INTRODUCTION
Problem statement
As of 2010, approximately 1.22 billion people worldwide, representing 21% of the global population, lived on less than $1.25 a day (World Bank) The focus on poverty reduction and improving living conditions has become a significant aspect of public policy globally In Vietnam, the poverty rate has seen a notable decline, dropping from 15.5% in 2006 to 10.7% in 2010, according to the General Statistics Office (GSO) However, the GSO's 2010 report indicated that poverty remains more prevalent in rural areas (13.2%) compared to urban areas (5.1%) Addressing the challenge of equitably distributing the benefits of economic growth, particularly to rural communities, is essential for reducing the inequality between rural and urban areas Therefore, enhancing support for the rural economy is crucial for fostering more equitable development.
Lack of access to funding for working capital and investment significantly contributes to poverty in developing countries, as noted by McCarty (2001) and Pham & Lensink (2002) Addressing credit constraints for impoverished rural households is a key focus of poverty alleviation strategies in developing nations, including Vietnam.
Farmers require access to credit to boost productivity and improve their living standards due to the seasonal nature of their work and associated uncertainties (Ololade & Ologunju, 2013) Microcredit has emerged as a vital tool for alleviating poverty and enhancing living conditions (ADB, 2000a; Morduch and Haley, 2002; Khandker, 2003) The role of agricultural credit is crucial for sustainable development globally, and the microfinance sector has gained prominence, particularly after receiving the Nobel Peace Prize in 2006 (Tra Pham and Robert Lensink, 2007) Numerous studies have highlighted microfinance's impact on poverty reduction by improving household welfare In Vietnam, the demand for rural credit surged following the agricultural decollectivization in 1986, leading to the expansion of microfinance and regulatory reforms that have significantly contributed to poverty alleviation over the past decades The Vietnamese government has implemented various microcredit programs through multiple channels, including banks, credit funds, money lenders, and input providers, to meet the diverse needs of clients.
Access to credit is crucial for impoverished families to enhance their productivity and living standards; however, many lack sufficient financial services Commercial banks often overlook these individuals due to inadequate collateral, prompting governments in developing countries to establish credit programs aimed at improving rural households' access to formal credit over the past forty years Despite these efforts, the lending mechanisms and credit markets remain heavily regulated, with government interventions like interest rate controls and credit quota allocations hindering their effectiveness.
Microfinance institutions in developing countries have struggled with sustainability, as highlighted by Robinson (2001) and Gonzalez Vega (2003) Agricultural development banks, established by commercial banks to provide credit to rural households deemed uncreditworthy, have often failed to meet their objectives of becoming sustainable credit providers and effectively serving the poor (Adams, Graham, and von Pischke 1984; Adams and Vogel 1986; Braverman and Guasch 1986) Additionally, lenders and borrowers face significant challenges due to risk management and transaction costs stemming from asymmetric information.
In the Vietnamese financial market, various forms of credit serve distinct borrower groups, yet many low-income households struggle to access these resources Due to credit rationing, poor households are often excluded from the formal credit sector, as highlighted by Stiglitz and Weiss (1981) To mitigate information asymmetry between borrowers and lenders, government microcredit programs, in collaboration with local People's Committees, aim to enhance the lending process and support the operation of the microcredit market.
The primary challenge for policymakers and program organizers is to enhance the efficiency and effectiveness of the microfinance system by addressing the gaps in service delivery and target demographics.
With data collected from The Vietnam Access to Resources Household Survey
The 2012 Vietnam Access to Resources Household Survey (VARHS) enhances the Vietnam Household Living Standards Survey (VHLSS) by conducting follow-up surveys on the same households, collecting detailed data on income, expenses, land use, agriculture, assets, investments, migration, climate change, and social welfare Supported by the University of Copenhagen, CIEM, ILSSA, and IPSARD, this research utilizes econometric techniques to analyze the factors influencing access to both formal and informal credit.
Research objective
This thesis aims to empirically examine the factors that affect households' access to various credit sectors and to explore the simultaneous relationship between participation in formal and informal credit markets.
Additionally, marginal effects of each independent variable on the choice of credit source are also shown in this research.
Research questions
This research is to answer two central questions:
What are determinants affecting the probability of household accessing to different types of credit sector?
Is there any evidence of a correlation between participating in formal credit and participating in informal credit at the same time?
Research Structure
This dissertation is structured into six chapters: Chapter one outlines the problem statement and research objectives Chapter two explores concepts related to demand credit and credit rationing theory, introducing key explanatory variables Chapter three delves into the history of microfinance and the government's role in it, while also examining the credit market structure in Vietnam Chapter four details the research framework, data description, and methodology employed Chapter five presents empirical models and their estimated results Finally, chapter six concludes with policy suggestions and discusses the limitations of the study.
LITERATURE REVIEW
Concept of credit
There are several and various definitions regarding the word credit as follows:
Credits are referred as loans which permit consuming in the present, in exchange for an agreement to make repayment at sometimes in the future (Pischie et al., 1983)
Obtaining credit was considered as the process of controlling over the use of money, goods and services based upon a promise to repay at a future day (Adegeye & Dittoh, 1985)
According to Ololade and Ologunju (2013), credit is defined as a method for the temporary transfer of assets to individuals or organizations that do not possess them This process necessitates evidence of a debt obligation in exchange for a loan, although transactions between friends or relatives based on trust are typically excluded from this requirement.
Microcredit which is a component of microfinance provides small loan to the poor for self –employment That generates income, helping them care for themselves and their family (The Microcredit Summit, 1997)
Microcredit aims to elevate income levels and enhance living standards in semi-urban and urban areas by offering small-scale financial services, including savings and credit, to rural households.
Microfinance is defined as small financial assistance provided to low-income individuals, encompassing various services such as savings, money transfers, payments, remittances, and insurance These services are designed to alleviate poverty and improve the financial well-being of underserved populations.
Microfinance is a development strategy that combines financial and social intermediation to support the impoverished (Legerwood, 1999) Microfinance institutions (MFIs) offer more than just loans; they also facilitate group formation, foster self-confidence, and provide training in financial literacy and management skills to empower group members.
Microcredit, often confused with microfinance, specifically refers to the provision of loans to individuals in poverty In contrast, microfinance encompasses a wider range of financial services, including loans, savings, and insurance, aimed at improving the financial stability of low-income populations.
Theory of demand for credit
The life cycle model proposed by Franco Modigliani in 1966 highlights that individuals struggle to maintain consistent consumption levels amid family size changes and future uncertainties To enhance lifetime utility, income must be strategically reallocated over time, as noted by Morduch in 1995 Consumers can finance their purchases through savings from past or present income, or by utilizing credit, which facilitates inter-temporal choices By borrowing, individuals gain increased spending power now, but they also incur the obligation to repay the loan along with interest in the future, as discussed by Soman and Cheema in 2002.
In 1986, Modigliani discussed the inter-temporal model of the life cycle hypothesis and the permanent income hypothesis, which elucidate individual consumption behavior His model assumes that borrowers have the opportunity to access loans in a perfect market.
According to the model by Chen and Chiivakul (2008), a household's consumption level is influenced not only by current income but also by lifetime characteristics, including their participation in the credit market Furthermore, it is suggested that current consumption is linked to anticipated future consumption, as consumers assess their long-term affordability, which is shaped by their savings and demand for loans (Hall).
The Cobb-Douglas production function, represented as Y = AL^α K^β, emphasizes the role of capital as a vital production input, highlighting its accessibility and affordability Profit generation in this model relies on labor (L) and capital (K) within a given technological framework (Zellne et al., 1966) This function illustrates the impact of capital and labor on production outcomes and how these factors influence income distribution (Felipe and Adams, 2005) Additionally, capital can be sourced from various credit channels, each offering different interest rates that affect the overall cost of capital.
Credit rationing theory
In 1981, Stiglitz and Weiss based on two assumptions to introduce theory of credit rationing:
(1) There is unable to differentiate level of risk associated with safe and risky borrower, and
(2) Loans are subjects to the ability of repayment ability of borrower at the end of investment period
In the microcredit market, two key issues arise from asymmetric information: adverse selection and moral hazard Adverse selection pertains to the challenges in the screening process, where transaction costs, such as interest rates, differ between reliable and unreliable borrowers Conversely, moral hazard occurs during the monitoring phase, as lenders are aware that the risk is distributed among them (Pham & Lensink, 2007).
In general, lenders judge borrower’s creditworthiness basing on a set of information that they have
The Stiglitz and Weiss (1981) model posits that while the expected returns of projects (E i) are uniform, the probability of success (𝑝i) varies among borrowers In this framework, E i s and E i f denote the returns on projects in instances of success and failure, respectively Furthermore, the bank extends loans (B) to all borrowers at a uniform interest rate (r).
A project is deemed feasible when its expected return exceeds the opportunity cost (C i) as outlined by Stiglitz and Weiss (1981) In successful scenarios, the project's return surpasses the total repayment amount owed to the bank, calculated as (1+r)*B, while in unsuccessful cases, the return falls short of the lender's repayment.
Additionally, in the case of success, the return to borrower is higher than opportunity cost: π(𝑝 i , r) = 𝑝 i [ E i s – (1+r)*B] ≥ C i (2.2)
Taking derivative of the function (2.3), (2.4) is obtained:
The first derivative of the function is negative, indicating that the expected return function is decreasing This means that as the probability of success increases, the expected return of the project diminishes.
From equation (2.2), using the implicit function theorem to differentiate r respects to p i:
Equation (2.5) shows that an increase in interest rate charged by lender leads to a decrease in probability and vice versa
Marginal borrowers are individuals who anticipate a zero return on their projects, represented by π(𝑝 i , r * ) = 0 When the interest rate (r) exceeds the threshold rate (r *), these marginal borrowers exit the market Consequently, new marginal borrowers encounter higher interest rates, illustrating the impact of this phenomenon.
Figure 2.1 Probability of success and expected returns to borrowers
Figure 2.1 demonstrates how interest rates and opportunity costs influence the expected returns for borrowers The expected returns for a marginal borrower, represented as E1, are determined by the probability of success, denoted as 𝑝 im As previously indicated, an increase in these factors can significantly affect the overall returns for borrowers.
Borrowers with 𝑝 i m’ ≤𝑝 i ≤𝑝 i m out of the market
Marginal borrowers exit the market when the expected return, E3, falls below the opportunity cost, leading to a decrease in the probability of success, p i m’, compared to the previous level, p i m.
According to Quach (2005), there is the same effect of opportunity cost on expected return
From the perspective of the bank, the expected return is: κ(𝑝 i, r)= 𝑝 i (1+r)B + (1-𝑝 i )E i f (2.6)
Differentiating function (2.6) with respects to p i :
The first derivative of the function is greater than zero It implies that an increase in probability leads to an increase in expected return to the bank
If the interest rate increases:
(1) The value of component (1+r)*B increase
(2) The value of p i decrease (according to function 2.5 shown above), which leads to lower expected return as result of withdraw of lower-risk borrowers (Stiglitz & Weiss, 1981)
Stiglitz and Weiss (1982) introduced the concept of the critical equilibrium interest rate, which plays a pivotal role in banking profitability When the current interest rate is below this critical level, banks can raise their rates without significantly losing borrowers, resulting in increased income However, if the interest rate exceeds the critical equilibrium, lower-risk borrowers may exit the credit market, leading to reduced profits for lenders Consequently, banks allocate credit at the critical equilibrium interest rate to maintain a balance between borrower retention and profitability.
Figure 2.2 illustrates the connection between lender returns and the interest rate (r) At the critical equilibrium interest rate (r ra), the supply of funds aligns with the demand, allowing banks to maximize expected returns without facing any rationing issues.
Figure 2.2 Return to the bank
Determinants of participation in microcredit programs
When households face insufficient income and wealth to support their consumption, they often resort to borrowing money (Kirchler et al., 2008) The loan acquisition process involves two key stages: initially, households seeking credit submit applications for a specific loan amount from their chosen credit sector Subsequently, lenders evaluate these applications, selecting those that meet their requirements based on the household's information and the lenders' available resources.
Stiglitz and Weiss (1981) showed that behavior of accessibility to credit is explained by demand theory (demand side) and rationing process of credit (supply side)
According to Vaessen (2002), credit accessibility in Northern Nicaragua is influenced by the interplay between household characteristics on the demand side and the attributes of financial institutions on the supply side Similarly, Duong Pham (2002) emphasizes the importance of these factors in shaping credit access.
E r focused on characteristics of both demand side and supply side in determining factor that influence on accessibility of rural credit in Vietnam
This analysis emphasizes the demand-side characteristics that influence the likelihood of accessing credit Key determinants of microcredit participation include factors such as age, gender, marital status, education, household size, income, savings amount, land size, agricultural activity, social networks, and geographic location.
The life cycle hypothesis suggests a negative correlation between age and the likelihood of obtaining a loan, a finding supported by research from M Ajugam and C Ramasamy (2007) Studies by Okurut (2006) and Mohamed (2003) further indicate that access to credit diminishes with age Younger individuals are generally more inclined to borrow than their older counterparts due to varying levels of personal risk (Fabbri and Padula, 2004; Zeller, 1994; Magri, 2002; Abdul-Muhmin, 2008; Del-Rio and Young, 2005) Additionally, younger people tend to engage in diverse spending activities, while older individuals may exhibit more conservative spending habits (Mpuga, 2008).
Research indicates a positive correlation between age and access to credit, with studies such as Tinh (2010) highlighting that the age of the household head significantly influences loan acquisition This finding is further supported by Tang et al (2010), reinforcing the notion that older individuals may have better access to credit opportunities.
Research by Banerjee et al (2010) and Bruno and Cre1pon et al (2011) indicates that a significant majority of borrowers in microcredit programs are male Additionally, studies by Nwaru (2011) and Bendig et al (2009) demonstrate that the demand for loans is negatively correlated with being female.
However, contrary to mentioned studies, Owuor George (2009) stated that being a female headed household increases probability of joining financial activities
Numerous studies have demonstrated that married individuals are more likely to secure loans compared to their unmarried counterparts, primarily due to their heightened financial needs (Kamleitneir & Kirchler, 2007; Bridges et al., 2004; Chen & Jensen, 1985; Duca & Rosenthal, 1994; Magri, 2002) Research conducted by Kenya National Fin Access (2009) further supports this, revealing that married individuals have the highest likelihood of participating in credit programs This trend is attributed to the challenges single households face in accessing credit, largely stemming from a lack of social networks (Ferede, 2012).
2.4.4 Level of household head’s education
Education significantly impacts access to microcredit programs, as highlighted by Tang et al (2010) Quach (2005) found a positive correlation between education level and loan demand Furthermore, higher education levels enhance awareness of the financial market system, making it a key determinant for accessing microcredit programs (Yehuala, 2008; Okunade, 2007; Vaessen, 2001).
Research by Khandker (2001, 2005) and Cuong H Nguyen (2007) indicates that households led by individuals with higher education levels are less likely to engage with microcredit programs in Vietnam These studies highlight that the majority of borrowers typically possess only primary or lower secondary education, suggesting a significant correlation between educational attainment and access to credit.
The influence of education on access to credit varies significantly across different sources of financing According to Bendig et al (2009), individuals with higher education levels are more likely to utilize formal financial institutions Conversely, there exists an inverse relationship between education and the reliance on informal loans.
The ideas of relationship between household size and accessibility of credit are different from studies to studies.
Research by Schreiner & Nagarajan (1998), Vaessen (2001), Ho (2004), and Quach (2005) shows a significant positive relationship between family size and household borrowing Larger households tend to demand more loans, indicating that credit allocation increases with the number of family members This finding is consistent with Nguyen's research, reinforcing the idea that household size influences borrowing behavior.
Bendig et al (2009) found that larger household sizes negatively affect the likelihood of obtaining credit, as families with more dependents, such as children and elderly members, tend to allocate a significant portion of their income to support these individuals, increasing the risk of default (Tang et al., 2010).
However, household size did not effect on getting loan in Greece (Mitrakos and Simiyiannis, 2009)
Income is a key indicator used to identify poverty, with the World Bank stating that households with higher incomes are not classified as poor Microcredit primarily targets low-income individuals, aiming to provide them with loans to enhance their earnings Consequently, those with higher incomes face a reduced likelihood of obtaining these loans This correlation is supported by research conducted by Pham & Lensink (2007) and Li et al (2011), which found that there is a negative relationship between household income and the probability of participating in credit programs, including microcredit.
It is also explained that marginal utility of consumption of poor household is high, leading to more demand for credit of the poor (Ferede, 2012)
Numerous studies suggest that high-income households have a greater likelihood of securing loans compared to their low-income counterparts (Crrok, 2001; Lin and Yang, 2005; Jappelli and Pistaferri, 2007) This disparity is often attributed to the assumption that high-income households typically possess mortgages (Ambrose et al., 2004).
The amount of savings serves as collateral, mitigating adverse selection and moral hazard in the lending process between lenders and borrowers Higher savings provide greater liquidity, enabling applicants with larger savings to access credit more easily This indicates a positive relationship between household savings and the likelihood of obtaining credit.
OVERVIEW OF MICROFINANCE SYSTEM
The role of government in microfinance
The state's influence on the microfinance sector can yield varied outcomes, including some negative consequences like restricting private sector service distribution However, it is widely acknowledged that the state significantly contributes to the growth of microfinance Its role can be categorized into three main functions: as a protector, a provider, and a promoter of the sector.
The protector role is crucial as it fosters confidence among low-income households and highlights the disparity between local credit recipients and financial institutions The state employs a legal framework as its primary tool to safeguard savers and regulate credit institutions, addressing issues of unfair competition To enhance the effectiveness of this legal environment, it is essential for the state to adopt a flexible rather than a rigid approach.
The government plays a direct role in providing microfinance services to impoverished individuals, but this involvement can lead to unfair competition for credit institutions due to preferential credit offerings To maintain a level playing field, the state should focus on supporting credit institutions and participate in payment services or savings initiatives, rather than offering preferential credit.
The state plays a crucial promotional role in the microfinance industry through both direct and indirect methods Indirectly, policies and investments serve as promotional tools that aim to create benefits for the sector, although they may not prioritize fair competition or strengthen the payment system Conversely, the direct approach involves the development of a comprehensive microfinance strategy, which includes establishing wholesale organizations and offering financial and technical assistance.
Overview of credit market in Vietnam
Vietnam's credit system is comprised of three main sectors: formal, semi-formal, and informal This coexistence enhances credit access for the poor and fosters competition among various credit suppliers in the rural market Additionally, there are notable differences in interest rates and lending practices across these sectors.
Figure 3.1 Microfinance Systems in Vietnam
The Vietnam Bank for Social Policies
Vietnam Bank for Agriculture and Rural Development ((VBARD)
The People’s Credit Funds (PCF)
Mi cr oc redi t S ys te m in V ie tn am
The formal sector in Vietnam includes state-owned organizations such as the Vietnam Bank for Social Policy (VBSP) and the Vietnam Bank for Agriculture and Rural Development (VBARD), along with the People Credit Funds (PCF) (World Bank, 2002).
According to the World Bank (2002), formal sector lending constitutes 73.5% of total lending in Vietnam, aiming to encompass the entire rural credit market This sector's primary advantage lies in its extensive network and strong ties with the Local People Committee, which is crucial to Vietnam's financial system However, despite its strengths, the formal sector has not fully met its established objectives (Phan, 2012), highlighting the need for improvements in its operations One key player in this sector is the Vietnam Bank for Social Policies (VBSP).
Established in 1995 and officially renamed the Vietnam Bank for Social Policies (VBSP) in 2002, this state-owned bank plays a crucial role in Vietnam's efforts to combat poverty Operating under the supervision of the State Bank of Vietnam, VBSP provides low-interest loans and accessible financial services to low-income households, aiming to bridge the economic gap and alleviate poverty Despite its intended mission, the bank has struggled to effectively serve the poor, with only half of its clients identified as impoverished as of 2001.
The fastest method for delivering loans to low-income individuals in Vietnam is through four major organizations: the War Veterans Union, the Farmers Union, the Youth Union, and the Women's Union These organizations are responsible for forming savings and credit groups, certifying poor households, and overseeing borrowers to ensure proper use of the loans.
As of July 2003, six programmes are offered loan by VBSP:
(1) Loan to low-income households
(3) Loan for job generation promotion (Resolution No.120/HDBT, dated
(4) Loan for foreign migrant worker
(5) Loan for difficult development areas
The outstanding loan of VBPS reached the amount of 72,660 billion VND with 7.5 million borrowers were provided (VBSP, 2009) b The Vietnam Bank for Agriculture and Rural Development (VBARD)
Established in 1988, the Vietnam Bank for Agriculture and Rural Development (VBARD) serves as the state policy bank, primarily providing credit to rural households engaged in various agricultural activities The lending process at VBARD requires collateral such as residential property, movable assets, goods, or land rights, catering to both rural households and local investment households (LIHs) Notably, access to VBARD for rural households surged from 9% in 1992 to 30% in 1994, highlighting its growing role in financial inclusion By the end of 2001, VBARD had become the largest commercial bank in Vietnam, boasting over 2,300 branches nationwide and serving 50% of the poor population seeking financial services.
VBARD primarily focuses on lending to state-owned enterprises (SOEs), viewing them as low-risk clients due to government support Despite being a key credit provider in rural areas, VBARD has not fully addressed the needs of the entire rural credit market, especially for impoverished households The underdevelopment of VBARD's operations can largely be attributed to biases in risk assessment and the complexities of the lending process.
There are three functions of VBARD:
(1) To provide credit to rural farmers and entrepreneurs with collateral
(2) To reduce transaction costs as well as increase its coverage of rural households, group lending methodology is used
VBARD offers loans to borrowers who cannot provide collateral by utilizing brokerage services from mass organizations In this system, members of the organization guarantee loan repayment to VBARD, ensuring financial support for those in need Additionally, the People's Credit Funds (PCF) play a crucial role in facilitating these lending opportunities.
In the late 1980s, following the collapse of Vietnam's rural credit cooperatives, the government established People's Credit Funds (PCFs) to support and reform the country's banking system The primary goal of PCFs is to restore public confidence in the formal rural credit system (Putzeys, 2002) Modeled on principles of self-help, self-organization, and self-management, PCFs specifically target farmers and small entrepreneurial households as their key clients (Putzeys, 2002).
By the end of 2004, over 900 individuals had access to PCFs across 53 provinces, as reported by BWTP However, the network primarily served economically prosperous areas with better infrastructure, resulting in limited coverage in poorer regions.
3.3.2 The semi-formal credit market
The semi-formal credit sector, which includes organizations like Women’s Union, Farmer’s Association, and Youth Union, is primarily funded by both international and national donors (Phan, 2012) While lacking a nationwide network, semi-formal credit providers offer distinct advantages over formal lenders by targeting the poor with subsidized credit NGOs, in particular, have emerged as effective channels for credit distribution, excelling in group lending and social intermediation, drawing on successful international practices (Dao, 2002; McCarty, 2001) Furthermore, the semi-formal sector plays a crucial role in various international initiatives aimed at poverty alleviation (World Bank, 2000, p 110).
Informal financial system consists of relatives, friends, neighbor, unregistered private money lenders, traders and rotating savings and credit associations (ROSCAs) (Quach, 2005) a Relatives, Friends and Neighbors
Informal credit typically comes with no interest and varies in repayment terms—monthly, quarterly, semi-annually, or annually—depending on the borrower's extra income and the rapport with the lender Friends, relatives, and neighbors often extend credit without requiring collateral or formal agreements, leveraging personal connections This type of credit is primarily intended to assist during emergencies or for personal expenses such as medical bills, funerals, and weddings, rather than for agricultural financing (Pham & Izumida).
Private money lenders, often affluent individuals in rural areas, offer various types of credit, including permanent, seasonal, and daily loans However, the exact number of these lenders and the extent of credit usage remains unquantified due to the informal nature of transactions between borrowers and lenders (Quach, 2005; Dao).
(2002), money lenders provide credit and negotiate for a rate about 3-10% per month c Rotating Savings Credit Associations (ROSCAs)
Rotating Savings Credit Associations (ROSCAs) serve as an informal credit channel for rural households in Vietnam, relying on individual social connections to provide credit across generations Despite their importance, ROSCAs operate outside the formal credit regulations due to insufficient social sanctions and a lack of rigorous screening processes (Phan, 2012) In Vietnam, these associations are known as "Hui" in the South and "Ho" in the North (Pham and Lensink, 2007) Members typically share common professions, such as farmers, traders, or war veterans, or belong to the same hamlet, fostering a sense of community and mutual support.
Two main forms of ROSCAs are applied to rotate the fund to participants:
Random allocation is a process where each participant contributes a fixed amount of money periodically, creating a collective fund This fund is then randomly distributed to one member in each cycle, with previous winners excluded from subsequent rounds The cycle continues until every participant has received the funds at least once, ensuring equitable distribution among all members of the group.
RESEARCH METHODOLOGY
Research process
Research process includes six steps as presented in Figure 4.1
The initial step in the research process is to assess the current role of microcredit as a crucial element in global development strategies This involves identifying existing challenges that need to be addressed to enhance its effectiveness and serves as the focal point of this study.
The second step in the research process is the literature review, where relevant literature pertaining to the research problem is examined This review highlights how previous studies were conducted and their findings, providing a comprehensive understanding of the problem statement.
The third step in process of research is to present model of research This section presents the frame of research with all variables in econometrics analysis
The fourth step in process of research is to collect data for variables related to econometrics model from VARHS source
The fifth step in the research process involves discussing the econometric methods utilized, specifically the bivariate probit model, which is employed to analyze the factors influencing access to credit in both the formal and informal sectors This section also presents the research findings.
The last step in process of research presents conclusion from empirical evidence after data analysis and recommends policy for improving operation of credit marker.
The data
The research utilizes data from the 2012 Vietnam Access to Resources Household Survey (VARHS), supported by the University of Copenhagen, CIEM, ILSSA, and CAP-IPSARD This comprehensive data collection involved a meticulously designed questionnaire that gathered extensive information on household characteristics, expenditures, income, wealth variables, demographics, education levels, employment, financing, and production activities The insights derived from VARHS are crucial for informing policymakers and contribute significantly to both Vietnamese and international research, particularly in the realm of rural development.
Table 4.1 presents various claim-type combinations represented by binary triples Each element of the triplet serves as a dummy variable for distinct credit sectors: Formal, Semi-formal, and Informal A value of 1 indicates a participant's involvement in a particular sector, while a value of 0 signifies non-participation.
Table 4.1 and Figure 4.2 present a comprehensive overview of household participation in the three credit sectors Out of 1,522 households involved in the rural credit market, 724 accessed formal credit, representing 47.57% of the sample This indicates that the probability of participation in the formal sector is 47.57%, while participation rates for the semi-formal and informal sectors are 4.47% and 27.14%, respectively.
Table 4.1 Summary of Participation in different credit sectors
(Formal, Semi-formal, Informal) Frequency Joint probability
Figure 4.2 Participation in credit sector
The estimations of unconditional and conditional for rural credit resources are calculated from Table 4.1 are presented in Table 4.2
Table 4.2 Conditional and Unconditional Credit Participation Probabilities
The semi-formal sector exhibits the lowest marginal probability at 8.55% within the sample, while the formal sector holds the highest marginal probability at 67.61% In contrast, the informal sector has a marginal probability of 46.39% Notably, the probability of simultaneously observing all three sectors is just 1.17%.
Formal onlySemi-formal onlyInformal onlyFormal & SemiformalSemi-formal & InformalFormal & InformalAll
The analysis reveals significant correlations among the three credit sectors Specifically, 66.67% of the sample engages in formal credit, while the likelihood of households participating as lenders in the semi-formal and informal sectors drops to 38% and 40%, respectively Additionally, households involved in either the semi-formal or informal credit sectors are less likely to participate in other credit sectors This indicates that the probability of participation in any given sector declines when alternatives are available, highlighting a clear relationship among different credit participation choices.
This research examines various variables, including age, gender, marital status, education level, household size, income, savings amount, land size, agricultural activities, social networks, and location Summary statistics for these explanatory variables are presented in Table 4.3.
Variables Mean Std Dev Min Max
Data Analysis Method
Household choices across various credit sectors may be interconnected, as it is common for households to utilize multiple forms of credit simultaneously Consequently, estimating probit equations for sector participation individually appears to be an inefficient method, as it overlooks the potential relationships between error terms.
The multivariate probit regression model is often favored over the multinomial logit regression model for estimating the impact of independent variables on various categories of dependent variables with unordered multiple choices This preference arises because the multivariate probit model simultaneously captures the effects of a set of explanatory variables on different choices while also accounting for the relationships between these decisions Furthermore, it permits the error terms, which represent unmeasured factors, to be freely correlated, enhancing the model's robustness and accuracy.
2005) There may be negative correlation which will be presented by the negative value of rho if two outcomes are substitutability, complementarity otherwise
The multivariate probit model differs from the univariate probit model by accounting for potential relationships among various credit sources and the correlation of unobserved disturbances in adoption functions Since the households observed are consistent across equations and there may be substitutability or complementarity among these sectors, the error terms reflecting unobservable factors in these estimations are likely to be correlated.
From the literature review, multivariate probit model is employed to investigate which determinants affecting on demand for different credit sources
The study reveals that households engaged in the semi-formal credit sector represent merely 10 percent of the total sample, indicating a minimal impact of semi-formal lending on other lending sectors Consequently, this research focuses on exploring the factors influencing access to both formal and informal credit sectors, while also examining the relationship between these two types of lending.
This study examines the correlation between formal and informal sector participation as dependent variables using a multivariate biprobit model Two binary variables are created to represent formal lenders (F) and informal lenders (IF) in order to analyze the factors influencing access to each credit sector.
Multivariate Probit regression utilizes Maximum Likelihood Estimation (MLE) alongside Likelihood Ratio (LR) and Wald tests to analyze the factors influencing household participation in various credit sectors This estimation involves two binary choice equations, employing the same explanatory variables for simplicity.
X ’ presents for explanatory variables including age, gender, marital_stt, edu, hhsize, income, savingamount, landsize, agriculture_act, network and location
F * , IF * are latent variables for the participation in formal credit sector and informal credit sector
F and IF which received value of 1 if these respective latent variable are greater than 0 and zero otherwise
In a probit estimation model, the indicator variable IF equals 1 when IF is greater than 0, and 0 otherwise The error terms εF and εIF are assumed to follow a multivariate normal distribution with a mean of zero and a covariance of 1, as the variance of the error term cannot be separately identified for each probit estimation (Greene, 2003).
There are four possible outcomes in bivariate probit model:
P 11 = Pr[F=1,IF=1] = ∫ −∞ β F X ′ ∫ −∞ β IF X ′ ϕ 2 (β F X ′ , β IF X ′ , ρ)d(β F X ′ )d(β IF X ′ )
(β F X ′ , β IF X ′ , ρ) = exp {−0.5[(βFX’)2 + ((βIFX’)2 + ρ(βFX’)( βIFX’)]}/(1 – ρ2)
RESULTS AND DISCUSSION
Estimation of determinants of microcredit participation
The model is estimated by carring out in Stata presented in Table 5.1
Table 5.1 Determinants of accessibility to formal and informal credit sector
Variables formal credit informal credit
Coefficient Std Error Coefficient Std Error
***significant at 1 percent, **significant at 5 percent, * significant at 10 percent
The study identifies that out of eleven explanatory variables—age, gender, marital status, education, household size, income, savings amount, land size, agricultural activities, network, and location—seven factors significantly influence a household's likelihood of participating in formal credit: age, education, household size, income, savings amount, land size, and location In contrast, the determinants for accessing the informal credit sector differ, with seven variables—age, gender, marital status, household size, income, savings amount, network, and location—proven to be statistically significant in determining participation in informal credit sources.
The relationship between age and the likelihood of obtaining formal credit is statistically significant at the 5% level, with older individuals having a higher probability of access, which declines as age decreases In contrast, informal lenders tend to favor younger borrowers over the elderly, aligning with studies indicating that access to rural credit is negatively related to age One reason for this trend is that younger individuals often lack the accumulated assets required by formal lenders for credit, leading them to seek loans from informal sources such as friends and relatives, who do not require collateral Conversely, older individuals typically possess sufficient assets to meet the collateral requirements set by formal lenders.
Younger individuals often have a stable income, which reduces the risk for informal lenders Additionally, these lenders provide opportunities for young family members, who represent the most productive generation in their careers.
Gender may not be a reliable indicator of formal credit borrowing likelihood, especially due to government policies aimed at reducing gender inequality in rural credit markets These policies ensure equal rights for applicants, promoting economic empowerment and development regardless of gender However, gender discrimination persists in informal lending, with a statistically significant negative impact on credit access for female-headed households Research indicates that informal lenders tend to favor male-headed households, aligning with Wabei's (2012) findings that women often lack the collateral available to men.
Estimates indicate a positive correlation between marital status and the likelihood of credit rationing in the informal sector Research by Ferede (2012) shows that single household heads have a reduced probability of accessing informal credit due to a lack of social networks This finding is supported by various studies, including those by Kirchler, Kamleitner, and Kirchler (2007), Bridges et al (2004), and Magri (2002).
Education does not influence the likelihood of obtaining informal credit, while it significantly impacts demand for formal credit, as evidenced by a 1% significance level Higher educational qualifications increase the probability of accessing the formal credit sector, aligning with existing literature that suggests borrowers with fewer years of schooling tend to have lower credit demand (Khandker, 2001; Cuong H Nguyen, 2007) Furthermore, higher education is associated with increased income and better job opportunities, making formally lenders more inclined to favor educated borrowers due to their lower default risk.
Research indicates that larger families are more likely to access formal credit, while they experience a significant decrease in participation in the informal credit sector This phenomenon is attributed to government policies aimed at supporting larger families, which often have more dependents such as children and elderly members, making them more susceptible to economic shocks.
Informal lenders operate without requiring collateral or formal documentation, relying instead on informal agreements between lenders and borrowers They assess the borrower's ability to repay by considering family size and dependents, as larger families with many children or elderly members who are not in the labor force pose a higher risk of default Consequently, informal lenders prefer to extend credit to families with fewer dependents, minimizing their lending risk.
The relationship between household income and credit allocation in both formal and informal sectors is significant, with differing impacts on each Research indicates that household income positively influences the demand for informal credit, particularly at a 5% significance level This suggests that higher income households have greater opportunities to borrow from the informal credit sector, aligning with findings from studies by Crrok (2001), Lin and Yang (2005), and Jappelli and Pistaferri (2007), among others.
A study from 2004 indicates that high-income households are more likely to access loans compared to low-income households In contrast, the informal credit sector shows a negative relationship between income and credit constraints in the formal sector The findings suggest that households with lower incomes are more inclined to borrow, as they experience a higher marginal utility of consumption Literature on rural credit highlights that microfinance primarily aims to provide capital for the poor to enhance their living conditions (World Bank, 2010), which results in lower chances of credit access for high-income applicants These research findings align with previous studies by Pham & Lensink (2007), Li et al (2011), and Ferede.
The estimation results indicate that the amount saved significantly influences credit participation in both formal and informal sectors at a 1 percent significance level Specifically, the negative coefficient for the savings variable in the informal sector suggests that households with larger savings are not perceived as poor In contrast, there is a positive relationship between savings and participation in formal credit This aligns with existing research, which highlights that formal credit often requires collateral; thus, borrowers with higher savings are more likely to secure loans from formal sources The findings of this study are consistent with the literature on this topic.
Table 5.1 reveals a statistically significant relationship between rural credit participation and land size at the 1 percent level Larger land holdings serve as collateral for credit borrowing, facilitating access to the credit market, particularly the formal sector, which requires collateral to mitigate risk This finding aligns with the research conducted by Vu (2002) and Zeller (2001).
In 2003, researchers including Ravi found that households with extensive land holdings require increased credit for agricultural production However, borrowing from the informal sector is significantly negatively correlated with land size at a 1% significance level This suggests that households with larger land sizes are often not perceived as poor (Quach).
The formal credit sector remains unaffected by social networks; however, social capital significantly influences participation in the informal credit sector, with a 5% level of significance The positive correlation between social networks and borrowing indicates that households involved in organizations are more likely to secure loans from informal sources This suggests that a household’s network acts as social collateral, essential for informal lending, which relies heavily on trust and personal relationships These findings align with the research of Okten and Osili (2004), Fafchamps (2000), Karlan (2009), Bui (2010), and Grootaert (1999).
The analysis indicates that rural households have a higher likelihood of accessing formal credit compared to their urban counterparts, with the location variable showing a statistically significant positive coefficient at the 1 percent level However, it is noteworthy that more urban households engage in the informal credit sector This trend can be attributed to the higher living standards in urban areas, which provide urban families with more disposable income to lend, thereby assisting those in need.
Estimation of conditional marginal effects
Table 5.2 Marginal effects for conditional probability of formal sector participation
Variables dy/dx Std Err P>|z| X
The likelihood of a household engaging in the formal sector is estimated at 40.18%, assuming it is already involved in the informal sector, as indicated by the calculation Pr(formal=1|informal=1) = 0.40 presented in Table 4.2.
Command predict(pcond1) is employed to compute conditional marginal effects for each independent variable on formal sector participation with given informal participation sector
The findings indicate that with each additional year of education for the household head, the likelihood of accessing the formal credit sector increases by 1.7%, assuming all other variables remain constant.
In term of marital_stt, the opportunity to get credit from formal source for married household head is 18.2 percent higher compared to that for unmarried household head
The location of a household significantly influences its access to formal sector credit, particularly for those engaged in the informal credit market Research indicates that households situated in rural areas have a 31.6 percent higher likelihood of receiving loans from formal lenders, assuming all other factors remain constant.
The derivative dy/dx for the income variable is -0.0002, indicating that when controlling for other variables at their average levels, the likelihood of accessing formal credit decreases by 0.02% for every 1,000,000 VND increase in household income, assuming the household is already engaged in the informal sector.
Savings serve as formal collateral for borrowing, establishing a positive link between savings and access to formal credit Specifically, for households engaged in the informal credit sector, an increase of 1,000,000 VND in savings enhances the likelihood of obtaining formal credit by 0.1 percent.
The likelihood of a household engaged in the informal sector gaining access to the formal sector rises by 0.3 percent for every additional 1000m² of land owned, assuming all other variables remain constant.
Similarly, marginal effects for conditional probability of informal participation sector with given formal participation sector, Pr(informal=1|formal=1) are shown as Table 5.3
Table 5.3 Marginal effects for conditional probability of informal sector
Variables dy/dx Std Err P>|z| X
The likelihood of a household engaging in the informal sector is estimated at 27 percent, particularly when it is already involved in the formal sector, as indicated by the probability Pr(informal=1|formal=1) = 0.27, as shown in Table 4.2.
Command predict(pcond2) is employed to compute conditional marginal effects for each independent variable on informal sector participation with given formal participation sector
The age of a household head negatively impacts participation in informal credit, especially when the household is already engaged in the formal sector Specifically, for each additional year of age, the likelihood of accessing informal credit decreases by 0.4 percent, assuming other variables remain constant.
Marital status significantly influences the likelihood of households engaging with informal credit providers Specifically, when a household is involved in the formal credit sector, married households are 18.7 percent more likely to seek assistance from informal sources compared to their unmarried counterparts, all else being equal.
Research by Yehuala (2008), Okunade (2007), and Vaessen (2001) indicates that an increase of one year in education correlates with a 0.9 percent rise in the likelihood of engaging with informal credit sources, assuming other variables remain constant.
The coefficient dy/dx for the household size variable is -0.02, indicating that, when controlling for other variables, the likelihood of informal participation decreases by 0.2% for each additional member in a household that is already engaged in the formal credit sector.
The relationship between social networks and location significantly influences household borrowing from the informal credit sector According to estimates, households engaged in the formal credit sector will see an increase of 5.7% and 6.1% in their access to informal credit as the independent variable shifts from zero to one.