14
Couple 65 21.5 72.3 6.2 14 42.9 71.4 14.3
Other 350 8.0 84.0 6.3 28 100 50 50
Total 2,645 9.8 84.7 4.4 258 83.7 42.3 27.9
p 0.003 0.009 0.149 0.000 0.043 0.006
(Continued).
ARTICLE
52
Table 3 Continued.
% of purchasers using:
Form of Owned % purchased/ Savings
ownership assets (n) constructed % inherited % other Purchased (n) Any savings Any credit and credit
Other real estate
Man 552 45.7 46.0 11.1 252 83.3 27.4 10.7
Woman 143 37.1 49.7 14.7 53 83 34 17
Couple 30 40.0 53.3 10.0 12 91.7 66.7 58.3
Other 79 20.3 69.6 8.9 16 81.3 25 6.3
Total 804 41.4 49.3 11.4 333 83.5 29.7 13.2
p 0.000 0.001 0.542 0.884 0.028 0.000
Business
Man 429 87.0 13.1 0.0 373 64.9 41.3 6.2
Woman 130 78.5 19.2 0.0 102 73.5 26.5 2.9
Couple 69 79.7 18.8 0.0 55 65.5 40 5.4
Other 102 70.6 29.4 0.0 72 73.6 40.3 13.9
Total 730 82.5 17.0 0.0 602 67.4 38.5 6.5
p 0.001 0.001 0.101 0.055 0.032
Notes: Respondents could indicate more than one form of acquisition. Statistical significance tests refer to Pearson Chi-sq test across all forms of ownership.
GENDERED PATHS TO ASSET ACCUMULATION
jointly. In Ecuador, couples are significantly more likely than individual
owners to acquire all forms of immovable property and businesses through
the market. In both Ghana and Karnataka, this pattern holds true for
dwellings and agricultural land; for other real estate and business, the share
of assets acquired through the market is greatest for those owned solely by
men.
Comparing just those assets owned individually by a man or a woman,
for most types of property, a greater percentage of those owned by men
are acquired through the market than those owned by women.15 These
differences are statistically significant in the case of dwellings in Karnataka
and Ghana, land in Ecuador, other real estate in Karnataka and Ghana, and
businesses in the case of Ecuador and Karnataka. Thus, we find differences
in the patterns of mode of acquisition by gender: a smaller share of
women’s assets is acquired through the market.
Overall, the assets owned by couples are most likely to have been acquired
through the market, followed by those owned solely by men and, lastly,
those owned solely by women. This smaller incidence of purchase by
women could be due to biases against them either in asset markets, and/or
in access to the means of financing their purchase.
HOW ARE ASSETS ACQUIRED THROUGH THE MARKET?
To purchase immovable property or businesses, people use some
combination of savings and credit. Tables 1–3 also provide information on
the percentage of purchased assets that were acquired using any savings
at all, any credit at all, or both savings and credit. (Note that these three
categories are not exclusive; someone who used both savings and credit
would be included in all three columns.)16
All four types of assets are much more likely to have been purchased
with savings rather than credit in all three countries. This reflects the
historically limited availability of access to credit for purchasing immovable
property and for business start-ups. The extent to which any credit has
been used varies across the three countries. Use of credit is more common
in Karnataka than in Ecuador, except in the case of other real estate. In
Ghana, the use of credit for acquiring any of these assets has been relatively
infrequent. The asset in Ecuador and Karnataka most likely to have been
purchased with any credit is the main residence; in Ghana, it is land or
businesses, although credit was used to acquire fewer than 5 percent of
these.
It is also more common in Karnataka for many of these assets to have been
purchased using both credit and savings than in the other two sites. For
example, 36 percent of the purchased dwellings in Karnataka were acquired
using both credit and savings, compared to only 19 percent in Ecuador and
less than 1 percent in Ghana. Thus, although it is much more common to
53
ARTICLE
Figure 3 Percentage of residences, agricultural parcels, and businesses that were
acquired with credit (partially or totally), by form of ownership
acquire immovable property and business through savings than credit in all
three countries, there are differences among them in the extent to which
credit has been used. Figure 3 summarizes the differences in the use of
credit to acquire the main residence, agricultural land, and businesses by
the form of property ownership.
In Ecuador, the residences and land parcels owned by couples are
significantly more likely to have been purchased with the use of credit
than those owned by individual owners (a pattern that holds for other
real estate; see Table 1). Among assets owned by individual owners, while
there is no gender gap in the use of credit to acquire a dwelling, other
real estate, or a business, individual female owners are significantly more
likely than individual male owners to have acquired their land with the use
of credit. However, relatively few women landowners have acquired their
parcels through the market.
In Ghana, the use of credit to acquire these four types of assets is so
limited that the differences by type of ownership are not significant overall.
However, examining the share of individually owned assets acquired via the
use of credit suggests that women more frequently than men have acquired
their main residence or a business with the use of credit.
In Karnataka, the dwellings owned by couples are equally as likely as those
owned by men to have been purchased with credit; the only statistically
significant difference is between men and women. In the case of land,
the parcels owned by couples are significantly more likely than those
owned by individuals to have been purchased with credit, although the
gender difference among individually owned assets is not significant. The
54
GENDERED PATHS TO ASSET ACCUMULATION
acquisition of other real estate follows a similar pattern to land. With
respect to businesses, the gender differences stand out, with businesses
owned by men and couples being more likely to have been started with
the use of credit than those owned by women.
In sum, our most important finding is how in the three sites, most
immovable assets and businesses that have been acquired through the
market have relied upon savings rather than credit, and how this holds for
both men and women. Most of the differences in the use of credit – with the
exception of Ghana where its use for asset acquisition has been infrequent –
are driven by the fact that couples tend to be more likely than individuals to
use credit to acquire any of these assets. Notwithstanding, only in Karnataka
is there a notable gender gap in the use of individual credit to acquire
assets. This raises the question of the extent to which couples are favored
over individual borrowers by financial institutions, given that they are more
likely to be composed of two income earners, and draws attention to the
importance of considering institutional variables.
CREDIT FOR ASSET ACCUMULATION
We turn now to the use of credit for the acquisition of assets and address this
from a different perspective. Rather than considering how specific assets
were acquired, we analyze current outstanding asset loans. For this analysis,
we consider loans to acquire a broader group of physical assets: besides
borrowing to acquire or enhance immovable property and businesses,17 we
include loans to purchase livestock, agricultural equipment, and consumer
durables.18 This analysis is only carried out for Ecuador and Ghana, since
differences in how the questions concerning outstanding loans were asked
do not make it possible to distinguish the purpose of many loans for
Karnataka.19
As Table 4 shows, in Ecuador a much larger share of respondents have
outstanding asset loans than in Ghana: 28 percent versus 12 percent. This
result is consistent with the data provided earlier that showed that very little
of the immovable property and businesses in Ghana has been acquired
through the use of credit. In Ecuador, there is a notable gender gap in
men’s favor in accessing credit for asset acquisition, while the incidence of
men and women accessing asset loans in Ghana is similar.
The asset loans included above may be from formal or informal sources.
In Ecuador, formal sector loans are defined as those from a private or
public bank, a savings and loan cooperative or credit union, department
stores, NGOs or other private institutions, housing developers, and credit
card debt. The informal sector includes loans from family, friends,
employers, informal credit groups, moneylenders, and other. Overall, most
asset loans, 81 percent, are from the formal sector (see Table 5). A slightly
higher share of loans that are held jointly, either by the couple or by two
55
ARTICLE
Table 4 Incidence of outstanding loans for asset acquisition
Country Women Men Total
Ecuador 24.1% 32.7% 27.8%
N 2,656 2,012 4,668
Ghana 12.2% 11.8% 12.1%
N 1,790 1,481 3,271
Note: The total refers to those answering the individual questionnaire in our respective surveys.
Table 5 Distribution of asset loans by person(s) responsible for repayment and
source of credit (%)
Formal sector Informal sector Total loans
Ecuador
Woman alone 78.9 21.1 100 n = 384
Man alone 78.4 21.6 100 n = 486
Couple 86.2 13.8 100 n = 318
Other joint 89.1 10.9 100 n = 64
Total 81.1 18.9 100 n = 1,252
Ghana
Woman alone 9.5 90.5 100 n = 201
Man alone 12.7 87.3 100 n = 150
Couple 27.3 72.7 100 n = 11
Other joint 28.6 71.4 100 n = 7
Total 11.7 88.3 100 n = 369
Notes: In Ghana, the source of loan could not be identified for all asset debt.
or more people, are from the formal sector than are those that are held
individually by either a man or a woman.
In Ghana, formal loans comprise those from banks, nongovernmental
organizations (NGOs), credit cooperatives, and microfinance institutions.
The majority of asset loans are from the informal sector, and only 12
percent are from the formal sector. The heavy reliance on informal
sector credit for asset acquisition in Ghana, compared to Ecuador, is
consistent with typical indicators of the level of financial development
across countries. In 2010, the ratio of private credit to GDP was 27.9 percent
in Ecuador and 13.8 percent in Ghana.20
Similar to Ecuador, jointly held loans in Ghana are more frequently from
the formal sector compared to individually held loans. We would expect
this result if formal institutions, such as private banks, prefer lending to
couples (or two relatives) if two income earners are considered more likely
to repay a loan than a sole borrower. These joint loans might also reflect
56
GENDERED PATHS TO ASSET ACCUMULATION
the fact that two people might better able to come up with the collateral
required for a bank loan. In the case of Ecuador, this tendency is also
favored by the marital regime. If a person is married, housing mortgages
and loans for vehicles are legally required to be signed by both spouses
since both are responsible for the debt, and would thus be reported as joint
loans. Mortgages and vehicle loans are more readily available in Ecuador
than in Ghana, and a larger share of households take out loans for these
purposes.21
Whereas in Ecuador the share of individual asset loans that are from
formal institutions is similar for men and women, there is a gender
gap favoring men in Ghana. These differences are largely explained
by the fact that Ecuador has a much more diversified formal sector
than Ghana. In Ghana, formal lending is concentrated in private banks,
whereas in Ecuador installment credit by department stores predominates
among formal sector loans, and savings and loan cooperatives are almost
as important a source of credit as private banks. Whereas the former
institutions are equally important sources for men and women, there is a
slight male bias in Ecuador among private as well as state banks as a source
of formal loans.22
Both countries have experienced the rapid growth of microfinance
institutions in recent years.23 In Ghana, nonetheless, our survey revealed
that loans from microfinance institutions and other private/NGO loan
programs still represent a very small share of total loans. In Ecuador, the
microfinance sector includes seven of the twenty-four private banks, all of
the savings and loans cooperatives, and other private/NGO loan programs.
Unfortunately, the microfinance activities of private banks cannot be
distinguished in our survey; other private/NGO loans also constitute a very
small share of total loans (Grown et al. 2015).
Our analysis of the credit requirements of private banks in both countries
suggests that these tend to favor clients who have stable wage employment
and/or a savings or checking account at their institutions.24 In both
countries, it is more common for wage workers to have a bank account,
since there is a growing tendency for paychecks to be deposited directly
into an account, making holding a bank account virtually a requirement
of the job. In Ecuador, 54.5 percent of male respondents, but only 23.5
percent of female respondents, reported themselves as wage workers; the
corresponding figures for Ghana were 20.2 and 5.7 percent. Hence, gender
differences in the rate of formal wage employment are likely to be reflected
in gender differences in the incidence of having a bank account and access
to formal credit. Indeed, women in Ghana are much less likely than women
in Ecuador to own formal accounts, and the gender gap in favor of men in
the ownership of formal accounts is wider in Ghana (Doss et al. 2012a).
Other requirements to opening a bank account include having a national
identification card (and in Ecuador, proof of having voted), proof of
57
ARTICLE
residence (such as a utility bill), and often, having a guarantor. These may
also create gender differentiated constraints, if women are less likely to have
an identification card or have utility services in their own names (or access
at all, in the case of Ghana), or if their social networks are less conducive to
producing viable guarantors for formal loans than men’s.
An econometric analysis of access to asset loans for Ecuador – where these
are predominantly from the formal sector – found that the likelihood of
having an asset loan for both men and women was positively associated
with having a formal bank account (Deere and Catanzarite 2017). Earning
an income increased the likelihood of having an asset loan for women,
but not for men, while being a member of a social network increased
it only for men. This latter result is likely related to the different kinds
of groups to which men and women belong in Ecuador, primarily trade
unions and sporting associations among men, and religious groups and
parents’ associations among women, with perhaps men’s groups more likely
to includemembers whomight be good guarantors of a loan. While earning
an income does not differentiate among men who are able to borrow to
acquire assets, belonging to a union reflects the importance of the form of
employment, in this case, formal sector employment.25
CONCLUSIONS
In analyzing the retrospective patterns of the acquisition of immovable
property and businesses in Ecuador, Ghana, and Karnataka, India, we find
that these differ by gender in ways that could have important implications
for policies to strengthen women’s property rights. Our analysis shows
that in all three countries women have acquired fewer assets through
the market than men. While inheritance is still the dominant means
of transferring immovable property, particularly agricultural land, asset
markets are increasingly important in Ecuador and Ghana, while less so
in Karnataka. This is especially true for the acquisition of dwellings in the
former two countries and business start-ups in all three. It is thus useful
to consider ways other than inheritance to increase women’s property
ownership because social norms and inheritance practices are slow to
change even when there is the political will to bring about legal change.
The broad policy question is the extent to which the market may facilitate
women’s accumulation of assets.
Our secondmajor finding pertains to the primacy of savings in facilitating
asset purchase. In all three countries, the main way that men and women
were able to buy immovable assets and businesses was using their personal
savings. Credit was also used but to a more limited extent, and showed
a much higher incidence in Ecuador and India than in Ghana. There
were substantial gender gaps in the use of credit for asset acquisition, most
58
GENDERED PATHS TO ASSET ACCUMULATION
notably in India, which points to the need for credit markets to widen both
reach and target populations and to expand in an inclusive manner.
Our analysis also indicates that couples were more likely than individual
owners to have relied upon credit to purchase assets. In Ecuador, given
the community of property marital regime, one might expect a relatively
high incidence of joint ownership of property by couples and that they
seek loans jointly to acquire assets. In contrast, in a country like India, with
separation of property, that couples are more likely than individuals to seek
credit for the acquisition of assets could point to a positive change in gender
relations, since it implies that women are being included by their spouses
in the process of purchase.
Our analysis of current borrowing for asset acquisition in Ecuador
and Ghana, however, suggests considerable efforts will be required to
fully incorporate women into credit markets. While financial deepening
and greater credit availability is associated with the process of economic
development, there is no guarantee that this process will be gender
neutral, particularly with respect to access to private bank credit, which
is usually cheaper than credit from other sources. An important insight
from Ecuador’s experience is that the expansion of savings and loan
cooperatives, as well as installment credit can play a key role in enhancing
women’s access to credit for the accumulation of assets. Also important
is women’s ownership of bank accounts, which could enable them to
accumulate savings and access formal credit.
Other types of microfinance institutions, besides cooperatives, can
potentially play an important role in expanding women’s access to formal
credit for the acquisition of assets, but they have been of relatively limited
reach thus far in the overall picture in both Ecuador and Ghana.26 Our
review of the literature on microfinance programs highlighted how in
the evaluation of these programs, insufficient attention has been given
to the role of microcredit in facilitating asset acquisition in general and,
specifically, whether loans taken out by women result in their ownership
of assets, either individually or jointly with their partners. The absence of
data on which household member acquired assets with credit is a serious
shortcoming since it stymies the analysis of whether women borrowers
themselves benefit from such programs by strengthening their fallback
position in household negotiations, or whether they are merely the conduit
for asset accumulation by other household members, especially their
partners.
A precondition for women borrowers to be able to utilize their loans
for asset accumulation is that they need to be able to control how their
loans are utilized, as stressed by Supriya Garikipati (2008) and others.
In the case of microentrepreneurs, the business environment must also
be sufficiently favorable to justify an investment in greater fixed capital
and stock, a condition not always met in the sectors in which women
59
ARTICLE
microentrepreneurs tend to be located (Klapper and Parker 2011). Finally,
women need access to loans that are large enough to purchase assets,
particularly larger assets. The evidence is mixed thus far as to whether there
is a glass ceiling on loan size (Agier and Szafarz 2013; Corsi and De Angelis
2017).
Our analysis is limited by the data available. While our surveys provide
the most detailed data available on individual asset ownership, they were
not designed to provide a full history of the accumulation of immovable
property and businesses. For a more complete analysis, a comprehensive
profile of all immoveable property and businesses acquired by individuals
over a given time horizon would be imperative, including information on
their location, marital status, employment status, and other individual and
household attributes at the time of purchase. Our findings point to the
importance of such data collection efforts in the future, given that markets
are increasingly becoming critical means of asset accumulation.
Our findings regarding the importance of savings in the acquisition of
assets support the growing emphasis in development policy and in the
literature on how efforts to promote financial inclusion should focus not
just on expanding access to credit but also to savings accounts. As Thorsten
Beck, Asli Demirgüç-Kunt, and Maria Soledad Martinez Peria (2008),
Felipe Kast and Dina Pomeranz (2014), and others argue, greater attention
must be given to the constraints that individuals face in opening a bank
account. These constraints are monetary, in terms of the costs of opening
and maintaining an account; institutional, in terms of the requirements
(such as multiple documents) to open an account; and cultural, in terms of
whether it is considered appropriate for women to participate in financial
markets.
Finally, our analysis also points to the importance for women of having
steady income-generating employment in order to be able to participate
in asset markets. Employment and the level of income thereby generated
directly influences the ability to save as well as access to credit geared to
asset accumulation.
Cheryl R. Doss
International Development – Oxford
3 Mansfield Rd., Oxford
United Kingdom of Great Britain and Northern Ireland
email: cheryl.doss@qeh.ox.ac.uk
Carmen Diana Deere
Center for Latin American Studies – University of Florida
PO Box 115530, Gainesville, FL, 32605
USA
email: deere@ufl.edu
60
GENDERED PATHS TO ASSET ACCUMULATION
Abena D. Oduro
Department of Economics – University of Ghana
P.O. Box LG57, Legon
Ghana
email: aoduro@ug.edu.gh
Hema Swaminathan
Indian Institute of Management Bangalore (IIMB)
Bannerghatta Road, Bangalore, Karnataka, 560076
India
email: h.swaminathan@gmail.com
Zachary Catanzarite
Department of Health, Behavior, and Society - Johns
Hopkins School of Public Health
615 North Wolfe Street, Baltimore, MD 21205
USA
email: zcatanz1@jhu.edu
J. Y. Suchitra
Indian Institute of Management Bangalore (IIMB)
Bannerghatta Road, Bangalore, Karnataka 560076
India
email: suchitrajy@gmail.com
NOTES ON CONTRIBUTORS
Cheryl R. Doss is Associate Professor and Senior Lecturer in Development
Economics in the Department of International Development at Oxford
University. She is a co-Principal Investigator of the Gender Asset Gap
project and has written extensively on issues of women’s asset ownership
and intrahousehold issues.
Carmen Diana Deere is Distinguished Professor Emerita of Latin American
Studies and Food & Resource Economics at the University of Florida,
and Honorary Professor Emerita of the Latin American Faculty of Social
Sciences (FLACSO-Ecuador) in Quito, where she teaches periodically. She
is a co-Principal Investigator of the Gender Asset Gap project, and directed
its Ecuador study.
Abena D. Oduro is Associate Professor of Economics and Director of the
Merian Institute for Advanced Studies in Africa at the University of Ghana.
Her current research interests are gender and assets, child marriage, and
poverty dynamics. She is a co-Principal Investigator of the Gender Asset Gap
project, and directed the Ghana study.
61
ARTICLE
Hema Swaminathan is Assistant Professor at the Indian Institute of
Management Bangalore. She is an empirical economist with broad interests
in understanding the causes and consequences of social and human
development with a focus on intrahousehold behavior. Her recent and
ongoing research focuses on wealth and asset distribution between men
and women and its implications for welfare outcomes. She is a co-Principal
Investigator of the Gender Asset Gap project, and directed the Karnataka,
India study.
Zachary Catanzarite is PhD student and Research Associate at Johns
Hopkins University School of Public Health.
J. Y. Suchitra is a researcher with the Centre for Public Policy at the Indian
Institute of Management Bangalore. Her work spans a range of issues on
social exclusion including gendered patterns of asset ownership.
ACKNOWLEDGMENTS
The authors are grateful to Caren Grown, a co-Principal Investigator of the
Gender Asset Gap project, who took the lead in the analysis of access to
credit. The Vanguard Charitable Endowment Program and UN Women
funded this research effort.
FUNDING
This work was supported by UNWomen; Vanguard Charitable Endowment
Program.
NOTES
1 Immovable property includes the principal residence, agricultural land, and other
real estate (the latter includes nonagricultural lots, dwellings other than the principal
residence, and buildings that are not part of a business).
2 For a summary on South Asia, see Bina Agarwal (1994); on Latin America, Carmen
Diana Deere and Magdalena León (2001); and on Africa, Elizabeth Cooper and Kate
Bird (2012).
3 A study of forty-four developing countries found that in thirteen, 10 percent or more
of the women sampled reported that they were not involved in the decision on how
to spend the income that they generated (World Bank 2011: Fig. 2.9).
4 A survey only in Central Uganda reports slightly different results; while 40 percent of
women had acquired land through inheritance, 43 percent of men had done so. But
43 percent of women had purchased land while only 36 percent of men had done so
(Kes, Jacobs, and Namy 2011).
5 This pattern has also been found in many developed countries. Susan Marlow and
Dean Patton (2005) argue that one of the reasons women tend to have smaller
businesses than men is that their personal savings are conditioned by their prior labor
market experience. Moreover, businesses that are initially undercapitalized also tend
to underperform.
62
GENDERED PATHS TO ASSET ACCUMULATION
6 Milford Bateman and Ha-Joon Chang (2012) provide a powerful political economy
analysis of why microcredit programs are unlikely to lead to sustainable poverty
reduction. For a systematic review of impact studies of microcredit on the well-being
of the poor see Maren Duvendack et al. (2011). They found that many of the studies
that claim a positive impact of microcredit are methodologically flawed. See Naila
Kabeer (2005) and Linda Mayoux (1999) for a good summary of the issues regarding
women’s empowerment and microfinance.
7 The systematic review resulted in fifteen studies, but two were of micro-savings
interventions. Overall, they find that microfinance programs have a limited impact
on wealth outcomes, defined as income, expenditure, savings, and assets. Non-wealth
outcomes in these same studies are examined in Marcel Korth et al. (2012).
8 Many of these studies shed little light on this relationship. For example, in a field
experiment in Nepal where a no-fee back account was offered to women, the take-
up (84 percent) and usage (80 percent) rates were quite high (Prina 2015). The
impact of the treatment, however, was only measured at the household level, and
had no statistically significant impact on the level of household assets (nor income or
expenditures) after one year. In this study, it is not clear if the women actually control
the bank account.
9 In the Ecuador questionnaire, these questions were located in the household
inventory. When the household consisted of a principal couple, they answered these
questions together whenever possible. In Ghana, questions about the acquisition of
the main residence was asked in the inventory, whereas questions about agricultural
land, other real estate, and businesses were asked in the individual questionnaire. In
Karnataka, these questions were in the individual section and were answered by the
person who was considered most knowledgeable about all of the owned assets. The
second respondent in the household was asked about the modes of acquisition only
of the assets that she or he owned.
10 Differences in how the questions concerning outstanding loans were asked do not
make it possible to distinguish the purpose of many of the loans in Karnataka.
11 A more detailed discussion of the patterns of asset ownership in these three countries
is provided in Cheryl R. Doss et al. (2014).
12 Henceforth, “purchase” is used to refer to both purchases as well as the construction
of the dwelling or another building.
13 “Inheritance” will be used to refer to property that is transferred as gifts, “en vivo,” as
well as upon death.
14 For some of the questions about asset acquisition, respondents could report multiple
modes of acquisition for a single piece of property. This was more common in
Karnataka than the other countries, especially for residences.
15 The significance tests for individually owned assets are not reported in Tables 1–3,
and are available from the authors.
16 Not shown is the category of “other” that includes government programs, for
example, that provide a grant that is then coupled with either savings or a loan.
Other also includes remittances that are sometimes also combined with a person’s
own savings or a loan.
17 We do not distinguish between borrowing to acquire a home, for example, and loans
to expand or improve the dwelling; similarly, in the case of businesses, we do not
distinguish between purchasing a business, borrowing to initiate one (start-up costs),
and expanding it by investing in physical or working capital.
18 All of these assets may serve as collateral, particularly for loans from savings and
loan cooperatives or the informal sector. Consumer durables may also provide the
potential means for income-generating activities should the need arise, or be pawned
or sold to meet debt repayment.
63
ARTICLE
19 See Caren Grown et al. (2015) for a detailed explanation on how information on
borrowing activities was gathered in the survey instruments and for an analysis of the
differences between borrowing for the purpose of asset accumulation and expenses.
In Ecuador, 67 percent, and in Ghana, 58 percent of the total loans reported were for
the purpose of asset accumulation (Grown et al. 2015: Table 8).
20 Based on the World Bank Financial Structure and Economic Development data base,
for private credit by deposit money banks and other financial institutions as a share
of GDP.
21 In Ecuador, 8 percent of households have a mortgage on their principal residence,
while 2.6 percent have borrowed to acquire other real estate and 2.9 percent, a
vehicle. In Ghana, less than 1 percent of households nationally have been able to
borrow for any of these purposes. Less than 1 percent of households in either country
have borrowed to acquire agricultural land (Grown et al. 2015: Table 2).
22 See Grown et al. (2015: Table 9), for the distribution of all formal and informal loans
by specific sources.
23 On Ghana, see Joseph Kimos Adjei, Thankom Arun, and Farhad Hossain (2009); on
Ecuador, see Deere and Catanzarite (2017).
24 This analysis is based on the study of the web pages of the main private banks as well
as interviews.
25 There was no significant difference in the likelihood of an asset loan by locale
of residence. Both Afro-Ecuadorian men and women were significantly more
likely to have asset loans than white/mestizos. Our subsequent qualitative research
suggests that this finding probably reflects the rapid expansion of department store
installment credit rather than microfinance institutions in the region where the
majority of Afro-Ecuadorians in the sample are located.
26 In the case of Ecuador, the data does not allow us to analyze the specific role
of microfinance banks within private bank credit. Loans from NGO microfinance
programs constitute less than 4 percent of the total loans.
REFERENCES
Adjei, Joseph Kimos, Thankom Arun, and Farhad Hossain. 2009. “The Role of
Microfinance in Asset-Building and Poverty Reduction: The Case of Sinapi Aba Trust
of Ghana.” BWPI Working Paper 87, Brooks World Poverty Institute, University of
Manchester, Manchester, UK.
Agarwal, Bina. 1994. A Field of One’s Own: Gender and Land Rights in South Asia. Cambridge:
Cambridge University Press.
Agier, Isabelle and Ariane Szafarz. 2013. “Microfinance and Gender: Is There a Glass
Ceiling on Loan Size?” World Development 42: 165–81.
Ali, Daniel Ayelew, Klaus Deininger, and Markus Goldstein. 2014. “Environmental
and Gender Impacts of Land Tenure Regularization in Africa: Pilot Evidence from
Rwanda.” Journal of Development Economics 110: 262–75.
Ashraf, Nava, Dean Karlan, and Wesley Yin. 2010. “Female Empowerment: Impact of a
Commitment Savings Product in the Philippines.” World Development 38(3): 333–44.
Barnes, Carolyn, Gary Gaile, and Richard Kibombo. 2001. The Impact of Three Microfinance
Programs in Uganda. Washington, DC: Assessing the Impact of Microenterprise
Services(AIMS).
Bateman, Milford and Ha-Joon Chang. 2012. “Microfinance and the Illusion of
Development: From Hubris to Nemesis in Thirty Years.” World Economic Review 1:
13–36.
64
GENDERED PATHS TO ASSET ACCUMULATION
Beck, Thorsten, Asli Demirgüç-Kunt, and Maria Soledad Martinez Peria. 2008. “Banking
Services for Everyone? Barriers to Bank Access and Use Around the World.” World
Bank Economic Review 22(3): 397–430.
Binaté Fofana, Namizata, Gerrit Antonides, Anke Niehof, and Johan van Ophem. 2015.
“How Microfinance Empowers Women in Côte d’Ivoire.” Review of Economics of the
Household 13(4): 1023–41.
Brannen, Conner. 2010. “An Impact Study of the Village Savings and Loan Association
(VSLA) Program in Zanzibar, Tanzania.” Honor’s thesis, Wesleyan University.
Cooper, Elizabeth and Kate Bird. 2012. “Inheritance: A Gendered and Intergenerational
Dimension of Poverty.” Development Policy Review 30(5): 527–41.
Corsi, Marcella and Marina De Angelis. 2017. “Gender Discrimination in Microfinance?
Some Evidence from Uganda.” Journal of Development Studies 53(5): 723–40.
Datta, Namita. 2006. “Joint Titling – A Win-Win Policy? Gender and Property Rights in
Urban Informal Settlements in Chandigarh, India.” Feminist Economics 12(1/2): 271–
98.
Davies, James B. and Anthony F. Shorrocks. 2000. “The Distribution of Wealth.” In
Handbook of Income Distribution, Vol. 1, edited by Anthony B. Atkinson and François
Bourguignon, 605–75. Amsterdam: North-Holland.
Deere, Carmen Diana. 1990. Household and Class Relations: Peasants and Landlords in
Northern Peru. Berkeley: University of California Press.
Deere, Carmen Diana and Zachary B. Catanzarite. 2017. “Who Borrows to Accumulate
Assets? Class, Gender and Indebtedness in Ecuador’s Credit Market.” CEPAL Review
122: 107–26. Economic Commission for Latin America and the Caribbean (ECLAC).
Deere, Carmen Diana and Cheryl R. Doss. 2006. “The Gender Asset Gap: What Do We
Know and Why Does It Matter?” Feminist Economics 12(1/2): 1–50.
Deere, Carmen Diana and Magdalena León. 2001. Empowering Women: Land and Property
Rights in Latin America. Pittsburgh: University of Pittsburgh Press.
Deere, Carmen Diana, Abena D. Oduro, Hema Swaminathan, and Cheryl Doss. 2013.
“Property Rights and the Gender Distribution of Wealth in Ecuador, Ghana and
India.” Journal of Economic Inequality 11(2): 249–65.
Doss, Cheryl R., Carmen Diana Deere, Abena D. Oduro, and Hema Swaminathan. 2012a.
“Women Who Save: A Comparative Analysis of Ecuador, Ghana and Karnataka India.”
Report prepared for United Nations Foundation.
________. 2014. “The Gender Asset and Wealth Gaps.” Development 57(3/4): 400–9.
Doss, Cheryl, Mai Truong, Gorrettie Nabanoga, and Justine Namaalwa. 2012b.
“Women, Marriage and Asset Inheritance in Uganda.” Development Policy Review 30(5):
597–616.
Dupas, Pascaline and Jonathan Robinson. 2013. “Savings Constraints and Micro-
enterprise Development: Evidence from a Field Experiment in Kenya.” American
Economic Journal: Applied Economics 5(1): 163–92.
Duvendack, Maren, Richard Palmer-Jones, James G. Copestake, Lee Hooper, Yoon Loke,
and Nitya Rao. 2011. “What is the Evidence of the Impact of Microfinance on the
Well-being of Poor People?” EPPI-Centre Report 1912, EPPI-Centre, Social Science
Research Unit, Institute of Education, University of London, London.
Elson, Diane. 1999. “Labor Markets as Gendered Institutions: Equality, Efficiency and
Empowerment Issues.” World Development 27(3): 611–27.
Fenrich, Jeanmarie and Tracy E. Higgins. 2001. “Promise Unfulfilled: Law, Culture,
and Women’s Inheritance Rights in Ghana.” Fordham International Law Journal 25(2):
259–341.
Friedemann-Sánchez, Greta. 2006. “Assets in Intrahousehold Bargaining Among
Women Workers in Colombia’s Cut-Flower Industry.” Feminist Economics 12(1/2):
247–69.
65
ARTICLE
Garikipati, Supriya. 2008. “The Impact of Lending to Women on Household
Vulnerability and Women’s Empowerment: Evidence from India.” World Development
36(12): 2620–42.
Grown, Caren, Carmen Diana Deere, Zachary Catanzarite, Abena D. Oduro, J. Y.
Suchitra, Hema Swaminathan, and Louis Boakye-Yiadom. 2015. “Who Borrows? An
Analysis of Gender, Debt, and Assets in Ecuador, Ghana, and Karnataka, India.”
Discussion Paper, UN Women, New York.
Kabeer, Naila. 2005. “Is Microfinance a ‘Magic Bullet’ for Women’s Empowerment?
Analysis of Findings from South Asia.” Economic and Political Weekly 40(44/45):
4709–18.
Kast, Felipe and Dina Pomeranz. 2014. “Saving More to Borrow Less: Experimental
Evidence from Access to Formal Savings Accounts in Chile.” NBER Working Paper
20239, National Bureau of Economic Research (NBER), Cambridge, MA.
Kes, Aslihan, Krista Jacobs, and Sophie Namy. 2011. Gender Differences in Asset Rights in
Central Uganda. Washington, DC: International Center for Research on Women.
Klapper, Leora F. and Simon C. Parker. 2011. “Gender and the Business Environment
for New Firm Creation.” World Bank Research Observer 26: 237–57.
Korth, Marcel, Ruth Stewwart, Carina Van Rooyen, and Thea De Wet. 2012.
“Microfinance: Development Intervention or Just Another Bank?” Journal of Agrarian
Change 12(4): 575–86.
Lastarria-Cornhiel, Susana. 1997. “Impact of Privatization on Gender and Property
Rights in Africa.” World Development 25(8): 1317–34.
Lastarria-Cornhiel, Susana, Julia A. Behrman, Ruth Meinzen-Dick, and Agnes R.
Quisumbing. 2014. “Gender Equity and Land: Toward Secure and Effective Access
for Rural Women.” In Gender in Agriculture: Closing the Knowledge Gap, edited by Agnes
R. Quisumbing, Ruth Meinzen-Dick, Terri L. Raney, André Croppenstedt, Julie A.
Behrman, and Amber Peterman, 117–44. New York: Springer.
Marlow, Susan and Dean Patton. 2005. “All Credit to Men? Entrepreneurship, Finance
and Gender.” Entrepreneurship Theory and Practice 29(6): 717–35.
Mayoux, Linda. 1999. “Questioning Virtuous Spirals: Microfinance and Women’s
Empowerment in Africa.” Journal of International Development 11: 957–84.
Prina, Silvia. 2015. “Banking the Poor via Savings Accounts: Evidence from a Field
Experiment.” Journal of Development Economics 115: 16–31.
Semyonov, Moshe and Noah Lewin-Epstein. 2013. “Ways to Richness: Determination of
Household Wealth in 16 Countries.” European Sociological Review 29(6): 1134–48.
Sircar, Ashok K. and Sohini Pal. 2014. “What is Preventing Women from Inheriting
Land? A Study of the Implementation of the Hindu Succession (Amendment) Act
2005 in Three States in India.” Paper presented at the 2014 World Bank Conference
on Land and Poverty, Washington, DC, March 24–7.
UN Women. 2015. Progress of the World’s Women 2015–2016: Transforming Economies,
Realizing Rights. New York: UN Women.
Van Rooyen, Carina, Ruth Stewart, and Thea DeWet. 2012. “The Impact of Microfinance
in Sub-Saharan Africa: A Systematic Review of the Evidence.”World Development 40(11):
2249–62.
World Bank. 2010. Women’s Economic Opportunities in the Formal Private Sector in Latin
America and the Caribbean. Washington, DC: The World Bank, the Inter-American
Development Bank, and GTZ.
World Bank. 2011. World Development Report 2012: Gender Equality and Development.
Washington, DC: The World Bank.
66