List of Abbreviations ANC Aggregate number of clients ANPC Aggregate number of poorest clients ATT Average treatment effect CHFS China Household Finance Survey CPI Consumer price index E
Introduction
Background
1.1.1Have we done enough against poverty?
Poverty is defined as a pronounced deprivation in wellbeing (Haughton & Khandker,
Traditional perspectives on wellbeing associate poverty primarily with insufficient income or consumption, defining the poor as individuals who do not meet an adequate minimum threshold of material resources (Haughton & Khandker, 2009) However, poverty extends beyond monetary measures, being causally linked to various deprivations that impact overall wellbeing Poor individuals often experience limited access to essential services such as healthcare, quality housing, education, and employment opportunities, as well as reduced political freedoms, highlighting the multidimensional nature of poverty (Gordon).
& Spicker, 1999; Haughton & Khandker, 2009) Measuring these variables and unravelling their complex connections is challenging because their effects on people’s lives can be devastating
Over the past 200 years, global absolute poverty has significantly declined, primarily driven by the rapid economic growth of India and China These two countries accounted for approximately 75% of the reduction in the world’s poor over a decade leading up to 2015 In India alone, over 360 million people have escaped poverty, highlighting the substantial progress made in reducing global poverty levels.
According to the World Bank, absolute or extreme poverty is defined as living on less than US$1.90 (or $3.10) per day, reflecting updated global standards since October 2015 Despite increasing income inequality, China has experienced unprecedented economic growth, leading to a dramatic reduction in extreme poverty Over 800 million people have escaped absolute poverty due to this rapid development.
1980 and 2015 2 According to the most recent estimates from the World Bank, China’s extreme poverty rate fell from 66.6 per cent in 1990 to 1.9 per cent in 2013 3
Despite progress, high poverty rates persist in regions like India, where in 2011, extreme poverty affected 21.2% of the population, or 268 million people Globally, the World Bank reports that using the $1.90 per day poverty line, the number of people in extreme poverty decreased from 1.85 billion (35%) in 1990 to 767 million (10.7%) in 2013 However, nearly half of these people live in Sub-Saharan Africa, where only a slight reduction of four million poor individuals occurred, leaving 389 million in extreme poverty in 2013—more than all other regions combined Although global poverty has declined over the decades, significant challenges remain, particularly in the poorest countries.
Figure 1.1: Poverty in China and India
Notes: The poverty headcount ratio is at US$1.90 a day (2011 purchasing power parity [PPP]) As a result of data unavailability, India has very few observations
Source: World Development Indicators from the World Bank
2 http://www.worldbank.org/en/country/china/overview#1
3 http://povertydata.worldbank.org/poverty/country/CHN
4 http://www.worldbank.org/en/topic/poverty/overview#1
Figure 1.2: Poverty around the world (2013)
1.1.2Gender and poverty: are women poorer than men?
The concept of the feminisation of poverty highlights that women often face higher rates of poverty compared to men, as first introduced by Pearce (1978) She observed that women constitute an increasing proportion of the economically disadvantaged This phenomenon can mean that women have a higher incidence of poverty than men, that their poverty rates are increasing over time relative to men, or that women's poverty is more severe Understanding these aspects is crucial for addressing gender disparities in economic vulnerability and developing targeted strategies to combat poverty among women.
Scholars have extensively studied the feminisation of poverty, identifying key factors such as parenthood, education, and employment as primary contributors For instance, research using U.S census data by Starrels, Bould, and Nicholas (1994) highlights that while gender, race, ethnicity, marital status, and employment all significantly influence poverty rates, race and gender are the most critical variables Notably, parenthood interacts with gender in a way that uniquely impacts women, reflecting the societal expectation of unpaid childcare and family responsibilities that disproportionately burden women and contribute to their higher poverty risk.
South Asia Latin America & Caribbean
The poverty headcount ratio at $1.90 a day (2011 PPP) highlights how limited income restricts individuals' participation in paid economic activities, leading to involuntary poverty Studies across multiple countries indicate that key factors such as parenthood, marital status, and employment significantly influence the gender-poverty gap Specifically, research by McLanahan, Garfinkel, and Casper (1994) showed these factors are crucial in both the US and other nations More recent research by Wilson (2012) emphasizes the role of employment, revealing that low wages caused by occupational segregation, discrimination, and insufficient work hours are primary contributors to women's poverty.
The feminisation of poverty remains an uncertain phenomenon globally due to limited data and measurement challenges There is a lack of regularly produced gender-disaggregated poverty data, and no internationally agreed-upon indicators specifically capture poverty differences between men and women Consequently, scholars often compare male- and female-headed households (FHHs) to assess whether poverty disproportionately affects women, with mixed findings For instance, research from ten developing countries by Quisumbing, Haddad, and Peña (1995) found weak evidence that FHHs dominate poverty, showing little difference between male- and female-headed households among the poor Similarly, Moghadam (2005) concluded that the link between FHHs and extreme poverty is inconclusive globally, with the strongest evidence in the US and some variation in other countries.
In summary, the assertion that women are universally poorer than men cannot be substantiated, but it is an undeniable truth that women are disadvantaged As Moghadam (2005, p 1) stated:
Women living in poverty face a double burden, suffering not only from economic hardship but also from gender inequality Recognizing poverty as a denial of human rights highlights the urgent need to address both issues simultaneously to promote gender equality and social justice Addressing the intersection of gender and poverty is essential for effective human rights advocacy and sustainable development.
1.1.3Role of financial inclusion—particularly microfinance—in reducing poverty and gender inequality
Over the past few decades, policymakers and regulators in both developing and developed countries have prioritized financial inclusion as a key component of financial sector development These efforts include implementing legislative measures and policies aimed at expanding access to financial services for underserved populations, promoting economic growth, and reducing income inequality.
The United Kingdom’s 'Financial Inclusion Task Force' was established in 2005 to address barriers to banking access, alongside initiatives like South Africa’s ‘Mzansi’ account and other alternative financial institutions such as microfinance Globally, the World Bank set a strategic goal to achieve universal financial access by 2020, recognizing that approximately two billion adults (38%) lack a basic bank account (Demirgüç-Kunt, Klapper, Singer, & Van Oudheusden, 2015) Microfinance institutions (MFIs) and programs have gained significant attention because they are effective tools for promoting financial inclusion—offering tailored financial services to the unbanked Unlike traditional banks, microfinance provides small loans to individuals with few or no assets, helping create job opportunities and generate income, especially among underserved populations.
Microfinance has experienced rapid growth across the developing world, with the Asia-Pacific region leading by 2013, hosting 144.17 million microfinance clients—about 18 times more than Sub-Saharan Africa, 35 times more than Latin America and the Caribbean, and 577 times more than Eastern Europe and Central Asia Notably, approximately 66.92% of these microfinance clients, or 96.8 million people, are classified as poor, underscoring the sector’s primary goal of poverty alleviation Microfinance's core objective is to improve access to financial services for the poor, enabling them to start small businesses like vegetable farming or pig raising, which can boost household income and health This transformative potential is the key driver behind microfinance’s global expansion and success.
Most of the poorest clients of microfinance institutions (MFIs) are women, highlighting microfinance’s potential to both reduce poverty and promote gender equality According to a Microcredit Summit Campaign report, microfinance plays a crucial role in empowering women and fostering socio-economic development.
In 2015, Microfinance Institutions (MFIs) reached over 211 million clients in 2013, including more than 114 million individuals living in extreme poverty, of whom 82.6% were women Providing women with access to credit is believed to enhance their decision-making power within households and encourages resource allocation that benefits the entire family Empowering women through microfinance plays a crucial role in advancing gender equality and promoting social and economic development.
Research aims and objectives
This thesis focuses on exploring the relationship between financial inclusion, poverty, and inequality Microfinance, as an effective tool for promoting financial inclusion (Dev, 2006), plays a crucial role in poverty alleviation The primary research question aims to examine how microfinance initiatives impact poverty reduction and contribute to narrowing inequality gaps.
As microfinance also aims to reduce gender inequality by targeting women (Cheston & Kuhn, 2002), the second research question is:
2 Does microfinance reduce gender inequality?
Limited reliable data on microfinance has led to a scarcity of recent research examining its impact on poverty and gender inequality at the macroeconomic level, as noted by Hermes (2014) Addressing these gaps through cross-country analysis can significantly enrich the existing literature by providing new insights into how microfinance influences broader economic and social outcomes.
Conversely, although financial inclusion has gained popularity around the world, only a limited number of studies have investigated whether financial inclusion has helped to
Currently, there is no comprehensive index measuring financial inclusion at the microeconomic level, which limits the ability to assess how financial services impact people's lives This gap highlights the need for improved metrics to better understand and enhance financial inclusion efforts Consequently, the third research question focuses on addressing this measurement challenge to facilitate more effective policymaking and resource allocation aimed at improving people's well-being For more detailed data and insights, visit https://stateofthecampaign.org/data-reported/.
3 Can we develop an index to measure financial inclusion at the microeconomic level?
With such an index, and considering the fact that household income is the most important determinant of poverty, this thesis then asks the final research question:
4 Does financial inclusion increase household income?
Methodology
This thesis employs a quantitative approach to explore key issues in financial inclusion, microfinance, poverty, and inequality Chapters 2–5 identify gaps in existing research by reviewing current literature, highlighting unresolved questions in these areas Following an extensive literature review, the study formulates four specific research questions to address these gaps To investigate these questions, the thesis develops three hypotheses that guide the empirical analysis and provide insights into the impact of microfinance on poverty alleviation and social inequality.
Hypothesis 1: Microfinance does not reduce poverty
Hypothesis 2: Microfinance does not reduce gender inequality
Hypothesis 3: Financial inclusion does not have a positive effect on household income
This thesis predicts that microfinance will reduce poverty and gender inequality, while increasing financial inclusion is expected to positively impact household income It collects data from diverse sources and employs advanced econometric methods, including fixed-effects analysis, Heckman two-step, quantile regression, and propensity score matching, to rigorously investigate four key research questions Multiple techniques across Chapters 2 to 5 ensure robust and reliable results, providing comprehensive insights into the effects of microfinance and financial inclusion.
Thesis structure
This thesis systematically addresses four key research questions through comprehensive analysis across Chapters 2 to 6 Chapter 2 explores the impact of microfinance on poverty reduction at the macroeconomic level, highlighting its potential to alleviate poverty through financial services targeted at underserved populations Emphasizing gender-focused initiatives, Chapter 3 examines how women’s participation in microfinance programs fosters advancements in gender equality In Chapter 4, the study develops the Multidimensional Financial Exclusion Index (MFEI) to identify the primary factors influencing household financial inclusion Using the MFEI, Chapter 5 investigates the relationship between financial inclusion and household income, demonstrating the importance of accessible financial services for economic empowerment Finally, Chapter 6 summarizes the main findings, discusses the study’s contributions, and offers policy recommendations to enhance microfinance and financial inclusion strategies.
Does Microfinance Reduce Poverty?
Introduction
Does microfinance, which is an effective tool in promoting financial inclusion (Dev,
The fundamental question in the debate on microfinance’s effectiveness is whether it truly helps poor people in developing and emerging countries to reduce poverty Poverty is a widely discussed topic across various disciplines, each offering different perspectives: demographers emphasize gender and family structure, historians attribute poverty to historical factors like triangular trade and colonization, and geographers focus on the geographical location of economies as key determinants.
7 Part of Chapter 2 has previously been published: Zhang, Q (2017) Does microfinance reduce poverty?
Most economists agree that location, historical, and cultural factors, along with macroeconomic policy, significantly influence economic development Evidence shows that economic growth is essential for reducing poverty, as supported by studies from Dollar & Kraay (2002) and Durlauf, Johnson, & Temple (2005) Additionally, factors such as democracy (Acemoglu & Robinson, 2000; Savoia, Easaw, & McKay, 2010), human capital accumulation (Gregorio & Lee, 2002; Jung & Thorbecke, 2003; Kappel, 2010), arable land (Tesfamicael, 2005), and income inequality (Besley & Burgess, 2003; Hermes, 2014; Soubbotina & Sheram, 2004) have been shown to positively impact poverty reduction and overall economic development.
Scholars debate the impact of various factors on poverty reduction, with some studies indicating that increased trade promotes poverty decline (Dollar & Kraay, 2004; Maertens & Swinnen, 2009), while others suggest trade openness may exacerbate poverty (Topalova, 2007; Wade, 2004) Similarly, research on financial development presents mixed findings: some argue that financial progress disproportionately benefits the poor (Beck, Demirgüç-Kunt, & Levine, 2007; Clarke, Xu, & Zou, 2006), whereas others contend that financial development could potentially increase poverty and inequality (Behrman, Birdsall, & Székely, 2001; Jalilian & Kirkpatrick, 2005).
Income inequality has garnered significant attention due to its profound impact on economic performance and social stability Mohammed (2017) highlights that in both developed and developing countries, the poorest half of the population often possesses less than 10% of national wealth, illustrating the wide income gap High income inequality hampers sustainable economic growth, weakens social cohesion and security, and fosters inequitable access to and utilization of global resources, undermining prospects for sustainable development and peaceful societies The growing income disparity across nations remains a critical challenge in today's world, with over 90% of respondents in seven Sub-Saharan African countries perceiving the wealth gap as a major issue, according to the 2014 Pew Global Attitudes Survey.
Microfinance, also known as MFIs or microfinance programmes, has received relatively little attention in the poverty literature Over the past few decades, various microfinance initiatives have been implemented worldwide, highlighting their potential role in poverty alleviation Despite growing interest, the impact of microfinance on reducing poverty remains understudied, making it a crucial area for further research and policy development.
The global microfinance sector primarily aims to alleviate poverty and reduce inequality by expanding access to financial services for the poor In 2013, the Asia-Pacific region alone had approximately 144.17 million microfinance clients, according to the Microcredit Summit Campaign, making it the largest market worldwide This number is nearly 18 times higher than in Sub-Saharan Africa, 35 times that of Latin America and the Caribbean, and 577 times greater than in Eastern Europe and Central Asia Of these clients, around 96.8 million people, or 66.92%, are classified as poor based on the international definition of poverty.
Microfinance has the potential to alleviate poverty while being financially sustainable, even generating profits, which explains its rapid global growth (Brau & Woller, 2004) Research indicates that microfinance positively impacts households’ economic and social welfare and plays a role in reducing poverty (Zhuang et al., 2009) However, most existing studies focus on micro-level, country-specific case studies, highlighting the need to analyze microfinance’s effects on poverty at the macroeconomic level This chapter aims to fill that gap by examining the broader, macroeconomic impacts of microfinance on poverty reduction.
Literature review
Recent empirical studies have increasingly focused on the relationship between financial development and social outcomes One stream of research explores how financial development contributes to reducing poverty and income inequality, highlighting its potential to promote inclusive growth Another key area of investigation examines microfinance as a crucial driver of financial development, emphasizing its role in empowering underserved populations These insights underscore the importance of financial sector growth in fostering economic stability and social equity.
2.2.1Financial development, poverty and income inequality
Numerous empirical studies have demonstrated that financial development plays a significant role in reducing poverty and income inequality (Bahmani-Oskooee & Zhang, 2015; Clarke, Zou, & Xu, 2003; Hamori & Hashiguchi, 2012; Jeanneney & Kpodar, 2011; Kappel, 2010; Li, Squire, & Zou, 1998) Cross-country evidence consistently shows that the positive effects of financial development on poverty alleviation are widely recognized, despite methodological challenges such as heterogeneity among countries and missing variables (Zhuang et al., 2009) Research using unbalanced panel data from 126 nations indicates that the choice of financial variables, income measures, and model specifications influences outcomes (Hamori & Hashiguchi, 2012) Specifically, Jeanneney and Kpodar (2011) found that financial development benefits the poor directly through better distributional effects and indirectly via economic growth, with the direct impact being more substantial Their findings suggest that the poor benefit more from financial development, although some may not fully utilize available credit or face increased financial instability, but overall, the advantages outweigh the costs.
Emerging research indicates an inverted U-curve relationship between financial development and income inequality, known as the Greenwood–Jovanovic (G–J) hypothesis (Greenwood & Jovanovic, 1990) According to this hypothesis, financial development initially exacerbates inequality and poverty but eventually reduces them as a country's average income rises This concept has been reinforced by subsequent theoretical studies, including those by Aghion and Bolton (1997) and Matsuyama, highlighting the dynamic impact of financial growth on income distribution during different stages of economic development.
(2000) In contrast, some empirical studies have supported the G–J hypothesis (Jalilian
Financial development plays a crucial role in promoting economic growth, reducing poverty, and addressing income inequality research by Jalilian and Kirkpatrick (2005) highlights that financial development supports poverty alleviation through growth when a certain level of economic development is reached Zhang and Chen (2015) find an inverted U-shaped relationship between financial development and income inequality, indicating that initial financial growth can increase inequality before eventually reducing it The fundamental functions of finance—such as providing saving services—enable the poor to securely accumulate funds for future investments and expenses, fostering economic stability Additionally, financial services like insurance help protect vulnerable populations from unexpected shocks and disasters, reducing their vulnerability and minimizing long-term income risks These mechanisms demonstrate how financial development contributes to growth, poverty reduction, and decreased income disparity.
Financial development plays a crucial role in reducing poverty and income inequality primarily through expanding access to financial services Empirical studies have shown that limited access to ongoing financial services poses a significant barrier, especially in developing countries, in efforts to combat poverty and bridge income gaps Providing affordable and widespread financial services can empower underserved populations, promote economic inclusion, and foster sustainable development.
Inequality tends to be relatively stable within countries but varies significantly across nations, largely due to capital market imperfections (Li et al., 1998) Adverse selection, asymmetric information, and moral hazard in financial markets lead to widespread credit constraints, especially in developing countries (Aghion & Bolton, 1997; Banerjee & Newman, 1993; Galor & Zeira, 1993) These credit constraints disproportionately affect the poor, who often lack the resources and collateral needed to access bank credit (Zhuang et al., 2009) Financial development plays a crucial role in removing these barriers, reducing transaction costs, and expanding access to financial services, which enables the poor to invest in human capital, start small businesses, and manage investments effectively.
Developing countries often lack access to ongoing financial services due to underdeveloped financial systems, as highlighted by Li et al (1998) Unlike developed nations, where collateral and credit-scoring systems facilitate borrowing, these tools are ineffective in many developing regions because many potential borrowers cannot provide collateral or have established credit records (Zhuang et al., 2009) Therefore, informal financial sectors, including microfinance, play a crucial role in bridging this gap and supporting financial inclusion in developing countries.
2.2.2Microfinance, poverty and income inequality
Several studies focus on the impact of microfinance in reducing poverty and income inequality, highlighting its potential benefits However, most of these research efforts are region-specific and rely on micro-level data, which makes it challenging to draw broad, generalizable conclusions The variation in research methods and measurement techniques further complicates the ability to establish definitive findings on microfinance’s overall effectiveness in alleviating poverty globally.
Numerous studies have demonstrated that microfinance effectively reduces poverty across various countries For instance, Ghalib, Malki, and Imai (2015) found that access to microfinance in Pakistan positively impacted household well-being based on survey data from 2008–2009 Similarly, Imai and Azam (2012), using a panel survey in Bangladesh from 1997–2004, showed that microfinance loans contributed to increased income and food consumption, supporting microfinance’s role in poverty alleviation Other researchers have reaffirmed these findings in Bangladesh (Chemin, 2008; Chowdhury, Ghosh, & Wright, 2005; Khandker, 1998; Nawaz, 2010) and extended the evidence to countries like Bolivia (Navajas et al., 2000), India (Imai, Arun, & Annim, 2010), Nigeria (Okpara, 2010), Sri Lanka (Shaw, 2004), Central America (Hiatt & Woodworth, 2006), and across Africa (Mosley & Rock, 2004).
Several studies highlight contrasting views on microfinance's impact Van Rooyen, Stewart, and De Wet (2012) examined how microcredit and micro-savings affect impoverished communities in Sub-Saharan Africa, revealing that in some cases, microfinance can exacerbate poverty, lower educational attainment, and disempower women.
(2005) surveyed the evidence from Asia and Latin America and found limited evidence that microfinance is reaching the core poor population in either region Chowdhury
Research by 2009 critically questioned the effectiveness of microfinance as a universal tool for poverty reduction, suggesting that its impact remains uncertain Experts like Chowdhury (2009) emphasize that public policies should prioritize growth-oriented and equity-enhancing programs, such as fostering broad-based productive employment Similarly, Littlefield, Morduch, and Hashemi (2003) argue that no single intervention, including microfinance, can eradicate poverty on its own Microfinance serves as a foundation for essential services like healthcare, education, and nutritional support, and improvements in these areas are sustainable only when households experience increased income and greater financial control Therefore, microfinance contributes to poverty alleviation through multiple tangible impacts, supporting broader development goals.
Limited macro-level data on microfinance has resulted in few recent studies examining its impact on poverty and income inequality globally For instance, Hisako and Shigeyuki (2009) analyzed 61 developing countries from 2005 to 2007, demonstrating that increased microfinance intensity—measured by the number of microfinance institutions and borrowers—can effectively reduce income inequality, highlighting microfinance’s potential as a tool for economic redistribution.
Research by (2009) highlighted the importance of focusing on the equalizing effects of microfinance for poorer countries but did not address endogeneity issues or control for time-invariant country characteristics using fixed effects To address these limitations, Imai, Gaiha, Thapa, and Annim (2012) conducted a comprehensive study using cross-country and panel data from 48 nations between 2003 and 2007, employing both OLS and IV techniques to assess microfinance's impact on poverty Their findings indicated that microfinance significantly reduces poverty at the macro level, supporting increased funding from development financial institutions and governments to microfinance institutions in developing countries Unlike Hisako and Shigeyuki (2009), who used different measures, Imai et al (2012) primarily relied on gross loan portfolios to evaluate microfinance activities across countries.
Hermes (2014), building on previous studies by Imai et al (2012) and Hisako and Shigeyuki (2009), examined the impact of microfinance activities on income inequality using cross-sectional data The study measured microfinance intensity through the number of active borrowers and the total value of loans issued, applying the IV estimation technique to address endogeneity issues The findings indicate that higher microfinance participation is associated with a modest reduction in income inequality, though the overall effect is relatively small Hermes (2014) concluded that microfinance should not be seen as a comprehensive solution for significantly reducing income inequality.
This chapter investigates the impact of microfinance on poverty reduction at the macro level across 106 countries from 1998 to 2013, providing comprehensive insights into the relationship between microfinance and poverty in developing and emerging economies Utilizing a unique cross-country panel dataset from the Microcredit Summit Campaign and the MIX Market, it contributes significantly to existing literature by offering robust, large-scale evidence Microfinance is measured both in depth—through the total number of clients and the poorest clients—and in size via gross loan portfolios, ensuring a thorough analysis The study also rigorously addresses potential endogeneity issues, such as sample selection bias and reverse causality, by applying the Heckman two-step method, thereby enhancing the validity of the findings.
Methodology and model
The methodology of this chapter follows previous studies that have examined the relationship between financial development and poverty (Clarke et al., 2006; Li et al.,
This chapter assesses financial development by focusing on the size and depth of microfinance, replacing traditional formal sector measures used in earlier studies (e.g., 1998) It adopts a similar methodological approach but emphasizes microfinance indicators to better capture financial inclusion The model is informed by recent empirical research such as Hermes (2014), Hisako & Shigeyuki (2009), and Imai et al (2012), ensuring alignment with current analytical standards The empirical specification incorporates these microfinance measures to analyze their impact on financial development effectively.
This study examines the impact of microfinance variables on poverty indicators across countries over time, modeled through the equation \( y_{it} = \beta M_{it} + \gamma X_{it} + \alpha_i + \lambda_t + \varepsilon_{it} \) The dependent variable \( y_{it} \) represents poverty levels for country \( i \) at time \( t \), while \( M_{it} \) captures the key microfinance factors influencing poverty Control variables are included in \( X_{it} \) to account for other influencing factors, and fixed effects \( \alpha_i \) and \( \lambda_t \) control for country-specific characteristics and global shocks like the Financial Crisis The error term \( \varepsilon_{it} \) accounts for unobserved random variations.
502 Bad GatewayUnable to reach the origin service The service may be down or it may not be responding to traffic from cloudflared
502 Bad GatewayUnable to reach the origin service The service may be down or it may not be responding to traffic from cloudflared
502 Bad GatewayUnable to reach the origin service The service may be down or it may not be responding to traffic from cloudflared
Estimating Equation (2.1) involves a selection problem due to a truncated sample, as poverty data is only available for specific years in most countries, leading to non-random samples and biased estimates of the relationship between poverty and microfinance To address this, a probit model is used in the first stage to estimate the probability of a country being included in the sample The selection criteria include gross domestic product (GDP) per capita and total population, since low GDP per capita indicates poverty and limited survey resources, while large populations pose challenges for household data collection.
In the second stage, the structural equation is estimated using lagged values of microfinance variables as instruments, addressing the challenge of finding suitable instruments that influence poverty indirectly without direct effects This approach is grounded in the principle that past values of independent variables impact current poverty levels The poverty measurements are regressed on these lagged microfinance variables, the inverse Mills ratio (IMR), and additional control variables to account for sample selection bias and endogeneity simultaneously.
Data
This chapter utilizes a comprehensive cross-country panel data set encompassing 106 countries from 1998 to 2013, selected based on data availability The appendix provides a detailed list of all included countries, along with the number of poverty observations from each nation.
Measuring poverty is complex, as it goes beyond income to include lack of basic human needs such as food, shelter, water, health, education, and opportunity, often linked to insecurity and limited freedom While many studies focus on economic measures like the poverty headcount ratio—which indicates the percentage of people living below the poverty line—and the poverty gap—which shows the average shortfall from the poverty line—these indicators provide valuable insights into poverty levels globally Using World Bank data, these metrics are reliably comparable across countries through purchasing power parity (PPP), making them essential tools for understanding and addressing poverty worldwide.
Microfinance data are primarily sourced from two major projects: the Microcredit Summit Campaign and the MIX Market The Microcredit Summit Campaign, established in 1997, is a prominent advocacy network that unites microfinance practitioners, donor agencies, NGOs, educational institutions, and international financial organizations to promote best practices and knowledge exchange in microfinance From 1997 to 2016, it played a pivotal role in advancing microfinance development globally Currently, the Campaign provides comprehensive, verified data at the firm, country, and regional levels, including figures on total and poorest clients, and the percentage of women among clients To ensure data quality, the Campaign implements a rigorous verification process for submitted information, enhancing the reliability of its microfinance reports.
Approximately 40% of reporting institutions are required to provide a third-party verifier to confirm the validity of key data, including the total number of borrowers, the number of the poorest borrowers, and the percentages of women among borrowers and the poorest borrowers To ensure data accuracy, all submitted information undergoes careful examination and adjustments to prevent double-counting of borrowers, enhancing the reliability of the reported figures The MIX Market, established in 2002 as a not-for-profit organization, collects and publishes industry, country, and region-level microfinance data, sourcing information from audits, financial statements, management reports, and direct inquiries to microfinance institutions (MFIs) All data entered into the MIX database are rigorously reviewed and validated against predefined business rules by MIX analysts and partners before publication, ensuring high-quality, accurate reporting of microfinance outreach and performance indicators.
In October 2015, the World Bank revised its definition of extreme poverty; however, this chapter continues to utilize the older indicators, as the World Bank still applies 2005 PPP exchange rates and poverty lines for key countries such as Bangladesh, Cambodia, Jordan, and Lao PDR Both the previous and current measures consistently reflect the percentage of the population living in poverty, ensuring comparability over time.
In 1997, the Campaign initiated a nine-year effort to reach 100 million of the world’s poorest families—particularly women—with access to credit, self-employment opportunities, and financial services by 2005 After the initial phase, the Campaign was relaunched in November 2006 with expanded objectives to assist 175 million of the poorest families through similar financial and business support programs Additionally, the submitted data undergoes rigorous quality assurance, with MIX’s database review system performing over 135 checks to ensure data accuracy and reliability.
This chapter assesses the depth and size of microfinance by analyzing the aggregate number of clients and the poorest clients from the Campaign, alongside the gross loan portfolio data from MIX While the Campaign provides insights into the number of borrowers served, MIX offers a comprehensive measure of the overall microfinance portfolio size According to Bauchet and Morduch (2010), the Campaign serves more borrowers on average than MIX, as many small and some large institutions report to the Campaign, highlighting differences in data strengths between the two sources.
The MIX database primarily attracts medium-sized financial institutions and offers a diverse array of financial indicators, such as gross loan portfolios, operational self-sufficiency, and total assets, with over 80% of institutions reporting these metrics In contrast, the Campaign mainly involves smaller institutions and provides only one financial indicator—operational self-sufficiency—which is reported by approximately 60% of institutions, indicating a weaker data scope A key limitation of the Campaign is its limited financial data, while the strength of MIX lies in its comprehensive reporting and larger dataset Notably, existing studies have not utilized a combined data set from both Campaign and MIX sources to test hypotheses, underscoring the need for macro-level research that integrates these data sets to yield more reliable and robust insights into the relationship between poverty and microfinance.
According to the Microcredit Summit Campaign Report 2015, 3,098 MFIs reported reaching 211.12 million borrowers in 2013 This was the largest number ever reported
Of these MFIs at the regional level, the Asia–Pacific region had the largest number (1,119), followed by Sub-Saharan Africa (1,045) and Latin America and the Caribbean
The Asia–Pacific region hosts the majority of microfinance clients, with 166.99 million clients overall, including 101.43 million living below US$1.25 per day—the poorest segment This region's poorest clientele is nearly ten times higher than in Latin America and the Caribbean, ten times that of Sub-Saharan Africa, thirty-one times that of Eastern Europe and Central Asia, and thirty-two times that of the Middle East and North Africa region Consequently, most microfinance clients are concentrated in Asia–Pacific, while North America and Western Europe have the fewest clients among the regions The participation rate of the poorest clients, defined as the ratio of the poorest to total clients, remains a key metric for assessing microfinance outreach and inclusivity.
The Campaign defines the poorest clients based on the World Bank’s standard of living beneath US$1.25 per day Participation rates among these clients vary significantly across regions, with the highest in the Asia–Pacific at 60.74%, and the lowest in Eastern Europe and Central Asia at just 2.03% Other regions show intermediate levels of participation: 54.73% in Sub-Saharan Africa, 24.71% in North America and Western Europe, 23.67% in the Middle East and North Africa, and 15.80% in Latin America and the Caribbean These figures indicate that the Asia–Pacific region has the highest engagement of the poorest clients, while Eastern Europe, Central Asia, and Latin America and the Caribbean have comparatively lower participation.
Figure 2.1: Total clients and poorest clients of MFIs in 2013 (region level)
Source: A uthor’s compilation using data from the Microcredit Summit Campaign Report 2015
This chapter incorporates key control variables from previous research that are linked to poverty, including GDP per capita to reflect economic development, CPI to account for inflation and macroeconomic stability, and private sector credit as a percentage of GDP to gauge financial sector development It also considers the democracy index to assess institutional quality, gross secondary school enrollment as a proxy for human capital, and the sum of exports and imports relative to GDP to measure economic openness Additionally, general government final consumption expenditure as a share of GDP is included to control for government spending effects.
U n it : m il li o n total clients total poorest clients poorest participation
This article utilizes variables sourced primarily from the World Bank's World Development Indicators database, ensuring comprehensive and reliable data The democracy index, however, is obtained from the Integrated Network for Societal Conflict Research (INSCR), providing specialized insights into democratic processes The democracy index is measured on an additive 11-point scale from 0 to 10, where 0 indicates the absence of democracy and 10 represents full democratic governance.
10 represents full democracy Table 2.1 presents the summary statistics for each variable and provides detailed descriptions of them
Variable Description Year Obs Mean Std Dev Min Max
PH125 Poverty headcount ratio at US$1.25 a day (PPP) (%) 1998–2013 592 14.75 19.33 0 87.72
PH2 Poverty headcount ratio at US$2 a day (PPP) (%) 1998–2013 592 26.74 26.46 0 95.41
PG125 Poverty gap at US$1.25 a day (PPP) (%) 1998–2013 592 5.637 8.774 0 52.76
PG2 Poverty gap at US$2 a day (PPP) (%) 1998–2013 592 11.36 13.86 0 67.58
GLP Gross loan portfolio (per client) 2003–2013 754 1,377 2,320 0 25,629
ANC Aggregate number of clients/total population (%) 1998–2008 825 0.0116 0.0252 0 0.274
ANPC Aggregate number of poorest clients/total population (%) 1998–2008 825 0.00726 0.0197 0 0.262
GDP per capita GDP/total population (constant 2005 US$) 1998–2013 1,673 2,467 2,512 129.8 15,423
Domestic credit Domestic credit to private sector by banks/GDP (%) 1998 – 2013 1,636 29.07 24.36 0.154 153.4
Education Gross secondary school enrolment (%) 1998–2013 1,233 63.16 27.27 5.132 114.6
Government General government final consumption expenditure/GDP (%) 1998–2013 1,582 14.47 5.513 2.058 42.51
Democracy Level of democracy in a country (0 – 10) 1998 – 2013 1,568 5.158 3.470 0 10
Openness Sum of exports and imports/GDP (%) 1998–2013 1,623 80.07 37.25 16.44 321.6
The gross loan portfolio per client is sourced from MIX Market (www.mixmarket.org), providing comprehensive data on microfinance activities The article also references the aggregate number of clients and the poorest clients as a percentage of the total population, compiled by the author from the Microcredit Summit Campaign dataset (http://www.microcreditsummit.org) The democracy index utilized is from INSCR (http://www.systemicpeace.org), offering insights into political stability and governance Additionally, data from the World Bank Development Indicators database (http://data.worldbank.org/data-catalog/world-development-indicators) is incorporated to support a broader understanding of socio-economic trends relevant to microfinance.
Empirical results
The empirical results of estimating Equation (2.1) are reported in Tables 2.2 and 2.3 Table 2.2 presents findings using the Heckman two-step method, controlling solely for sample selection bias, whereas Table 2.3 displays results from the two-stage model that accounts for both sample selection bias and endogeneity Microfinance is measured differently across columns: Columns (1), (4), (7), and (10) utilize the aggregate number of clients relative to the total population (ANC) as the measure of microfinance depth, while columns (2), (5), and (8) employ alternative measures.
(11) use the aggregate number of poorest clients/total population (ANPC) and columns
The article utilizes gross loan portfolio (GLP) as the primary size measure in columns (3), (6), (9), and (12) To assess poverty, the analysis differentiates between two measures: the poverty headcount ratio in columns (1)–(6) and the poverty gap in columns (7)–(12) These metrics provide a comprehensive understanding of how microfinance portfolios relate to poverty levels, enhancing the relevance for SEO by including key terms such as "gross loan portfolio," "poverty headcount ratio," and "poverty gap."
The Heckman two-step method results, accounting for sample selection bias, are summarized in Table 2.2, highlighting the importance of controlling for IMR, which is significant across all columns Consistently, all 12 columns demonstrate a negative relationship between the three microfinance variables and the two poverty measures Notably, in columns (1), (3), (4), (7), and (10), this negative effect is statistically significant for both ANC and GLP at various levels, exemplified by column (4) where a coefficient of −1.18 indicates that a 10% increase in microfinance participation (ANC) correlates with a 0.18 unit reduction in the poverty gap at US$2 a day, holding other factors constant Conversely, the ANPC variable from the Campaign shows no significant association with the poverty indicators across all columns.
The results from the combined Heckman two-step method demonstrate that sample selection bias is a significant concern, as indicated by the consistent significance of the IMR across nearly all columns The regressions confirm that microfinance is negatively associated with poverty, regardless of the measures used, with a 10% increase in ANPC linked to a 0.085 decrease in the poverty headcount ratio at US$2 per day Similarly, a 10% increase in GLP correlates with a 0.126 reduction in the poverty headcount ratio at US$1.25 per day Notably, the poverty-reducing impact varies among microfinance variables, with GLP generally exhibiting a larger coefficient compared to ANC and ANPC, indicating a stronger effect on alleviating poverty.
Table 2.3 reveals that several control variables, such as per capita income, inflation, education, democracy, and openness, are significantly linked to poverty, with signs aligning with prior expectations Increased education correlates with a reduction in the poverty headcount ratio, decreasing by approximately 1.5 units when education rises by 10 units, all else being equal Similarly, higher levels of democracy are associated with lower poverty rates, with a one-unit increase in democracy leading to a decline of about 1.22 units in the poverty headcount ratio at US$2 per day These findings support the notion that poverty diminishes as education and democratic governance strengthen, consistent with established political economy theories (Gradstein, Milanovic, & Ying, 2001).
Table 2.4 analyzes the impact of sample selection bias and endogeneity on the bias in estimating the poverty–microfinance relationship Controlling only for sample selection bias using the Heckman method results in certain coefficients, but including both sample selection bias and endogeneity makes most coefficients significant, indicating more accurate estimates The comparison reveals that failure to address these biases can lead to inaccurate conclusions about the strength of the poverty–microfinance relationship, highlighting the importance of comprehensive bias control.
This chapter evaluates the robustness of the model by employing various microfinance and poverty measurement methods Microfinance is assessed through both depth and size indicators, utilizing data from two extensive projects: ANC and ANPC from the Campaign for microfinance depth, and GLP from MIX for microfinance size Poverty is measured using two distinct indicators derived from World Bank data, ensuring a comprehensive analysis of the relationship between microfinance initiatives and poverty reduction.
Microfinance significantly reduces poverty, as evidenced by the negative association between microfinance indicators and the poverty headcount ratio and poverty gap across columns (1)–(12) of Table 2.3 Notably, the GLP variable has a greater impact on alleviating poverty compared to ANC and ANPC, highlighting its effectiveness The control variables reinforce these findings, demonstrating the robustness of the model Overall, the results confirm that microfinance initiatives play a crucial role in poverty reduction.
Table 2.2: Results of Heckman two-step method, corrected for sample selection bias
Poverty headcount ratio at US$1.25 a day (PPP) (%)
Poverty headcount ratio at US$2 a day (PPP) (%)
Poverty gap at US$1.25 a day (PPP) (%)
Poverty gap at US$2 a day (PPP) (%)
Country fixed effects Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
Year fixed effects Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
Table 2.3: Results of combined Heckman two-step method, corrected for sample selection bias and endogeneity
Dependent variables Poverty headcount ratio at
Poverty headcount ratio at US$2 a day (PPP) (%)
Poverty gap at US$1.25 a day (PPP) (%)
Poverty gap at US$2 a day (PPP) (%)
Country fixed effects Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
Year fixed effects Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
Notes: Numbers in square brackets are t-values ***, ** and * represent significance at the 1%, 5% and 10% levels, respectively
Table 2.4: Step by step approach
Poverty headcount ratio at US$1.25 a day (PPP) (%)
Poverty headcount ratio at US$2 a day (PPP) (%)
Combined Heckman two-step method
Combined Heckman two-step method
Poverty gap at US$1.25 a day (PPP) (%)
Poverty gap at US$2 a day (PPP) (%)
Combined Heckman two-step method
Combined Heckman two-step method
Notes: Numbers in square brackets are t-values ***, ** and * represent significance at the 1%, 5% and 10% levels, respectively.
Conclusion
Understanding the impact of microfinance on poverty from a macro perspective is essential due to the rising number of microfinance institutions (MFIs) and growing interest from development agencies, governments, and stakeholders This chapter provides an overview of various microfinance programs and discusses practical implications for leveraging microfinance to effectively alleviate poverty.
This chapter uses a unique data set covering 106 countries for the 16-year period from
From 1998 to 2013, a study examined the relationship between microfinance and poverty in developing and emerging countries, utilizing macro-level data from the Microcredit Summit Campaign and the MIX Market The findings reveal a significant and negative correlation between microfinance and poverty, confirming previous research (Hermes, 2014; Hisako & Shigeyuki, 2009; Imai et al., 2012) This negative relationship remains consistent when replacing the poverty headcount ratio with the poverty gap, both before and after controlling for sample selection bias and endogeneity, demonstrating the robustness of the results Additionally, the study indicates that microfinance not only reduces the incidence of poverty but also lessens its depth and severity.
Microfinance is not a universal solution for poverty alleviation, as country-specific and cultural factors influence its effectiveness (Chowdhury, 2009) While failures such as small loans plunging households into deeper debt have been documented (Bateman, 2013), evidence generally supports its role as an effective tool in reducing poverty in many developing and emerging countries Microfinance enables the poor to participate in self-employment and income-generating activities, fostering financial independence and helping them escape poverty Governments and international agencies should continue promoting microfinance, recognizing its limitations and the need for a comprehensive approach to address entrenched global poverty.
According to the World Bank, the poverty gap measures the average shortfall from the poverty line, expressed as a percentage of the poverty line, capturing both the intensity and prevalence of poverty This indicator reflects not only how many people are living in poverty but also the depth of their poverty Imai et al (2012) confirmed similar findings in their research, highlighting the significance of the poverty gap as a crucial measure for assessing poverty severity Incorporating the poverty gap into social and economic analyses provides a comprehensive understanding of poverty dynamics and informs targeted interventions.
Asia and Pacific Middle East and North Africa Sub-Saharan Africa
Pakistan (6) Democratic Republic of Congo (1)
Latin America and Caribbean Eastern Europe and Central Asia Ghana (2)
Brazil (13) Bosnia and Herzegovina (3) Liberia (1)
Jamaica (4) Poland (13) Sao Tome and Principe (2)
Based on the World Bank data, the poverty headcount ratio at US$1.25 a day varies across countries, with the number of observations indicated in parentheses The analysis shows that all four poverty indicators consistently reflect similar trends in poverty levels within these nations This uniformity underscores the reliability of these measures in assessing extreme poverty globally Accurate measurement of poverty is essential for targeted policy interventions aimed at reducing income inequality and improving living standards worldwide.
Does Microfinance Improve Gender Equality?
Introduction
Gender equality, as defined by the United Nations (UN), involves ensuring equal rights, responsibilities, and opportunities for women, men, girls, and boys, emphasizing the consideration of diverse needs and priorities without suggesting sameness Despite progress made under the UN’s Millennium Development Goals (MDGs), women and girls worldwide still face discrimination and violence Achieving true gender equality remains a critical global challenge, requiring ongoing efforts to address these disparities.
17 Part of Chapter 3 has been published: Zhang, Q & Posso, A (2017) Microfinance and gender inequality: Cross-country evidence Applied Economics Letters, 24(20), 1494–1498
Gender equality is recognized as a vital component of global development, highlighted by its inclusion as one of the United Nations' 17 Sustainable Development Goals (SDGs) in the 2030 Agenda This comprehensive framework aims to eradicate all forms of poverty and promote inclusive progress worldwide.
Economists have long believed that economic growth was a key strategy to reduce gender inequality, with the 1960s and early 1970s viewing it as a universal solution (Inglehart & Norris, 2003) During the 1980s and 1990s, there was continued optimism that economic growth would automatically benefit women in developing countries (Beneria & Bisnath, 2001; Mundial, 2001) However, by the end of the 20th century, it became clear that growth alone did not eliminate gender disparities, as countries like Kuwait and Saudi Arabia, despite their high GDP per capita, still faced significant legal and social discrimination against women, exemplified by Saudi Arabia’s delay in allowing women to drive until 2017 This realization highlighted the complexity of gender inequality, prompting a move towards comprehensive strategies that include economic development, legal reforms, cultural change, and political institutions Among these strategies, microfinance has gained increasing critical attention as an effective tool to address gender disparities.
Over the past 30 years, microfinance has proven to be an effective development tool that provides sustainable, tailored financial services to the poor, especially women, to improve their welfare (McCarter, 2006) In 2013, 3,098 microfinance institutions (MFIs) served over 211 million clients worldwide, with more than 114 million living in extreme poverty; notably, 82.6% of these clients were women While some studies suggest microfinance may promote gender equality, most are country- or region-specific and rely on micro-level data This chapter aims to advance the understanding by analyzing the macroeconomic impact of microfinance on gender equality.
19 Here, GDP per capita is based on PPP (constant 2011 international $) provided by the World Bank (http://data.worldbank.org/data-catalog/world-development-indicators)
20 The Microcredit Summit Campaign uses the World Bank’s definition of ‘extreme poverty’ to mean those living on less than US$1.90 per day PPP (the recently updated international poverty line).
Literature review
Providing women with access to credit can enhance their decision-making power within households, potentially leading to increased spending on reproductive health and education Microfinance empowers women to negotiate better participation in labor markets, which can help reduce gender inequality While case studies across developing countries offer mixed evidence, research suggests that improved access to credit may strengthen women’s bargaining power and promote gender equality in health, education, and employment.
Several studies have found that microfinance plays a positive role in reducing gender inequality For example, using data from a large household survey conducted in rural
Research by Pitt et al (2006) in Bangladesh (1998–1999) found that women’s participation in group-based microfinance programs significantly enhances their decision-making power within the household Similarly, Swain and Wallentin (2009) observed comparable positive effects in India, indicating that microfinance initiatives play a crucial role in promoting women’s empowerment across different regions Incorporating microfinance into development strategies can effectively strengthen women’s agency and influence household dynamics.
Microfinance alone does not automatically reduce gender inequality, as its effectiveness depends on supplementary interventions such as education and improved access to waged employment (Kabeer, 2005) To effectively address gender disparities, microfinance initiatives must be integrated with these complementary strategies Additionally, Kabeer (2005) emphasizes the importance of considering specific contextual factors, since microfinance institutions (MFIs) operate in diverse environments, which can influence their impact on gender equality Similar caution is advised by Kabeer (2001) and Mahmood, highlighting the variability in results based on different settings and interventions.
(2011) and Ngo and Wahhaj (2012) argued that gender inequality ultimately depends on context and cultural norms, which determine autonomy over production Mahmood
In 2011, it was highlighted that providing women with business-related training is essential for their entrepreneurial success The lack of training offered by Microfinance Institutions (MFIs) has been identified as a key factor contributing to the low number of women starting new businesses with their loans Without financial and entrepreneurial autonomy, women often find their funds appropriated by men, which exacerbates gender inequality Additionally, Ngo and Wahhaj support these findings, emphasizing the critical role of capacity building in empowering women entrepreneurs.
Research by 2012 indicates that microfinance institutions (MFIs) with varying levels of innovativeness can have diverse effects on households, sometimes leading to a decline in female borrowers’ welfare Women receiving business-related training involving their husbands are more likely to be empowered than those trained in independent household activities Microfinance's impact on education is complex, as it can both increase resource constraints and demand for schooling while also raising opportunity costs due to more productive alternatives (Maldonado, González-Vega, & Romero, 2003) Gender inequality tends to worsen when girls are disproportionately affected, with studies by Agier and Szafarz (2013) and Brana (2013) revealing that women face a glass ceiling in larger projects, facing harsher borrowing conditions despite no bias in loan assessments For example, Agier and Szafarz’s analysis of 34,000 loan applications from Brazil showed that the gender gap in loan sizes widens with project scale Similarly, Brana’s study of 3,640 microcredit applicants in France found that women are more likely to be on state benefits, and gender remains a key factor influencing credit amounts, highlighting how microfinance can hinder gender equality.
(2013) argued that it is in this sense that MFIs reinforce gender inequalities
This chapter investigates the effect of microfinance on gender equality at the macroeconomic level using a panel of 64 developing economies over the period 2003–
This chapter offers the first rigorous global analysis of microfinance’s impact on gender equality in developing countries, demonstrating that women’s participation in microfinance is generally linked to reduced gender inequality Utilizing a macroeconomic approach based on cross-country data, the study provides clearer insights into how microfinance contributes to promoting gender equality Additionally, by incorporating regional interactions in the analysis, the research highlights that cultural factors significantly influence the relationship between microfinance and gender equality across different regions.
Methodology and model
The cross-national relationship between microfinance and gender inequality that is used in this chapter builds on existing macroeconomic models (Beer, 2009; Forsythe, Korzeniewicz, & Durrant, 2000):
The model 𝐺𝐼𝑖𝑡 = 𝛽𝑃𝑊𝐵𝑖𝑡+ 𝛾𝑿𝑖,𝑡+𝛼𝑖+ 𝜆𝑡+ 𝜀𝑖𝑡 captures the relationship between gender inequality (measured on a 0–1 scale) and the proportion of women borrowers in microfinance, alongside control variables like national income and democracy Country-specific factors such as geographic features are accounted for through country fixed effects (𝛼𝑖), while global shocks affecting all countries, like the financial crisis, are controlled via year fixed effects (𝜆𝑡) Since the dependent variable is measured continuously within a strict range, fixed-effects models are appropriate for estimation, with the variable expressed in natural logarithm to interpret coefficients as elasticities or semi-elasticities.
To study whether unobserved country-level characteristics affect the relationship between microfinance and gender inequality in the model, see Equation (3.2):
𝐺𝐼𝑖𝑡 = 𝛾𝑿𝑖,𝑡+ 𝜃1𝑃𝑊𝐵𝑖𝑡𝑹𝑖 + 𝛼𝑖 + 𝜆𝑡+ 𝜀𝑖𝑡, (3.2) where 𝑹𝑖 is a vector of regional dummy variables for Latin America and the Caribbean (LAC), Middle East and North Africa (MENA), South Asia (SA), Europe and Central
Asia (ECA), East Asia and the Pacific (EAP) and Sub-Saharan Africa (SSA) The equations are estimated using heteroscedasticity robust standard errors 21
Data
International comparative gender inequality indices, such as the Gender-related Development Index (GDI) and the Gender Inequality Index (GII), provide valuable insights into gender disparities worldwide The GDI, introduced by the UN in 2003, assesses inequalities in health, knowledge, and living standards—mirroring the Human Development Index (HDI)—with higher values indicating lower gender disparities In contrast, the GII, introduced in 2010, excludes income imputations to reduce measurement errors and focuses on reproductive health, empowerment, and labor market participation, with lower values signifying less gender inequality Despite some biases, especially towards elites in indicators like parliamentary representation, the GII offers a relevant comparative measure of gender disparities Data availability for the GDI spans 2003 to 2009, while the GII is available from 2010 to 2014, and both indices are measured on a 0–1 scale, where higher GDI scores indicate reduced gender inequality, and lower GII scores reflect the same.
Two large-scale data collection projects provide data for the key variable PWB: the
Microcredit Summit Campaign and the MIX Market This study employs MIX data for
21 A test for potential endogeneity was conducted before estimating Equations (3.1) and (3.2) It is possible that providing funds to women could be easier in countries with greater gender equality Further,
GI and GNP per capita may be endogenous, as improvements in GI can promote economic development by increasing female labor participation However, Hausman-endogeneity tests indicate these variables can be considered exogenous The test results favor the null hypothesis that PWB is exogenous, supported by two reasons: first, the MIX database covers over 80% of global institutions, whereas the Campaign data includes only 60%, and second, the Campaign data contains numerous missing values and typographical errors at the firm level, complicating accurate country-level aggregation (Bauchet & Morduch, 2010).
In contrast, MIX produces readily available country-level indicators The key variable,
PWB, is logged for ease of interpretation
This model, aligned with other macroeconomic gender inequality frameworks, accounts for economic development by incorporating gross national income (GNI) per capita (logged), sourced from the World Bank’s World Development Indicators It also considers democracy as a key variable, measured on an 11-point scale from 0 (no democracy) to 10 (full democracy), following methodologies by Inglehart and Norris (2003) and Beer (2009), with data obtained from INSCR Table 3.1 provides detailed summary statistics of these variables to support comprehensive analysis.
Obs Mean Std Dev Min Max
Notes: The GII is available from 2010 to 2014 and the GDI is available from 2003 to 2009.
Empirical results
Increasing the proportion of women borrowers is associated with a reduction in gender inequality, as shown in Columns (1) and (4) of Table 3.2, which present the results of Equation (3.1) A 10 percent rise in the share of women borrowers leads to a 0.38 percent improvement in the Gender Development Index (GDI), while the same increase corresponds to a 0.15 percent decrease in the Gender Inequality Index (GII), holding other factors constant.
Columns (2) and (5) display the results from Equation (3.2), highlighting that the main findings are driven by economies in the ECA and MENA regions An increase in the proportion of women borrowers is associated with a decrease in gender inequality, particularly in these regions This significant effect may be attributed to the conservative nature of these societies, where even small increases in women borrowers can substantially reduce gender disparity Additionally, as microfinance is still emerging in ECA and MENA, its impact on gender inequality remains positive and impactful Moreover, MFIs in these regions often employ innovative strategies, such as targeted training programs, to further reduce gender inequality.
Country and year FE? Yes Yes Yes Yes Yes Yes
Notes: Numbers in square brackets are t-values ***, ** and * represent significance at the 1%, 5% and
The proportion of women borrowers in the South Asia (SA) region is associated with worsening gender inequality, aligning with previous regional studies highlighting that increased female borrowing can marginalize men, leading to project sabotage, fund misappropriation, and violence Research by Leach and Sitaram (2002) and Agier and Szafarz (2013) illustrates that microfinance may contribute to a glass ceiling effect, hindering female entrepreneurs undertaking large projects and exacerbating gender disparities This negative impact may be intensified in the SA context, where microfinance's emphasis aimed at reducing gender inequality has sometimes produced unintended adverse effects, as suggested by Kabeer (2001, 2005) and Mahmood (2011) In contrast, the association between women borrowers and improved gender equality remains inconclusive in Sub-Saharan Africa (SSA), potentially due to differences between the Gender Inequality Index (GII), which emphasizes health, education, political representation, and labor participation, and the Gender Development Index (GDI), which focuses more on income levels (United Nations Development Programme, 2010).
Given that both ECA and MENA regions are predominantly Muslim, columns (3) and (6) employ models that include Muslim-nation dummy variables interacting with the proportion of women borrowers to analyze gender dynamics The findings indicate that cultural factors influence the interaction between microfinance and gender inequality; however, Islam itself does not account for this relationship These results highlight the complex role of local cultural contexts in shaping microfinance outcomes related to women's participation.
Increasing GNI per capita is associated with a reduction in gender inequality when using the Gender Inequality Index (GII) as the dependent variable However, this correlation does not hold for the Gender Development Index (GDI), likely because GDI measures income disparities specifically between men and women.
22 Muslim-country dummies are obtained from Grim and Karim (2011), who defined a nation as Muslim if at least 50 per cent of its population identified with that religion.
Conclusion
This study explores the impact of microfinance on gender equality from a macroeconomic perspective, analyzing data from 64 developing and emerging countries between 2003 and 2014 Key measures of gender inequality are based on UN indices, namely the Gender Development Index (GDI) and the Gender Inequality Index (GII) The analysis focuses on the proportion of women borrowers in microfinance programs, derived from data provided by the MIX Market, a microfinance auditing firm Although the findings do not definitively confirm a direct negative relationship, they suggest that microfinance access for women has the potential to reduce gender inequality by empowering women and enhancing their decision-making power within households and society.
Microfinance does not automatically empower women, as country-specific and cultural factors significantly influence the gender inequality–microfinance relationship In the MENA and ECA regions, this relationship is shaped by these unobserved societal characteristics rather than religion, with Islam not being a key determinant Factors such as cultural conservatism, gender norms, and the state of the microfinance industry affect how microfinance interacts with gender equality Many microfinance firms recognize the challenges women face in working outside the home and often support their efforts to start small businesses at home, sometimes through pooled resources across households Future research should further explore these cultural and societal influences on microfinance's role in reducing gender inequality.
Gender inequality, measured through composite indices of health, education, and income, suggests that increased access to credit for women can lead to improved outcomes across these areas Greater microcredit availability in developing countries empowers women by expanding their access to education, healthcare, and income opportunities As a result, promoting microfinance initiatives in developing nations is a positive step toward reducing gender disparities and fostering women’s economic and social empowerment.
Governments and international organizations in developing countries should continue supporting microcredit institutions to empower women, recognizing their positive impact However, it is essential to understand that microfinance alone does not automatically lead to women's empowerment; country-specific and cultural factors influence its effectiveness Careful consideration of these contextual factors is crucial when evaluating the role of microcredit in addressing gender inequality in the developing world.