Effects of Multiple Borrowing on Household Debt

Abstract

Rural financial markets have grown rapidly in the economy of Bangladesh and it has the largest proportion of microcredit borrowers in terms of population in the world. Currently, 65% of the rural population has access to formal finance which includes microfinance. According to the academic literature, this rapid growth of rural finance has contributed to economic development and the reduction of poverty in the last two decades. At the same time, however, the level of indebtedness of rural households has increased. One of the reasons for rising household debt is the ease of borrowing from a growing number of lending institutions. Borrowing more than one loan simultaneously is called “multiple borrowing”. This paper reviews the literature regarding whether multiple borrowing: (a) is responsible for over-indebtedness, and / (b) helps many households access additional credit access that enables them to absorb future income shocks.

Introduction

Microfinance has received much attention in development research due to its potential socioeconomic benefits to poor people. Poor people have been excluded from the credit facility of the traditional banking system due to their (low-income people) lack of collateral, the lack of information banks have about the poor people (agency problem) and higher transaction costs for banks to serve them (Armendariz & Morduch, 2010). Microfinance covers people, extensively and worldwide, who are mostly poor and probably would not have access to credit from the formal banking sector (Yunus, 2000).  For many recipients, it is a lifeline, and very often it is the only way for them to establish a business. Despite its overwhelming success in reaching the poor, some studies found that microfinance has a negative impact considering overall borrowers’ welfare.

Bangladesh’s experience with microfinance has been phenomenal. The contribution of microfinance is about 8.9 percent to 11.9 percent to the GDP of Bangladesh. Even the contribution is higher in the rural area because microfinance heavily focused on rural communities (Raihan et al., 2017). According to the Micro Credit Regulatory Authority (MRA), in 2017, there were some 24.85 million members of Microfinance Institutions (MFIs) who received an annual disbursement of BDT 1,046.12 billion with an outstanding balance of BDT 583.62 billion. There were some 783 registered MFIs in June 2017, compared with only a few in the early of nineties.

The contribution of microcredit in poverty reduction, women empowerment and fend off income shock is highly recognized. Microfinance not only contributes to the economic growth of poor borrowers but also help to adopt recovery after a natural disaster and help to empower women (Hermes & Lensink, 2011). Besides these achievements, microfinance helps to reduce income inequality and increase the happiness of borrowers. Successful microcredit borrowers have self-esteem, social recognition, expectations on future economic perspectives and enhance trustworthiness (Becchetti & Conzo, 2010)

Despite the successful journey of microcredit, many researchers allege that too much credit or too many MFIs are neither good for borrowers nor good for the economy as a whole. A recent study by the Institute for Inclusive Finance and Development (InM) based on a panel survey of 2007-2013 shows that multiple borrowing has increased manifold over the years in Bangladesh. The survey found that 51 percent of microcredit borrowers used multiple credit. But the growth of multiple borrowing has not yet caused loan recoveries of MFIs to fall (Khandker, Koolwal & Badruddoza, 2013). And multiple borrowers do not necessarily being trapped in poverty or debt (Khandker & Samad, 2013). Though the number of multiple borrowers increased over time, but at the same time multiple borrowers accumulate higher net assets over time than non-multiple borrowers (Khalily & Faridi, 2011).

A number of studies conducted on the relationship between multiple borrowing and over-indebtedness in different countries. One study in Bolivia concluded that there is no association between multiple borrowing and over-indebtedness (Gonzalez, 2008). Another study conducted by (Breza & Kinnan, 2018) has found that there is a significant causal impact in the reduction of daily wages and consumption to decrease the supply of microcredit in the different district level of India. On the other hand, findings from some other countries suggest that over-indebtedness and multiple borrowing may be positively correlated. For example, a study conducted in Sri Lanka found that alongside the growth of MFIs and increasing levels of multiple borrowing, a number of MFIs have experienced high levels of turnover among their borrowers, deteriorating portfolio quality and weakening financial performance (Tilakaratna & Hulme, 2015). Over-indebtedness and multiple borrowing are also a persistent problem for a rural household in Thailand (Bezawit & Hermann, 2017). However, I was unable to find any research papers on the bad impact of multiple borrowing on the microcredit borrowers in Bangladesh.  

The purpose of this study is to review the phenomenon of over-indebtedness among microcredit borrowers in Bangladesh. The next subsection of the paper defines the concept of multiple borrowing. Subsection 2 presents the relevant literature on the multiple borrowing practice among the MFIs clients in Bangladesh and subsection 2.1 explain the probable causes of multiple borrowing. In subsection 3 describe the present status of microcredit in the world and in subsection 3.1 I focus on the present status of microcredit in Bangladesh. In section 4 describes the impact of multiple borrowing from MFIs and over-indebtedness in Bangladesh. Finally, in section 5 have a concluding remark.

Definition of Multiple Borrowing and Over-indebtedness

There is a debate over the definition of multiple borrowing in the literature. I can define multiple borrowing as the practice of borrowing from different or the same sources simultaneously. When households take multiple loans from one financial institution, several financial institutions or both formal and informal credit sources including MFIs simultaneously, we can define the households as multiple borrowers. In Appendix 1 we present the different forms of credit lending models that households may borrow and MFIs follows throughout the world.  Households could use different microcredit products to be regarded as multiple borrowers e.g. investment loans for cow breeding, education loan, working capital loan, fishing cultivation loan, poultry loan and the like from the same or a different financial institutions. Alternatively, households may take a loan from one MFIs to service debt owed to another lender. For example, households may adopt multiple loans from both formal and informal credit sources as a substitution strategy to pay down expensive loans (Chen et al., 2010; Gine, 2011; Wampfler et al., 2014). Finally, households may borrow from relatives and neighbors to mitigate income shock or meet short term family demands. In nutshell, a household is identified as multiple borrowers if the household has multiple active loans outstanding simultaneously regardless of the source of the loan.

There is also a controversy in the literature over what constitutes over-indebtedness and how it ought to be measured.  A number of authors conclude that finding a single optimal measure that captures every aspect of over-indebtedness seems is very difficult (Betti et al., 2007; D’Alessio & Iezzi, 2013; Schicks & Rosenberg, 2011). However, recent studies found a debt service ratio (DSR) is a better measurement of over-indebtedness. The DSR indicator is defined as the proportion of annual gross income that a household must devote to service its annual debt obligation (ECB, 2013). The DSR considers both the annual interest payment and the principal repayment as a debt servicing cost, while annual gross income includes household earnings from all income-generating activities such as crop and livestock production, self-employment, off-farm employment, and return on assets (ECB, 2013). Based on the DSR indicator, a household is considered to be over-indebted when its annual debt repayment obligation in relation to income surpasses a certain threshold, commonly set at 40% or 50% (Banbula, Kotula, Przeworska, & Strzelecki, 2016; Bryan, Taylor, & Veliziotis, 2010; Disney et al., 2008; D’Alessio & Iezzi, 2013;Muthitacharoen, Nuntramas, & Chotewattanakul, 2015; OXERA, 2004).

Since our analysis takes place at the household level rather than individual level, throughout the paper when we refer to borrowers we mean borrower households rather than individual borrowers. The same applies when we discuss multiple borrowers, so-called households and not individuals.

Review of Literature on Multiple Borrowing

Several studies focus on the importance of multiple borrowing and also find evidence to suggest that multiple borrowing may be linked to the phenomenon of borrower default in different countries. However, a number of studies on Bangladesh did not find any long terms bad consequences of multiple MFIs borrowing on the borrower. According to a study by InM and Palli Karma Sahayak Foundation (PKSF) on overlapping loan among microcredit borrowers in Pathrail union in Tangail district of Bangladesh, 59 percent of the households had multiple borrowing and 31 percent of the individual members had multiple borrowing. The study found that default in previous loans forced households to take multiple loans, and the intensity of households multiple borrowing practice were higher in the villages where more MFIs were present. However, the study did not find any evidence of over-indebtedness due to multiple borrowing. Alternatively, the study concludes that there was an increasing trend in net assets and net savings on an average among the multiple household borrowers. In a recent literature, (Osmani, 2017) have evidence that those who have fallen in to debt trap due to multiple MFIs borrowing have been bottom of pile to begin with- they started their journey in life with far fewer assets and less education compared to non-trapped households. This trap may involve borrowing from MFIs, but this is because borrowing is one of the shock absorbing strategies through which households try to fend off the prospect of suffering in the face of income shocks. Sometimes this shock absorbing strategy will succeed, in which case, microcredit would have helped the borrowers to avoid a debt trap. Alternatively, this strategy may fail and the borrowers might fall into a trap. If the multiple borrowing households fall in to debt trap, it’s not because of multiple borrowing; the shocks would have been too worsened relative to the meagre resources possessed by vulnerable households, even with the support of microcredit. The access of these borrowing had also provided the poor households with means to insure against unanticipated fund needs (Rabbani & Khalily, 2012). Multiple borrowing mechanisms can be invaluable as the borrowing essentially provides household the ability to cope with unanticipated shocks through self-insurance. Multiple MFIs borrowers are not necessarily trapped either in poverty or debt trap. Moreover, participants borrowed and afterwards accumulated debt, they accumulated more assets than debt over time consequently, debt-asset ratios have in fact declined as a result of multiple borrowing facilities (Shahidur & Samad, 2013). On the other hand, borrowers who borrow from multiple sources are to pay a larger amount of their income in installments so borrowers who borrow from single source are leading better lives than those who borrow from multiple sources in terms of living expenses on food, clothing, quality of shelter, and level of health care (Chaudhury & Matin, 2002).

Multiple borrowing arises due to information asymmetry and it is analyzed in the context of the ‘principal-agent theory’. In the context of the principal-agent theory, multiple loan contracting involves hidden information that represents a moral hazard problem.  Microcredit borrowers (agents) receive multiple loans from different lenders (principals), hiding their other loan contracts from each lender. Such hidden loan contracts entail a negative externality because the unknown debt increases the level of indebtedness and reduces the probability that each loan will be repaid (McIntosh & Wydick, 2007). In the theoretical model, (McIntosh & Wydick, 2007) multiple borrowing occurs in a microcredit market with information asymmetry regarding borrowers’ level of indebtedness and a large number of credit providers. In their model, they showed that impatient borrowers with a high rate of time preference receive separate hidden multiple loans from different lenders. Therefore,  the probability of default for each loan increases because the true probability of repayment no longer depends only on one lender’s own lending but also on other unknown amounts of loan borrowed elsewhere.

Causes of Multiple Borrowing

From the extant literature, it became evident that the causes of the multiple borrowing practice could emanate from both supply- and demand-side factors. We have categorized the whole causes considering the need of borrowers who demand the credit and MFIs who supply the credit.

Multiple Borrowing

                                         Demand Side                                       Supply Side

1. Seasonal income shock             1.Excessive competition among MFI
2. Family medical emergency      2.  Inadequate loan disbursement
3. Family occasion                       3. Lack of monitoring
4. Previous loan repayment
5. Easy access of loan
6. Climate change and vulnerability
7. Capital injection in new business or existing business

Supply‐side factors

According to the MRA report (2017), it has been observed that MFIs are more aggressive in expanding the lending activities to capture the market and increase the market share. According to the report MRA, the annual average growth of borrower was 6% while the annual growth rate was 5% for new MFI creation. However, the total number of clients did not grow at the same rate as branch expansion or new MFI creation. The number of clients grew by only 5% per annum, although the loan outstanding increased by 23% per annum. This comparatively wide gap between loan outstanding and client growth is an indication of multiple borrowing (Mia, 2017). Expanding credit services through the expansion of branches to capture the market is a common practice among the big MFIs due to their high growth strategy.). MFIs’ upscaling intentions, unhealthy competition among MFIs, high interest rate cause of multiple lending in Ghana (Koomson & Peprah, 2018. The competition in the microfinance sector in Bangladesh is most likely to be localized competition. I have observed that so many firms are competing in a very certain areas like the Dhaka, Chittagong, and Rangpur districts of Bangladesh, while in other several places only a few MFIs are in operation; this reflects the regional disparity and various competition levels. The following geographical map shows the concentration of MFIs in the different district of Bangladesh. From the map, it shows that more than 60 MFIs work only in 3 districts out of 64 districts of Bangladesh.

Map: Geographical Concentration of NGO-MFIs in Bangladesh

Source: MRA-MIS-2010

Demand‐side factors

A group of low‐income people is believed to constitute the main clients of MFIs. Reasonably, this high percentage of poor people lack sufficient means for their daily consumption needs. Rural incomes tend to be subject to some degree of seasonality, whereas rational households would prefer to maintain a smooth level of consumption to the extent possible. This could result in a mismatch between the time pattern of income flow and the time pattern of consumption demand. If such mismatches occur more than once a year, it may be sensible to take an overlapping loan in order to maintain a steady level of consumption when income flow falls short of demand, without waiting for a previous loan to be paid off fully, and to pay off both the loans when income flow rises again. Although MFI loans are disbursed mainly for forming new or expanding existing micro‐enterprises a major part of the loans will most likely be spent on consumption expenditure and only a small percentage will be invested in micro‐enterprises (Mallick, 2012; Rahman, 1999; Sinha & Matin, 1998).

Setting up a new micro‐enterprise or expanding an existing micro‐enterprise requires a significant amount of financing, but MFI loans are so small in size, which motivates borrowers to take multiple loans (Srinivasan & Kamath, 2009). A recent investigation of multiple borrowing by (Mpogole et al., 2012) in Iringa, Tanzania found that the prevalence of multiple borrowing is very high there. They argued that the amounts of loans disbursed by MFIs were insufficient, and clients were allegedly borrowing from other MFIs and money lenders to meet family and business obligations. This finding is consistent with a study by (Krishnaswamy, 2007) in the Indian microfinance sector. In the first place, many enterprising borrowers might find it necessary to take overlapping loans when they wish to scale up their economic activities but cannot do so because MFIs won’t give them a large amount of money they need. We can call the situation as a credit constraint. According to (Khandker, Faruqee & Samad, 2016) 55 percent of the microcredit borrowers has credit constraint which is shown in the Table-1. In that case, a borrower might choose to borrow from more than one MFI at around the same time in order to achieve the desired scale of operation in their business.

Table1: Extent of Multiple-Source of Borrowing by Credit Constraint (%)

Source: World Bank (WB) - Bangladesh Institute of Development Studies (BIDS) survey, 1991-92 and 1998-99, and WB-InM survey 2010-11

Finally, overlapping can also occur in the face of large shocks – whether they are climatic shocks that destroy output and assets, or personal shocks such as illness or death of an income-earning member of the family, or a daughter’s wedding, and so on. Poor people often experience sudden reductions in income e.g. illness in the household or large unexpected expenses due to accidents, medical expenses, wedding expense or funeral obligations, etc. Other shocks—e.g. natural disasters or manmade conflicts that destroy livelihoods can affect many borrowers at the same time. Those shocks are also responsible for receiving multiple loans. In Table 2, shows the use of overlapping loan for different economic purposes. The table shows that almost 68 percent of overlapping loans uses for income generating and consumption smoothing purposes. Moreover, we have also found in the literature that people take multiple loans for an extra source of credit when they find lenders more flexible and reliable (Morduch and Rutherford, 2003).

Table 2: Percentage of overlapping loan used for different economic purposes

Source: (InM Survey, 2017)

Present Status of Microcredit in World

Microfinance has built a solid track record as a critical tool in the fight against poverty. Literature support that financial services for poor people are a powerful instrument for reducing poverty, enabling them to build assets, increasing incomes, and reducing their vulnerability to economic stress. Microfinance services such as savings, loans, and money transfers enable poor families to invest in enterprises, better nutrition, improved living conditions, and the health and education of borrowers’ children. It has also been a powerful catalyst for empowering women. The evolution of the microfinance industry has been driven by many factors which include the transformation of microfinance providers, the sizable supply gap for basic financial services, the expansion of funding sources supporting the industry and the use of technology.

According to the Microfinance Barometer Report, 2018 MFIs reached an estimated 139 million low-income and underserved clients with loans totaling an estimated 114 billion dollars. These levels represented a growth of 5.6% in total borrowers and a growth of 15.6% in the loan portfolio. While the loan portfolio growth was stronger than the 2016 results (plus 6.2 points), expansion in outreach to new borrowers slowed by half in 2017 compared to the 9.6% growth experienced in 2016, representing a rising average loan balance per borrower. Despite this slowdown in outreach to new clients, some MFIs are already positioning themselves to take advantage of an increasing digital user base with mobile money. Yet microfinance still reaches less than 20 percent of its potential market among the world’s three billion or poorer. Nearly three billion people in developing countries have little or no access to formal financial services.

Source: Microfinance Barometer Report, 2018

Map 2: Key figures of global financial inclusions, 2018

                                          Source: Microfinance Barometer Report, 2018

Asia and Microfinance

According to the Microfinance Barometer Report, 2018 South Asia continues to lead global outreach, accounting for nearly two-thirds of global borrowers (60%) which is shown in Table-3. That being said, the region’s growth in borrowers has slowed for a second year, going from 13.4% in 2016 to 6.6% in 2017. This result is overshadowed by significantly slower growth in the region’s largest market, India, as fallout from the November 2016 demonetization decree. Demonetization withdrew over 80% of the value of the currency in circulation and unsettled MFIs’ cash-based disbursement and reimbursement businesses. As a result, borrower growth slowed from nearly 20% in the prior year to 5.8% in 2017. Map 2 shows that despite a slowdown in Cambodia, one of its lead markets, East Asia and the Pacific led regional growth in 2017 with 10.6% growth in borrowers and 18.1% in the loan portfolio. Indonesia, the Philippines, and Myanmar all experienced an increase in borrowers greater than 15%. In Cambodia, on the other hand, the cooling effects of an interest rate cap from March 2017 led to an estimated 5% decline in borrowers by the end of 2017.

Africa and Microfinance

According to the Microfinance Barometer Report, 2018, in Africa, overall outreach to borrowers remained largely flat between 2016 and 2017, growing by 0.4%, with a similar trend in the loan portfolio outstanding (+3.5%). Several countries in West Africa, such as Benin, Senegal, and Mali, experienced single-digit growth in borrowers. In East Africa, Kenyan MFIs faced a challenging environment for lending in 2017, with prolonged elections contributing to uncertainty and an interest rate cap announcement that put a brake on lending; the combination resulted in a contraction of the borrower based by over 18% during 2017.

Microfinance and Latin America

Finally, growth in Latin American and the Caribbean, the largest regional portfolio by portfolio value (44%), slowed considerably in 2017, registering just 1.1% aggregate growth in borrowers (compared to +8.1% in 2016). Two of the largest markets, Mexico and Peru, experienced opposite trends: the total borrower outreach in Mexico shrank by 3.8% and Peruvian MFIs increased their client base by 9.5% which is shown in Table 3.

Present Status of Microcredit in Bangladesh

 Microfinance has become an integral part of Bangladesh’s economy. Microfinance targets and reaches the poor, especially women, as well as small producers and entrepreneurs, who have often limited access to formal financial credit (Khandker, Khalily & Samad, 2016).

Microfinance plays a much more important role now than it did in the 1990s when its main purpose was limited to savings mobilization and credit disbursement. Nowadays, microfinance’s role as an instrument for initiating productive employment is appreciated. After taking MFIs’ credit facilities 92 percent of women in Bangladesh involve in income generating activities while this number was only 2.5 percent before the taking credit from the MFIs (Nawaz, 2019). As its roles have become more diversified over time, microfinance has become more appealing to policymakers as an instrument for promoting social development rather than simply microcredit for employment generation and poverty reduction.

Loan Performance

Microfinance loans are well targeted in the sense that their ultimate objective is to gradually alleviate poverty from the society. The Table 4 shows that MFIs disbursed 1,046.12 billion BDT in 2016-17 which was 33% greater than the previous years. The number of borrowers rose to 24.85 million and 93% of them are women. It covers 14% of the total population of Bangladesh. There are different types of loan in this sector including general microcredit, ultra-poor loan, micro