Vous êtes sur la page 1sur 21

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/235443327

IMPACT OF MICROFINANCE ON RURAL POOR: AN EMPIRICAL


INVESTIGATION FROM INDIA

Article · December 2009

CITATION READS

1 6,326

2 authors, including:

Bimal Sahoo
Indian Institute of Technology Kharagpur
49 PUBLICATIONS   176 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Defining and Measuring Informal and Decent Jobs in India View project

All content following this page was uploaded by Bimal Sahoo on 26 May 2014.

The user has requested enhancement of the downloaded file.


Published in: Microfinance: performance Evaluation and Enterprise Development”
(ed.) by Lazar D. and Deo M. , Allied Publishers Pvt. Ltd, Chennai, 2009
IMPACT OF MICROFINANCE ON RURAL POOR: AN EMPIRICAL
INVESTIGATION FROM INDIA
S. Mohapatra1
B.K Sahoo2
Abstract:
This paper examines the impact of microfinance on income and employment in
general, and on socially disadvantage group (SC and ST) in particular. The study
primarily based on the field survey conducted by the authors in two district of Orissa.
Both participants and non-participants are taken into consideration for better
understanding of the impact of microfinance. Comparison of member and non-
members socio-economic condition is reported and it was found that non-participants
are the most vulnerable and are at a disadvantage position than their counterparts.
Probit-model is applied to examine the household specific factors explaining
participation in the programme. It was found that land holding increases the
probability of participation, where as SC or ST status of the household reduce the
probability. Ordinary Least Square regression analysis was carried out to determine
the determinants of household monthly income, and to find the impact of microfinance
on income and employment. The result suggests that, land holding, education, and
participation has positively related to household income. However, ST/SC enters
negatively. Caste of the participants, amount borrowed individually, and monitoring
of the by SHPI explained much of the increase/decrease in income and employment.
Principally, the study found that ST/SC population participation in the programme is
limited and those have joined the programme also received little benefits from the
programme. The programme has left behind the poorest section of the society, those
who need at most attention. Monitoring of the group activities and providing suitable
training facility is critical to materialize the benefits of microfinance

1. Introduction
Credit plays an important role in economic development. It is an important
input for productive activity. It meets the capital requirements for new startups or
expansion of existing production lines. It also meets the requirements of ongoing
production activity. Besides its role in meeting capital requirements in the productive
sectors, credit also plays an important role in day-to-day life as it helps in
consumption smoothening, i.e. bridging the gap between present consumption and
present income, especially in rural areas. There has been a longstanding interest

1
Research Scholar, CSRD, JNU, New Delhi-67, Email- simantini21@gmail.com
2
Research Scholar, Dept. of HSS, IIT Roorkee, Uttarakhanda, Pin-247 667, Email-
bimalkishore.sahoo@gmail.com

1
among development economists and practitioners in the contribution that finance
makes to development. The issue of causality between financial development and
economic growth has been at the center of this debate. Almost a century ago,
Schumpeter (1911) argued that financial intermediation through the banking system
plays a pivotal role in economic development by affecting the allocation of savings,
thereby improving productivity, technical change and the rate of economic growth.

Studies on poverty reveal that lack of access to capital – whether monetary,


educational, or health – is one of the major reasons for continued poverty especially in
rural sector (Verner, 2006; Hulme, Moore and Shepherd, 2001; Nandal, 2005). The
timely availability of credit in the right quantity and at an affordable cost can help to
reduce the high incidence of poverty and thus go a long way in contributing to the
well being of the people, especially in the lower rung of rural society. Stiglitz, (1994)
argued that market failure is a fundamental cause of poverty. Particularly asymmetric
information and high, fixed costs of small-scale lending limit the access of the poor to
formal finance. Expanding the supply of financial services that can be accessed by the
poor can contribute directly to poverty reduction through increase in income and
employment.

The received wisdom from a long time was that lending to poor households is
doomed to failure. Because in lending to the poor costs are too high, risks are too
great and savings propensities are too low. Also the poor households have little to
offer as collateral. Not long ago, the norm was heavily subsidized credit provided by
government banks with repayment rate of 70-80 percent at best. For example in
Bangladesh loans targeted to poor households by traditional banks had repayment
rates of 51.6 per cent in 1980 (Khalily and Meyer, 1993). Similarly, by 1986
repayment rates sank to 41 per cent for subsidized credit delivered as part of India’s
high profile Integrated Rural Development Programme (IRDP) (Pulley,1989). These
programmes offered highly subsidized credit with the view that poor household can’t
afford to borrow at higher interest rates. But the costs to provide this type of credit
quickly mounted and the programmes soon bogged down government budgets, giving
little incentives for banks to expand. Gaiha et al. (2001) argued that benefit to the
rural poor of two major anti-poverty programmes in India (Rural Public Works and
Integrated Rural Development Programme) are likely to be limited, given their gross
mistargeting. Larger sections among the poor were not covered and moreover, the

2
non-poor were the majority among the participants. Similar outcomes were also
reported by Dreze (1990) in his study on IRDP in the state of Uttar Pradesh.
Similalarly, Adams and Von Pischke, (1992), subsidized credit was often diverted to
politically favoured non-poor households. The repeated failures appeared to suggest
that poor households are neither credit-worthy nor able to save much.

The experience of Bangladesh’s Grameen Bank, however, was an exception to


this. Over the last three decades the Grameen Bank has pioneered a credit delivery
system in rural Bangladesh providing banking services to the poor villagers and
focusing on women. The idea for the Grameen Bank came neither from academy nor
from the ideas that started in high income countries and then spread globally3.

In India the adaptation of microfinance approach has assumed the form of


Self-Help Group- Bank linkage programme. Encouraged by the results of the studies
on SHGs, NABARD began exploring the possibilities of establishing linkage between
such groups and banks. As a result SHG-Bank linkage programme is operating in
India as a major microfinance programme since 1992-93 (NABARD, 2003).
Microfinance programmes are generally seen as small loans to poor people for
self-employment projects that generate income allowing them to care for themselves.
Micro credit has been defined by the micro credit summit held in 1997 as
“Programmes that provide credit for self employment and other financial and business
services (including savings and technical assistance) to very poor persons”4.

Microfinance services include savings, consumption loans and insurance in


particular other than micro credit for micro enterprises. In other words, microfinance

3
Yunus described (as quoted in Jonathan Morduch, 1999) the beginning in the following
words:“Bangladesh had a terrible famine in 1974. I was teaching economics in a Bangladesh
University at that time. You can guess how difficult it is to teach the elegant theories of economics
when people are dying of hunger all around you. These theories appeared like cruel jokes. I became a
drop out from formal economics. I wanted to learn economics from the poor in the villages next door to
the university campus.” Yunus found that most villagers were not able to obtain credit at reasonable
rates. So he started lending them from his own pocket, allowing them to buy materials for projects like
weaving bamboo stools and making pots. Ten years after of this event, Yunus set up the bank i.e.
Grameen Bank drawing lessons from informal financial institutions to lend to the groups of poor
households. The small loans were used for rice processing, livestock rising and traditional crafts.
Groups were formed voluntarily and while loan was sanctioned to an individual, all the members of the
group were held responsible for loan repayment.
4
The summit has accepted the definition of poor and “poorest” determined by the “Policy Advisory
Group” (PAG) of the World Bank’s Consultative Group to Assist the Poorest. The PAG defines the
poor as those people living below the poverty line established by each country and the ‘poorest’ as
people in the bottom of that group.

3
includes a range of financial services that seek to meet the needs of poor people. It
protects them from fluctuation in incomes and other shocks, and helps to promote
their incomes and livelihoods, whereas, micro credit precisely caters to the credit
requirement of the poor for the production purpose. While most micro-credit
providers emphasize investment of working or fixed capital in micro-enterprises, the
reality is that many clients use the credit for consumption purpose. Such consumption
smoothing can allow household to cope more effectively, but also it runs the risk of
pushing them further into debt if they can’t repay the loan out of enhanced income
streams. To avoid such circumstances, financial products such as savings, insurance
and consumption loans etc, are the promises of the microfinance.
Microfinance allows poor people to protect, diversify, and increase their
source of income which is essential path out of poverty and hunger. The ability to
borrow a small amount of money to take advantage of business opportunity, to pay for
school fee, or to bridge a cash-flow gap, can be a first step in breaking the cycle of
poverty. Similarly poor households will use a safe, convenient savings account to
accumulate enough cash to buy assets such as inventory for small business enterprise,
to fix a leaky roof, to pay for health care, or to send more children to schools (CGAP,
2003).
This paper attempts to explore the impact of microfinance on income and
employment in general, and in particularly for socially disadvantage groups, namely,
Schedule Tribe and Schedule Cast, by primary survey. Rest of the paper is arranged in
the following order. The coming section, present the data sources and methodology
applied in this study. This is followed by a brief, review of literature on microfinance.
Section-4 reported the socio-economic profile of sample households. Section-5
presents the comparison op participants and non-participants in some key socio-
economic indicators. This is followed by the impact assessment of microfinance
through regression technique. Final section presents the conclusion.

2. Data Source and Methodology

The present study is based on primary field survey as sufficient secondary level data
is not available to understand the functioning of the programme and to know its
impact on rural poor households in the backward region of India. The survey was
conducted in two district of Orissa, namely, Bhadrak and Angul. In primary field
survey data was collected through semi-structure schedule, group discussion and

4
observation method. The study examine only one model of SHG-based microfinance
programme i.e. a facilitator promotes the SHGs and they get financed by the banks5.

Probability and non probability sampling procedure is followed to select the


sample. Two districts of Orissa are chosen randomly, namely, Bhadrak and Angul.
The idea of selecting two districts is to compare the regional variation of impact of
microfinance programme on rural poverty. Then one block of each district is selected
randomly. From each district, purposively, villages are selected; keeping in mind the
programme should have intervened at least four year prior to the survey. Then from
each village, purposively those groups are selected which had longest history in
microfinance. This is because, the benefits of microfinance need time to materialize.
For better understanding of the microfiance impact on the members both participants
and non-participants (control group) households were covered. Inclusion of non-
participants in the analysis is imperative to the study, because in one hand, it acts as
the control group, and in the other hand, their input on, ‘why they did not participate
in the programme’ gives insights to the fact that how this programme is different from
other poverty eradication programmes.

For this study, total of 240 households covered, out of which 150 are
participants and 90 are non-participants. Various suitable statistical and econometric
tools are applied for the analysis of data. Regression analysis, t-test, chi-square test is
conducted for the study.

3. Review of literature

Micro lending has progressed to the greatest extent in the Asian region. Some
of the innovative approaches that have been used successfully in the Asian region are
by Grameen Bank’s credit delivery system in Bangladesh, Thailand’s Bank of
Agriculture and Agricultural Cooperatives, the Association of Cambodia Local
Economic Development Agencies, Buro-Tangail of Bangladesh, the SEWA bank in
India and Amanah Ikhtiar Malaysia. Microfinance has been extensively examined
over the past 10 to 15 years, and the resulting literature is now quite substantial.
Studies on the social aspects of Microfinance are a plenty. The findings range
from extreme positive to skeptical conclusions. However, most of the studies have

5
Though theoretically there are three models of SHG-Bank linkage programme in India, the pilot study
reveals that operation of other two models is very limited in the proposed study area.

5
acclaimed the potential of the programme for reducing vulnerability and income
poverty of the participants. There is substantial evidence in support of a beneficial
effect on income and reductions in vulnerability (Wright 2000; UNICEF 1997;
Khandker 1998,2001). Empirical findings are supportive of the point that the poorest
can benefit from microfinance from both an economic and social well-being, and that
this can be done without jeopardizing the viability of the financial institutions (Wright
2000; Zaman 2000; McCulloch and Baulch 2000).
Rajasekhar and Madheswaran (2005) examine the functioning of microfinance
programmes and analyse economic and social benefits of the programme based on a
study of the project areas of two NGOs in Karnataka and Andhra Pradesh. The study
found that economic benefits of the programme are region specific. The programme
did not result in large amount of loans and income in one study are (3 villages in
Krishna district, Andhra Pradesh) while this result in substantial economic benefits to
the members of another area (one village in Kolar district of Karnataka). They found
that the NGO could not facilitate linkage between SHGs and banks in the former
whereas the NGO was successful in doing so in the second case. Further, in the
former region, most members had fairly stable employment opportunities, thus they
faced high opportunity costs in switching to other occupations. Thus, success of
microfinance programme seems context and region specific. Further, they found that
microfinance benefits were not significant in case of members belonging to the
landless and SC/ST categories. They conclude “the microfinance programme do
provide access to credit for the poor and enable them to undertake income generation
programmes. Given that the formal banks have not very well succeeded in the past in
improving the access to credit for the rural households, the strategy of supporting the
formation and nurturing of micro-finance groups is to be supported”.

While analysing the poverty reduction role of microfinance and the pathways
which facilitate the ascend of the poor, Robinson (2001) concluded that microfinance
services help low income people to (i) improve household and enterprise
management, (ii) increase productivity, (iii) smoothen income flows and consumption
cost, (iv) enlarge and diversify their micro-business and increase their incomes.
Zaman (2000) examined the extent to which micro-credit reduces poverty and
vulnerability through a case study of Bangaladesh Rural Advancement Committee
(BRAC), one of the largest providers of microcredit to the poor in Bangladesh.

6
Household consumption data collected from 1,072 households was used to show that
the largest effect on poverty arises when a moderate-poor BRAC client borrows more
that 10,000 taka (US$200) in cumulative loans. Different control groups and
estimation techniques were used to illustrate this point. Zaman discusses several ways
by which membership in microcredit programmes reduces vulnerability-by (i)
smoothening consumption, (ii) building assets, (iii) providing emergency assistance
during natural disasters, and (iv) contributing to female empowerment.
Cohen (1999) has examined the role of microfinance in reducing the
vulnerabilities of the rural poor, especially its role in enhancing their risk coping
capabilities, in detail with data from microfinance programmes in Latin America. The
study found that the programmes outreach plays more of a role in helping clients
protect against risks ahead of time than in case of coping with shocks after they occur.
Versluvysen (1999), in a cross country study on the programme impact of
microfinance, concluded that poor households that have had to microfinance services
show significant increases in asset accumulation, which provide them with both the
safety-net against misadventure and the resources for self-help investments. Increased
household income improves nutrition, and improves the probability that poor children
from poor families will go to school.
Susan and Rogaly (1997) state that providing microfinance can give poor
people the means to protect their livelihoods against shocks as well as to build up and
to diversify their livelihood activities by investing in productive activities. Kabeer
(2001) observed that in Bangladesh, the Grameen Programme had resulted in
redefinition of the intra household power relations and aggravated gender conflicts.
The above literature review presents a scenario of microfinance activities in
India and also to a lesser extent from the global perspective. The literature also
reported how microfinance is playing a greater role in poverty reduction so as to
contributing to the economic development.
With this background of microfinance programme in India and some extent in
the global environment, the following section presented the socio-economic condition
of sample households for this study.

7
4. Socio-Economic profile of sample households
Data about the socio-economic background and economic activity related aspects of
participants and non-participants was also collected. Some of the socio-economic
indicators are presented bellow.
4.1 Distribution of Individuals in sample households by education

Table-1: Education Table- 1 shows the level of education


Education level participants non-participants
No education 81(16.9%) 45(22.6%) of individual members of the sample
Below primary 159(33.2%) 58(29.14%)
Primary 90(18.8%) 35(17.56%)
households. Of the total 479 persons
Middle 91(19.0%) 23(11.65%) from the participant households, about
High school 12(2.5%) 14(7%)
Inter mediate 11(2.3%) 0(0%) one sixth have no education and only
Graduate 2(0.4%) 1(0.5%)
Not applicable 33(6.9%) 23(11.6%) about 25% of them have received
Total 479(100%) 199(100%)
education beyond primary level.
χ2 =23.74472, d.f. = 7 Source: Field Survey
Figures in parenthesis is percentage However, some of the members of the
participant household have received education intermediate and above. From the non-
participant households 22.6% have no education either from formal schooling or from
non-formal sources. Only one person has received education beyond high school.
4.2 Caste composition of Participant and non-participant households
Table-2 : Caste It is observed from Table-2 it is evident that a
No of No of
caste major proportion of the participants (40%)
participants non-participants
General 60 (40) 14(15.55) belong to general caste, and smaller
S.C. 40(26.67) 42(46.67)
percentages (14.67%) are Scheduled Tribes
S.T 22(14.67) 15(16.67)
O.B.C. 28(18.67) 19(21.11) (S.Ts).In case of non-participants in
Total 150(100) 90(100)
χ2 =17.80 d.f. =3 Source: Field Survey microfinance (NPM), 15.55% are in general
Figures in parenthesis are percentage
castes, 46.67% are S.Cs, 16.67% are S.Ts and
21.11% are in O.B.C category. Thus, there are a much greater proportion of SCs and
STs among those poor who, as yet are not availing of the facility of microfinance
services or do not have access to it.
4.3 Major Occupation of Sample Household
The data about the major occupations of the households shows that many of
them have more than one major occupation for their livelihood, however, only the
first major occupation of households is reported in Table-3.

Table-3: Occupation
Occupation Participant Participant

8
Wage labour 52(34.67) 79 (87.78) In a rural set up wage labour and
Cultivation 72 (48.00) 9(10.00)
cultivation dominate as the major
Government Service 10(6.67) 1 (1.11)
Trading 16(10.67) 1(1.11) occupation, and the present sample is
Total 150 (100.00) 90(100.00)
not an exception. It is significant that
χ2 =64.17, d.f. =3 , Source: Field Survey
Figures in parenthesis are percentage although only 34.67% of participant
households depend on wage labour, as many as 87.78% of the non-participants have
to fall back on wage labour. Only about 10% of the participants are doing business as
their major occupation in the form of shop keeper, rice business etc. whereas about
7% of households earn their livelihood as government employees.
4.4 Monthly household income of the participants and non-participants

Table-4: Household’s Monthly Income The monthly average household


Income (in Rs.) Participant Non-participant
400-600 9 (6.00) 2(2.22) income of the participants varies a
600-800 16 (10.67) 37(41.11)
800-1000 41(27.33) 24(26.67)
great deal from household to
1000-1200 40(26.67) 14(15.55) household. We can observe from the
1200-1400 19(12.67) 13(14.44)
1400-1600 8(5.33) 0 (0) table that about 6% households have
>1600 17(11.33) 0(0)
Total 150(100) 90(100%) monthly income as low as Rs. 400-
χ2 =41.66, d.f. = 6 Source: Field Survey
600, and 10.67% have between
Figures in parenthesis are percentage
Rs.600-800. Majority of the households (54%) earn a monthly income between Rs.
800-1200. In case of non-participants, it is clear that they exhibit still lower economic
achievements, more than two thirds of them earning no more than Rs. 1,000/- per
month as household income.
4.5 Land holdings of the participants and non-participants

It is evident from Table-5 that many of the households were landless. Among the
participants, 26.67% households have no land and 34.67% households have land but
less than 1 acre.

Table-5: Land Holding In case of non-participants, these


Land(In Acres) Participant Non-participant
No land 40(26.67) 59(65.55) percentages are 65.55% and 21.11%
<1 52(34.67) 19 (21.11)
1-2 21(14.00) 9(1.00)
respectively. Only a handful of
2-3 11(7.33) 3(3.33) households among participants have
3-4 5(3.33) 0(0)
4-5 2(1.33) 0(0) land more than six acres and none of the
5-6 3(2.00) 0(0)
6 and above 16(10.66) 0(0) non-participating households have land
Total 150(100) 90(100%)
more than three acres.
χ2 =41.97, d.f. = 8 6 Source: Field Survey
Figures in parenthesis are percentage

9
4.6 Dwelling structure of the participants and non-participants
Table-6: Dwelling structure While visiting the study villages,
Dwelling structure Participant Non-participant
Katcha/Thatch 120(80) 72(80.00) most of the houses were found to be
Katcha/Tile 5(3.33) 0(0)
Semi-pucca 4(2.66) 0(0)
constructed with katcha/thatch.
Pucca (weaker 20(13.33) 18(20.00) Table-6 reported the dwelling
Section housing
scheme) structure of both the participants and
Pucca 1(0.006) 0(0)
Total 150(100) 90(100) the non-participants. 80% of the
χ2 =43.97, d.f. = 4 Source: Field Survey;
Figures in parenthesis are percentage participants have house which is
constructed of katcha/thatch, 3.33% houses are katcha/tile, 2.66% houses are semi-
pucca and 13.33% houses are pucca but they are constructed through weaker sector
housing schemes. 80% of the non-participants have katcha/thatch house and other
20% have pucca house but it is constructed through weaker sector housing schemes.

4.7 Latrine type of the sample households


Table-7: Latrine Type Access to latrine enlightens not only
Latrine type Participants Non-participants
No latrine 144 (96) 90(100%) economic well being of a rural
Septic tank 3(2) 0 ()
Service latrine 3(2) 0()
household but it is perceived as an
Total 150 (100) 90(100%) important facilitator for women. From
χ2 =3.69, d.f. = 3; Source: Field Survey;
Figures in parenthesis are percentage the Table-7 it is seen that 96% of the
participant households have no latrine in their house and only three house have septic
tank and three houses has service latrine. None of the houses of the non-participants
have this facility. It shows that while at least some of the participants have awareness
about the safe sanitation, none of the non-participants are aware of this fact, or unable
to afford this facility.

4.8 Lighting system of the sample households


Table-8: Lighting system It is seen from the table that whereas
Lighting system Participant Non-participant
Electricity 72 (48) 17(18.89) 48% of participants have access to
Oil, kerosene 78(52) 73(81.11)
Total 150 (100) 90(100)
electricity for lighting, this is the case
χ2 =20.43, d.f. = 1; Source: Field Survey for only 19% of non-participants.
Figures in parenthesis are percentage

4.9 Drinking Water Source of Sample Households


Table-9: Drinking Water Source

10
Source Participant Non-participant The sample households depend on
Individual Tap/Well 19 (12.67) 4 (4.44) river and ground water for drinking
Community Well/Tap 93 (62.00) 36 (40.00)
River 38 (25.53) 50 (55.55) purposes. It is seen from Table-9
Total 150 (100) 90 (100) 12.67% of the participant
2
χ =23.04, d.f. =2; Source: Field Survey
Figures in parenthesis are percentage households have personal well/tap
and 62% acquire drinking water from common hand pump/well; and 25.53% drink
river water. In case of non-participants a handful has personal drinking water, and
55.55% non-participants drink the river water.
4.10 Ownership of durable assets by sample households
Of the participants of microfinance, many of the households were found to
possess almost all of the household assets and also its value is much more than non-
participant households. But in case of non-participants, they possess only some basic
assets for their livelihood such as radio, cycle and watches. Thus there appears to be a
noticeable difference in possession of durable household assets among the participant
and the non-participant households. With these socio-economic backdrop, the present
paper attempts to examine the impact of microfinance on income, and employment.
Table-10: Ownership of durable assets
Participants Non-participants
Total value No of
Assets Total value No of
in Rs households (b)/(c) (d)/(e)
(a) in Rs (d) households (e)
(b) (c)
Radio 12450 63 (42) 197.62 3000 18 (20) 166.67
Cycle 44550 120 (80) 371.25 28600 58 (64.44) 493.10
Fan 28350 48 (32) 590.63 1200 6(6.67) 200.00
B/w TV 25500 15(10) 1700.00 0 0 (0%) 0
Colour TV 74,000 6(4) 12333.33 0 0 (0%) 0
Sewing 4000 2(1.33) 0 0(0%) 0
Machine 2000.00
Cooker 800 2(1.33) 400.00 0 0(0%) 0
Watch 22340 83(55.33) 269.16 2700 38(42.22) 123.68
2
χ =56.73, d.f 7; Source: Field Survey

5. Comparison of Participants and non-participants


A comparison of the participants and the non-participants poor from the same
locations potentially reveals about (a) access of the poor to the microfinance
programme in terms of the socio-economic characteristics that can be taken as given
(exogenous) in the short term for which the programme has been in operation; and (b)
impact of the programme on those economic outcomes which can change in the short

11
run. In this effort, the present section draws some concluding remarks on the statistics
given in the previous section.
It is observed that χ2 (χ2 =17.80) value of cast composition of participants and
non participants is greater than the critical value of the χ2, therefore, it can be
conclude that there is significantly greater preponderance of SCs/ STs among the non-
participants compared to participants (Table-2). It appears that the SCs/ STs are at
relative disadvantage in having access to microfinance programme.
Rural economies are characterized by primary economic activities, which are
based on land and natural resources, and the degree of access to these resources by the
households. Size of land held, therefore, plays an important role in determining the
welfare of the people. More crucially, land is the prime marketable collateral in the
rural areas. Availability of credit and its terms are very much dependent of the size of
land held by the borrowers. In short, land, not only defines the live hood structure of
borrowers, but also influences the source, terms and quantum of credit received by the
poor.
It is observed from the primary field investigation the likelihood of
landholding are in support of participants (χ2 =41.97, d.f. = 8, Table-5). The non-
participants are found to be more often landless than their counterparts. Even though
land holdings are generally small among both the participant and the non-participant
households, except a few exceptions where landholding of participants is quite high.
However, the average land holding of participants is significantly higher compared to
non-participants (Table-11). The hypothesis that mean land holding is equal for the
two groups is contentedly rejected at 1%. It is observed from Table-11 that there is
high variation (measured by coefficient of variation) in landholding among the
participants and also non-participants.

Table-11: Participation in microfinance and land holding


Item Participant Non-participant
Mean (acres) 1.48 0.26
Standard deviation (acres) 2.06 0.37
Coefficient of variation 139.19 142.31
No. of observations 150 90
T- Value = 7.53. Sources: Field Survey

12
In terms of some of the basic amenities such as, type of dwelling structures,
lighting system, sources of drinking water, and latrine use, it is observed that the odds
are in against the poor non-participants (high value of χ2 against the respective critical
values), except for latrine type where both participants and non-participants have
similar access (low value of χ2) (Table-6 to Table-9).
It is evident from the occupational structure, that there are more than twice as
high wage earners among the non-participants (about 88%) compare to the
participants (about 35%). There are significantly fewer cultivators among the non-
participants (10%) compared to the participants (48%) (Table-3).
Looking at educational achievements, the participants are significantly better
educated compared to the non-participants (Table-1). However the rates of dropouts
from formal education system (schools) are similar among the children of participants
(29.4%) and non-participants (33.8%). It is slightly higher for the non-participants.
Looking at the ownership pattern of common durable assets, the participants
own greater variety of assets. Further, the mean value of assets per households is
significantly higher for the participants (Rs. 802.99) compared to the non-participants
(Rs. 417.66) as evidenced from.
Comparing current monthly household incomes among these two groups’
divulge that mean income is higher among the participants (Rs. 1272.23) compared to
the non-participants (Rs. 986.67), and the difference is statistically significant as it is
evident from t-statistic (Table-12). However, the distribution of income incase of
participants is highly skewed (Coefficient of variation 87.40) compare to the non-
participants (Coefficient of variation 35.19). This suggests that among the participants
there are household having very high income and very low income. To further study
impact of participation in the microfinance programme, as well as other variables
which may be prima facie important in explaining income, and employment, present
study applied regression technique, and details is reported in the following section.

Table -12: Participation in microfinance and monthly household income


Item Participants Non-participants
Mean (Rs.) 1272.23 986.67
Standard deviation (Rs.) 1112.05 347.21
Coefficient of variation (%) 87.40 35.19
Number of observations 150 90
T Value = 74.51; Source: Field Survey

13
6. Impact of microfinance
6.1 Determinants of Participation
Prior to examine the impact of microfinance on household per-capita income,
increase in income and employment, it is pertinent to examine what are the factors
determine the participation in the programme. For this purpose, binary Probit-model
with Maximum Likelihood Estimates is carried out. The estimable regression
equation is:
Y   0   1 X 1   2 X 2   3 X 3   4 X 4   5 X 5   6 D1   7 D2   8 D3   9 D4  10 D 5   ---1

Where, Y is the binary variable which takes the value 1 for participation
otherwise 0. X1 is the average household income per month (HHI), X2 is land owned
by the household (Land), X3 is value of livestock owned by the household
(Livestock), X4 is highest education level in the house hold (Education), measured in
terms of years of schooling, and X5 is dependency ratio in the household (DR). D1, D2,
D3, D4, and D5 are dummy variable for occupation-1, occupation-2, caste general,
caste ST/SC and district. D1=1 if the household main occupation is wage labour
otherwise zero (wage). D2=1 if the household main occupation in cultivation
otherwise zero (cultivation). D3=1 for general cast otherwise zero (Gen). D4=1 for
ST/SC otherwise zero (ST/SC). D5= is 1 for Angul district otherwise zero (DIST).
The regression result is presented in the Table-13.  is the error term.
It is seen from the Table-13 that Log likelihood function is statistically
significant at one percent level of significance implying collectively all the
coefficients are statistically significant. The coefficient of land owned by household
enters positively and statistically significant, thus, it can be concluded that the
probability of participation in the microfinance programme increases with the rise in
land holding. This is probably because of because, microfinance programme needs
saving in each month, and access to land can provides stable income to the household.

Table-12: Regression result Binomial Probit Model


(Maximum Likelihood Estimates)
Variables Coefficients S.E. Z-stat
Constant -0.341 0.676 -0.504
HHI 0.001 0.000 1.241
Land 1.132* 0.344 3.293
Livestock 0.317** 0.165 1.924

14
Education 0.220* 0.048 4.609 This fact is supported by the fact
DR -0.300** 0.137 -2.191
Wage -1.199* 0.268 -4.468 that the coefficient for occupation
Cultivation 1.449* 0.373 3.882 dummy, for cultivation as main
Gen 1.056* 0.376 2.806
ST/SC -0.117** 0.094 -1.882 occupation enters positively and
DIST -0.018 0.261 -0.069 statistically significant. Besides
Log likelihood function -73.737*
Restricted log likelihood -158.775* that land is viewed as a social
No. of observation 240
S.E. Standard error
status of the household, which may
*, and ** indicates 1% and 5% level of significance
influence their participation. Other
main source of income in rural household is livestock, and it is evident from the table
that, household having more livestock (in terms of value) has higher probability of
participating in the programme. The coefficients of dummy variables for castes (Gen
and ST/SC) illustrate that the household’s probability of participation in the
programme increased if it belongs to the general category, where as it reduces if the
household belongs to ST/SC groups. This signifies that, microfinance programme has
not competent to by-pass the pre-justice and social obstacles attached to the ST/SC
population, and to encourage their participation. For households, whose main
occupation is wage labour, the probability of participation has come down. This is
may be the fact that in rural area, wage earning is very much sporadic and hence these
households believe they can not afford to offer the monthly saving installment.
Similarly, high dependency ratio has decreases the probability of participation. Now
it is interesting to know the factors, including participation, are explaining household
income.

6.2 Impact on Household Income and Employment


Y   0   1 X 1   2 X 2   3 X 3   4 X 4   5 D1   6 D2   7 D3   8 D4   9 D 5   ---2

Where, Y is log of average per-capita income per month of household. X1, is


cultivatable land owned by the household (Land), X2 is value of livestock owned by
the household (Livestock), X3 is highest education level in the house hold (Education)
measured in terms of years of schooling, and X4 is dependency ratio in the household
(DR). D1, D2, D3, and D4, are dummy variables as explained earlier. D5 is the dummy
variable for participation which takes the value of 1 for participation and zero for non-
participation. The regression result is presented in the Table-13.  is the classical

15
regression error term and assumed to follows a normal distribution with mean zero
and variance one.

Table- 13:Regression result of equation:2 The OLS regression result exhibits


Variables Coefficients S.E.
Constant 355.440* 70.018
statistically significant F-vale, close to
Land 61.642* 7.905 two D-W statistics, and reasonably
Livestock 12.599** 5.053
Education 7.590*** 4.718 high R-square vale (considering the
DR -43.105* 12.351 number of qualitative data), it can be
Wage -5.227 31.281
Cultivation 0.071 33.152 said that the model is good fit. It is
Gen 32.448 36.787
observed that, cultivatable land, value
ST/SC -1.111 32.261
Participation 21.289* 5.456 of livestock, education and
F-value 8.85*
participation in the programme has
R2 0.274
Adj. R2 0.243 positive bearing on per-capita
D-W Stat 2.1
No. of observation 240 household income, as these variables
S.E. Standard error;
*, and ** 1% and 5% level of significance
coefficients are statistically significant
and positive. The coefficient of participation needs careful investigation. The robust
statistically significant positive coefficient on the face can be reported as participation
has increased the per-capita household income. However, as it is explained earlier,
participants have better access to amenities like water, housing, light, education, and
they have higher land holding and better equipped with durable assets, and most of
the participants have cultivation as main source of income, it is natural that the
participants will have higher income. Therefore, the participation dummy has entered
positively. Now, it will pertinent to examine the factors explaining increase in income
and employment of the participants.
In the rural setting employment has very decisive function for the over all
development of the household. Employment through out the year will assure
continuous (and rescannable) flow of income, and this is important for sustainable
development. It is due to the fact that continuous flow of income will ensure good
quality of food and the household is less vulnerable to extraneous contingencies. Even
continuous flow of income will ensure elucidation to the children. Thus, development
can be sustainable. Thus, it is important to look into microfinance potential of
increasing employment of the participants. For this purpose regression analysis is
applied and the estimable regression equation is given as:
Y   0  1 X 1   2 X 2   3 X 3   4 D1   5 D2   6 D3   7 D4   8 D 5   9 D6   ----3

16
Where, Y is the increase in income or employment. X1 is cultivable land
owned by the household (Land), X2 is education level of the participants (Education),
measured in terms of years of schooling, and X3 is the amount borrowed individually
(amount borrowed). D1, D2, D3, D4, and D5 are dummy variable as explained in
equation-1. D6 is the dummy variable which takes the value 1 if the household
borrowed individually otherwise zero. The OLS regression result is reported in table-
14.
Table-14: Regression result of Equation-14
Variables Dependant variable Dependant variable
Increase in Income Increase in employment
Coefficients S.E Coefficients .E.

Constant 2109.482** 598.410 160.968 29.619

Land -102.086 81.542 -13.712* 4.036

Education 66.952 66.211 3.277 3.277

Wage 186.052 443.591 3.705* 1.956

Cultivation 295.685 389.807 -0.785 19.294

Gen 1188.617** 459.312 -20.703 22.734

ST/SC -1242.459* 418.948 -31.133*** 18.736

Amount 2.441* 0.494 0.187* .024


Borrow
Borrowed 1228.870* 631.688 106.538* 31.266
Individually
Dist -1039.469** 477.073 -103.194 23.613

F-value 9.466* 18.968*

R2 0.340 0.551

Adj. R2 0.32 0.522

D-W Stat 1.891 1.87

No. of 150 150


observation
S.E. Standard error; *, **, and *** indicates 1%, 5%, and 10% level of significance respectively
For both the dependent variables F-value is statistically significant at 1% level
of significance. D-W statistics is close to two and the R-square value is reasonable
high, suggesting the model is good fit. It is evident from statistically significant and
positive coefficients of amount borrowed and borrowed individually that, when the
household borrowed it individually impact on income and employment is high. This is
because; household to use the fund according to their productive needs. Increase in
employment is due to the fact that, individually taken loans are invested in petty

17
business, which has high potential of employment generation in the rural areas.
Household having main occupation wage earning has performed better in case of
employment generation. This is due to the fact that they invest the borrowed money to
small enterprises, where as household having main occupation as cultivation, invest
the borrowed amount primarily for purchase of agricultural inputs. It is clear that,
ST/SC participant’s income and employment does not increase. This is primary
because, most of the ST/SC participants who are taken loan from the group did not
invest in the productive purpose because social rigidities operate as a hindrance to
start up new business.
The coefficient of district dummy enters negatively both the explained
variables, however only incase of increase in income it is statistically significant. This
indicates that, in Angul district participant’s income has not gone up. This is because,
in case of Bhadrak district, Non Governmental Organization, takes initiative and
motivate people to takes up of productive business, provides training facility for
taking up suitable business activity, and monitor group activities, where as in case of
Angul, all the groups are formed by Integrated Child Development Scheme (ICDS)
and after formation the organization does not monitor the functioning and
performance of groups.
7. Conclusion
The study found that there is significantly greater preponderance of
SCs/ STs among the non-participants compared to participants. The non-participants
are found to be more often landless than their counterparts. In terms of some of the
basic amenities such as, type of dwelling structures, lighting system, sources of
drinking water, and latrine use, it is observed that the odds are in against the poor non-
participants. Mean household income is higher among the participants compared to
the non-participants, and the difference is statistically significant
The participant households are better equipped with durable assets,
landholding, and educational achievement. From the Probit-model regression it is
found that probability of participation in the microfinance programme increases with
the rise in land holding, and household’s probability of participation in the
programme increased if it belongs to the general category, where as it reduces if the
household belongs to ST/SC groups. This signifies that, microfinance programme has

18
not competent to by-pass the pre-justice and social obstacles attached to the ST/SC
population, and to encourage their participation.
It is observed that, cultivatable land, value of livestock, education and
participation in the programme has positive bearing on per-capita household income,
the coefficient of participation needs careful investigation. The robust statistically
significant positive coefficient on the face can be reported as participation has
increased the per-capita household income. However, as the participants have better
access to amenities like water, housing, light, education, and they have higher land
holding and better equipped with durable assets, and most of the participants have
cultivation as main source of income, it is natural that they will have higher income.
Present study found that when the household borrowed it individually impact on
income and employment is high. Household having main occupation wage earning
has performed better in case of employment generation. It is also evident that, ST/SC
participant’s income and employment does not increase. This is primary because,
most of the ST/SC participants who are taken loan from the group did not invest in the
productive purpose because social rigidities operate as a hindrance to start up new
business. The study also found that participants from Bhadrak distrci has performed
better that that of Angul district.
To sum up, ST/SC population participation in the programme is limited and
those have joined the programme also received little benefits from the programme.
The programme has leave behind the poorest section of the society, those who need at
most attention. Monitoring of the group activities and providing suitable training
facility is critical to materialize the benefits of microfinance.
Reference:
Adam, D., Graham, D., Von Pischke, J.D. (1984) “Undermining rural development
with cheap credit”, Boulder, Westside Press.

CGAP, (2002) “Microfinance and the millennium development goals”, Donor Brief
No. 9, December.

CGAP, (2003) “Is Microfinance an effective strategy to reach the Millennium


Development Goals?” Discussion Paper No.5, January.

Cohen, Monique (1999) “Opening Up the Impact Assessment Agenda”. AIMS paper
prepared for the Latin American Microfinance Meeting, October.

19
Kabeer, Naila (2001), “Conflicts over Credit: Re-evaluating the Empowerment Potential
of Loans to Women in Rural Bangladesh”, World Development, vol.29, No.1,
pp.63-84

Khalily, M.A. and R. Meyer (1993) “The political economy of rural loan recovery:
Evidence from Bangladesh, Saving and Development Vol.17 No.1

Khandker, Shahidur R. (1998) “Fighting Poverty with Microcredit: Experience in


Bangladesh”. New York: Oxford University Press.

Khandker, Shahidur R. (2001) “Does Micro-finance really Benefit the Poor? Evidence
from Bangladesh”, paper presented at Asia and Pacific Forum on Poverty:
Reforming Policies and Institutions for Poverty Reduction, Manila, Asian
Development Bank.

Morduch, Jonathan (1999) “The Microfinance Promise” Journal of Economic


Literature, Vol. 37, December, pp.1569-1614

NABARD (2003) “Banking with the Poor-A Handbook on Self-Help Groups”

Pulley, Robert (1989) “Making the Poor Credit Worthy: A Case Study of the
Integrated Rural Development Programme in India”, World Bank Discussion
Paper 58, Washington D.C.

Rajsekhar, D. and S.Madheswaran (2005) “Economic and Social Benefits of Micro


finance Programmes: An Econometric Analysis”. By B.B.Bhattacharya and
Arup Mitra (2005), Studies in Macro Economic and welfare. New Delhi,
Academic Foundation. P.p.383-404.

Robinson, M.S.(2001) “Microfinance Revolution-Sustainable Development for the Poor”,


Washington D.C.,World Bank
Versluvysen, Eugene (1999) Defying the Odds: Banking for the Poor, Connecticut:
Kumarian Press

Wright, Graham A.N. (2000), Microfinance Systems: Designing Quality Financial


Services for the Poor, London and New York: Zed Books Ltd and Dhaka: The
University Press Limited

Zaman, Hassan (2000) “Assessing the poverty and vulnerability Impact of Micro-
credit in Bangladesh: A Case Study of BRAC”, Working Paper No. 2145, World
Bank.

20

View publication stats