Does Microfinance Cause or Reduce Suicides?
The need for support groups for males[1]
Arvind Ashta[2], Saleh Khan[3] & Philipp Otto[4]
15th May, 2011
ABSTRACT
In a background ofborrower stress and suicides by microfinance borrowers in India, a review of literature indicates that poor countries should have lower suicides but that increase in suicides accompanies a country’s development. Empirical evidence on India as well as international comparison is presented.We find that suicides, especially male suicides, seem to accompany microfinance growth and penetration.This could therefore point towards specific forms of egoistic as well as anomic tendencies increasing suicide rates, possibly alongwith changing gender roles in the family structure. The major recommendation stresses the requirement of (new) support groups for male population.
Keywords: microfinance, impact, suicide, stress, India,
Does Microfinance Cause or Reduce Suicides?
Policy recommendations for reducing borrower stress
INTRODUCTION
For some time now, a spate of suicides by microfinance borrowers has been making the news (Microfinance Focus, 2010;Hulme, 2000). Already in 2006, there were warnings that the pressures of hitting outreach numbers at any cost, pressures attributed to the governments of Andhra Pradesh and Tamil Nadu, was creating immense stress on NGOs who feared their subsidies would be cut if they don't expand sufficiently (Guerin et al., 2006). Indeed, Andhra Pradesh (along with Karnataka) is the state with the most saturated microfinance market in India. Does this quick expansion come with a cost?
Microcredit is the lending of small amounts of money to the poor and financially excluded to enable them to increase income and smooth consumption. The sector has experienced fast growth rates and has been considered as an important tool of economic development, in conjunction with other initiatives such as education, health, sanitation, infrastructure and public governance. Based on over 3552 reporting MFIS, the microcredit summit campaign indicates that the sector has an outreach of 155 million people in 2007 (Daley-Harris, 2009). Table 1 provides details of median MFIs from a sample of over one thousand MFIs reporting to Microfinance Information Exchange (MIX)[5]. The median MFI has about a hundred employees who lend to about 10,000 people. The average loan size is USD 523. The MFI lends at about 28% per annum and makes a positive return on both equity and assets. The portfolio at risk (30 days) is about 4.76%. A Nobel Peace Prize has been awarded to an MFI institution and its founder in 2006 to indicate the importance of this initiative. Millions of people are working in this sector, often as volunteers or working at lower than market wages and returns, because they believe that microfinance can create a difference at the grassroots level by making people responsible (as opposed to handouts). And, now, one wonders whether the initiative has not gone awry?
Table 1: Characteristics of a Median MFI institution (2009)
Institutional characteristics / All / IndiaNumber of MFIs reporting / 1132 to 1146* / 88
Personnel / 104 / 287
Number of MFIs / 1146 / 88
Number of active borrowers / 10,487 / 65,008
Percent of women borrowers / 63.39% / 100.00%
Gross loan portfolio USD / 4,902,526 / 8,405,779
Average loan balance per borrower USD / 523 / 144
Number of MFIs / 1141 / 88
Return on assets / 1.50% / 1.81%
Return on equity / 7.27% / 10.54%
Yield on gross portfolio (nominal) / 27.93% / 23.97%
Portfolio at risk > 30 days / 4.76% / 0.51%
source:
*All 1146 did not provide information for all indicators.
Recently, Microfinance Focus (2010) has reported that there have been 54 suicides by microfinance borrowers in the State of Andhra Pradesh alone. When there are one or two suicides it could be chance or coincidence, but adding up to 54? Who can be liable?
The English law of torts distinguishes between strict liability and vicarious liability. Strict liability is liability that does not depend on actual negligence or intent to harm, but that is based on the breach of an absolute duty to make something safe. The question here is when does microfinance become dangerous? Which practices are to be avoided? What features are required to make microloans safe? Who is responsible for safety?
This brings us to the concept of vicarious liability (the etymology of which is that the vicar is an agent through whom the Pope exercises authority). This is liability that a supervisory party (such as an employer) bears because of the actionable conduct of a subordinate or associate (such as an employee) because of the relationship between the two parties.
Putting the two concepts together, if certain microfinance practices can now be considered dangerous, and if the CEOs of microfinance firms take no action to account for the involved risk, they could be held to be vicariously strictly liable.
In this context, one can understand that the Andhra Pradesh government has passed an ordinance (2010) limiting interest rates at a maximum of the loan amount and curtailing unethical collection practices.
As can be seen from Table 1, the median MFI in India in 2009 was almost three times larger in terms of employees than the median worldwide. The median Indian MFI has six times more borrowers, thus indicating that employees also serve more borrowers. Although the average loans amounts are much smaller in India, only about $ 150, the interest yields are much lower at around 24% instead of 28%. Despite this, Indian MFIs have a higher return on equity.
Why are Indian MFis more efficient than global MFIs? Essentially, this is the result of the peculiarity of the Indian model which is based on Self-Help Group (SHG) and Joint Liability Group (JLG). It is estimated that 58% of Indian Microfinance outstanding loan portfolio is based on Bank-SHG linkage model and another 34% is based on NBFI lending, primarily to JLG. The balance 8% is with the not-for profits (trusts, societies) (Malegam, 2011). In the Bank SHG linkage model developed by NABARD in 1992, 5 to 20 women get together. They pool their savings and give loans to each other. The banks supplement the credit available. This Bank-SHG model has about 65 million customers (70%).
Indian MFIs, (with 27 million customers) operate primarily in the NBFI – JLG model which is similar. The considerable costs in developing SHGs is incurred by the government. If MFIs piggy-bank on this State subsidized SHGs, and create JLGs with the same women borrowers as in the SHG, they lower their transaction costs considerably. Moreover, since 75% of the financial needs of NBFI MFIs is provided by banks and financial institutions like SIDBI, this may also lower the total costs of MFIs. All these features make Indian MFI more cost-effective and more profitable than those in the rest of the world, even though yields are lower than global interest rates.
These yield figures correspond with statements in the press that the microfinance industry in India has been charging interest rates with a stated APR of between 25% and 30%. However, more detailed calculations by the Malegam Committee (Malegam, 2011) incorporating compulsory deposits indicate that Indian MFIs are charging an average of 37% of the mean outstanding portfolio. Of these, 13% is the cost of borrowing, 8% is the cost of operating staff, other overheads were about 6% and loan loss provision was 2%, leaving a profit margin of about 11% (some rounding off differences seem to persist in their report).
In view of the furor created in the press, Indian MFIs have responded by reducing interest rates by one or two percentage points. For example, the largest Indian MFI, SKS, reduced its interest rates from 26.5% to 24.5% (Microfinance Focus (2010)). SKS has been in the news this year for its Initial Public Offering which was a big success but which invited criticism for its financial approach to microfinance. It is now also in the news for the high number of suicides by its borrowers in Andhra Pradesh. The question is whether a reduction in interest rate is enough to cage the beast leading to suicides.
In this paper, we go beyond the Indian question. We first look at the literature on suicides. Then, we analyze longitudinal and cross-sectional empirical data to check for relationship with suicides. Based on this, four related questions are put into context. First, do rapid growth and sustainability aims make MFIs forget the human angle? Second, is social pressure necessary to make people repay? Third, could these people be saved without Microfinance or would they die even earlier? Fourth, finally, what can authorities do, and, can different angles of action be proposed? We start with a brief background literature review of research on suicides.
The reader should keep in mind that MIX data for India and RBI terminology often limits microfinance on NBFIs and NGOs. The Bank-SHG linkage model is considered apart. However, this Bank-SHG sector is 54% of total lending to the poor. Therefore, if suicides in India are increasing, it may be as much due to the public sector initiatives of the banks as due to the MFIs. We will revert to this remark in the discussion.
BACKGROUND ON SUICIDES
Suicide is defined as the act of purposefully killing oneself. Why do people suicide? The French sociologist Emile Durkheim (1858-1917) was the first to raise his pen to answer this question in 1897 in a three volume book called "Le Suicide" (Durkheim, 1897). He categorized the causes as egoist, altruist, anomic and fatalistic. These could be considered as isolation from groups, not being able to meet group norms and targets, lack of social direction due to social upheaval, and oppression, respectively. Examples for each could be found in today's world. For example, Belarus, the world leader in suicides and many former Soviet countries have a very high suicide rate and this could be considered “anomic suicide”. Japan and South Korea are very high performing countries and the failure to meet these exacting standards may cause people to commit “altruistic suicide”. In general, Durkheim found higher suicide rates for males than for females, for single people than for couples, for parentless than for parents, for Protestants than Catholics or Jews, and for economically more active and urban communities than for rural communities. Thus, social solidarity, and not just economic solidarity keeps people happier. This also explains why richer people, more financially independent, have a higher propensity to commit suicide.
The major causes of suicides could be regrouped as mental disorders or afflicted problems. Mental disorders could be the result of a depression or other mood disorders, of alcoholism or other drug addiction, and of schizophrenia or other personality disorders (Nashold and Naor, 1981). The main afflicted problems are financial difficulties, interpersonal conflicts like divorce, and unemployment. These are at least the most researched non-medical causes of suicide. Of course, some of these could be rather country specific. For example, in Ireland , of those who had attempted suicide 50% had a history of child abuse, 55% had a history of depression, and 67% had experienced learning difficulties (Birchard, 2001).The suicide statistics from different countries highlight different variables.
There is high correlation between suicide and unemployment in the US (Yang and Lester, 1992, Yang and Lester, 1994, Yang and Lester, 1995, Yang and Stack, 1992), Japan (Watanabe et al., 2006, Koo and Cox, 2008, McCurry, 2006), and some other countries (Yang and Lester, 1996). There is an inverse relation between suicides and female participation in the labor force (Yang and Stack, 1992). It is also found that job losses double the suicide rate for both men as well as women and increases the deaths due to alcoholism (Eliason and Storrie, 2009). In the US, suicides also increase with age (McKeown et al., 2006), perhaps because the cost of maintaining good health increases (Koo and Cox, 2008). Also unemployment creates more impact on suicide rates for older people (Schapiro and Ahlburg, 1982). Further on, some evidence indicates that higher IQ may be linked to higher suicides (Voracek, 2005). On the contrary, it has been reported that married (and even unmarried people) have significantly lower suicide rates than divorced or widowed people (Smith et al., 1988).
Even in the absence of unemployment and no divorce, there would still be some “natural rate of suicide”, which differs from country to country. Certainly, this rate depends on the method used to calculate it. A recent study finds that the natural rate varies between 5.0 in Poland to 24.6 in Russia using regional data of 11 countries, while its range could be from 2.9 in Norway to 25.0 in Japan using time series data of 13 countries (Yang and Lester, 2009).
Although poverty is not considered by itself a reason for suicide, it may render a person fragile and limit his capacity to cope with his mental fragility. In fact, the presence of several risk factors may increase the probability of suicide (Watanabe et al., 2006).In Japan, a country with traditionally high suicide rates, there is a strong cultural relation between social stigma and suicides. There is a causal relationship between suicides and personal bankruptcy (Watanabe et al., 2006). Many of these suicides may be due to joint liability debt contracts taken by the self-employed (Chen et al., 2007).
China, India and Japan account for about 40% of the 1 million suicides in the world. Their joint population is about 35% of the world's population. Putting this together, poor Indians may be fairly suicide prone to start with. Financial difficulties may push them across the threshold. To investigate if this is the case, we look at suicide rates.Suicides rate vary from country to country. Belarus has the highest suicide rate followed by South Korea. Japan's male suicide rate is double that of the US male suicide rate (Koo and Cox, 2008). In terms of suicide rate, India is 43rd among 106 countries with an annual suicide rate of about 10 per 100,000 people, slightly higher for men and slightly lower (than the average) for women. Even within a country, suicide rates may vary. For example, US suicide rates in 2003 average 10.5, but it varies from 14.7 in the West to 8.6 in the North East (McKeown et al., 2006).
Longitudinal data indicates the overall suicide trends vary with periods and regions. In the US, death rates came down between 1950 to 1993 for young children largely due to better medicine but not for adolescents. But suicide rates increased, especially for minorities and poor people (Singh and Yu, 1996a, Singh and Yu, 1996b). This is a continuation of increasing suicide trend for younger people (below 35), while for older people, suicide rates decreased from 1940 to 1982 (Schapiro and Ahlburg, 1982). In Ireland for example suicide rates quadrupled for young adults from 1980 to 2000 (Birchard, 2001).
According to Jain (2009), suicides are growing in number in the rural countryside of India, with more than half the suicides in just five States: Maharashtra, Andhra Pradesh, Karnataka, Madhya Pradesh and Chhattisgarh. The causes include desertification of farms (because children go to the city leaving their parents without social networks), the vagaries of nature creating a high risk environment, and the global agricultural surplus with inelastic demand (creating economic hardships). Mechanization and the associate debt burden have added to the stress of farmers. In Durkheim's terms, the suicides seem to be caused by anomic reasons (changing social structure) as well as egoistic reasons (increased autonomy).
Another study (Gruère et al., 2008) examines whether genetically modified cotton production is linked to farmer suicides using different data sets. The conclusion depends on the data set which is used. However, the study indicates that in Andhra Pradesh, moneylenders are the main source of debt (53.4% of loans taken). This indebtedness with high interest rates coupled with crop failure is the cause of stress. We know that ever since Emile Durkheim (1897), regulators have been conscious that an increase in economic activity leads to more stress and more suicide of the egoistic kind (ie. people tend to become more independent). Nevertheless, if the vast majority of regulators with different ideologies across the world and across decades have gone in for development, it is because other kinds of satisfactions, including the lowering of other kinds of deaths, are possible with economic development. Thus, the debate on the GMO-suicide link is just a case in point. The same could be true of Microfinance.
Andhra Pradesh has a population of about 85 million people. Expected suicides of 8500 per year or 700 per month would be regarded as "normal". Andhra Pradesh is also a fairly saturated microfinance market. Therefore, 54 suicides in a month in Microfinance may not be significantly different from the rest. SKS and Spandana, the two biggest Microfinance Institutions (MFIs) in India, jointly have over 10 million microfinance borrowers from all over India. There are about 6.7 million microfinance borrowers in Andhra Pradesh (Ratwatte, 2010). With a suicide rate of 10 per 100 000, a 670 suicides per year would be within the expected statistical rate. This works out to 56 suicides a month. Thus, the 54 suicides in October 2010 seem to be about normal for India. Except the fact that suicide rates for women should be lower and microfinance should be affecting mainly women.